Re: Known and unfixed active data loss bug in MM + XFS with large folios since Dec 2021 (any kernel from 6.1 upwards)

[Jens Axboe <axboe@kernel.dk>]

On 9/17/24 7:25 AM, Matthew Wilcox wrote:
> On Tue, Sep 17, 2024 at 01:13:05PM +0200, Chris Mason wrote:
>> On 9/17/24 5:32 AM, Matthew Wilcox wrote:
>>> On Mon, Sep 16, 2024 at 10:47:10AM +0200, Chris Mason wrote:
>>>> I've got a bunch of assertions around incorrect folio->mapping and I'm
>>>> trying to bash on the ENOMEM for readahead case.  There's a GFP_NOWARN
>>>> on those, and our systems do run pretty short on ram, so it feels right
>>>> at least.  We'll see.
>>>
>>> I've been running with some variant of this patch the whole way across
>>> the Atlantic, and not hit any problems.  But maybe with the right
>>> workload ...?
>>>
>>> There are two things being tested here.  One is whether we have a
>>> cross-linked node (ie a node that's in two trees at the same time).
>>> The other is whether the slab allocator is giving us a node that already
>>> contains non-NULL entries.
>>>
>>> If you could throw this on top of your kernel, we might stand a chance
>>> of catching the problem sooner.  If it is one of these problems and not
>>> something weirder.
>>>
>>
>> This fires in roughly 10 seconds for me on top of v6.11.  Since array seems
>> to always be 1, I'm not sure if the assertion is right, but hopefully you
>> can trigger yourself.
> 
> Whoops.
> 
> $ git grep XA_RCU_FREE
> lib/xarray.c:#define XA_RCU_FREE        ((struct xarray *)1)
> lib/xarray.c:   node->array = XA_RCU_FREE;
> 
> so you walked into a node which is currently being freed by RCU.  Which
> isn't a problem, of course.  I don't know why I do that; it doesn't seem
> like anyone tests it.  The jetlag is seriously kicking in right now,
> so I'm going to refrain from saying anything more because it probably
> won't be coherent.

Based on a modified reproducer from Chris (N threads reading from a
file, M threads dropping pages), I can pretty quickly reproduce the
xas_descend() spin on 6.9 in a vm with 128 cpus. Here's some debugging
output with a modified version of your patch too, that ignores
XA_RCU_FREE:

node ffff8e838a01f788 max 59 parent 0000000000000000 shift 0 count 0 values 0 array ffff8e839dfa86a0 list ffff8e838a01f7a0 ffff8e838a01f7a0 marks 0 0 0
WARNING: CPU: 106 PID: 1554 at lib/xarray.c:405 xas_alloc.cold+0x26/0x4b

which is:

XA_NODE_BUG_ON(node, memchr_inv(&node->slots, 0, sizeof(void *) * XA_CHUN  K_SIZE));

and:

node ffff8e838a01f788 offset 59 parent ffff8e838b0419c8 shift 0 count 252 values 0 array ffff8e839dfa86a0 list ffff8e838a01f7a0 ffff8e838a01f7a0 marks 0 0 0

which is:

XA_NODE_BUG_ON(node, node->count > XA_CHUNK_SIZE);

and for this particular run, 2 threads spinning:

rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
rcu: 	Tasks blocked on level-1 rcu_node (CPUs 16-31): P1555
rcu: 	Tasks blocked on level-1 rcu_node (CPUs 64-79): P1556
rcu: 	(detected by 97, t=2102 jiffies, g=7821, q=293800 ncpus=128)
task:reader          state:R  running task     stack:0     pid:1555  tgid:1551  ppid:1      flags:0x00004006
Call Trace:
 <TASK>
 ? __schedule+0x37f/0xaa0
 ? sysvec_apic_timer_interrupt+0x96/0xb0
 ? asm_sysvec_apic_timer_interrupt+0x16/0x20
 ? xas_load+0x74/0xe0
 ? xas_load+0x10/0xe0
 ? xas_find+0x162/0x1b0
 ? find_lock_entries+0x1ac/0x360
 ? find_lock_entries+0x76/0x360
 ? mapping_try_invalidate+0x5d/0x130
 ? generic_fadvise+0x110/0x240
 ? xfd_validate_state+0x1e/0x70
 ? ksys_fadvise64_64+0x50/0x90
 ? __x64_sys_fadvise64+0x18/0x20
 ? do_syscall_64+0x5d/0x180
 ? entry_SYSCALL_64_after_hwframe+0x4b/0x53
 </TASK>
task:reader          state:R  running task     stack:0     pid:1556  tgid:1551  ppid:1      flags:0x00004006

The reproducer takes ~30 seconds, and will lead to anywhere from 1..N
threads spinning here.

Now for the kicker - this doesn't reproduce in 6.10 and onwards. There
are only a few changes here that are relevant, seemingly, and the prime
candidates are:

commit a4864671ca0bf51c8e78242951741df52c06766f
Author: Kairui Song <kasong@tencent.com>
Date:   Tue Apr 16 01:18:55 2024 +0800

    lib/xarray: introduce a new helper xas_get_order

and the followup filemap change:

commit 6758c1128ceb45d1a35298912b974eb4895b7dd9
Author: Kairui Song <kasong@tencent.com>
Date:   Tue Apr 16 01:18:56 2024 +0800

    mm/filemap: optimize filemap folio adding

and reverting those two on 6.10 hits it again almost immediately. Didn't
look into these commit, but looks like they inadvertently also fixed
this corruption issue.

-- 
Jens Axboe