From: Mel Gorman <mgorman@suse.de>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Andrea Arcangeli <aarcange@redhat.com>,
Ingo Molnar <mingo@kernel.org>
Cc: Rik van Riel <riel@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Hugh Dickins <hughd@google.com>,
Thomas Gleixner <tglx@linutronix.de>,
Paul Turner <pjt@google.com>,
Lee Schermerhorn <Lee.Schermerhorn@hp.com>,
Alex Shi <lkml.alex@gmail.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
Linux-MM <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>, Mel Gorman <mgorman@suse.de>
Subject: [PATCH 15/46] mm: numa: Create basic numa page hinting infrastructure
Date: Wed, 21 Nov 2012 10:21:21 +0000 [thread overview]
Message-ID: <1353493312-8069-16-git-send-email-mgorman@suse.de> (raw)
In-Reply-To: <1353493312-8069-1-git-send-email-mgorman@suse.de>
Note: This patch started as "mm/mpol: Create special PROT_NONE
infrastructure" and preserves the basic idea but steals *very*
heavily from "autonuma: numa hinting page faults entry points" for
the actual fault handlers without the migration parts. The end
result is barely recognisable as either patch so all Signed-off
and Reviewed-bys are dropped. If Peter, Ingo and Andrea are ok with
this version, I will re-add the signed-offs-by to reflect the history.
In order to facilitate a lazy -- fault driven -- migration of pages, create
a special transient PAGE_NUMA variant, we can then use the 'spurious'
protection faults to drive our migrations from.
The meaning of PAGE_NUMA depends on the architecture but on x86 it is
effectively PROT_NONE. Actual PROT_NONE mappings will not generate these
NUMA faults for the reason that the page fault code checks the permission on
the VMA (and will throw a segmentation fault on actual PROT_NONE mappings),
before it ever calls handle_mm_fault.
[dhillf@gmail.com: Fix typo]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
---
include/linux/huge_mm.h | 10 +++++
mm/huge_memory.c | 21 ++++++++++
mm/memory.c | 104 +++++++++++++++++++++++++++++++++++++++++++++--
3 files changed, 132 insertions(+), 3 deletions(-)
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index b31cb7d..a13ebb1 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -159,6 +159,10 @@ static inline struct page *compound_trans_head(struct page *page)
}
return page;
}
+
+extern int do_huge_pmd_numa_page(struct mm_struct *mm, unsigned long addr,
+ pmd_t pmd, pmd_t *pmdp);
+
#else /* CONFIG_TRANSPARENT_HUGEPAGE */
#define HPAGE_PMD_SHIFT ({ BUILD_BUG(); 0; })
#define HPAGE_PMD_MASK ({ BUILD_BUG(); 0; })
@@ -195,6 +199,12 @@ static inline int pmd_trans_huge_lock(pmd_t *pmd,
{
return 0;
}
+
+static inline int do_huge_pmd_numa_page(struct mm_struct *mm, unsigned long addr,
+ pmd_t pmd, pmd_t *pmdp);
+{
+}
+
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
#endif /* _LINUX_HUGE_MM_H */
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 3aaf242..7224efd 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1017,6 +1017,27 @@ out:
return page;
}
+/* NUMA hinting page fault entry point for trans huge pmds */
+int do_huge_pmd_numa_page(struct mm_struct *mm, unsigned long addr,
+ pmd_t pmd, pmd_t *pmdp)
+{
+ struct page *page;
+
+ spin_lock(&mm->page_table_lock);
+ if (unlikely(!pmd_same(pmd, *pmdp)))
+ goto out_unlock;
+
+ page = pmd_page(pmd);
+ pmd = pmd_mknonnuma(pmd);
+ set_pmd_at(mm, haddr, pmdp, pmd);
+ VM_BUG_ON(pmd_numa(*pmdp));
+ update_mmu_cache_pmd(vma, addr, pmdp);
+
+out_unlock:
+ spin_unlock(&mm->page_table_lock);
+ return 0;
+}
+
int zap_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma,
pmd_t *pmd, unsigned long addr)
{
diff --git a/mm/memory.c b/mm/memory.c
index 73834e7..277b6d8 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3448,6 +3448,95 @@ static int do_nonlinear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte);
}
+int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long addr, pte_t pte, pte_t *ptep, pmd_t *pmd)
+{
+ struct page *page;
+ spinlock_t *ptl;
+
+ /*
+ * The "pte" at this point cannot be used safely without
+ * validation through pte_unmap_same(). It's of NUMA type but
+ * the pfn may be screwed if the read is non atomic.
+ *
+ * ptep_modify_prot_start is not called as this is clearing
+ * the _PAGE_NUMA bit and it is not really expected that there
+ * would be concurrent hardware modifications to the PTE.
+ */
+ ptl = pte_lockptr(mm, pmd);
+ spin_lock(ptl);
+ if (unlikely(!pte_same(*ptep, pte)))
+ goto out_unlock;
+ pte = pte_mknonnuma(pte);
+ set_pte_at(mm, addr, ptep, pte);
+ update_mmu_cache(vma, addr, ptep);
+
+ page = vm_normal_page(vma, addr, pte);
+ if (!page) {
+ pte_unmap_unlock(ptep, ptl);
+ return 0;
+ }
+
+out_unlock:
+ pte_unmap_unlock(ptep, ptl);
+ return 0;
+}
+
+/* NUMA hinting page fault entry point for regular pmds */
+int do_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma,
+ unsigned long addr, pmd_t *pmdp)
+{
+ pmd_t pmd;
+ pte_t *pte, *orig_pte;
+ unsigned long _addr = addr & PMD_MASK;
+ unsigned long offset;
+ spinlock_t *ptl;
+ bool numa = false;
+
+ spin_lock(&mm->page_table_lock);
+ pmd = *pmdp;
+ if (pmd_numa(pmd)) {
+ set_pmd_at(mm, _addr, pmdp, pmd_mknonnuma(pmd));
+ numa = true;
+ }
+ spin_unlock(&mm->page_table_lock);
+
+ if (!numa)
+ return 0;
+
+ /* we're in a page fault so some vma must be in the range */
+ BUG_ON(!vma);
+ BUG_ON(vma->vm_start >= _addr + PMD_SIZE);
+ offset = max(_addr, vma->vm_start) & ~PMD_MASK;
+ VM_BUG_ON(offset >= PMD_SIZE);
+ orig_pte = pte = pte_offset_map_lock(mm, pmdp, _addr, &ptl);
+ pte += offset >> PAGE_SHIFT;
+ for (addr = _addr + offset; addr < _addr + PMD_SIZE; pte++, addr += PAGE_SIZE) {
+ pte_t pteval = *pte;
+ struct page *page;
+ if (!pte_present(pteval))
+ continue;
+ if (!pte_numa(pteval))
+ continue;
+ if (addr >= vma->vm_end) {
+ vma = find_vma(mm, addr);
+ /* there's a pte present so there must be a vma */
+ BUG_ON(!vma);
+ BUG_ON(addr < vma->vm_start);
+ }
+ if (pte_numa(pteval)) {
+ pteval = pte_mknonnuma(pteval);
+ set_pte_at(mm, addr, pte, pteval);
+ }
+ page = vm_normal_page(vma, addr, pteval);
+ if (unlikely(!page))
+ continue;
+ }
+ pte_unmap_unlock(orig_pte, ptl);
+
+ return 0;
+}
+
/*
* These routines also need to handle stuff like marking pages dirty
* and/or accessed for architectures that don't do it in hardware (most
@@ -3486,6 +3575,9 @@ int handle_pte_fault(struct mm_struct *mm,
pte, pmd, flags, entry);
}
+ if (pte_numa(entry))
+ return do_numa_page(mm, vma, address, entry, pte, pmd);
+
ptl = pte_lockptr(mm, pmd);
spin_lock(ptl);
if (unlikely(!pte_same(*pte, entry)))
@@ -3554,9 +3646,11 @@ retry:
barrier();
if (pmd_trans_huge(orig_pmd)) {
- if (flags & FAULT_FLAG_WRITE &&
- !pmd_write(orig_pmd) &&
- !pmd_trans_splitting(orig_pmd)) {
+ if (pmd_numa(*pmd))
+ return do_huge_pmd_numa_page(mm, address,
+ orig_pmd, pmd);
+
+ if ((flags & FAULT_FLAG_WRITE) && !pmd_write(orig_pmd)) {
ret = do_huge_pmd_wp_page(mm, vma, address, pmd,
orig_pmd);
/*
@@ -3568,10 +3662,14 @@ retry:
goto retry;
return ret;
}
+
return 0;
}
}
+ if (pmd_numa(*pmd))
+ return do_pmd_numa_page(mm, vma, address, pmd);
+
/*
* Use __pte_alloc instead of pte_alloc_map, because we can't
* run pte_offset_map on the pmd, if an huge pmd could
--
1.7.9.2
next prev parent reply other threads:[~2012-11-21 10:30 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-21 10:21 [PATCH 00/46] Automatic NUMA Balancing V4 Mel Gorman
2012-11-21 10:21 ` [PATCH 01/46] x86: mm: only do a local tlb flush in ptep_set_access_flags() Mel Gorman
2012-11-21 10:21 ` [PATCH 02/46] x86: mm: drop TLB flush from ptep_set_access_flags Mel Gorman
2012-11-21 10:21 ` [PATCH 03/46] mm,generic: only flush the local TLB in ptep_set_access_flags Mel Gorman
2012-11-21 10:21 ` [PATCH 04/46] x86/mm: Introduce pte_accessible() Mel Gorman
2012-11-21 10:21 ` [PATCH 05/46] mm: Only flush the TLB when clearing an accessible pte Mel Gorman
2012-11-21 10:21 ` [PATCH 06/46] mm: Count the number of pages affected in change_protection() Mel Gorman
2012-11-21 10:21 ` [PATCH 07/46] mm: Optimize the TLB flush of sys_mprotect() and change_protection() users Mel Gorman
2012-11-21 10:21 ` [PATCH 08/46] mm: compaction: Move migration fail/success stats to migrate.c Mel Gorman
2012-11-21 10:21 ` [PATCH 09/46] mm: migrate: Add a tracepoint for migrate_pages Mel Gorman
2012-11-21 10:21 ` [PATCH 10/46] mm: compaction: Add scanned and isolated counters for compaction Mel Gorman
2012-11-21 10:21 ` [PATCH 11/46] mm: numa: define _PAGE_NUMA Mel Gorman
2012-11-21 10:21 ` [PATCH 12/46] mm: numa: pte_numa() and pmd_numa() Mel Gorman
2012-11-21 10:21 ` [PATCH 13/46] mm: numa: Support NUMA hinting page faults from gup/gup_fast Mel Gorman
2012-11-21 10:21 ` [PATCH 14/46] mm: numa: split_huge_page: transfer the NUMA type from the pmd to the pte Mel Gorman
2012-11-21 10:21 ` Mel Gorman [this message]
2012-11-21 10:21 ` [PATCH 16/46] mm: mempolicy: Make MPOL_LOCAL a real policy Mel Gorman
2012-11-21 10:21 ` [PATCH 17/46] mm: mempolicy: Add MPOL_MF_NOOP Mel Gorman
2012-11-21 10:21 ` [PATCH 18/46] mm: mempolicy: Check for misplaced page Mel Gorman
2012-11-21 10:21 ` [PATCH 19/46] mm: migrate: Introduce migrate_misplaced_page() Mel Gorman
2012-11-21 10:21 ` [PATCH 20/46] mm: mempolicy: Use _PAGE_NUMA to migrate pages Mel Gorman
2012-11-21 10:21 ` [PATCH 21/46] mm: mempolicy: Add MPOL_MF_LAZY Mel Gorman
2012-11-21 10:21 ` [PATCH 22/46] mm: mempolicy: Implement change_prot_numa() in terms of change_protection() Mel Gorman
2012-11-21 10:21 ` [PATCH 23/46] mm: mempolicy: Hide MPOL_NOOP and MPOL_MF_LAZY from userspace for now Mel Gorman
2012-11-21 10:21 ` [PATCH 24/46] mm: numa: Add fault driven placement and migration Mel Gorman
2012-11-21 10:21 ` [PATCH 25/46] mm: sched: numa: Implement constant, per task Working Set Sampling (WSS) rate Mel Gorman
2012-11-21 10:21 ` [PATCH 26/46] sched, numa, mm: Count WS scanning against present PTEs, not virtual memory ranges Mel Gorman
2012-11-21 10:21 ` [PATCH 27/46] mm: sched: numa: Implement slow start for working set sampling Mel Gorman
2012-11-21 10:21 ` [PATCH 28/46] mm: numa: Add pte updates, hinting and migration stats Mel Gorman
2012-11-21 10:21 ` [PATCH 29/46] mm: numa: Migrate on reference policy Mel Gorman
2012-11-21 10:21 ` [PATCH 30/46] mm: numa: Migrate pages handled during a pmd_numa hinting fault Mel Gorman
2012-11-21 10:21 ` [PATCH 31/46] mm: numa: Structures for Migrate On Fault per NUMA migration rate limiting Mel Gorman
2012-11-21 10:21 ` [PATCH 32/46] mm: numa: Rate limit the amount of memory that is migrated between nodes Mel Gorman
2012-11-21 10:21 ` [PATCH 33/46] mm: numa: Rate limit setting of pte_numa if node is saturated Mel Gorman
2012-11-21 10:21 ` [PATCH 34/46] sched: numa: Slowly increase the scanning period as NUMA faults are handled Mel Gorman
2012-11-21 10:21 ` [PATCH 35/46] mm: numa: Introduce last_nid to the page frame Mel Gorman
2012-11-21 10:21 ` [PATCH 36/46] mm: numa: Use a two-stage filter to restrict pages being migrated for unlikely task<->node relationships Mel Gorman
2012-11-21 18:25 ` Ingo Molnar
2012-11-21 19:15 ` Mel Gorman
2012-11-21 19:39 ` Mel Gorman
2012-11-21 19:46 ` Rik van Riel
2012-11-22 0:05 ` Ingo Molnar
2012-11-21 10:21 ` [PATCH 37/46] mm: numa: Add THP migration for the NUMA working set scanning fault case Mel Gorman
2012-11-21 11:24 ` Mel Gorman
2012-11-21 12:21 ` Mel Gorman
2012-11-21 10:21 ` [PATCH 38/46] sched: numa: Introduce tsk_home_node() Mel Gorman
2012-11-21 10:21 ` [PATCH 39/46] sched: numa: Make find_busiest_queue() a method Mel Gorman
2012-11-21 10:21 ` [PATCH 40/46] sched: numa: Implement home-node awareness Mel Gorman
2012-11-21 10:21 ` [PATCH 41/46] sched: numa: Introduce per-mm and per-task structures Mel Gorman
2012-11-21 10:21 ` [PATCH 42/46] sched: numa: CPU follows memory Mel Gorman
2012-11-21 10:21 ` [PATCH 43/46] sched: numa: Rename mempolicy to HOME Mel Gorman
2012-11-21 10:21 ` [PATCH 44/46] sched: numa: Consider only one CPU per node for CPU-follows-memory Mel Gorman
2012-11-21 10:21 ` [PATCH 45/46] balancenuma: no task swap in finding placement Mel Gorman
2012-11-21 10:21 ` [PATCH 46/46] Simple CPU follow Mel Gorman
2012-11-21 16:53 ` [PATCH 00/46] Automatic NUMA Balancing V4 Mel Gorman
2012-11-21 17:03 ` Ingo Molnar
2012-11-21 17:20 ` Mel Gorman
2012-11-21 17:33 ` Ingo Molnar
2012-11-21 18:02 ` Mel Gorman
2012-11-21 18:21 ` Ingo Molnar
2012-11-21 19:01 ` Mel Gorman
2012-11-21 23:27 ` Ingo Molnar
2012-11-22 9:32 ` Mel Gorman
2012-11-22 9:05 ` Ingo Molnar
2012-11-22 9:43 ` Mel Gorman
2012-11-22 12:56 ` Mel Gorman
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=1353493312-8069-16-git-send-email-mgorman@suse.de \
--to=mgorman@suse.de \
--cc=Lee.Schermerhorn@hp.com \
--cc=a.p.zijlstra@chello.nl \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lkml.alex@gmail.com \
--cc=mingo@kernel.org \
--cc=pjt@google.com \
--cc=riel@redhat.com \
--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).