On 2/18/19 4:55 PM, Michal Hocko wrote:
> [Sorry for an excessive quoting in the previous email]
> [Cc Pavel - the full report is http://lkml.kernel.org/r/20190218052823.GH29177@shao2-debian[]
>
> On Mon 18-02-19 08:08:44, Michal Hocko wrote:
>> On Mon 18-02-19 13:28:23, kernel test robot wrote:
> [...]
>>> [   40.305212] PGD 0 P4D 0
>>> [   40.308255] Oops: 0000 [#1] PREEMPT SMP PTI
>>> [   40.313055] CPU: 1 PID: 239 Comm: udevd Not tainted 5.0.0-rc4-00149-gefad4e4 #1
>>> [   40.321348] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
>>> [   40.330813] RIP: 0010:page_mapping+0x12/0x80
>>> [   40.335709] Code: 5d c3 48 89 df e8 0e ad 02 00 85 c0 75 da 89 e8 5b 5d c3 0f 1f 44 00 00 53 48 89 fb 48 8b 43 08 48 8d 50 ff a8 01 48 0f 45 da <48> 8b 53 08 48 8d 42 ff 83 e2 01 48 0f 44 c3 48 83 38 ff 74 2f 48
>>> [   40.356704] RSP: 0018:ffff88801fa87cd8 EFLAGS: 00010202
>>> [   40.362714] RAX: ffffffffffffffff RBX: fffffffffffffffe RCX: 000000000000000a
>>> [   40.370798] RDX: fffffffffffffffe RSI: ffffffff820b9a20 RDI: ffff88801e5c0000
>>> [   40.378830] RBP: 6db6db6db6db6db7 R08: ffff88801e8bb000 R09: 0000000001b64d13
>>> [   40.386902] R10: ffff88801fa87cf8 R11: 0000000000000001 R12: ffff88801e640000
>>> [   40.395033] R13: ffffffff820b9a20 R14: ffff88801f145258 R15: 0000000000000001
>>> [   40.403138] FS:  00007fb2079817c0(0000) GS:ffff88801dd00000(0000) knlGS:0000000000000000
>>> [   40.412243] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>> [   40.418846] CR2: 0000000000000006 CR3: 000000001fa82000 CR4: 00000000000006a0
>>> [   40.426951] Call Trace:
>>> [   40.429843]  __dump_page+0x14/0x2c0
>>> [   40.433947]  is_mem_section_removable+0x24c/0x2c0
>> This looks like we are stumbling over an unitialized struct page again.
>> Something this patch should prevent from. Could you try to apply [1]
>> which will make __dump_page more robust so that we do not blow up there
>> and give some more details in return.
>>
>> Btw. is this reproducible all the time?
> And forgot to ask whether this is reproducible with pending mmotm
> patches in linux-next.


Do you mean the below patch? I can reproduce the problem too.

Best Regards,
Rong Chen


>
>> I will have a look at the memory layout later today.
> [    0.059335] No NUMA configuration found
> [    0.059345] Faking a node at [mem 0x0000000000000000-0x000000001ffdffff]
> [    0.059399] NODE_DATA(0) allocated [mem 0x1e8c3000-0x1e8c5fff]
> [    0.073143] Zone ranges:
> [    0.073175]   DMA32    [mem 0x0000000000001000-0x000000001ffdffff]
> [    0.073204]   Normal   empty
> [    0.073212] Movable zone start for each node
> [    0.073240] Early memory node ranges
> [    0.073247]   node   0: [mem 0x0000000000001000-0x000000000009efff]
> [    0.073275]   node   0: [mem 0x0000000000100000-0x000000001ffdffff]
> [    0.073309] Zeroed struct page in unavailable ranges: 98 pages
> [    0.073312] Initmem setup node 0 [mem 0x0000000000001000-0x000000001ffdffff]
> [    0.073343] On node 0 totalpages: 130942
> [    0.073373]   DMA32 zone: 1792 pages used for memmap
> [    0.073400]   DMA32 zone: 21 pages reserved
> [    0.073408]   DMA32 zone: 130942 pages, LIFO batch:31
>
> We have only a single NUMA node with a single ZONE_DMA32. But there is a
> hole in the zone and the first range before the hole is not section
> aligned. We do zero some unavailable ranges but from the number it is no
> clear which range it is and 98. [0x60fff, 0xfffff) is 96 pages. The
> patch below should tell us whether we are covering all we need. If yes
> then the hole shouldn't make any difference and the problem must be
> somewhere else.
>
> ---
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 35fdde041f5c..c60642505e04 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -6706,10 +6706,13 @@ void __init zero_resv_unavail(void)
>   	pgcnt = 0;
>   	for_each_mem_range(i, &memblock.memory, NULL,
>   			NUMA_NO_NODE, MEMBLOCK_NONE, &start, &end, NULL) {
> -		if (next < start)
> +		if (next < start) {
> +			pr_info("zeroying %llx-%llx\n", PFN_DOWN(next), PFN_UP(start));
>   			pgcnt += zero_pfn_range(PFN_DOWN(next), PFN_UP(start));
> +		}
>   		next = end;
>   	}
> +	pr_info("zeroying %llx-%lx\n", PFN_DOWN(next), max_pfn);
>   	pgcnt += zero_pfn_range(PFN_DOWN(next), max_pfn);
>   
>   	/*