From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Andrew Morton <akpm@linux-foundation.org>,
David Miller <davem@davemloft.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
Benjamin Herrenschmidt <benh@kernel.crashing.org>
Subject: [PATCH 25/50] memblock: NUMA allocate can now use early_pfn_map
Date: Tue, 13 Jul 2010 00:10:19 -0700 [thread overview]
Message-ID: <1279005044-24777-26-git-send-email-yinghai@kernel.org> (raw)
In-Reply-To: <1279005044-24777-1-git-send-email-yinghai@kernel.org>
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We now provide a default (weak) implementation of memblock_nid_range()
which uses the early_pfn_map[] if CONFIG_ARCH_POPULATES_NODE_MAP
is set. Sparc still needs to use its own method due to the way
the pages can be scattered between nodes.
This implementation is inefficient due to our main algorithm and
callback construct wanting to work on an ascending addresses bases
while early_pfn_map[] would rather work with nid's (it's unsorted
at that stage). But it should work and we can look into improving
it subsequently, possibly using arch compile options to chose a
different algorithm alltogether.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
include/linux/memblock.h | 3 +++
mm/memblock.c | 28 +++++++++++++++++++++++++++-
2 files changed, 30 insertions(+), 1 deletions(-)
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 15da7d9..b69c243 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -47,6 +47,9 @@ extern long memblock_remove(phys_addr_t base, phys_addr_t size);
extern long __init memblock_free(phys_addr_t base, phys_addr_t size);
extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
+/* The numa aware allocator is only available if
+ * CONFIG_ARCH_POPULATES_NODE_MAP is set
+ */
extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid);
extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
diff --git a/mm/memblock.c b/mm/memblock.c
index b4870cf..a9c15a5 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -15,6 +15,7 @@
#include <linux/init.h>
#include <linux/bitops.h>
#include <linux/poison.h>
+#include <linux/pfn.h>
#include <linux/memblock.h>
struct memblock memblock;
@@ -451,11 +452,36 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
/*
* Additional node-local allocators. Search for node memory is bottom up
* and walks memblock regions within that node bottom-up as well, but allocation
- * within an memblock region is top-down.
+ * within an memblock region is top-down. XXX I plan to fix that at some stage
+ *
+ * WARNING: Only available after early_node_map[] has been populated,
+ * on some architectures, that is after all the calls to add_active_range()
+ * have been done to populate it.
*/
phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
{
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+ /*
+ * This code originates from sparc which really wants use to walk by addresses
+ * and returns the nid. This is not very convenient for early_pfn_map[] users
+ * as the map isn't sorted yet, and it really wants to be walked by nid.
+ *
+ * For now, I implement the inefficient method below which walks the early
+ * map multiple times. Eventually we may want to use an ARCH config option
+ * to implement a completely different method for both case.
+ */
+ unsigned long start_pfn, end_pfn;
+ int i;
+
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ get_pfn_range_for_nid(i, &start_pfn, &end_pfn);
+ if (start < PFN_PHYS(start_pfn) || start >= PFN_PHYS(end_pfn))
+ continue;
+ *nid = i;
+ return min(end, PFN_PHYS(end_pfn));
+ }
+#endif
*nid = 0;
return end;
--
1.6.4.2
WARNING: multiple messages have this Message-ID (diff)
From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Andrew Morton <akpm@linux-foundation.org>,
David Miller <davem@davemloft.net>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org
Subject: [PATCH 25/50] memblock: NUMA allocate can now use early_pfn_map
Date: Tue, 13 Jul 2010 00:10:19 -0700 [thread overview]
Message-ID: <1279005044-24777-26-git-send-email-yinghai@kernel.org> (raw)
Message-ID: <20100713071019.p1kjhaXfXiTTX0veV2DncNOrokudIAHypGkGRdQn69Q@z> (raw)
In-Reply-To: <1279005044-24777-1-git-send-email-yinghai@kernel.org>
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
We now provide a default (weak) implementation of memblock_nid_range()
which uses the early_pfn_map[] if CONFIG_ARCH_POPULATES_NODE_MAP
is set. Sparc still needs to use its own method due to the way
the pages can be scattered between nodes.
This implementation is inefficient due to our main algorithm and
callback construct wanting to work on an ascending addresses bases
while early_pfn_map[] would rather work with nid's (it's unsorted
at that stage). But it should work and we can look into improving
it subsequently, possibly using arch compile options to chose a
different algorithm alltogether.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
include/linux/memblock.h | 3 +++
mm/memblock.c | 28 +++++++++++++++++++++++++++-
2 files changed, 30 insertions(+), 1 deletions(-)
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index 15da7d9..b69c243 100644
--- a/include/linux/memblock.h
+++ b/include/linux/memblock.h
@@ -47,6 +47,9 @@ extern long memblock_remove(phys_addr_t base, phys_addr_t size);
extern long __init memblock_free(phys_addr_t base, phys_addr_t size);
extern long __init memblock_reserve(phys_addr_t base, phys_addr_t size);
+/* The numa aware allocator is only available if
+ * CONFIG_ARCH_POPULATES_NODE_MAP is set
+ */
extern phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid);
extern phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align);
diff --git a/mm/memblock.c b/mm/memblock.c
index b4870cf..a9c15a5 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -15,6 +15,7 @@
#include <linux/init.h>
#include <linux/bitops.h>
#include <linux/poison.h>
+#include <linux/pfn.h>
#include <linux/memblock.h>
struct memblock memblock;
@@ -451,11 +452,36 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
/*
* Additional node-local allocators. Search for node memory is bottom up
* and walks memblock regions within that node bottom-up as well, but allocation
- * within an memblock region is top-down.
+ * within an memblock region is top-down. XXX I plan to fix that at some stage
+ *
+ * WARNING: Only available after early_node_map[] has been populated,
+ * on some architectures, that is after all the calls to add_active_range()
+ * have been done to populate it.
*/
phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
{
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+ /*
+ * This code originates from sparc which really wants use to walk by addresses
+ * and returns the nid. This is not very convenient for early_pfn_map[] users
+ * as the map isn't sorted yet, and it really wants to be walked by nid.
+ *
+ * For now, I implement the inefficient method below which walks the early
+ * map multiple times. Eventually we may want to use an ARCH config option
+ * to implement a completely different method for both case.
+ */
+ unsigned long start_pfn, end_pfn;
+ int i;
+
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ get_pfn_range_for_nid(i, &start_pfn, &end_pfn);
+ if (start < PFN_PHYS(start_pfn) || start >= PFN_PHYS(end_pfn))
+ continue;
+ *nid = i;
+ return min(end, PFN_PHYS(end_pfn));
+ }
+#endif
*nid = 0;
return end;
--
1.6.4.2
next prev parent reply other threads:[~2010-07-13 7:14 UTC|newest]
Thread overview: 103+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-13 7:09 [PATCH -v24 00/50] Use memblock with x86 Yinghai Lu
2010-07-13 7:09 ` Yinghai Lu
2010-07-13 7:09 ` [PATCH 01/50] x86, numa: fix boot without RAM on node0 again Yinghai Lu
2010-07-13 7:09 ` Yinghai Lu
2010-07-13 7:09 ` [PATCH 02/50] x86,mm: fix 32bit numa sparsemem Yinghai Lu
2010-07-13 7:09 ` Yinghai Lu
2010-07-13 7:09 ` [PATCH 03/50] lmb: rename to memblock Yinghai Lu
2010-07-13 7:09 ` [PATCH 04/50] memblock: Rename memblock_region to memblock_type and memblock_property to memblock_region Yinghai Lu
2010-07-13 7:09 ` Yinghai Lu
2010-07-13 7:09 ` [PATCH 05/50] memblock: No reason to include asm/memblock.h late Yinghai Lu
2010-07-13 7:09 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 06/50] memblock: Introduce for_each_memblock() and new accessors, and use it Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 07/50] memblock: Remove nid_range argument, arch provides memblock_nid_range() instead Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 08/50] memblock: Factor the lowest level alloc function Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 09/50] memblock: Expose MEMBLOCK_ALLOC_ANYWHERE Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 10/50] memblock: Introduce default allocation limit and use it to replace explicit ones Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 11/50] memblock: Remove rmo_size, burry it in arch/powerpc where it belongs Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 12/50] memblock: Change u64 to phys_addr_t Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 13/50] memblock: Remove unused memblock.debug struct member Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 14/50] memblock: Remove memblock_type.size and add memblock.memory_size instead Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 15/50] memblock: Move memblock arrays to static storage in memblock.c and make their size a variable Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 16/50] memblock: Add debug markers at the end of the array Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 17/50] memblock: Make memblock_find_region() out of memblock_alloc_region() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 18/50] memblock: Define MEMBLOCK_ERROR internally instead of using ~(phys_addr_t)0 Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 19/50] memblock: Move memblock_init() to the bottom of the file Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 20/50] memblock: split memblock_find_base() out of __memblock_alloc_base() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 21/50] memblock: Move functions around into a more sensible order Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 22/50] memblock: Add array resizing support Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 23/50] memblock: Add arch function to control coalescing of memblock memory regions Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 24/50] memblock: Add "start" argument to memblock_find_base() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu [this message]
2010-07-13 7:10 ` [PATCH 25/50] memblock: NUMA allocate can now use early_pfn_map Yinghai Lu
2010-07-13 7:10 ` [PATCH 26/50] memblock: Separate memblock_alloc_nid() and memblock_alloc_try_nid() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 27/50] memblock: Make memblock_alloc_try_nid() fallback to MEMBLOCK_ALLOC_ANYWHERE Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 28/50] memblock: Add debugfs files to dump the arrays content Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 29/50] memblock: Prepare x86 to use memblock to replace early_res Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 30/50] memblock: Print new doubled array location info Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 31/50] memblock: Export MEMBLOCK_ERROR again Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 32/50] memblock: Prepare to include linux/memblock.h in core file Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 33/50] memblock: Add ARCH_DISCARD_MEMBLOCK to put memblock code to .init Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 34/50] memblock: Add memblock_find_in_range() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 35/50] x86, memblock: Add memblock_x86_find_in_range_size() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 36/50] bootmem, x86: Add weak version of reserve_bootmem_generic Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 37/50] x86, memblock: Add memblock_x86_to_bootmem() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 38/50] x86,memblock: Add memblock_x86_reserve_range/memblock_x86_free_range Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 39/50] x86, memblock: Add get_free_all_memory_range() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 40/50] x86, memblock: Add memblock_x86_register_active_regions() and memblock_x86_hole_size() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 41/50] memblock: Add find_memory_core_early() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 42/50] x86, memblock: Add memblock_x86_find_in_range_node() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 43/50] x86, memblock: Add memblock_x86_free_memory_in_range() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 44/50] x86, memblock: Add memblock_x86_memory_in_range() Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 45/50] x86, memblock: Use memblock_debug to control debug message print out Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 20:37 ` Bjorn Helgaas
2010-07-13 20:40 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 46/50] x86: Use memblock to replace early_res Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 47/50] x86: Replace e820_/_early string with memblock_ Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 48/50] x86: Remove not used early_res code Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 49/50] x86, memblock: Use memblock_memory_size()/memblock_free_memory_size() to get correct dma_reserve Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
2010-07-13 7:10 ` [PATCH 50/50] x86: remove old bootmem code Yinghai Lu
2010-07-13 7:10 ` Yinghai Lu
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=1279005044-24777-26-git-send-email-yinghai@kernel.org \
--to=yinghai@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=benh@kernel.crashing.org \
--cc=davem@davemloft.net \
--cc=hannes@cmpxchg.org \
--cc=hpa@zytor.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--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).