Commit-ID: 2532fc318db0e1fe68e01407ee27634c76916e44
Gitweb: http://git.kernel.org/tip/2532fc318db0e1fe68e01407ee27634c76916e44
Author: Tang Chen
AuthorDate: Thu, 25 Aug 2016 16:35:14 +0800
Committer: Thomas Gleixner
CommitDate: Wed, 21 Sep 2016 21:18:38 +0200
x86/numa: Online memory
Hi Rafael, Len,
Would you please help to review the ACPI part of this patch-set ?
Thanks.
On 01/25/2016 02:08 PM, Tang Chen wrote:
[Problem]
cpuid <-> nodeid mapping is firstly established at boot time. And workqueue
caches
the mapping in wq_numa_possible_cpumask in wq_numa_init() a
Hi tj,
I have resent the patch-set and to Rafael and Len.
Thanks.
On 01/22/2016 05:26 AM, Tejun Heo wrote:
Hello,
Most changes being in ACPI, I think it probably would be a good idea
to cc Rafael and Len Brown.
Thanks.
For now, x86 does not support memory-less node. A node without memory
will not be onlined, and the cpus on it will be mapped to the other
online nodes with memory in init_cpu_to_node(). The reason of doing this
is to ensure each cpu has mapped to a node with memory, so that it will
be able to alloc
<-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/include/asm/mpspec.h
s step 4.
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/ia64/kernel/acpi.c | 2 +-
arch/x86/kernel/acpi/b
control if disabled processors are ignored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_core.c | 57 +++
2 files changed, 40 insertions(+), 22 deletions(-)
diff --git a/drivers/
all possible cpus at
boot time.
x86, acpi, cpu-hotplug: Introduce cpuid_to_apicid[] array to store
persistent cpuid <-> apicid mapping.
x86, acpi, cpu-hotplug: Enable MADT APIs to return disabled apicid.
x86, acpi, cpu-hotplug: Set persistent cpuid <-> nodeid mapping when
27;
apicid.
This is also done by introducing an extra parameter to these apis to let the
caller
control if disabled cpus are ignored.
4. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step
27;
apicid.
This is also done by introducing an extra parameter to these apis to let the
caller
control if disabled cpus are ignored.
4. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step
<-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/include/asm/mpspec.h
control if disabled processors are ignored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_core.c | 57 +++
2 files changed, 40 insertions(+), 22 deletions(-)
diff --git a/drivers/
s step 4.
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/ia64/kernel/acpi.c | 2 +-
arch/x86/kernel/acpi/b
all possible cpus at
boot time.
x86, acpi, cpu-hotplug: Introduce cpuid_to_apicid[] array to store
persistent cpuid <-> apicid mapping.
x86, acpi, cpu-hotplug: Enable MADT APIs to return disabled apicid.
x86, acpi, cpu-hotplug: Set persistent cpuid <-> nodeid mapping when
For now, x86 does not support memory-less node. A node without memory
will not be onlined, and the cpus on it will be mapped to the other
online nodes with memory in init_cpu_to_node(). The reason of doing this
is to ensure each cpu has mapped to a node with memory, so that it will
be able to alloc
e
persistent cpuid <-> apicid mapping.
x86, acpi, cpu-hotplug: Enable MADT APIs to return disabled apicid.
x86, acpi, cpu-hotplug: Set persistent cpuid <-> nodeid mapping when
booting.
Tang Chen (1):
x86, memhp, numa: Online memory-less nodes
From: Gu Zheng
This patch finishes step 4.
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/ia64/kernel/acpi.c
27;
apicid.
This is also done by introducing an extra parameter to these apis to let the
caller
control if disabled cpus are ignored.
4. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step
ns -ENODEV.
After this patch, a new parameter will be added to MADT APIs so that caller
is able to control if disabled processors are ignored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_core.c | 57 +
from local node, it will automatically fall back to the proper zones
in the zonelists.
Signed-off-by: Tang Chen
---
arch/x86/mm/numa.c | 27 +--
include/linux/mmzone.h | 1 +
mm/page_alloc.c| 2 +-
3 files changed, 15 insertions(+), 15 deletions(-)
diff -
d <-> nodeid persistent.
cpuid <-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by: Gu Zheng
Signed-off-by: Ta
h mem=0x20a000, Memory of
Node 2 and 3 are ignored and the Nodes become memoryless node.
OK, I'm using initrd overwrite. It could also fake the memory-less node.
Thanks.
Thanks,
Yasuaki Ishimatsu
On Thu, 19 Nov 2015 12:22:10 +0800
Tang Chen wrote:
[Problem]
cpuid <-> nodeid ma
On 11/24/2015 06:04 AM, Tejun Heo wrote:
Hello,
On Thu, Nov 19, 2015 at 12:22:10PM +0800, Tang Chen wrote:
[Solution]
There are four mappings in the kernel:
1. nodeid (logical node id) <-> pxm
2. apicid (physical cpu id) <-> nodeid
3. cpuid (logical cpu id) <->
Hi Shi,
Would you please share where did you add the printk debug info ?
Thanks. :)
On 11/20/2015 04:17 PM, Xishi Qiu wrote:
I find the direct mapping count in /proc/meminfo is error.
The value should be equal to the size of init_memory_mapping which
showed in boot log.
I add some print to sh
Hi,
Sorry for the terrible delay for this patch-set.
But unfortunately, they are still not fully tested for the memory-less
node case.
Please help to review first. Will soon do the tests.
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a messa
For now, x86 does not support memory-less node. A node without memory
will not be onlined, and the cpus on it will be mapped to the other
online nodes with memory in init_cpu_to_node(). The reason of doing this
is to ensure each cpu has mapped to a node with memory, so that it will
be able to alloc
d <-> nodeid persistent.
cpuid <-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by: Gu Zheng
Signed-off-by: Ta
From: Gu Zheng
This patch finishes step 4.
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/ia64/kernel/acpi.c
ns -ENODEV.
After this patch, a new parameter will be added to MADT APIs so that caller
is able to control if disabled processors are ignored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_core.c | 57 +
27;
apicid.
This is also done by introducing an extra parameter to these apis to let the
caller
control if disabled cpus are ignored.
4. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step
:
x86, acpi, cpu-hotplug: Enable acpi to register all possible cpus at
boot time.
x86, acpi, cpu-hotplug: Introduce cpuid_to_apicid[] array to store
persistent cpuid <-> apicid mapping.
x86, acpi, cpu-hotplug: Enable MADT APIs to return disabled apicid.
x86, acpi, cpu-hotpl
On 10/21/2015 05:18 PM, Tang Chen wrote:
On 10/21/2015 05:00 PM, Vlastimil Babka wrote:
On 10/21/2015 04:20 AM, Changsheng Liu wrote:
在 2015/10/15 0:18, Vlastimil Babka 写道:
On 10/12/2015 08:58 AM, Changsheng Liu wrote:
From: Changsheng Liu
After the user config CONFIG_MOVABLE_NODE
On 10/21/2015 05:00 PM, Vlastimil Babka wrote:
On 10/21/2015 04:20 AM, Changsheng Liu wrote:
在 2015/10/15 0:18, Vlastimil Babka 写道:
On 10/12/2015 08:58 AM, Changsheng Liu wrote:
From: Changsheng Liu
After the user config CONFIG_MOVABLE_NODE,
When the memory is hot added, should_add_memory
On 10/14/2015 11:28 AM, Xishi Qiu wrote:
On 2015/10/14 10:50, Tang Chen wrote:
Hi, Qiu
The patch seems OK to me. Only one little concern below.
On 10/12/2015 09:37 AM, Xishi Qiu wrote:
On 2015/10/9 23:41, Yasuaki Ishimatsu wrote:
On Thu, 8 Oct 2015 10:21:05 +0800
Xishi Qiu wrote:
If
Hi, Qiu
The patch seems OK to me. Only one little concern below.
On 10/12/2015 09:37 AM, Xishi Qiu wrote:
On 2015/10/9 23:41, Yasuaki Ishimatsu wrote:
On Thu, 8 Oct 2015 10:21:05 +0800
Xishi Qiu wrote:
If kernelcore was not specified, or the kernelcore size is zero
(required_movablecore >=
Hi Ishimatsu,
On 10/10/2015 02:13 AM, Yasuaki Ishimatsu wrote:
Hi Tang,
On Fri, 9 Oct 2015 15:59:38 +0800
Tang Chen wrote:
Hi,
I don't mean to offend, but I really think it is not necessary to do this.
hot-added memory will be added to ZONE_NORMAL by default. You can
modify it whe
Liu
Signed-off-by: Xiaofeng Yan
Tested-by: Dongdong Fan
Cc: Wang Nan
Cc: Zhang Yanfei
Cc: Dave Hansen
Cc: Yinghai Lu
Cc: Tang Chen
Cc: Hu Tao
Cc: Lai Jiangshan
Cc: Yasuaki Ishimatsu
Cc: Gu Zheng
Cc: Toshi Kani
Cc: Xishi Qiu
Signed-off-by: Andrew Morton
---
Documentation/memory-
Seems OK to me.
Thanks.
On 09/30/2015 05:33 PM, Xishi Qiu wrote:
If user set "movablecore=xx" to a large number, corepages will overflow,
this patch fix the problem.
Signed-off-by: Xishi Qiu
---
mm/page_alloc.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/mm/page_alloc.c b/mm/page_
On 09/27/2015 01:56 AM, Tejun Heo wrote:
On Sat, Sep 26, 2015 at 05:52:09PM +0800, Tang Chen wrote:
+static int allocate_logical_cpuid(int apicid)
+{
+ int i;
+
+ /*
+* cpuid <-> apicid mapping is persistent, so when a cpu is up,
+* check if the kernel has all
Hi, tj,
On 09/27/2015 01:53 AM, Tejun Heo wrote:
Hello, Tang.
On Sat, Sep 26, 2015 at 05:31:07PM +0800, Tang Chen wrote:
@@ -307,13 +307,19 @@ static inline struct page *alloc_pages_node(int nid,
gfp_t gfp_mask,
if (nid < 0)
nid = numa_node_id();
+
Hi tj,
On 09/11/2015 03:55 AM, Tejun Heo wrote:
Hello,
So, overall, I think this is the right way to go although I have no
idea whether the acpi part is okay.
Thank you very much for reviewing. :)
+/*
+ * Current allocated max logical CPU ID plus 1.
+ * All allocated CPU ID should be in [
Hi Rafael,
On 09/11/2015 07:10 AM, Rafael J. Wysocki wrote:
On Thursday, September 10, 2015 12:27:46 PM Tang Chen wrote:
..
Can you please avoid using the same (or at least very similar changelog)
for multiple patches in the series? That doesn't help a lot.
OK, will update the co
Hi, Christoph, tj,
On 09/11/2015 08:14 AM, Christoph Lameter wrote:
On Thu, 10 Sep 2015, Tejun Heo wrote:
Why not just update node_data[]->node_zonelist in the first place?
Also, what's the synchronization rule here? How are allocators
synchronized against node hot [un]plugs?
Also, shouldn't
Hi, tj
On 09/11/2015 03:29 AM, Tejun Heo wrote:
Hello,
On Thu, Sep 10, 2015 at 12:27:45PM +0800, Tang Chen wrote:
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index ad35f30..1a1324f 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -307,13 +307,19 @@ static inline struct
e this array to make cpuid <-> nodeid persistent.
cpuid <-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by:
PUs on memory-less nodes
inside alloc_pages_node() and alloc_pages_exact_node() to avoid memory
allocation
failure.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/mm/numa.c | 3 +--
include/linux/gfp.h | 8 +++-
2 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/ar
these CPUs successful. This is done
outside alloc_pages_node() and alloc_pages_exact_node(), when the kernel boots.
In this patch, we calculate best near online node for all nodes at node hotplug
time,
and store them in an array so that they could be obtained insid
27;
apicid.
This is also done by introducing an extra parameter to these apis to let the
caller
control if disabled cpus are ignored.
4. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step
Will call this function earlier in next coming patches.
So simply move its definition forward. And also, add comments for it.
Signed-off-by: Tang Chen
---
arch/x86/mm/numa.c | 47 +--
1 file changed, 29 insertions(+), 18 deletions(-)
diff --git a
disabled processors and returns -ENODEV.
After this patch, a new parameter will be added to MADT APIs so that caller
is able to control if disabled processors are ignored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_
id mapping for all possible CPUs, which has been done in
patch 4.
2. Setup cpuid <-> nodeid mapping for all possible CPUs.
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Si
at
boot time.
x86, acpi, cpu-hotplug: Introduce apicid_to_cpuid[] array to store
persistent cpuid <-> apicid mapping.
x86, acpi, cpu-hotplug: Enable MADT APIs to return disabled apicid.
x86, acpi, cpu-hotplug: Set persistent cpuid <-> nodeid mapping when
booting.
Tang Che
On 08/24/2015 07:24 PM, Xishi Qiu wrote:
..
[ 2007.584000] On node 5 totalpages: 0
[ 2007.585000] Built 5 zonelists in Node order, mobility grouping on. Total
pages: 32588823
[ 2007.594000] Policy zone: Normal
[ 2007.598000] init_memory_mapping: [mem 0x600-0x607]
And als
On 08/25/2015 09:09 AM, Xishi Qiu wrote:
On 2015/8/24 17:22, Xishi Qiu wrote:
On 2015/8/24 1:06, Tang Chen wrote:
The commit below adds hot-added memory range to memblock, after
creating pgdat for new node.
commit f9126ab9241f66562debf69c2c9d8fee32ddcc53
Author: Xishi Qiu
Date: Fri Aug
On 08/24/2015 05:17 PM, Xishi Qiu wrote:
On 2015/8/23 18:28, Tang Chen wrote:
Hi Shi,
Sorry for the late reply. I hope it won't be too late.
NON-ACK by me, I think.
I noticed that your first has been merged. But it won't fix the problem
these code intended to fix.
After yo
on. Total
pages: 32588823
[ 2007.594000] Policy zone: Normal
[ 2007.598000] init_memory_mapping: [mem 0x600-0x607]
The solution is simple, just add the memory range to memblock a little earlier,
before hotadd_new_pgdat().
Signed-off-by: Tang Chen
---
mm/memory_hotplug.c | 10
Hi Shi,
Sorry for the late reply. I hope it won't be too late.
NON-ACK by me, I think.
I noticed that your first has been merged. But it won't fix the problem
these code intended to fix.
After your patch 1, zone's spanned/present won't be set to 0 because:
free_area_init_node()
|--> get_pfn_
On 08/18/2015 02:59 PM, Jiang Liu wrote:
...
}
@@ -739,6 +746,22 @@ void __init init_cpu_to_node(void)
if (!node_online(node))
node = find_near_online_node(node);
Hi Liu,
If cpu-less, memory-less and normal node will all be online anyway,
I think we don't ne
On 08/17/2015 11:18 AM, Jiang Liu wrote:
This is the third version to enable memoryless node support on x86
platforms. The previous version (https://lkml.org/lkml/2014/7/11/75)
blindly replaces numa_node_id()/cpu_to_node() with numa_mem_id()/
cpu_to_mem(). That's not the right solution as pointe
Hi Liu,
On 08/17/2015 11:19 AM, Jiang Liu wrote:
..
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index b3a1a5d77d92..5d7ad70ace0d 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -2069,6 +2069,9 @@ config USE_PERCPU_NUMA_NODE_ID
def_bool y
depends on NUMA
+conf
but I think "if (ret)" is enough.
The code looks good.
Reviewed-by: Tang Chen
Thanks.
+ release_memory_resource(res);
+ return ret;
+}
EXPORT_SYMBOL_GPL(add_memory);
#ifdef CONFIG_MEMORY_HOTREMOVE
--
To unsubscribe from this list: send the line &qu
wrote:
On 2015/8/4 11:36, Tang Chen wrote:
Hi TJ,
Sorry for the late reply.
On 07/16/2015 05:48 AM, Tejun Heo wrote:
..
so in initialization pharse makes no sense any more. The best near online
node for each cpu should be cached somewhere.
I'm not really following. Is this because th
On 08/04/2015 04:05 PM, Jiang Liu wrote:
..
But,
1) in cpu_up(), it will try to online a node, and it doesn't check if
the node has memory.
2) in try_offline_node(), it offlines CPUs first, and then the memory.
This behavior looks a little wired, or let's say it is ambiguous. It
seems tha
Hi TJ,
Sorry for the late reply.
On 07/16/2015 05:48 AM, Tejun Heo wrote:
..
so in initialization pharse makes no sense any more. The best near online
node for each cpu should be cached somewhere.
I'm not really following. Is this because the now offline node can
later come online and we'd
On 07/16/2015 06:13 AM, Tejun Heo wrote:
Hello,
On Tue, Jul 07, 2015 at 05:30:20PM +0800, Tang Chen wrote:
[Solution]
To fix this problem, we establish cpuid <-> nodeid mapping for all the possible
cpus at boot time, and make it invariable. And according to init_cpu_to_node(),
3-157
NUMA node4 CPU(s): 62-76,190-204
NUMA node5 CPU(s): 78-92,206-220
Signed-off-by: Tang Chen
Reviewed-by: Yasuaki Ishimatsu
---
arch/x86/mm/numa.c | 6 --
include/linux/memblock.h | 2 ++
mm/memblock.c| 2 +-
3 files changed, 7 insertions(+), 3 deletions(-)
di
3-157
NUMA node2 CPU(s): 31-45,159-173
NUMA node3 CPU(s): 46-60,174-188
NUMA node4 CPU(s): 62-76,190-204
NUMA node5 CPU(s): 78-92,206-220
Tang Chen (2):
memblock: Make memblock_overlaps_region() return bool.
mem-hotplug: Handle node hole when initializing numa_meminfo.
a
-by: Tang Chen
---
include/linux/memblock.h | 2 +-
mm/memblock.c| 10 +-
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/include/linux/memblock.h b/include/linux/memblock.h
index cc4b019..d312ae3 100644
--- a/include/linux/memblock.h
+++ b/include/linux
On 07/16/2015 05:20 AM, Tejun Heo wrote:
On Wed, Jul 01, 2015 at 11:16:54AM +0800, Tang Chen wrote:
...
- /* and there's no empty block */
- if (bi->start >= bi->end)
+ /* and there's no empty or non-exist block */
+
On 07/16/2015 05:20 AM, Tejun Heo wrote:
On Wed, Jul 01, 2015 at 11:16:54AM +0800, Tang Chen wrote:
...
- /* and there's no empty block */
- if (bi->start >= bi->end)
+ /* and there's no empty or non-exist block */
+
On 07/15/2015 01:35 PM, Jiang Liu wrote:
On 2015/7/15 11:33, Tang Chen wrote:
Hi Mika,
On 07/07/2015 07:14 PM, Mika Penttilä wrote:
I think you forgot to reserve CPU 0 for BSP in cpuid mask.
Sorry for the late reply.
I'm not familiar with BSP. Do you mean in get_cpuid(),
I should re
hanks.
--Mika
On Tue, Jul 7, 2015 at 12:30 PM, Tang Chen wrote:
From: Gu Zheng
In this patch, we introduce a new static array named apicid_to_cpuid[],
which is large enough to store info for all possible cpus.
And then, we modify the cpuid calculation. In generic_processor_info(),
it simply find
nored.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
drivers/acpi/acpi_processor.c | 5 +++-
drivers/acpi/processor_core.c | 57 +++
2 files changed, 40 insertions(+), 22 deletions(-)
diff --git a/drivers/acpi/acpi_processor.c b/drivers/acpi/acpi_proces
() and alloc_pages_exact_node().
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/include/asm/topology.h | 2 ++
arch/x86/mm/numa.c | 57 ++---
include/linux/gfp.h | 12 -
3 files changed, 50 insertions(+), 21 deletions(-)
From: Gu Zheng
This patch set the persistent cpuid <-> nodeid mapping for all enabled/disabled
processors at boot time via an additional acpi namespace walk for processors.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/ia64/kernel/acpi.c | 2 +-
arch/x86/kerne
cpuid <-> apicid mapping is established at local apic registeration time.
But non-present or disabled cpus are ignored.
In this patch, we establish all possible cpuid <-> apicid mapping when
registering local apic.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/include
apis to let the
caller
control if disabled cpus are ignored.
5. Establish all possible cpuid <-> nodeid mapping.
This is done via an additional acpi namespace walk for processors.
This patch finished step 1.
Signed-off-by: Gu Zheng
Signed-off-by: Tang Chen
---
arch/x86/kernel/ap
[Problem]
cpuid <-> nodeid mapping is firstly established at boot time. And workqueue
caches
the mapping in wq_numa_possible_cpumask in wq_numa_init() at boot time.
When doing node online/offline, cpuid <-> nodeid mapping is
established/destroyed,
which means, cpuid <-> nodeid mapping will chan
On 07/07/2015 12:42 AM, Yasuaki Ishimatsu wrote:
On Fri, 3 Jul 2015 09:26:05 +0800
Tang Chen wrote:
On 07/02/2015 11:02 PM, Yasuaki Ishimatsu wrote:
Hi Tang,
On my box, if I run lscpu, the output looks like this:
NUMA node0 CPU(s): 0-14,128-142
NUMA node1 CPU(s): 15-29,143-157
max_pfn is removed,
but holes between nodes are not removed.
I think libraries are not able to handle this problem since nodes are
set online in kernel.
Seeing from user space, there is no hole.
Thanks.
Thanks,
Yasuaki Ishimatsu
On Wed, 1 Jul 2015 15:55:30 +0800
Tang Chen wrote:
On 07/01/
On 07/01/2015 02:25 PM, Xishi Qiu wrote:
On 2015/7/1 11:16, Tang Chen wrote:
When parsing SRAT, all memory ranges are added into numa_meminfo.
In numa_init(), before entering numa_cleanup_meminfo(), all possible
memory ranges are in numa_meminfo. And numa_cleanup_meminfo() removes
all ranges
nline.
In this patch, we use memblock_overlaps_region() to check if ranges in
numa_meminfo overlap with ranges in memory_block. Since memory_block contains
all available memory at boot time, if they overlap, it means the ranges
exist. If not, then remove them from numa_meminfo.
Signed-off-by:
Thanks for the sharing. Will do more tests. :)
On 11/14/2014 07:39 AM, Wanpeng Li wrote:
Hi Tang,
On Tue, Nov 11, 2014 at 01:35:29PM +0800, Tang Chen wrote:
Hi Wanpeng,
Sorry for the late.
I think I have totally missed this thread.
I opened lockdep and RCU debug, and tried on 3.18-rc1
Sorry, please ignore this series. Sent by mistake.
On 11/12/2014 04:36 PM, Tang Chen wrote:
There are two problems when calculating node meminfo:
1. When hot-adding a node without onlining any page, MemTotal corrupted.
# hot-add node2 (memory not onlined)
# cat /sys/device/system/node/node2
in patch 2/2.
2. Add Cc: sta...@vger.kernel.org # 3.16+ since it is urgent.
Change log v1 -> v2:
1. Replace patch 2/2 with a new one. It provides the simplest way to
fix problem 2.
Tang Chen (2):
mem-hotplug: Reset node managed pages when hot-adding a new pgdat.
mem-hotplug: Reset nod
_pages() to
reset_all_zones_managed_pages()
2. Make reset_node_managed_pages() non-static
3. Call reset_node_managed_pages() in hotadd_new_pgdat() after pgdat is
initialized
Signed-off-by: Tang Chen
Signed-off-by: Yasuaki Ishimatsu
Cc: sta...@vger.kernel.org # 3.16+
---
include/linux/bootmem.h | 1 +
mm/bootmem.c
vable
..
spanned 8388608
present 8388608
managed 0
When we online memory on node2:
# cat /proc/zoneinfo
..
Node 2, zone Movable
..
spanned 8388608
present 16777216
managed 8388608
Signed-off-by: Tang Chen
Reviewed-by: Yasuaki Ishimats
emUsed: 0 kB
..
This patch-set fixes them.
Tang Chen (2):
mem-hotplug: Reset node managed pages when hot-adding a new pgdat.
mem-hotplug: Fix wrong check for zone->pageset initialization in
online_pages().
include/linux/bootmem.h | 1 +
include/linux/mm.h |
Hi Wanpeng,
I think I have totally missed this thread.
I opened lockdep and RCU debug, and tried on 3.18-rc1. But I didn't get
the warning.
My steps are:
1. Use numactl to bind a qemu process to node1.
2. Offline all node1 memory. And the qemu process is still running.
Would you please tell m
_pages() to
reset_all_zones_managed_pages()
2. Make reset_node_managed_pages() non-static
3. Call reset_node_managed_pages() in hotadd_new_pgdat() after pgdat is
initialized
Signed-off-by: Tang Chen
Signed-off-by: Yasuaki Ishimatsu
---
include/linux/bootmem.h | 1 +
mm/bootmem.c| 9 +
mm/memory_hotplu
provides the simplest way to
fix problem 2.
Tang Chen (2):
mem-hotplug: Reset node managed pages when hot-adding a new pgdat.
mem-hotplug: Reset node present pages when hot-adding a new pgdat.
include/linux/bootmem.h | 1 +
mm/bootmem.c| 9 +
mm/memory_hotplug.c
.
Signed-off-by: Tang Chen
Reviewed-by: Yasuaki Ishimatsu
---
mm/memory_hotplug.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 8aba12b..26eac61 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -1067,6 +1067,
_pages() to
reset_all_zones_managed_pages()
2. Make reset_node_managed_pages() non-static
3. Call reset_node_managed_pages() in hotadd_new_pgdat() after pgdat is
initialized
Signed-off-by: Tang Chen
Signed-off-by: Yasuaki Ishimatsu
---
include/linux/bootmem.h | 1 +
mm/bootmem.c| 9 +
mm/memory_hotplu
provides the simplest way to
fix problem 2.
Tang Chen (2):
mem-hotplug: Reset node managed pages when hot-adding a new pgdat.
mem-hotplug: Reset node present pages when hot-adding a new pgdat.
include/linux/bootmem.h | 1 +
mm/bootmem.c| 9 +
mm/memory_hotplug.c
.
Signed-off-by: Tang Chen
---
mm/memory_hotplug.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 8aba12b..26eac61 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -1067,6 +1067,14 @@ out:
On 11/05/2014 09:01 AM, Kamezawa Hiroyuki wrote:
> ..
> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> index 3ab01b2..bc0de0f 100644
> --- a/mm/memory_hotplug.c
> +++ b/mm/memory_hotplug.c
> @@ -1013,9 +1013,13 @@ int __ref online_pages(unsigned long pfn, unsigned
> long nr_pages, in
On 11/04/2014 04:10 PM, Yasuaki Ishimatsu wrote:
> (2014/10/31 18:46), Tang Chen wrote:
>> When we are doing memory hot-add, the following functions are called:
>>
>> add_memory()
>> |--> hotadd_new_pgdat()
>> |--> free_area_init_node()
Hi all,
I thinks these two problems are very important and should be merged ASAP.
Would you please help to have a look at it ?
Thanks.
On 10/31/2014 05:46 PM, Tang Chen wrote:
There are two problems when calculating node meminfo:
1. When hot-adding a node without onlining any page, MemTotal
_pages() to
reset_all_zones_managed_pages()
2. Make reset_node_managed_pages() non-static
3. Call reset_node_managed_pages() in hotadd_new_pgdat() after pgdat is
initialized
Signed-off-by: Tang Chen
Signed-off-by: Yasuaki Ishimatsu
---
include/linux/bootmem.h | 1 +
mm/bootmem.c| 9 +
mm/memory_hotplu
emUsed: 0 kB
..
This patch-set fixes them.
Tang Chen (2):
mem-hotplug: Reset node managed pages when hot-adding a new pgdat.
mem-hotplug: Fix wrong check for zone->pageset initialization in
online_pages().
include/linux/bootmem.h | 1 +
include/linux/mm.h |
1 - 100 of 1031 matches
Mail list logo