* [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
@ 2014-01-13 21:56 Prarit Bhargava
2014-01-13 22:17 ` KOSAKI Motohiro
2014-01-14 1:45 ` Dave Young
0 siblings, 2 replies; 14+ messages in thread
From: Prarit Bhargava @ 2014-01-13 21:56 UTC (permalink / raw)
To: linux-kernel
Cc: Prarit Bhargava, Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
x86, Len Brown, Rafael J. Wysocki, Linn Crosetto, Pekka Enberg,
Yinghai Lu, Andrew Morton, Toshi Kani, Tang Chen, Wen Congyang,
Vivek Goyal, kosaki.motohiro, dyoung, linux-acpi, linux-mm
When booting a kexec/kdump kernel on a system that has specific memory hotplug
regions the boot will fail with warnings like:
[ 2.939467] swapper/0: page allocation failure: order:9, mode:0x84d0
[ 2.946564] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
3.10.0-65.el7.x86_64 #1
[ 2.954532] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS
QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011
[ 2.964926] 0000000000000000 ffff8800341bd8c8 ffffffff815bcc67
ffff8800341bd950
[ 2.973224] ffffffff8113b1a0 ffff880036339b00 0000000000000009
00000000000084d0
[ 2.981523] ffff8800341bd950 ffffffff815b87ee 0000000000000000
0000000000000200
[ 2.989821] Call Trace:
[ 2.992560] [<ffffffff815bcc67>] dump_stack+0x19/0x1b
[ 2.998300] [<ffffffff8113b1a0>] warn_alloc_failed+0xf0/0x160
[ 3.004817] [<ffffffff815b87ee>] ?
__alloc_pages_direct_compact+0xac/0x196
[ 3.012594] [<ffffffff8113f14f>] __alloc_pages_nodemask+0x7ff/0xa00
[ 3.019692] [<ffffffff815b417c>] vmemmap_alloc_block+0x62/0xba
[ 3.026303] [<ffffffff815b41e9>] vmemmap_alloc_block_buf+0x15/0x3b
[ 3.033302] [<ffffffff815b1ff6>] vmemmap_populate+0xb4/0x21b
[ 3.039718] [<ffffffff815b461d>] sparse_mem_map_populate+0x27/0x35
[ 3.046717] [<ffffffff815b400f>] sparse_add_one_section+0x7a/0x185
[ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
[ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
[ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
[ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
[ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
[ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
[ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
[ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
[ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
[ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
[ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
[ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
[ 3.131987] [<ffffffff81a1fd58>] ? acpi_sleep_proc_init+0x2a/0x2a
[ 3.138889] [<ffffffff810020e2>] do_one_initcall+0xe2/0x190
[ 3.145210] [<ffffffff819e20c4>] kernel_init_freeable+0x17c/0x207
[ 3.152111] [<ffffffff819e18d0>] ? do_early_param+0x88/0x88
[ 3.158430] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
[ 3.164264] [<ffffffff8159feae>] kernel_init+0xe/0x180
[ 3.170097] [<ffffffff815cca2c>] ret_from_fork+0x7c/0xb0
[ 3.176123] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
[ 3.181956] Mem-Info:
[ 3.184490] Node 0 DMA per-cpu:
[ 3.188007] CPU 0: hi: 0, btch: 1 usd: 0
[ 3.193353] Node 0 DMA32 per-cpu:
[ 3.197060] CPU 0: hi: 42, btch: 7 usd: 0
[ 3.202410] active_anon:0 inactive_anon:0 isolated_anon:0
[ 3.202410] active_file:0 inactive_file:0 isolated_file:0
[ 3.202410] unevictable:0 dirty:0 writeback:0 unstable:0
[ 3.202410] free:872 slab_reclaimable:13 slab_unreclaimable:1880
[ 3.202410] mapped:0 shmem:0 pagetables:0 bounce:0
[ 3.202410] free_cma:0
because the system has run out of memory at boot time. This occurs
because of the following sequence in the boot:
Main kernel boots and sets E820 map. The second kernel is booted with a
map generated by the kdump service using memmap= and memmap=exactmap.
These parameters are added to the kernel parameters of the kexec/kdump
kernel. The kexec/kdump kernel has limited memory resources so as not
to severely impact the main kernel.
The system then panics and the kdump/kexec kernel boots (which is a
completely new kernel boot). During this boot ACPI is initialized and the
kernel (as can be seen above) traverses the ACPI namespace and finds an
entry for a memory device to be hotadded.
ie)
[ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
[ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
[ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
[ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
[ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
[ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
[ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
[ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
[ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
[ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
[ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
[ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
At this point the kernel adds page table information and the the kexec/kdump
kernel runs out of memory.
This can also be reproduced with a "regular" kernel by using the
memmap=exactmap and mem=X parameters on the main kernel and booting.
This patchset resolves the problem by adding a kernel parameter,
acpi_no_memhotplug, to disable ACPI memory hotplug. ACPI memory hotplug
should also be disabled by default when a user specified a memory mapping with
"memmap=exactmap" or "mem=X".
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Len Brown <lenb@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Linn Crosetto <linn@hp.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Toshi Kani <toshi.kani@hp.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Wen Congyang <wency@cn.fujitsu.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: kosaki.motohiro@gmail.com
Cc: dyoung@redhat.com
Cc: Toshi Kani <toshi.kani@hp.com>
Cc: linux-acpi@vger.kernel.org
Cc: linux-mm@kvack.org
---
Documentation/kernel-parameters.txt | 3 +++
arch/x86/kernel/e820.c | 4 ++++
drivers/acpi/acpi_memhotplug.c | 18 ++++++++++++++++++
include/linux/memory_hotplug.h | 9 +++++++++
4 files changed, 34 insertions(+)
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index b9e9bd8..ea93f75 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -343,6 +343,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
no: ACPI OperationRegions are not marked as reserved,
no further checks are performed.
+ acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kexec
+ and kdump kernels.
+
add_efi_memmap [EFI; X86] Include EFI memory map in
kernel's map of available physical RAM.
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 174da5f..3c431fe 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -20,6 +20,7 @@
#include <linux/firmware-map.h>
#include <linux/memblock.h>
#include <linux/sort.h>
+#include <linux/memory_hotplug.h>
#include <asm/e820.h>
#include <asm/proto.h>
@@ -834,6 +835,8 @@ static int __init parse_memopt(char *p)
return -EINVAL;
e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1);
+ set_acpi_no_memhotplug();
+
return 0;
}
early_param("mem", parse_memopt);
@@ -857,6 +860,7 @@ static int __init parse_memmap_one(char *p)
#endif
e820.nr_map = 0;
userdef = 1;
+ set_acpi_no_memhotplug();
return 0;
}
diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c
index 551dad7..d104a7d 100644
--- a/drivers/acpi/acpi_memhotplug.c
+++ b/drivers/acpi/acpi_memhotplug.c
@@ -361,7 +361,25 @@ static void acpi_memory_device_remove(struct acpi_device *device)
acpi_memory_device_free(mem_device);
}
+static bool acpi_no_memhotplug;
+
+void set_acpi_no_memhotplug(void)
+{
+ acpi_no_memhotplug = true;
+ pr_info_once("ACPI: Memory Hotplug Disabled\n");
+}
+
void __init acpi_memory_hotplug_init(void)
{
+ if (acpi_no_memhotplug)
+ return;
+
acpi_scan_add_handler_with_hotplug(&memory_device_handler, "memory");
}
+
+static int __init disable_acpi_memory_hotplug(char *str)
+{
+ set_acpi_no_memhotplug();
+ return 1;
+}
+__setup("acpi_no_memhotplug", disable_acpi_memory_hotplug);
diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index 4ca3d95..3cdb6e0 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -12,6 +12,15 @@ struct pglist_data;
struct mem_section;
struct memory_block;
+#ifdef CONFIG_ACPI_HOTPLUG_MEMORY
+/* set flag to disable ACPI memory hotplug */
+extern void set_acpi_no_memhotplug(void);
+#else
+static inline void set_acpi_no_memhotplug(void)
+{
+}
+#endif
+
#ifdef CONFIG_MEMORY_HOTPLUG
/*
--
1.7.9.3
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-13 21:56 [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug Prarit Bhargava
@ 2014-01-13 22:17 ` KOSAKI Motohiro
2014-01-13 23:41 ` Prarit Bhargava
2014-01-14 1:45 ` Dave Young
1 sibling, 1 reply; 14+ messages in thread
From: KOSAKI Motohiro @ 2014-01-13 22:17 UTC (permalink / raw)
To: Prarit Bhargava
Cc: LKML, Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Toshi Kani, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On Mon, Jan 13, 2014 at 4:56 PM, Prarit Bhargava <prarit@redhat.com> wrote:
> When booting a kexec/kdump kernel on a system that has specific memory hotplug
> regions the boot will fail with warnings like:
>
> [ 2.939467] swapper/0: page allocation failure: order:9, mode:0x84d0
> [ 2.946564] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
> 3.10.0-65.el7.x86_64 #1
> [ 2.954532] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS
> QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011
> [ 2.964926] 0000000000000000 ffff8800341bd8c8 ffffffff815bcc67
> ffff8800341bd950
> [ 2.973224] ffffffff8113b1a0 ffff880036339b00 0000000000000009
> 00000000000084d0
> [ 2.981523] ffff8800341bd950 ffffffff815b87ee 0000000000000000
> 0000000000000200
> [ 2.989821] Call Trace:
> [ 2.992560] [<ffffffff815bcc67>] dump_stack+0x19/0x1b
> [ 2.998300] [<ffffffff8113b1a0>] warn_alloc_failed+0xf0/0x160
> [ 3.004817] [<ffffffff815b87ee>] ?
> __alloc_pages_direct_compact+0xac/0x196
> [ 3.012594] [<ffffffff8113f14f>] __alloc_pages_nodemask+0x7ff/0xa00
> [ 3.019692] [<ffffffff815b417c>] vmemmap_alloc_block+0x62/0xba
> [ 3.026303] [<ffffffff815b41e9>] vmemmap_alloc_block_buf+0x15/0x3b
> [ 3.033302] [<ffffffff815b1ff6>] vmemmap_populate+0xb4/0x21b
> [ 3.039718] [<ffffffff815b461d>] sparse_mem_map_populate+0x27/0x35
> [ 3.046717] [<ffffffff815b400f>] sparse_add_one_section+0x7a/0x185
> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
> [ 3.131987] [<ffffffff81a1fd58>] ? acpi_sleep_proc_init+0x2a/0x2a
> [ 3.138889] [<ffffffff810020e2>] do_one_initcall+0xe2/0x190
> [ 3.145210] [<ffffffff819e20c4>] kernel_init_freeable+0x17c/0x207
> [ 3.152111] [<ffffffff819e18d0>] ? do_early_param+0x88/0x88
> [ 3.158430] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
> [ 3.164264] [<ffffffff8159feae>] kernel_init+0xe/0x180
> [ 3.170097] [<ffffffff815cca2c>] ret_from_fork+0x7c/0xb0
> [ 3.176123] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
> [ 3.181956] Mem-Info:
> [ 3.184490] Node 0 DMA per-cpu:
> [ 3.188007] CPU 0: hi: 0, btch: 1 usd: 0
> [ 3.193353] Node 0 DMA32 per-cpu:
> [ 3.197060] CPU 0: hi: 42, btch: 7 usd: 0
> [ 3.202410] active_anon:0 inactive_anon:0 isolated_anon:0
> [ 3.202410] active_file:0 inactive_file:0 isolated_file:0
> [ 3.202410] unevictable:0 dirty:0 writeback:0 unstable:0
> [ 3.202410] free:872 slab_reclaimable:13 slab_unreclaimable:1880
> [ 3.202410] mapped:0 shmem:0 pagetables:0 bounce:0
> [ 3.202410] free_cma:0
>
> because the system has run out of memory at boot time. This occurs
> because of the following sequence in the boot:
>
> Main kernel boots and sets E820 map. The second kernel is booted with a
> map generated by the kdump service using memmap= and memmap=exactmap.
> These parameters are added to the kernel parameters of the kexec/kdump
> kernel. The kexec/kdump kernel has limited memory resources so as not
> to severely impact the main kernel.
>
> The system then panics and the kdump/kexec kernel boots (which is a
> completely new kernel boot). During this boot ACPI is initialized and the
> kernel (as can be seen above) traverses the ACPI namespace and finds an
> entry for a memory device to be hotadded.
>
> ie)
>
> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>
> At this point the kernel adds page table information and the the kexec/kdump
> kernel runs out of memory.
>
> This can also be reproduced with a "regular" kernel by using the
> memmap=exactmap and mem=X parameters on the main kernel and booting.
>
> This patchset resolves the problem by adding a kernel parameter,
> acpi_no_memhotplug, to disable ACPI memory hotplug. ACPI memory hotplug
> should also be disabled by default when a user specified a memory mapping with
> "memmap=exactmap" or "mem=X".
>
> Signed-off-by: Prarit Bhargava <prarit@redhat.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: x86@kernel.org
> Cc: Len Brown <lenb@kernel.org>
> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
> Cc: Linn Crosetto <linn@hp.com>
> Cc: Pekka Enberg <penberg@kernel.org>
> Cc: Yinghai Lu <yinghai@kernel.org>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Toshi Kani <toshi.kani@hp.com>
> Cc: Tang Chen <tangchen@cn.fujitsu.com>
> Cc: Wen Congyang <wency@cn.fujitsu.com>
> Cc: Vivek Goyal <vgoyal@redhat.com>
> Cc: kosaki.motohiro@gmail.com
> Cc: dyoung@redhat.com
> Cc: Toshi Kani <toshi.kani@hp.com>
> Cc: linux-acpi@vger.kernel.org
> Cc: linux-mm@kvack.org
I think we need a knob manually enable mem-hotplug when specify memmap. But
it is another story.
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-13 22:17 ` KOSAKI Motohiro
@ 2014-01-13 23:41 ` Prarit Bhargava
2014-01-14 1:11 ` Yasuaki Ishimatsu
0 siblings, 1 reply; 14+ messages in thread
From: Prarit Bhargava @ 2014-01-13 23:41 UTC (permalink / raw)
To: KOSAKI Motohiro
Cc: LKML, Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Toshi Kani, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On 01/13/2014 05:17 PM, KOSAKI Motohiro wrote:
> On Mon, Jan 13, 2014 at 4:56 PM, Prarit Bhargava <prarit@redhat.com> wrote:
>> When booting a kexec/kdump kernel on a system that has specific memory hotplug
>> regions the boot will fail with warnings like:
>>
>> [ 2.939467] swapper/0: page allocation failure: order:9, mode:0x84d0
>> [ 2.946564] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
>> 3.10.0-65.el7.x86_64 #1
>> [ 2.954532] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS
>> QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011
>> [ 2.964926] 0000000000000000 ffff8800341bd8c8 ffffffff815bcc67
>> ffff8800341bd950
>> [ 2.973224] ffffffff8113b1a0 ffff880036339b00 0000000000000009
>> 00000000000084d0
>> [ 2.981523] ffff8800341bd950 ffffffff815b87ee 0000000000000000
>> 0000000000000200
>> [ 2.989821] Call Trace:
>> [ 2.992560] [<ffffffff815bcc67>] dump_stack+0x19/0x1b
>> [ 2.998300] [<ffffffff8113b1a0>] warn_alloc_failed+0xf0/0x160
>> [ 3.004817] [<ffffffff815b87ee>] ?
>> __alloc_pages_direct_compact+0xac/0x196
>> [ 3.012594] [<ffffffff8113f14f>] __alloc_pages_nodemask+0x7ff/0xa00
>> [ 3.019692] [<ffffffff815b417c>] vmemmap_alloc_block+0x62/0xba
>> [ 3.026303] [<ffffffff815b41e9>] vmemmap_alloc_block_buf+0x15/0x3b
>> [ 3.033302] [<ffffffff815b1ff6>] vmemmap_populate+0xb4/0x21b
>> [ 3.039718] [<ffffffff815b461d>] sparse_mem_map_populate+0x27/0x35
>> [ 3.046717] [<ffffffff815b400f>] sparse_add_one_section+0x7a/0x185
>> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
>> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
>> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
>> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
>> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
>> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
>> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
>> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
>> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
>> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>> [ 3.131987] [<ffffffff81a1fd58>] ? acpi_sleep_proc_init+0x2a/0x2a
>> [ 3.138889] [<ffffffff810020e2>] do_one_initcall+0xe2/0x190
>> [ 3.145210] [<ffffffff819e20c4>] kernel_init_freeable+0x17c/0x207
>> [ 3.152111] [<ffffffff819e18d0>] ? do_early_param+0x88/0x88
>> [ 3.158430] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
>> [ 3.164264] [<ffffffff8159feae>] kernel_init+0xe/0x180
>> [ 3.170097] [<ffffffff815cca2c>] ret_from_fork+0x7c/0xb0
>> [ 3.176123] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
>> [ 3.181956] Mem-Info:
>> [ 3.184490] Node 0 DMA per-cpu:
>> [ 3.188007] CPU 0: hi: 0, btch: 1 usd: 0
>> [ 3.193353] Node 0 DMA32 per-cpu:
>> [ 3.197060] CPU 0: hi: 42, btch: 7 usd: 0
>> [ 3.202410] active_anon:0 inactive_anon:0 isolated_anon:0
>> [ 3.202410] active_file:0 inactive_file:0 isolated_file:0
>> [ 3.202410] unevictable:0 dirty:0 writeback:0 unstable:0
>> [ 3.202410] free:872 slab_reclaimable:13 slab_unreclaimable:1880
>> [ 3.202410] mapped:0 shmem:0 pagetables:0 bounce:0
>> [ 3.202410] free_cma:0
>>
>> because the system has run out of memory at boot time. This occurs
>> because of the following sequence in the boot:
>>
>> Main kernel boots and sets E820 map. The second kernel is booted with a
>> map generated by the kdump service using memmap= and memmap=exactmap.
>> These parameters are added to the kernel parameters of the kexec/kdump
>> kernel. The kexec/kdump kernel has limited memory resources so as not
>> to severely impact the main kernel.
>>
>> The system then panics and the kdump/kexec kernel boots (which is a
>> completely new kernel boot). During this boot ACPI is initialized and the
>> kernel (as can be seen above) traverses the ACPI namespace and finds an
>> entry for a memory device to be hotadded.
>>
>> ie)
>>
>> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
>> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
>> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
>> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
>> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
>> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
>> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
>> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
>> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
>> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>>
>> At this point the kernel adds page table information and the the kexec/kdump
>> kernel runs out of memory.
>>
>> This can also be reproduced with a "regular" kernel by using the
>> memmap=exactmap and mem=X parameters on the main kernel and booting.
>>
>> This patchset resolves the problem by adding a kernel parameter,
>> acpi_no_memhotplug, to disable ACPI memory hotplug. ACPI memory hotplug
>> should also be disabled by default when a user specified a memory mapping with
>> "memmap=exactmap" or "mem=X".
>>
>> Signed-off-by: Prarit Bhargava <prarit@redhat.com>
>> Cc: Thomas Gleixner <tglx@linutronix.de>
>> Cc: Ingo Molnar <mingo@redhat.com>
>> Cc: "H. Peter Anvin" <hpa@zytor.com>
>> Cc: x86@kernel.org
>> Cc: Len Brown <lenb@kernel.org>
>> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
>> Cc: Linn Crosetto <linn@hp.com>
>> Cc: Pekka Enberg <penberg@kernel.org>
>> Cc: Yinghai Lu <yinghai@kernel.org>
>> Cc: Andrew Morton <akpm@linux-foundation.org>
>> Cc: Toshi Kani <toshi.kani@hp.com>
>> Cc: Tang Chen <tangchen@cn.fujitsu.com>
>> Cc: Wen Congyang <wency@cn.fujitsu.com>
>> Cc: Vivek Goyal <vgoyal@redhat.com>
>> Cc: kosaki.motohiro@gmail.com
>> Cc: dyoung@redhat.com
>> Cc: Toshi Kani <toshi.kani@hp.com>
>> Cc: linux-acpi@vger.kernel.org
>> Cc: linux-mm@kvack.org
>
> I think we need a knob manually enable mem-hotplug when specify memmap. But
> it is another story.
>
> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
As mentioned, self-NAK. I have seen a system that I needed to specify
memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
option in the next version of the patch.
P.
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-13 23:41 ` Prarit Bhargava
@ 2014-01-14 1:11 ` Yasuaki Ishimatsu
2014-01-14 1:41 ` Toshi Kani
0 siblings, 1 reply; 14+ messages in thread
From: Yasuaki Ishimatsu @ 2014-01-14 1:11 UTC (permalink / raw)
To: Prarit Bhargava
Cc: KOSAKI Motohiro, LKML, Thomas Gleixner, Ingo Molnar,
H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Toshi Kani, Tang Chen, Wen Congyang, Vivek Goyal,
Dave Young, linux-acpi, linux-mm
(2014/01/14 8:41), Prarit Bhargava wrote:
>
>
> On 01/13/2014 05:17 PM, KOSAKI Motohiro wrote:
>> On Mon, Jan 13, 2014 at 4:56 PM, Prarit Bhargava <prarit@redhat.com> wrote:
>>> When booting a kexec/kdump kernel on a system that has specific memory hotplug
>>> regions the boot will fail with warnings like:
>>>
>>> [ 2.939467] swapper/0: page allocation failure: order:9, mode:0x84d0
>>> [ 2.946564] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
>>> 3.10.0-65.el7.x86_64 #1
>>> [ 2.954532] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS
>>> QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011
>>> [ 2.964926] 0000000000000000 ffff8800341bd8c8 ffffffff815bcc67
>>> ffff8800341bd950
>>> [ 2.973224] ffffffff8113b1a0 ffff880036339b00 0000000000000009
>>> 00000000000084d0
>>> [ 2.981523] ffff8800341bd950 ffffffff815b87ee 0000000000000000
>>> 0000000000000200
>>> [ 2.989821] Call Trace:
>>> [ 2.992560] [<ffffffff815bcc67>] dump_stack+0x19/0x1b
>>> [ 2.998300] [<ffffffff8113b1a0>] warn_alloc_failed+0xf0/0x160
>>> [ 3.004817] [<ffffffff815b87ee>] ?
>>> __alloc_pages_direct_compact+0xac/0x196
>>> [ 3.012594] [<ffffffff8113f14f>] __alloc_pages_nodemask+0x7ff/0xa00
>>> [ 3.019692] [<ffffffff815b417c>] vmemmap_alloc_block+0x62/0xba
>>> [ 3.026303] [<ffffffff815b41e9>] vmemmap_alloc_block_buf+0x15/0x3b
>>> [ 3.033302] [<ffffffff815b1ff6>] vmemmap_populate+0xb4/0x21b
>>> [ 3.039718] [<ffffffff815b461d>] sparse_mem_map_populate+0x27/0x35
>>> [ 3.046717] [<ffffffff815b400f>] sparse_add_one_section+0x7a/0x185
>>> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
>>> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
>>> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
>>> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
>>> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
>>> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
>>> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>>> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>>> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
>>> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
>>> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
>>> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>>> [ 3.131987] [<ffffffff81a1fd58>] ? acpi_sleep_proc_init+0x2a/0x2a
>>> [ 3.138889] [<ffffffff810020e2>] do_one_initcall+0xe2/0x190
>>> [ 3.145210] [<ffffffff819e20c4>] kernel_init_freeable+0x17c/0x207
>>> [ 3.152111] [<ffffffff819e18d0>] ? do_early_param+0x88/0x88
>>> [ 3.158430] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
>>> [ 3.164264] [<ffffffff8159feae>] kernel_init+0xe/0x180
>>> [ 3.170097] [<ffffffff815cca2c>] ret_from_fork+0x7c/0xb0
>>> [ 3.176123] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
>>> [ 3.181956] Mem-Info:
>>> [ 3.184490] Node 0 DMA per-cpu:
>>> [ 3.188007] CPU 0: hi: 0, btch: 1 usd: 0
>>> [ 3.193353] Node 0 DMA32 per-cpu:
>>> [ 3.197060] CPU 0: hi: 42, btch: 7 usd: 0
>>> [ 3.202410] active_anon:0 inactive_anon:0 isolated_anon:0
>>> [ 3.202410] active_file:0 inactive_file:0 isolated_file:0
>>> [ 3.202410] unevictable:0 dirty:0 writeback:0 unstable:0
>>> [ 3.202410] free:872 slab_reclaimable:13 slab_unreclaimable:1880
>>> [ 3.202410] mapped:0 shmem:0 pagetables:0 bounce:0
>>> [ 3.202410] free_cma:0
>>>
>>> because the system has run out of memory at boot time. This occurs
>>> because of the following sequence in the boot:
>>>
>>> Main kernel boots and sets E820 map. The second kernel is booted with a
>>> map generated by the kdump service using memmap= and memmap=exactmap.
>>> These parameters are added to the kernel parameters of the kexec/kdump
>>> kernel. The kexec/kdump kernel has limited memory resources so as not
>>> to severely impact the main kernel.
>>>
>>> The system then panics and the kdump/kexec kernel boots (which is a
>>> completely new kernel boot). During this boot ACPI is initialized and the
>>> kernel (as can be seen above) traverses the ACPI namespace and finds an
>>> entry for a memory device to be hotadded.
>>>
>>> ie)
>>>
>>> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
>>> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
>>> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
>>> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
>>> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
>>> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
>>> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>>> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
>>> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
>>> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
>>> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
>>> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>>>
>>> At this point the kernel adds page table information and the the kexec/kdump
>>> kernel runs out of memory.
>>>
>>> This can also be reproduced with a "regular" kernel by using the
>>> memmap=exactmap and mem=X parameters on the main kernel and booting.
>>>
>>> This patchset resolves the problem by adding a kernel parameter,
>>> acpi_no_memhotplug, to disable ACPI memory hotplug. ACPI memory hotplug
>>> should also be disabled by default when a user specified a memory mapping with
>>> "memmap=exactmap" or "mem=X".
>>>
>>> Signed-off-by: Prarit Bhargava <prarit@redhat.com>
>>> Cc: Thomas Gleixner <tglx@linutronix.de>
>>> Cc: Ingo Molnar <mingo@redhat.com>
>>> Cc: "H. Peter Anvin" <hpa@zytor.com>
>>> Cc: x86@kernel.org
>>> Cc: Len Brown <lenb@kernel.org>
>>> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
>>> Cc: Linn Crosetto <linn@hp.com>
>>> Cc: Pekka Enberg <penberg@kernel.org>
>>> Cc: Yinghai Lu <yinghai@kernel.org>
>>> Cc: Andrew Morton <akpm@linux-foundation.org>
>>> Cc: Toshi Kani <toshi.kani@hp.com>
>>> Cc: Tang Chen <tangchen@cn.fujitsu.com>
>>> Cc: Wen Congyang <wency@cn.fujitsu.com>
>>> Cc: Vivek Goyal <vgoyal@redhat.com>
>>> Cc: kosaki.motohiro@gmail.com
>>> Cc: dyoung@redhat.com
>>> Cc: Toshi Kani <toshi.kani@hp.com>
>>> Cc: linux-acpi@vger.kernel.org
>>> Cc: linux-mm@kvack.org
>>
>> I think we need a knob manually enable mem-hotplug when specify memmap. But
>> it is another story.
>>
>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
>
> As mentioned, self-NAK. I have seen a system that I needed to specify
> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> option in the next version of the patch.
Your following first patch is simply and makes sense.
http://marc.info/?l=linux-acpi&m=138922019607796&w=2
Thanks,
Yasuaki Ishiamtsu
>
> P.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 1:11 ` Yasuaki Ishimatsu
@ 2014-01-14 1:41 ` Toshi Kani
2014-01-14 2:43 ` Yasuaki Ishimatsu
` (2 more replies)
0 siblings, 3 replies; 14+ messages in thread
From: Toshi Kani @ 2014-01-14 1:41 UTC (permalink / raw)
To: Yasuaki Ishimatsu
Cc: Prarit Bhargava, KOSAKI Motohiro, LKML, Thomas Gleixner,
Ingo Molnar, H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
:
> >> I think we need a knob manually enable mem-hotplug when specify memmap. But
> >> it is another story.
> >>
> >> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> >
> > As mentioned, self-NAK. I have seen a system that I needed to specify
> > memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> > option in the next version of the patch.
>
>
> Your following first patch is simply and makes sense.
>
> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
>
In this option, it also requires changing kexec-tools to specify the new
option for kdump. It won't be simpler.
Thanks,
-Toshi
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-13 21:56 [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug Prarit Bhargava
2014-01-13 22:17 ` KOSAKI Motohiro
@ 2014-01-14 1:45 ` Dave Young
1 sibling, 0 replies; 14+ messages in thread
From: Dave Young @ 2014-01-14 1:45 UTC (permalink / raw)
To: Prarit Bhargava
Cc: linux-kernel, Thomas Gleixner, Ingo Molnar, H. Peter Anvin, x86,
Len Brown, Rafael J. Wysocki, Linn Crosetto, Pekka Enberg,
Yinghai Lu, Andrew Morton, Toshi Kani, Tang Chen, Wen Congyang,
Vivek Goyal, kosaki.motohiro, linux-acpi, linux-mm
On 01/13/14 at 04:56pm, Prarit Bhargava wrote:
> When booting a kexec/kdump kernel on a system that has specific memory hotplug
> regions the boot will fail with warnings like:
>
> [ 2.939467] swapper/0: page allocation failure: order:9, mode:0x84d0
> [ 2.946564] CPU: 0 PID: 1 Comm: swapper/0 Not tainted
> 3.10.0-65.el7.x86_64 #1
> [ 2.954532] Hardware name: QCI QSSC-S4R/QSSC-S4R, BIOS
> QSSC-S4R.QCI.01.00.S013.032920111005 03/29/2011
> [ 2.964926] 0000000000000000 ffff8800341bd8c8 ffffffff815bcc67
> ffff8800341bd950
> [ 2.973224] ffffffff8113b1a0 ffff880036339b00 0000000000000009
> 00000000000084d0
> [ 2.981523] ffff8800341bd950 ffffffff815b87ee 0000000000000000
> 0000000000000200
> [ 2.989821] Call Trace:
> [ 2.992560] [<ffffffff815bcc67>] dump_stack+0x19/0x1b
> [ 2.998300] [<ffffffff8113b1a0>] warn_alloc_failed+0xf0/0x160
> [ 3.004817] [<ffffffff815b87ee>] ?
> __alloc_pages_direct_compact+0xac/0x196
> [ 3.012594] [<ffffffff8113f14f>] __alloc_pages_nodemask+0x7ff/0xa00
> [ 3.019692] [<ffffffff815b417c>] vmemmap_alloc_block+0x62/0xba
> [ 3.026303] [<ffffffff815b41e9>] vmemmap_alloc_block_buf+0x15/0x3b
> [ 3.033302] [<ffffffff815b1ff6>] vmemmap_populate+0xb4/0x21b
> [ 3.039718] [<ffffffff815b461d>] sparse_mem_map_populate+0x27/0x35
> [ 3.046717] [<ffffffff815b400f>] sparse_add_one_section+0x7a/0x185
> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
> [ 3.131987] [<ffffffff81a1fd58>] ? acpi_sleep_proc_init+0x2a/0x2a
> [ 3.138889] [<ffffffff810020e2>] do_one_initcall+0xe2/0x190
> [ 3.145210] [<ffffffff819e20c4>] kernel_init_freeable+0x17c/0x207
> [ 3.152111] [<ffffffff819e18d0>] ? do_early_param+0x88/0x88
> [ 3.158430] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
> [ 3.164264] [<ffffffff8159feae>] kernel_init+0xe/0x180
> [ 3.170097] [<ffffffff815cca2c>] ret_from_fork+0x7c/0xb0
> [ 3.176123] [<ffffffff8159fea0>] ? rest_init+0x80/0x80
> [ 3.181956] Mem-Info:
> [ 3.184490] Node 0 DMA per-cpu:
> [ 3.188007] CPU 0: hi: 0, btch: 1 usd: 0
> [ 3.193353] Node 0 DMA32 per-cpu:
> [ 3.197060] CPU 0: hi: 42, btch: 7 usd: 0
> [ 3.202410] active_anon:0 inactive_anon:0 isolated_anon:0
> [ 3.202410] active_file:0 inactive_file:0 isolated_file:0
> [ 3.202410] unevictable:0 dirty:0 writeback:0 unstable:0
> [ 3.202410] free:872 slab_reclaimable:13 slab_unreclaimable:1880
> [ 3.202410] mapped:0 shmem:0 pagetables:0 bounce:0
> [ 3.202410] free_cma:0
>
> because the system has run out of memory at boot time. This occurs
> because of the following sequence in the boot:
>
> Main kernel boots and sets E820 map. The second kernel is booted with a
> map generated by the kdump service using memmap= and memmap=exactmap.
> These parameters are added to the kernel parameters of the kexec/kdump
> kernel. The kexec/kdump kernel has limited memory resources so as not
> to severely impact the main kernel.
>
> The system then panics and the kdump/kexec kernel boots (which is a
> completely new kernel boot). During this boot ACPI is initialized and the
> kernel (as can be seen above) traverses the ACPI namespace and finds an
> entry for a memory device to be hotadded.
>
> ie)
>
> [ 3.053720] [<ffffffff815a1e9f>] __add_pages+0xaf/0x240
> [ 3.059656] [<ffffffff81047359>] arch_add_memory+0x59/0xd0
> [ 3.065877] [<ffffffff815a21d9>] add_memory+0xb9/0x1b0
> [ 3.071713] [<ffffffff81333b9c>] acpi_memory_device_add+0x18d/0x26d
> [ 3.078813] [<ffffffff81309a01>] acpi_bus_device_attach+0x7d/0xcd
> [ 3.085719] [<ffffffff8132379d>] acpi_ns_walk_namespace+0xc8/0x17f
> [ 3.092716] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.100004] [<ffffffff81309984>] ? acpi_bus_type_and_status+0x90/0x90
> [ 3.107293] [<ffffffff81323c8c>] acpi_walk_namespace+0x95/0xc5
> [ 3.113904] [<ffffffff8130a6d6>] acpi_bus_scan+0x8b/0x9d
> [ 3.119933] [<ffffffff81a2019a>] acpi_scan_init+0x63/0x160
> [ 3.126153] [<ffffffff81a1ffb5>] acpi_init+0x25d/0x2a6
>
> At this point the kernel adds page table information and the the kexec/kdump
> kernel runs out of memory.
>
> This can also be reproduced with a "regular" kernel by using the
> memmap=exactmap and mem=X parameters on the main kernel and booting.
>
> This patchset resolves the problem by adding a kernel parameter,
> acpi_no_memhotplug, to disable ACPI memory hotplug. ACPI memory hotplug
> should also be disabled by default when a user specified a memory mapping with
> "memmap=exactmap" or "mem=X".
>
> Signed-off-by: Prarit Bhargava <prarit@redhat.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: x86@kernel.org
> Cc: Len Brown <lenb@kernel.org>
> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
> Cc: Linn Crosetto <linn@hp.com>
> Cc: Pekka Enberg <penberg@kernel.org>
> Cc: Yinghai Lu <yinghai@kernel.org>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Toshi Kani <toshi.kani@hp.com>
> Cc: Tang Chen <tangchen@cn.fujitsu.com>
> Cc: Wen Congyang <wency@cn.fujitsu.com>
> Cc: Vivek Goyal <vgoyal@redhat.com>
> Cc: kosaki.motohiro@gmail.com
> Cc: dyoung@redhat.com
> Cc: Toshi Kani <toshi.kani@hp.com>
> Cc: linux-acpi@vger.kernel.org
> Cc: linux-mm@kvack.org
> ---
> Documentation/kernel-parameters.txt | 3 +++
> arch/x86/kernel/e820.c | 4 ++++
> drivers/acpi/acpi_memhotplug.c | 18 ++++++++++++++++++
> include/linux/memory_hotplug.h | 9 +++++++++
> 4 files changed, 34 insertions(+)
>
> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
> index b9e9bd8..ea93f75 100644
> --- a/Documentation/kernel-parameters.txt
> +++ b/Documentation/kernel-parameters.txt
> @@ -343,6 +343,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
> no: ACPI OperationRegions are not marked as reserved,
> no further checks are performed.
>
> + acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kexec
> + and kdump kernels.
> +
> add_efi_memmap [EFI; X86] Include EFI memory map in
> kernel's map of available physical RAM.
>
> diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
> index 174da5f..3c431fe 100644
> --- a/arch/x86/kernel/e820.c
> +++ b/arch/x86/kernel/e820.c
> @@ -20,6 +20,7 @@
> #include <linux/firmware-map.h>
> #include <linux/memblock.h>
> #include <linux/sort.h>
> +#include <linux/memory_hotplug.h>
>
> #include <asm/e820.h>
> #include <asm/proto.h>
> @@ -834,6 +835,8 @@ static int __init parse_memopt(char *p)
> return -EINVAL;
> e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1);
>
> + set_acpi_no_memhotplug();
> +
> return 0;
> }
> early_param("mem", parse_memopt);
> @@ -857,6 +860,7 @@ static int __init parse_memmap_one(char *p)
> #endif
> e820.nr_map = 0;
> userdef = 1;
> + set_acpi_no_memhotplug();
> return 0;
> }
>
> diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c
> index 551dad7..d104a7d 100644
> --- a/drivers/acpi/acpi_memhotplug.c
> +++ b/drivers/acpi/acpi_memhotplug.c
> @@ -361,7 +361,25 @@ static void acpi_memory_device_remove(struct acpi_device *device)
> acpi_memory_device_free(mem_device);
> }
>
> +static bool acpi_no_memhotplug;
> +
> +void set_acpi_no_memhotplug(void)
> +{
> + acpi_no_memhotplug = true;
> + pr_info_once("ACPI: Memory Hotplug Disabled\n");
> +}
> +
> void __init acpi_memory_hotplug_init(void)
> {
> + if (acpi_no_memhotplug)
> + return;
> +
> acpi_scan_add_handler_with_hotplug(&memory_device_handler, "memory");
> }
> +
> +static int __init disable_acpi_memory_hotplug(char *str)
> +{
> + set_acpi_no_memhotplug();
> + return 1;
> +}
> +__setup("acpi_no_memhotplug", disable_acpi_memory_hotplug);
> diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
> index 4ca3d95..3cdb6e0 100644
> --- a/include/linux/memory_hotplug.h
> +++ b/include/linux/memory_hotplug.h
> @@ -12,6 +12,15 @@ struct pglist_data;
> struct mem_section;
> struct memory_block;
>
> +#ifdef CONFIG_ACPI_HOTPLUG_MEMORY
> +/* set flag to disable ACPI memory hotplug */
> +extern void set_acpi_no_memhotplug(void);
> +#else
> +static inline void set_acpi_no_memhotplug(void)
> +{
> +}
> +#endif
> +
> #ifdef CONFIG_MEMORY_HOTPLUG
>
> /*
> --
> 1.7.9.3
>
Acked-by: Dave Young <dyoung@redhat.com>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 1:41 ` Toshi Kani
@ 2014-01-14 2:43 ` Yasuaki Ishimatsu
2014-01-14 10:58 ` Prarit Bhargava
2014-01-14 15:17 ` Toshi Kani
2014-01-14 11:05 ` Prarit Bhargava
2014-01-14 14:36 ` Vivek Goyal
2 siblings, 2 replies; 14+ messages in thread
From: Yasuaki Ishimatsu @ 2014-01-14 2:43 UTC (permalink / raw)
To: Toshi Kani, Prarit Bhargava
Cc: KOSAKI Motohiro, LKML, Thomas Gleixner, Ingo Molnar,
H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
(2014/01/14 10:41), Toshi Kani wrote:
> On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
> :
>>>> I think we need a knob manually enable mem-hotplug when specify memmap. But
>>>> it is another story.
>>>>
>>>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
>>>
>>> As mentioned, self-NAK. I have seen a system that I needed to specify
>>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
>>> option in the next version of the patch.
>>
>>
>> Your following first patch is simply and makes sense.
>>
>> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
>>
>
> In this option, it also requires changing kexec-tools to specify the new
> option for kdump. It won't be simpler.
Hmm.
I use memm= boot option and hotplug memory for memory hot-remove.
At least, the patch cannot be accepted.
Thanks,
Yasuaki Ishimatsu
>
> Thanks,
> -Toshi
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 2:43 ` Yasuaki Ishimatsu
@ 2014-01-14 10:58 ` Prarit Bhargava
2014-01-14 15:17 ` Toshi Kani
1 sibling, 0 replies; 14+ messages in thread
From: Prarit Bhargava @ 2014-01-14 10:58 UTC (permalink / raw)
To: Yasuaki Ishimatsu
Cc: Toshi Kani, KOSAKI Motohiro, LKML, Thomas Gleixner, Ingo Molnar,
H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On 01/13/2014 09:43 PM, Yasuaki Ishimatsu wrote:
> (2014/01/14 10:41), Toshi Kani wrote:
>> On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
>> :
>>>>> I think we need a knob manually enable mem-hotplug when specify memmap. But
>>>>> it is another story.
>>>>>
>>>>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
>>>>
>>>> As mentioned, self-NAK. I have seen a system that I needed to specify
>>>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
>>>> option in the next version of the patch.
>>>
>>>
>>> Your following first patch is simply and makes sense.
>>>
>>> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
>>>
>>
>> In this option, it also requires changing kexec-tools to specify the new
>> option for kdump. It won't be simpler.
>
> Hmm.
> I use memm= boot option and hotplug memory for memory hot-remove.
> At least, the patch cannot be accepted.
Thanks for the information Yasuaki. I will resubmit my first patch that only
adds the kernel parameter.
P.
>
> Thanks,
> Yasuaki Ishimatsu
>
>>
>> Thanks,
>> -Toshi
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>
>
>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 1:41 ` Toshi Kani
2014-01-14 2:43 ` Yasuaki Ishimatsu
@ 2014-01-14 11:05 ` Prarit Bhargava
2014-01-14 14:37 ` Vivek Goyal
2014-01-14 15:26 ` Vivek Goyal
2014-01-14 14:36 ` Vivek Goyal
2 siblings, 2 replies; 14+ messages in thread
From: Prarit Bhargava @ 2014-01-14 11:05 UTC (permalink / raw)
To: Toshi Kani
Cc: Yasuaki Ishimatsu, KOSAKI Motohiro, LKML, Thomas Gleixner,
Ingo Molnar, H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On 01/13/2014 08:41 PM, Toshi Kani wrote:
> On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
> :
>>>> I think we need a knob manually enable mem-hotplug when specify memmap. But
>>>> it is another story.
>>>>
>>>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
>>>
>>> As mentioned, self-NAK. I have seen a system that I needed to specify
>>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
>>> option in the next version of the patch.
>>
>>
>> Your following first patch is simply and makes sense.
>>
>> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
>>
>
> In this option, it also requires changing kexec-tools to specify the new
> option for kdump. It won't be simpler.
It will be simpler for the kernel and those of us who have to debug busted e820
maps ;)
Unfortunately I may not be able to give you the automatic disable. I did
contemplate adding a !is_kdump_kernel() to the ACPI memory hotplug init call,
but it seems like that is unacceptable as well.
P.
>
> Thanks,
> -Toshi
>
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 1:41 ` Toshi Kani
2014-01-14 2:43 ` Yasuaki Ishimatsu
2014-01-14 11:05 ` Prarit Bhargava
@ 2014-01-14 14:36 ` Vivek Goyal
2 siblings, 0 replies; 14+ messages in thread
From: Vivek Goyal @ 2014-01-14 14:36 UTC (permalink / raw)
To: Toshi Kani
Cc: Yasuaki Ishimatsu, Prarit Bhargava, KOSAKI Motohiro, LKML,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Tang Chen, Wen Congyang, Dave Young, linux-acpi, linux-mm
On Mon, Jan 13, 2014 at 06:41:29PM -0700, Toshi Kani wrote:
> On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
> :
> > >> I think we need a knob manually enable mem-hotplug when specify memmap. But
> > >> it is another story.
> > >>
> > >> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> > >
> > > As mentioned, self-NAK. I have seen a system that I needed to specify
> > > memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> > > option in the next version of the patch.
> >
> >
> > Your following first patch is simply and makes sense.
> >
> > http://marc.info/?l=linux-acpi&m=138922019607796&w=2
> >
>
> In this option, it also requires changing kexec-tools to specify the new
> option for kdump. It won't be simpler.
I am thinking that instead of modifying kexec-tools, it can be made
part of the documentation so that user is expected to pass this parameter.
In fedora, we can modify /etc/sysconfig/kdump and add the parameter by
default so that user's don't have to worry about passing it.
Thanks
Vivek
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 11:05 ` Prarit Bhargava
@ 2014-01-14 14:37 ` Vivek Goyal
2014-01-14 15:26 ` Vivek Goyal
1 sibling, 0 replies; 14+ messages in thread
From: Vivek Goyal @ 2014-01-14 14:37 UTC (permalink / raw)
To: Prarit Bhargava
Cc: Toshi Kani, Yasuaki Ishimatsu, KOSAKI Motohiro, LKML,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Tang Chen, Wen Congyang, Dave Young, linux-acpi, linux-mm
On Tue, Jan 14, 2014 at 06:05:15AM -0500, Prarit Bhargava wrote:
>
>
> On 01/13/2014 08:41 PM, Toshi Kani wrote:
> > On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
> > :
> >>>> I think we need a knob manually enable mem-hotplug when specify memmap. But
> >>>> it is another story.
> >>>>
> >>>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> >>>
> >>> As mentioned, self-NAK. I have seen a system that I needed to specify
> >>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> >>> option in the next version of the patch.
> >>
> >>
> >> Your following first patch is simply and makes sense.
> >>
> >> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
> >>
> >
> > In this option, it also requires changing kexec-tools to specify the new
> > option for kdump. It won't be simpler.
>
> It will be simpler for the kernel and those of us who have to debug busted e820
> maps ;)
>
> Unfortunately I may not be able to give you the automatic disable. I did
> contemplate adding a !is_kdump_kernel() to the ACPI memory hotplug init call,
> but it seems like that is unacceptable as well.
Yep, I don't think hotplug feature should be tied to it being a kdump
kernel.
Thanks
Vivek
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 2:43 ` Yasuaki Ishimatsu
2014-01-14 10:58 ` Prarit Bhargava
@ 2014-01-14 15:17 ` Toshi Kani
1 sibling, 0 replies; 14+ messages in thread
From: Toshi Kani @ 2014-01-14 15:17 UTC (permalink / raw)
To: Yasuaki Ishimatsu
Cc: Prarit Bhargava, KOSAKI Motohiro, LKML, Thomas Gleixner,
Ingo Molnar, H. Peter Anvin, the arch/x86 maintainers, Len Brown,
Rafael J. Wysocki, Linn Crosetto, Pekka Enberg, Yinghai Lu,
Andrew Morton, Tang Chen, Wen Congyang, Vivek Goyal, Dave Young,
linux-acpi, linux-mm
On Tue, 2014-01-14 at 11:43 +0900, Yasuaki Ishimatsu wrote:
> (2014/01/14 10:41), Toshi Kani wrote:
> > On Tue, 2014-01-14 at 10:11 +0900, Yasuaki Ishimatsu wrote:
> > :
> >>>> I think we need a knob manually enable mem-hotplug when specify memmap. But
> >>>> it is another story.
> >>>>
> >>>> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> >>>
> >>> As mentioned, self-NAK. I have seen a system that I needed to specify
> >>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> >>> option in the next version of the patch.
> >>
> >>
> >> Your following first patch is simply and makes sense.
> >>
> >> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
> >>
> >
> > In this option, it also requires changing kexec-tools to specify the new
> > option for kdump. It won't be simpler.
>
> Hmm.
> I use memm= boot option and hotplug memory for memory hot-remove.
> At least, the patch cannot be accepted.
Do you mean mem=nG option? I am curious to know how memory hot-remove
works in this case. Don't acpi_memhotplug add all the memory ranges at
boot, and defeat the mem=nG option?
Thanks,
-Toshi
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 11:05 ` Prarit Bhargava
2014-01-14 14:37 ` Vivek Goyal
@ 2014-01-14 15:26 ` Vivek Goyal
2014-01-14 15:31 ` Toshi Kani
1 sibling, 1 reply; 14+ messages in thread
From: Vivek Goyal @ 2014-01-14 15:26 UTC (permalink / raw)
To: Prarit Bhargava
Cc: Toshi Kani, Yasuaki Ishimatsu, KOSAKI Motohiro, LKML,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Tang Chen, Wen Congyang, Dave Young, linux-acpi, linux-mm
On Tue, Jan 14, 2014 at 06:05:15AM -0500, Prarit Bhargava wrote:
[..]
> >>> As mentioned, self-NAK. I have seen a system that I needed to specify
> >>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> >>> option in the next version of the patch.
> >>
> >>
> >> Your following first patch is simply and makes sense.
> >>
> >> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
> >>
> >
> > In this option, it also requires changing kexec-tools to specify the new
> > option for kdump. It won't be simpler.
>
> It will be simpler for the kernel and those of us who have to debug busted e820
> maps ;)
>
> Unfortunately I may not be able to give you the automatic disable. I did
> contemplate adding a !is_kdump_kernel() to the ACPI memory hotplug init call,
> but it seems like that is unacceptable as well.
I think everybody agrees that there has to be a stand alone command line
option to disable memory hotplug.
Whether to tie it into memmap=exactmap and mem=X is the contentious bit.
So I would suggest that just post a patch to disable memory hotplut using
a command line and later more patches can go in if people strongly feel
the need to tie it into memmap=exactmap.
In the mean time, we will modify /etc/sysconfig/kdump to pass
acpi_no_memhotplug so that user does not have to worry about passing this
parameter and kexec-tools will not have to be modified either.
Thanks
Vivek
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug
2014-01-14 15:26 ` Vivek Goyal
@ 2014-01-14 15:31 ` Toshi Kani
0 siblings, 0 replies; 14+ messages in thread
From: Toshi Kani @ 2014-01-14 15:31 UTC (permalink / raw)
To: Vivek Goyal
Cc: Prarit Bhargava, Yasuaki Ishimatsu, KOSAKI Motohiro, LKML,
Thomas Gleixner, Ingo Molnar, H. Peter Anvin,
the arch/x86 maintainers, Len Brown, Rafael J. Wysocki,
Linn Crosetto, Pekka Enberg, Yinghai Lu, Andrew Morton,
Tang Chen, Wen Congyang, Dave Young, linux-acpi, linux-mm
On Tue, 2014-01-14 at 10:26 -0500, Vivek Goyal wrote:
> On Tue, Jan 14, 2014 at 06:05:15AM -0500, Prarit Bhargava wrote:
>
> [..]
> > >>> As mentioned, self-NAK. I have seen a system that I needed to specify
> > >>> memmap=exactmap & had hotplug memory. I will only keep the acpi_no_memhotplug
> > >>> option in the next version of the patch.
> > >>
> > >>
> > >> Your following first patch is simply and makes sense.
> > >>
> > >> http://marc.info/?l=linux-acpi&m=138922019607796&w=2
> > >>
> > >
> > > In this option, it also requires changing kexec-tools to specify the new
> > > option for kdump. It won't be simpler.
> >
> > It will be simpler for the kernel and those of us who have to debug busted e820
> > maps ;)
> >
> > Unfortunately I may not be able to give you the automatic disable. I did
> > contemplate adding a !is_kdump_kernel() to the ACPI memory hotplug init call,
> > but it seems like that is unacceptable as well.
>
> I think everybody agrees that there has to be a stand alone command line
> option to disable memory hotplug.
>
> Whether to tie it into memmap=exactmap and mem=X is the contentious bit.
> So I would suggest that just post a patch to disable memory hotplut using
> a command line and later more patches can go in if people strongly feel
> the need to tie it into memmap=exactmap.
>
> In the mean time, we will modify /etc/sysconfig/kdump to pass
> acpi_no_memhotplug so that user does not have to worry about passing this
> parameter and kexec-tools will not have to be modified either.
Fine by me. Thanks for modifying /etc/sysconfig/kdump file.
-Toshi
--
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>
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2014-01-14 15:37 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-01-13 21:56 [PATCH] x86, acpi memory hotplug, add parameter to disable memory hotplug Prarit Bhargava
2014-01-13 22:17 ` KOSAKI Motohiro
2014-01-13 23:41 ` Prarit Bhargava
2014-01-14 1:11 ` Yasuaki Ishimatsu
2014-01-14 1:41 ` Toshi Kani
2014-01-14 2:43 ` Yasuaki Ishimatsu
2014-01-14 10:58 ` Prarit Bhargava
2014-01-14 15:17 ` Toshi Kani
2014-01-14 11:05 ` Prarit Bhargava
2014-01-14 14:37 ` Vivek Goyal
2014-01-14 15:26 ` Vivek Goyal
2014-01-14 15:31 ` Toshi Kani
2014-01-14 14:36 ` Vivek Goyal
2014-01-14 1:45 ` Dave Young
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox