[patch 00/31] Latest numa/core patches, v15

[patch 00/31] Latest numa/core patches, v15

[Ingo Molnar]

Hi,

This is the latest iteration of our numa/core patches, which 
implements adaptive NUMA affinity balancing.

Changes in this version:

    https://lkml.org/lkml/2012/11/12/315

Performance figures:

    https://lkml.org/lkml/2012/11/12/330

Any review feedback, comments and test results are welcome!

For testing purposes I'd suggest using the latest tip:master 
integration tree, which has the latest numa/core tree merged:

   git pull git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git master

(But you can also directly use the tip:numa/core tree as well.)

Thanks,

    Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 00/31] Latest numa/core patches, v15

[Ingo Molnar]

Hi,

This is the latest iteration of our numa/core tree, which
implements adaptive NUMA affinity balancing.

Changes in this version:

    https://lkml.org/lkml/2012/11/12/315

Performance figures:

    https://lkml.org/lkml/2012/11/12/330

Any review feedback, comments and test results are welcome!

For testing purposes I'd suggest using the latest tip:master
integration tree, which has the latest numa/core tree merged:

   git pull git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git master

(But you can also directly use the tip:numa/core tree as well.)

Thanks,

    Ingo

----------------------->
Andrea Arcangeli (1):
  numa, mm: Support NUMA hinting page faults from gup/gup_fast

Gerald Schaefer (1):
  sched, numa, mm, s390/thp: Implement pmd_pgprot() for s390

Ingo Molnar (3):
  mm/pgprot: Move the pgprot_modify() fallback definition to mm.h
  sched, mm, x86: Add the ARCH_SUPPORTS_NUMA_BALANCING flag
  mm: Allow the migration of shared pages

Lee Schermerhorn (3):
  mm/mpol: Add MPOL_MF_NOOP
  mm/mpol: Check for misplaced page
  mm/mpol: Add MPOL_MF_LAZY

Peter Zijlstra (16):
  sched, numa, mm: Make find_busiest_queue() a method
  sched, numa, mm: Describe the NUMA scheduling problem formally
  mm/thp: Preserve pgprot across huge page split
  mm/mpol: Make MPOL_LOCAL a real policy
  mm/mpol: Create special PROT_NONE infrastructure
  mm/migrate: Introduce migrate_misplaced_page()
  mm/mpol: Use special PROT_NONE to migrate pages
  sched, numa, mm: Introduce sched_feat_numa()
  sched, numa, mm: Implement THP migration
  sched, numa, mm: Add last_cpu to page flags
  sched, numa, mm, arch: Add variable locality exception
  sched, numa, mm: Add the scanning page fault machinery
  sched, numa, mm: Add adaptive NUMA affinity support
  sched, numa, mm: Implement constant, per task Working Set Sampling (WSS) rate
  sched, numa, mm: Count WS scanning against present PTEs, not virtual memory ranges
  sched, numa, mm: Implement slow start for working set sampling

Ralf Baechle (1):
  sched, numa, mm, MIPS/thp: Add pmd_pgprot() implementation

Rik van Riel (6):
  mm/generic: Only flush the local TLB in ptep_set_access_flags()
  x86/mm: Only do a local tlb flush in ptep_set_access_flags()
  x86/mm: Introduce pte_accessible()
  mm: Only flush the TLB when clearing an accessible pte
  x86/mm: Completely drop the TLB flush from ptep_set_access_flags()
  sched, numa, mm: Add credits for NUMA placement

---

 CREDITS                                  |    1 +
 Documentation/scheduler/numa-problem.txt |  236 +++++++++++
 arch/mips/include/asm/pgtable.h          |    2 +
 arch/s390/include/asm/pgtable.h          |   13 +
 arch/sh/mm/Kconfig                       |    1 +
 arch/x86/Kconfig                         |    1 +
 arch/x86/include/asm/pgtable.h           |    7 +
 arch/x86/mm/pgtable.c                    |    8 +-
 include/asm-generic/pgtable.h            |    4 +
 include/linux/huge_mm.h                  |   19 +
 include/linux/hugetlb.h                  |    8 +-
 include/linux/init_task.h                |    8 +
 include/linux/mempolicy.h                |    8 +
 include/linux/migrate.h                  |    7 +
 include/linux/migrate_mode.h             |    3 +
 include/linux/mm.h                       |  122 ++++--
 include/linux/mm_types.h                 |   10 +
 include/linux/mmzone.h                   |   14 +-
 include/linux/page-flags-layout.h        |   83 ++++
 include/linux/sched.h                    |   46 ++-
 include/uapi/linux/mempolicy.h           |   16 +-
 init/Kconfig                             |   23 ++
 kernel/bounds.c                          |    2 +
 kernel/sched/core.c                      |   68 +++-
 kernel/sched/fair.c                      | 1032 ++++++++++++++++++++++++++++++++++++++++---------
 kernel/sched/features.h                  |    8 +
 kernel/sched/sched.h                     |   38 +-
 kernel/sysctl.c                          |   45 ++-
 mm/huge_memory.c                         |  253 +++++++++---
 mm/hugetlb.c                             |   10 +-
 mm/memory.c                              |  129 ++++++-
 mm/mempolicy.c                           |  206 ++++++++--
 mm/migrate.c                             |   81 +++-
 mm/mprotect.c                            |   64 ++-
 mm/pgtable-generic.c                     |    9 +-
 35 files changed, 2200 insertions(+), 385 deletions(-)
 create mode 100644 Documentation/scheduler/numa-problem.txt
 create mode 100644 include/linux/page-flags-layout.h

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Mel Gorman]

On Tue, Nov 13, 2012 at 06:13:23PM +0100, Ingo Molnar wrote:
> Hi,
> 
> This is the latest iteration of our numa/core tree, which
> implements adaptive NUMA affinity balancing.
> 
> Changes in this version:
> 
>     https://lkml.org/lkml/2012/11/12/315
> 
> Performance figures:
> 
>     https://lkml.org/lkml/2012/11/12/330
> 
> Any review feedback, comments and test results are welcome!
> 

For the purposes of review and testing, this is going to be hard to pick
apart and compare. It doesn't apply against 3.7-rc5 and when trying to
resolve the conflicts it quickly becomes obvious that the series depends
on other scheduler patches such as

sched: Add an rq migration call-back to sched_class
sched: Introduce temporary FAIR_GROUP_SCHED dependency for load-tracking

This is not a full list, it was just the first I hit. What are the other
scheduler patches you are depend on? Knowing that will probably help pick
apart some of the massive patches like "sched, numa, mm: Add adaptive
NUMA affinity support" which is a massive monolithic patch I have not even
attempted to read yet but the diffstat for it alone says a lot.

7 files changed, 901 insertions(+), 197 deletions(-)

-- 
Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Ingo Molnar]

* Mel Gorman <mgorman@suse.de> wrote:

> On Tue, Nov 13, 2012 at 06:13:23PM +0100, Ingo Molnar wrote:
> > Hi,
> > 
> > This is the latest iteration of our numa/core tree, which
> > implements adaptive NUMA affinity balancing.
> > 
> > Changes in this version:
> > 
> >     https://lkml.org/lkml/2012/11/12/315
> > 
> > Performance figures:
> > 
> >     https://lkml.org/lkml/2012/11/12/330
> > 
> > Any review feedback, comments and test results are welcome!
> > 
> 
> For the purposes of review and testing, this is going to be 
> hard to pick apart and compare. It doesn't apply against 
> 3.7-rc5 [...]

Because the scheduler changes are highly non-trivial it's on top 
of the scheduler tree:

   git pull git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core

I just tested the patches, they all apply cleanly, with zero 
fuzz and offsets.

Thanks,

	Ingo

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Mel Gorman]

On Wed, Nov 14, 2012 at 08:52:22AM +0100, Ingo Molnar wrote:
> 
> * Mel Gorman <mgorman@suse.de> wrote:
> 
> > On Tue, Nov 13, 2012 at 06:13:23PM +0100, Ingo Molnar wrote:
> > > Hi,
> > > 
> > > This is the latest iteration of our numa/core tree, which
> > > implements adaptive NUMA affinity balancing.
> > > 
> > > Changes in this version:
> > > 
> > >     https://lkml.org/lkml/2012/11/12/315
> > > 
> > > Performance figures:
> > > 
> > >     https://lkml.org/lkml/2012/11/12/330
> > > 
> > > Any review feedback, comments and test results are welcome!
> > > 
> > 
> > For the purposes of review and testing, this is going to be 
> > hard to pick apart and compare. It doesn't apply against 
> > 3.7-rc5 [...]
> 
> Because the scheduler changes are highly non-trivial it's on top 
> of the scheduler tree:
> 
>    git pull git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core
> 
> I just tested the patches, they all apply cleanly, with zero 
> fuzz and offsets.
> 

The actual numa patches don't apply on top of that but at least the
conflicts are obvious to resolve. I'll queue up a test to run overnight
but in the meantime, does the current implementation of the NUMA patches
*depend* on any of those scheduler patches? Normally I would say it'd be
obvious from the series except in this case it just isn't because of the
monolithic nature of some of the patches.

-- 
Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Mel Gorman]

On Wed, Nov 14, 2012 at 08:52:22AM +0100, Ingo Molnar wrote:
> 
> * Mel Gorman <mgorman@suse.de> wrote:
> 
> > On Tue, Nov 13, 2012 at 06:13:23PM +0100, Ingo Molnar wrote:
> > > Hi,
> > > 
> > > This is the latest iteration of our numa/core tree, which
> > > implements adaptive NUMA affinity balancing.
> > > 
> > > Changes in this version:
> > > 
> > >     https://lkml.org/lkml/2012/11/12/315
> > > 
> > > Performance figures:
> > > 
> > >     https://lkml.org/lkml/2012/11/12/330
> > > 
> > > Any review feedback, comments and test results are welcome!
> > > 
> > 
> > For the purposes of review and testing, this is going to be 
> > hard to pick apart and compare. It doesn't apply against 
> > 3.7-rc5 [...]
> 
> Because the scheduler changes are highly non-trivial it's on top 
> of the scheduler tree:
> 
>    git pull git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core
> 
> I just tested the patches, they all apply cleanly, with zero 
> fuzz and offsets.
> 

My apologies about the merge complaint. I used the wrong baseline and
the problem was on my side. The series does indeed apply cleanly once
the scheduler patches are pulled in too.

-- 
Mel Gorman
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Alex Shi]

had caught a ops on my 2 sockets SNB EP server. but can not reproduce it.
send out as a reminder:
on tip/master, head : a7b7a8ad4476bb641c8455a4e0d7d0fd3eb86f90

Oops: 0000 [#1] SMP
[   21.967103] Modules linked in: iTCO_wdt iTCO_vendor_support
i2c_i801 igb microcode lpc_ich ioatdma i2c_core joydev mfd_core hed
dca ipv6 isci libsas scsi_transport_sas
[   21.967109] CPU 7
[   21.967109] Pid: 754, comm: systemd-readahe Not tainted
3.7.0-rc5-tip+ #20 Intel Corporation S2600CP/S2600CP
[   21.967115] RIP: 0010:[<ffffffff8114987f>]  [<ffffffff8114987f>]
__fd_install+0x2d/0x4f
[   21.967117] RSP: 0018:ffff8808187f7de8  EFLAGS: 00010246
[   21.967118] RAX: ffff881018bfb700 RBX: ffff88081c2f5d80 RCX: ffff880818dfc620
[   21.967120] RDX: ffff881019b10000 RSI: 00000000ffffffff RDI: ffff88081c2f5e00
[   21.967122] RBP: ffff8808187f7e08 R08: ffff88101b37e008 R09: ffffffff811644a6
[   21.967123] R10: ffff880818005e00 R11: ffff880818005e00 R12: 00000000ffffffff
[   21.967125] R13: 0000000000000000 R14: 00000000fffffff2 R15: 0000000000000000
[   21.967128] FS:  00007ffa79ead7e0(0000) GS:ffff88081fce0000(0000)
knlGS:0000000000000000
[   21.967130] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   21.967131] CR2: ffff881819b0fff8 CR3: 000000081be54000 CR4: 00000000000407e0
[   21.967133] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   21.967135] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[   21.967137] Process systemd-readahe (pid: 754, threadinfo
ffff8808187f6000, task ffff880818dfc620)
[   21.967138] Stack:
[   21.967145]  ffff880818005e00 ffff88101b37e000 ffff880818005e00
00007fff57d29378
[   21.967150]  ffff8808187f7e18 ffffffff811498c6 ffff8808187f7ed8
ffffffff81167a7c
[   21.967155]  ffff880818dfc620 ffff880818dfc620 ffff880818004d00
ffff880818005e40
[   21.967156] Call Trace:
[   21.967162]  [<ffffffff811498c6>] fd_install+0x25/0x27
[   21.967168]  [<ffffffff81167a7c>] fanotify_read+0x38d/0x475
[   21.967176]  [<ffffffff8106716e>] ? remove_wait_queue+0x3a/0x3a
[   21.967181]  [<ffffffff81133e21>] vfs_read+0xa9/0xf0
[   21.967186]  [<ffffffff811422cb>] ? poll_select_set_timeout+0x63/0x81
[   21.967189]  [<ffffffff81133ec1>] sys_read+0x59/0x7e
[   21.967195]  [<ffffffff814bd699>] system_call_fastpath+0x16/0x1b
[   21.967222] Code: 66 66 90 55 48 89 e5 41 55 49 89 d5 41 54 41 89
f4 53 48 89 fb 48 8d bf 80 00 00 00 41 53 e8 69 ce 36 00 48 8b 43 08
48 8b 50 08 <4a> 83 3c e2 00 74 02 0f 0b 48 8b 40 08 4e 89 2c e0 66 83
83 80
[   21.967226] RIP  [<ffffffff8114987f>] __fd_install+0x2d/0x4f
[   21.967227]  RSP <ffff8808187f7de8>
[   21.967228] CR2: ffff881819b0fff8

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH 00/31] Latest numa/core patches, v15

[Linus Torvalds]

On Fri, Nov 16, 2012 at 10:45 PM, Alex Shi <lkml.alex@gmail.com> wrote:
> had caught a ops on my 2 sockets SNB EP server. but can not reproduce it.
> send out as a reminder:
> on tip/master, head : a7b7a8ad4476bb641c8455a4e0d7d0fd3eb86f90

This is an independent bug, nothing to do with the NUMA stuff. Fixed
in my tree now (commit 3587b1b097d70).

Of course, it's entirely possible that the NUMA patches are subtly
buggy and helped trigger the fanotify OVERFLOW event that had this
particular bug. But the oops itself is due to a real bug.

           Linus

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>