From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: mingo@kernel.org
Cc: riel@redhat.com, oleg@redhat.com, pjt@google.com,
akpm@linux-foundation.org, torvalds@linux-foundation.org,
tglx@linutronix.de, Lee.Schermerhorn@hp.com,
linux-kernel@vger.kernel.org, Paul Mundt <lethal@linux-sh.org>
Subject: Re: [PATCH 18/19] sched, numa: Per task memory placement for big processes
Date: Wed, 08 Aug 2012 23:35:32 +0200 [thread overview]
Message-ID: <1344461732.16728.19.camel@laptop> (raw)
In-Reply-To: <20120731192809.428855038@chello.nl>
On Tue, 2012-07-31 at 21:12 +0200, Peter Zijlstra wrote:
> +#ifdef CONFIG_NUMA
> + /*
> + * XXX fold this into flags for 64bit or so...
> + */
> + int nid_last;
> +#endif
Something like the below? I still ought to update all the various
comments about page flag layout etc..
Also, that #warning gives a very noisy build indeed, I guess we should
either make it silent or increase the page frame size for those
configs.. 32bit NUMA is quite rare for normal people (sorry Paul) :)
---
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -611,10 +611,19 @@ static inline pte_t maybe_mkwrite(pte_t
#define NODES_WIDTH 0
#endif
+#if NODES_WIDTH && (SECTIONS_WIDTH+ZONES_WIDTH+2*NODES_SHIFT <= BITS_PER_LONG - NR_PAGEFLAGS)
+#define LAST_NID_WIDTH NODES_SHIFT
+#else
+#warning "faking page_xchg_last_nid"
+#define LAST_NID_NOT_IN_PAGE_FLAGS
+#define LAST_NID_WIDTH 0
+#endif
+
/* Page flags: | [SECTION] | [NODE] | ZONE | ... | FLAGS | */
#define SECTIONS_PGOFF ((sizeof(unsigned long)*8) - SECTIONS_WIDTH)
#define NODES_PGOFF (SECTIONS_PGOFF - NODES_WIDTH)
#define ZONES_PGOFF (NODES_PGOFF - ZONES_WIDTH)
+#define LAST_NID_PGOFF (ZONES_PGOFF - LAST_NID_WIDTH)
/*
* We are going to use the flags for the page to node mapping if its in
@@ -632,6 +641,7 @@ static inline pte_t maybe_mkwrite(pte_t
#define SECTIONS_PGSHIFT (SECTIONS_PGOFF * (SECTIONS_WIDTH != 0))
#define NODES_PGSHIFT (NODES_PGOFF * (NODES_WIDTH != 0))
#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_WIDTH != 0))
+#define LAST_NID_PGSHIFT (LAST_NID_PGOFF * (LAST_NID_WIDTH != 0))
/* NODE:ZONE or SECTION:ZONE is used to ID a zone for the buddy allocator */
#ifdef NODE_NOT_IN_PAGE_FLAGS
@@ -653,6 +663,7 @@ static inline pte_t maybe_mkwrite(pte_t
#define ZONES_MASK ((1UL << ZONES_WIDTH) - 1)
#define NODES_MASK ((1UL << NODES_WIDTH) - 1)
#define SECTIONS_MASK ((1UL << SECTIONS_WIDTH) - 1)
+#define LAST_NID_MASK ((1UL << LAST_NID_WIDTH) - 1)
#define ZONEID_MASK ((1UL << ZONEID_SHIFT) - 1)
static inline enum zone_type page_zonenum(const struct page *page)
@@ -691,6 +702,39 @@ static inline int page_to_nid(const stru
}
#endif
+#ifdef LAST_NID_NOT_IN_PAGE_FLAGS
+static inline int page_xchg_last_nid(struct page *page, int nid)
+{
+ return nid; /* fakin' it */
+}
+
+static inline int page_last_nid(struct page *page)
+{
+ return page_to_nid(page);
+}
+#else
+static inline int page_xchg_last_nid(struct page *page, int nid)
+{
+ unsigned long old_flags, flags;
+ int last_nid;
+
+ old_flags = flags = page->flags;
+ last_nid = (flags >> LAST_NID_PGSHIFT) & LAST_NID_MASK;
+
+ flags &= ~(LAST_NID_MASK << LAST_NID_PGSHIFT);
+ flags |= (nid & LAST_NID_MASK) << LAST_NID_PGSHIFT;
+
+ (void)cmpxchg(&page->flags, old_flags, flags);
+
+ return last_nid;
+}
+
+static inline int page_last_nid(struct page *page)
+{
+ return (page->flags >> LAST_NID_PGSHIFT) & LAST_NID_MASK;
+}
+#endif /* LAST_NID_NOT_IN_PAGE_FLAGS */
+
static inline struct zone *page_zone(const struct page *page)
{
return &NODE_DATA(page_to_nid(page))->node_zones[page_zonenum(page)];
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -176,12 +176,6 @@ struct page {
*/
void *shadow;
#endif
-#ifdef CONFIG_NUMA
- /*
- * XXX fold this into flags for 64bit or so...
- */
- int nid_last;
-#endif
}
/*
* The struct page can be forced to be double word aligned so that atomic ops
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1366,6 +1366,7 @@ static void __split_huge_page_refcount(s
page_tail->mapping = page->mapping;
page_tail->index = page->index + i;
+ page_xchg_last_nid(page, page_last_nid(tail_page));
page_tail->nid_last = page->nid_last;
BUG_ON(!PageAnon(page_tail));
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -2265,10 +2265,9 @@ int mpol_misplaced(struct page *page, st
* task_tick_numa().
*/
if (multi && (pol->flags & MPOL_F_HOME)) {
- if (page->nid_last != polnid) {
- page->nid_last = polnid;
+ int last_nid = page_xchg_last_nid(page, polnid);
+ if (last_nid != polnid)
goto out;
- }
}
if (curnid != polnid)
next prev parent reply other threads:[~2012-08-08 21:35 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-31 19:12 [PATCH 00/19] sched-numa rewrite Peter Zijlstra
2012-07-31 19:12 ` [PATCH 01/19] task_work: Remove dependency on sched.h Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 02/19] mm/mpol: Remove NUMA_INTERLEAVE_HIT Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-08-09 21:41 ` Andrea Arcangeli
2012-08-10 0:50 ` Andi Kleen
2012-07-31 19:12 ` [PATCH 03/19] mm/mpol: Make MPOL_LOCAL a real policy Peter Zijlstra
2012-07-31 20:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 04/19] mm, thp: Preserve pgprot across huge page split Peter Zijlstra
2012-07-31 20:53 ` Rik van Riel
2012-08-09 21:42 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 05/19] mm, mpol: Create special PROT_NONE infrastructure Peter Zijlstra
2012-07-31 20:55 ` Rik van Riel
2012-08-09 21:43 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 06/19] mm/mpol: Add MPOL_MF_LAZY Peter Zijlstra
2012-07-31 21:04 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 07/19] mm/mpol: Add MPOL_MF_NOOP Peter Zijlstra
2012-07-31 21:06 ` Rik van Riel
2012-08-09 21:44 ` Andrea Arcangeli
2012-10-01 9:36 ` Michael Kerrisk
2012-10-01 9:45 ` Ingo Molnar
2012-07-31 19:12 ` [PATCH 08/19] mm/mpol: Check for misplaced page Peter Zijlstra
2012-07-31 21:13 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 09/19] mm, migrate: Introduce migrate_misplaced_page() Peter Zijlstra
2012-07-31 21:16 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 10/19] mm, mpol: Use special PROT_NONE to migrate pages Peter Zijlstra
2012-07-31 21:24 ` Rik van Riel
2012-08-09 21:44 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 11/19] sched, mm: Introduce tsk_home_node() Peter Zijlstra
2012-07-31 21:30 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 12/19] mm/mpol: Make mempolicy home-node aware Peter Zijlstra
2012-07-31 21:33 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 13/19] sched: Introduce sched_feat_numa() Peter Zijlstra
2012-07-31 21:34 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 14/19] sched: Make find_busiest_queue() a method Peter Zijlstra
2012-07-31 21:34 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 15/19] sched: Implement home-node awareness Peter Zijlstra
2012-07-31 21:52 ` Rik van Riel
2012-08-09 21:51 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 16/19] sched, numa: NUMA home-node selection code Peter Zijlstra
2012-07-31 21:52 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 17/19] sched, numa: Detect big processes Peter Zijlstra
2012-07-31 21:53 ` Rik van Riel
2012-07-31 19:12 ` [PATCH 18/19] sched, numa: Per task memory placement for " Peter Zijlstra
2012-07-31 21:56 ` Rik van Riel
2012-08-08 21:35 ` Peter Zijlstra [this message]
2012-08-09 21:57 ` Andrea Arcangeli
2012-07-31 19:12 ` [PATCH 19/19] mm, numa: retry failed page migrations Peter Zijlstra
2012-08-02 20:40 ` Christoph Lameter
2012-08-08 17:17 ` [PATCH 00/19] sched-numa rewrite Andrea Arcangeli
2012-08-08 18:43 ` Rik van Riel
2012-08-17 18:08 ` Andrea Arcangeli
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=1344461732.16728.19.camel@laptop \
--to=a.p.zijlstra@chello.nl \
--cc=Lee.Schermerhorn@hp.com \
--cc=akpm@linux-foundation.org \
--cc=lethal@linux-sh.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--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 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.