mm/mm_init.c | 4 ++++ 1 file changed, 4 insertions(+)
Currently when CONFIG_DEFERRED_STRUCT_PAGE_INIT is not set, the reserved
pages are initialized to MIGRATE_MOVABLE by default in memmap_init.
Reserved memory mainly store the metadata of struct page. When
HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=Y and hugepages are allocated,
the HVO will remap the vmemmap virtual address range to the page which
vmemmap_reuse is mapped to. The pages previously mapping the range will
be freed to the buddy system.
Before this patch:
when CONFIG_DEFERRED_STRUCT_PAGE_INIT is not set, the freed memory was
placed on the Movable list;
When CONFIG_DEFERRED_STRUCT_PAGE_INIT=Y, the freed memory was placed on
the Unmovable list.
After this patch, the freed memory is placed on the Movable list
regardless of whether CONFIG_DEFERRED_STRUCT_PAGE_INIT is set.
Eg:
Tested on a virtual machine(1000GB):
Intel(R) Xeon(R) Platinum 8358P CPU
After vm start:
echo 500000 > /proc/sys/vm/nr_hugepages
cat /proc/meminfo | grep -i huge
HugePages_Total: 500000
HugePages_Free: 500000
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 1024000000 kB
cat /proc/pagetypeinfo
before：
Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
…
Node 0, zone Normal, type Unmovable 51 2 1 28 53 35 35 43 40 69 3852
Node 0, zone Normal, type Movable 6485 4610 666 202 200 185 208 87 54 2 240
Node 0, zone Normal, type Reclaimable 2 2 1 23 13 1 2 1 0 1 0
Node 0, zone Normal, type HighAtomic 0 0 0 0 0 0 0 0 0 0 0
Node 0, zone Normal, type Isolate 0 0 0 0 0 0 0 0 0 0 0
Unmovable ≈ 15GB
after：
Free pages count per migrate type at order 0 1 2 3 4 5 6 7 8 9 10
…
Node 0, zone Normal, type Unmovable 0 1 1 0 0 0 0 1 1 1 0
Node 0, zone Normal, type Movable 1563 4107 1119 189 256 368 286 132 109 4 3841
Node 0, zone Normal, type Reclaimable 2 2 1 23 13 1 2 1 0 1 0
Node 0, zone Normal, type HighAtomic 0 0 0 0 0 0 0 0 0 0 0
Node 0, zone Normal, type Isolate 0 0 0 0 0 0 0 0 0 0 0
Signed-off-by: Hua Su <suhua.tanke@gmail.com>
---
mm/mm_init.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/mm/mm_init.c b/mm/mm_init.c
index 4ba5607aaf19..6dbf2df23eee 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -722,6 +722,10 @@ static void __meminit init_reserved_page(unsigned long pfn, int nid)
if (zone_spans_pfn(zone, pfn))
break;
}
+
+ if (pageblock_aligned(pfn))
+ set_pageblock_migratetype(pfn_to_page(pfn), MIGRATE_MOVABLE);
+
__init_single_page(pfn_to_page(pfn), pfn, zid, nid);
}
#else
--
2.34.1
Hello,
kernel test robot noticed "kernel_BUG_at_include/linux/mm.h" on:
commit: 0a19e28247d042d639e5a46c3698adeda268a7a2 ("[PATCH] memblock: uniformly initialize all reserved pages to MIGRATE_MOVABLE")
url: https://github.com/intel-lab-lkp/linux/commits/Hua-Su/memblock-uniformly-initialize-all-reserved-pages-to-MIGRATE_MOVABLE/20241021-131358
base: https://git.kernel.org/cgit/linux/kernel/git/akpm/mm.git mm-everything
patch link: https://lore.kernel.org/all/20241021051151.4664-1-suhua.tanke@gmail.com/
patch subject: [PATCH] memblock: uniformly initialize all reserved pages to MIGRATE_MOVABLE
in testcase: boot
config: x86_64-randconfig-012-20241023
compiler: gcc-12
test machine: qemu-system-x86_64 -enable-kvm -cpu SandyBridge -smp 2 -m 16G
(please refer to attached dmesg/kmsg for entire log/backtrace)
+------------------------------------------+------------+------------+
| | a8883372ec | 0a19e28247 |
+------------------------------------------+------------+------------+
| boot_successes | 18 | 0 |
| boot_failures | 0 | 18 |
| kernel_BUG_at_include/linux/mm.h | 0 | 18 |
| Oops:invalid_opcode:#[##]SMP_PTI | 0 | 18 |
| RIP:page_zone | 0 | 18 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 18 |
+------------------------------------------+------------+------------+
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@intel.com>
| Closes: https://lore.kernel.org/oe-lkp/202410251024.eb4a89f1-oliver.sang@intel.com
[ 0.262363][ T0] ------------[ cut here ]------------
[ 0.262921][ T0] kernel BUG at include/linux/mm.h:1637!
[ 0.263532][ T0] Oops: invalid opcode: 0000 [#1] SMP PTI
[ 0.264140][ T0] CPU: 0 UID: 0 PID: 0 Comm: swapper Tainted: G T 6.12.0-rc3-00235-g0a19e28247d0 #1
[ 0.265300][ T0] Tainted: [T]=RANDSTRUCT
[ 0.265762][ T0] RIP: 0010:page_zone (include/linux/mm.h:1858)
[ 0.266284][ T0] Code: 43 08 89 ee 48 89 df 31 d2 5b 5d 41 5c 41 5d 41 5e e9 f1 08 02 00 48 8b 07 48 ff c0 75 0e 48 c7 c6 27 2e 99 ac e8 42 73 fd ff <0f> 0b 48 8b 07 48 c1 e8 3e 48 69 c0 40 06 00 00 48 05 c0 63 6c ad
All code
========
0: 43 08 89 ee 48 89 df rex.XB or %cl,-0x2076b712(%r9)
7: 31 d2 xor %edx,%edx
9: 5b pop %rbx
a: 5d pop %rbp
b: 41 5c pop %r12
d: 41 5d pop %r13
f: 41 5e pop %r14
11: e9 f1 08 02 00 jmpq 0x20907
16: 48 8b 07 mov (%rdi),%rax
19: 48 ff c0 inc %rax
1c: 75 0e jne 0x2c
1e: 48 c7 c6 27 2e 99 ac mov $0xffffffffac992e27,%rsi
25: e8 42 73 fd ff callq 0xfffffffffffd736c
2a:* 0f 0b ud2 <-- trapping instruction
2c: 48 8b 07 mov (%rdi),%rax
2f: 48 c1 e8 3e shr $0x3e,%rax
33: 48 69 c0 40 06 00 00 imul $0x640,%rax,%rax
3a: 48 05 c0 63 6c ad add $0xffffffffad6c63c0,%rax
Code starting with the faulting instruction
===========================================
0: 0f 0b ud2
2: 48 8b 07 mov (%rdi),%rax
5: 48 c1 e8 3e shr $0x3e,%rax
9: 48 69 c0 40 06 00 00 imul $0x640,%rax,%rax
10: 48 05 c0 63 6c ad add $0xffffffffad6c63c0,%rax
[ 0.268346][ T0] RSP: 0000:fffffffface03dc0 EFLAGS: 00010046
[ 0.268988][ T0] RAX: 0000000000000000 RBX: 0000000000000007 RCX: 0000000000000000
[ 0.269844][ T0] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[ 0.270685][ T0] RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
[ 0.271486][ T0] R10: 0000000000000000 R11: 6d75642065676170 R12: 0000000000000001
[ 0.272336][ T0] R13: 0000000000159400 R14: fffff7bb05650000 R15: ffff9bfa1ffff178
[ 0.273172][ T0] FS: 0000000000000000(0000) GS:ffff9bfcefa00000(0000) knlGS:0000000000000000
[ 0.274145][ T0] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 0.274817][ T0] CR2: ffff9bfcfffff000 CR3: 000000015c2b2000 CR4: 00000000000000b0
[ 0.275658][ T0] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 0.276502][ T0] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 0.277343][ T0] Call Trace:
[ 0.277691][ T0] <TASK>
[ 0.277992][ T0] ? __die_body (arch/x86/kernel/dumpstack.c:421)
[ 0.278448][ T0] ? die (arch/x86/kernel/dumpstack.c:449)
[ 0.278838][ T0] ? do_trap (arch/x86/kernel/traps.c:156 arch/x86/kernel/traps.c:197)
[ 0.279276][ T0] ? page_zone (include/linux/mm.h:1858)
[ 0.279720][ T0] ? page_zone (include/linux/mm.h:1858)
[ 0.280170][ T0] ? do_error_trap (arch/x86/kernel/traps.c:218)
[ 0.280648][ T0] ? page_zone (include/linux/mm.h:1858)
[ 0.281095][ T0] ? exc_invalid_op (arch/x86/kernel/traps.c:316)
[ 0.281597][ T0] ? page_zone (include/linux/mm.h:1858)
[ 0.282041][ T0] ? asm_exc_invalid_op (arch/x86/include/asm/idtentry.h:621)
[ 0.282582][ T0] ? page_zone (include/linux/mm.h:1858)
[ 0.283027][ T0] set_pfnblock_flags_mask (mm/page_alloc.c:408)
[ 0.283583][ T0] reserve_bootmem_region (mm/mm_init.c:729 mm/mm_init.c:765)
[ 0.284142][ T0] free_low_memory_core_early (mm/memblock.c:2192 mm/memblock.c:2205)
[ 0.284736][ T0] ? swiotlb_init_io_tlb_pool+0x86/0x133
[ 0.285419][ T0] memblock_free_all (mm/memblock.c:2252)
[ 0.285925][ T0] mem_init (arch/x86/mm/init_64.c:1360)
[ 0.286332][ T0] mm_core_init (mm/mm_init.c:2658)
[ 0.286790][ T0] start_kernel (init/main.c:965)
[ 0.287272][ T0] x86_64_start_reservations (arch/x86/kernel/head64.c:381)
[ 0.287850][ T0] x86_64_start_kernel (arch/x86/kernel/ebda.c:57)
[ 0.288377][ T0] common_startup_64 (arch/x86/kernel/head_64.S:414)
[ 0.288899][ T0] </TASK>
[ 0.289213][ T0] Modules linked in:
[ 0.289626][ T0] ---[ end trace 0000000000000000 ]---
[ 0.290175][ T0] RIP: 0010:page_zone (include/linux/mm.h:1858)
[ 0.290680][ T0] Code: 43 08 89 ee 48 89 df 31 d2 5b 5d 41 5c 41 5d 41 5e e9 f1 08 02 00 48 8b 07 48 ff c0 75 0e 48 c7 c6 27 2e 99 ac e8 42 73 fd ff <0f> 0b 48 8b 07 48 c1 e8 3e 48 69 c0 40 06 00 00 48 05 c0 63 6c ad
All code
========
0: 43 08 89 ee 48 89 df rex.XB or %cl,-0x2076b712(%r9)
7: 31 d2 xor %edx,%edx
9: 5b pop %rbx
a: 5d pop %rbp
b: 41 5c pop %r12
d: 41 5d pop %r13
f: 41 5e pop %r14
11: e9 f1 08 02 00 jmpq 0x20907
16: 48 8b 07 mov (%rdi),%rax
19: 48 ff c0 inc %rax
1c: 75 0e jne 0x2c
1e: 48 c7 c6 27 2e 99 ac mov $0xffffffffac992e27,%rsi
25: e8 42 73 fd ff callq 0xfffffffffffd736c
2a:* 0f 0b ud2 <-- trapping instruction
2c: 48 8b 07 mov (%rdi),%rax
2f: 48 c1 e8 3e shr $0x3e,%rax
33: 48 69 c0 40 06 00 00 imul $0x640,%rax,%rax
3a: 48 05 c0 63 6c ad add $0xffffffffad6c63c0,%rax
Code starting with the faulting instruction
===========================================
0: 0f 0b ud2
2: 48 8b 07 mov (%rdi),%rax
5: 48 c1 e8 3e shr $0x3e,%rax
9: 48 69 c0 40 06 00 00 imul $0x640,%rax,%rax
10: 48 05 c0 63 6c ad add $0xffffffffad6c63c0,%rax
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20241025/202410251024.eb4a89f1-oliver.sang@intel.com
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
From: Mike Rapoport (Microsoft) <rppt@kernel.org>
On Mon, 21 Oct 2024 13:11:51 +0800, Hua Su wrote:
> Currently when CONFIG_DEFERRED_STRUCT_PAGE_INIT is not set, the reserved
> pages are initialized to MIGRATE_MOVABLE by default in memmap_init.
>
> Reserved memory mainly store the metadata of struct page. When
> HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=Y and hugepages are allocated,
> the HVO will remap the vmemmap virtual address range to the page which
> vmemmap_reuse is mapped to. The pages previously mapping the range will
> be freed to the buddy system.
>
> [...]
Applied to for-next branch of memblock.git tree, thanks!
[1/1] memblock: uniformly initialize all reserved pages to MIGRATE_MOVABLE
commit: ad48825232a91a382f665bb7c3bf0044027791d4
tree: https://git.kernel.org/pub/scm/linux/kernel/git/rppt/memblock
branch: for-next
--
Sincerely yours,
Mike.
© 2016 - 2024 Red Hat, Inc.