Re: PROBLEM: Power9: kernel oops on memory hotunplug from ppc64le guest

[Nicholas Piggin]

Bharata B Rao's on May 20, 2019 2:25 pm:
> On Mon, May 20, 2019 at 12:02:23PM +1000, Michael Ellerman wrote:
>> Bharata B Rao <bharata@xxxxxxxxxxxxx> writes:
>> > On Thu, May 16, 2019 at 07:44:20PM +0530, srikanth wrote:
>> >> Hello,
>> >> 
>> >> On power9 host, performing memory hotunplug from ppc64le guest results in
>> >> kernel oops.
>> >> 
>> >> Kernel used : https://github.com/torvalds/linux/tree/v5.1 built using
>> >> ppc64le_defconfig for host and ppc64le_guest_defconfig for guest.
>> >> 
>> >> Recreation steps:
>> >> 
>> >> 1. Boot a guest with below mem configuration:
>> >> Â <maxMemory slots='32' unit='KiB'>33554432</maxMemory>
>> >> Â <memory unit='KiB'>8388608</memory>
>> >> Â <currentMemory unit='KiB'>4194304</currentMemory>
>> >> Â <cpu>
>> >> ÂÂÂ <numa>
>> >> ÂÂÂÂÂ <cell id='0' cpus='0-31' memory='8388608' unit='KiB'/>
>> >> ÂÂÂ </numa>
>> >> Â </cpu>
>> >> 
>> >> 2. From host hotplug 8G memory -> verify memory hotadded succesfully -> now
>> >> reboot guest -> once guest comes back try to unplug 8G memory
>> >> 
>> >> mem.xml used:
>> >> <memory model='dimm'>
>> >> <target>
>> >> <size unit='GiB'>8</size>
>> >> <node>0</node>
>> >> </target>
>> >> </memory>
>> >> 
>> >> Memory attach and detach commands used:
>> >> ÂÂÂ virsh attach-device vm1 ./mem.xml --live
>> >> ÂÂÂ virsh detach-device vm1 ./mem.xml --live
>> >> 
>> >> Trace seen inside guest after unplug, guest just hangs there forever:
>> >> 
>> >> [ÂÂ 21.962986] kernel BUG at arch/powerpc/mm/pgtable-frag.c:113!
>> >> [ÂÂ 21.963064] Oops: Exception in kernel mode, sig: 5 [#1]
>> >> [ÂÂ 21.963090] LE PAGE_SIZE=64K MMU=Radix MMU=Hash SMP NR_CPUS=2048 NUMA
>> >> pSeries
>> >> [ÂÂ 21.963131] Modules linked in: xt_tcpudp iptable_filter squashfs fuse
>> >> vmx_crypto ib_iser rdma_cm iw_cm ib_cm ib_core libiscsi scsi_transport_iscsi
>> >> ip_tables x_tables autofs4 btrfs zstd_decompress zstd_compress lzo_compress
>> >> raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx
>> >> xor raid6_pq multipath crc32c_vpmsum
>> >> [ÂÂ 21.963281] CPU: 11 PID: 316 Comm: kworker/u64:5 Kdump: loaded Not
>> >> tainted 5.1.0-dirty #2
>> >> [ÂÂ 21.963323] Workqueue: pseries hotplug workque pseries_hp_work_fn
>> >> [ÂÂ 21.963355] NIP:Â c000000000079e18 LR: c000000000c79308 CTR:
>> >> 0000000000008000
>> >> [ÂÂ 21.963392] REGS: c0000003f88034f0 TRAP: 0700ÂÂ Not tainted (5.1.0-dirty)
>> >> [ÂÂ 21.963422] MSR:Â 800000000282b033 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE>Â CR:
>> >> 28002884Â XER: 20040000
>> >> [ÂÂ 21.963470] CFAR: c000000000c79304 IRQMASK: 0
>> >> [ÂÂ 21.963470] GPR00: c000000000c79308 c0000003f8803780 c000000001521000
>> >> 0000000000fff8c0
>> >> [ÂÂ 21.963470] GPR04: 0000000000000001 00000000ffe30005 0000000000000005
>> >> 0000000000000020
>> >> [ÂÂ 21.963470] GPR08: 0000000000000000 0000000000000001 c00a000000fff8e0
>> >> c0000000016d21a0
>> >> [ÂÂ 21.963470] GPR12: c0000000016e7b90 c000000007ff2700 c00a000000a00000
>> >> c0000003ffe30100
>> >> [ÂÂ 21.963470] GPR16: c0000003ffe30000 c0000000014aa4de c00a0000009f0000
>> >> c0000000016d21b0
>> >> [ÂÂ 21.963470] GPR20: c0000000014de588 0000000000000001 c0000000016d21b8
>> >> c00a000000a00000
>> >> [ÂÂ 21.963470] GPR24: 0000000000000000 ffffffffffffffff c00a000000a00000
>> >> c0000003ffe96000
>> >> [ÂÂ 21.963470] GPR28: c00a000000a00000 c00a000000a00000 c0000003fffec000
>> >> c00a000000fff8c0
>> >> [ÂÂ 21.963802] NIP [c000000000079e18] pte_fragment_free+0x48/0xd0
>> >> [ÂÂ 21.963838] LR [c000000000c79308] remove_pagetable+0x49c/0x5b4
>> >> [ÂÂ 21.963873] Call Trace:
>> >> [ÂÂ 21.963890] [c0000003f8803780] [c0000003ffe997f0] 0xc0000003ffe997f0
>> >> (unreliable)
>> >> [ÂÂ 21.963933] [c0000003f88037b0] [0000000000000000] (null)
>> >> [ÂÂ 21.963969] [c0000003f88038c0] [c00000000006f038]
>> >> vmemmap_free+0x218/0x2e0
>> >> [ÂÂ 21.964006] [c0000003f8803940] [c00000000036f100]
>> >> sparse_remove_one_section+0xd0/0x138
>> >> [ÂÂ 21.964050] [c0000003f8803980] [c000000000383a50]
>> >> __remove_pages+0x410/0x560
>> >> [ÂÂ 21.964093] [c0000003f8803a90] [c000000000c784d8]
>> >> arch_remove_memory+0x68/0xdc
>> >> [ÂÂ 21.964136] [c0000003f8803ad0] [c000000000385d74]
>> >> __remove_memory+0xc4/0x110
>> >> [ÂÂ 21.964180] [c0000003f8803b10] [c0000000000d44e4]
>> >> dlpar_remove_lmb+0x94/0x140
>> >> [ÂÂ 21.964223] [c0000003f8803b50] [c0000000000d52b4]
>> >> dlpar_memory+0x464/0xd00
>> >> [ÂÂ 21.964259] [c0000003f8803be0] [c0000000000cd5c0]
>> >> handle_dlpar_errorlog+0xc0/0x190
>> >> [ÂÂ 21.964303] [c0000003f8803c50] [c0000000000cd6bc]
>> >> pseries_hp_work_fn+0x2c/0x60
>> >> [ÂÂ 21.964346] [c0000003f8803c80] [c00000000013a4a0]
>> >> process_one_work+0x2b0/0x5a0
>> >> [ÂÂ 21.964388] [c0000003f8803d10] [c00000000013a818]
>> >> worker_thread+0x88/0x610
>> >> [ÂÂ 21.964434] [c0000003f8803db0] [c000000000143884] kthread+0x1a4/0x1b0
>> >> [ÂÂ 21.964468] [c0000003f8803e20] [c00000000000bdc4]
>> >> ret_from_kernel_thread+0x5c/0x78
>> >> [ÂÂ 21.964506] Instruction dump:
>> >> [ÂÂ 21.964527] fbe1fff8 f821ffd1 78638502 78633664 ebe90000 7fff1a14
>> >> 395f0020 813f0020
>> >> [ÂÂ 21.964569] 7d2907b4 7d2900d0 79290fe0 69290001 <0b090000> 7c0004ac
>> >> 7d205028 3129ffff
>> >> [ÂÂ 21.964613] ---[ end trace aaa571aa1636fee6 ]---
>> >> [ÂÂ 21.966349]
>> >> [ÂÂ 21.966383] Sending IPI to other CPUs
>> >> [ÂÂ 21.978335] IPI complete
>> >> [ÂÂ 21.981354] kexec: Starting switchover sequence.
>> >> I'm in purgatory
>> >
>> > git bisect points to
>> >
>> > commit 4231aba000f5a4583dd9f67057aadb68c3eca99d
>> > Author: Nicholas Piggin <npiggin@xxxxxxxxx>
>> > Date:   Fri Jul 27 21:48:17 2018 +1000
>> >
>> >     powerpc/64s: Fix page table fragment refcount race vs speculative references
>> >
>> >     The page table fragment allocator uses the main page refcount racily
>> >     with respect to speculative references. A customer observed a BUG due
>> >     to page table page refcount underflow in the fragment allocator. This
>> >     can be caused by the fragment allocator set_page_count stomping on a
>> >     speculative reference, and then the speculative failure handler
>> >     decrements the new reference, and the underflow eventually pops when
>> >     the page tables are freed.
>> >
>> >     Fix this by using a dedicated field in the struct page for the page
>> >     table fragment allocator.
>> >
>> >     Fixes: 5c1f6ee9a31c ("powerpc: Reduce PTE table memory wastage")
>> >     Cc: stable@xxxxxxxxxxxxxxx # v3.10+
>> 
>> That's the commit that added the BUG_ON(), so prior to that you won't
>> see the crash.
> 
> Right, but the commit says it fixes page table page refcount underflow by
> introducing a new field &page->pt_frag_refcount. Now we are hitting the underflow
> for this pt_frag_refcount.

The fixed underflow is caused by a bug (race on page count) that got 
fixed by that patch. You are hitting a different underflow here. It's
not certain my patch caused it, I'm just trying to reproduce now.

> 
> BTW, if I go below this commit, I don't hit the pagecount
> 
> VM_BUG_ON_PAGE(page_ref_count(page) == 0, page);
> 
> which is in pte_fragment_free() path.

Do you have CONFIG_DEBUG_VM=y?

Thanks,
Nick