Message-ID: <db1a4681-1882-4e0a-b96f-a793e8fffb56@suse.cz>
Date: Tue, 4 Mar 2025 11:26:07 +0100
From: Vlastimil Babka <vbabka@...e.cz>
To: Hannes Reinecke <hare@...e.com>, Hannes Reinecke <hare@...e.de>,
 Matthew Wilcox <willy@...radead.org>, Boris Pismenny <borisp@...dia.com>,
 John Fastabend <john.fastabend@...il.com>, Jakub Kicinski <kuba@...nel.org>
Cc: Sagi Grimberg <sagi@...mberg.me>,
 "linux-nvme@...ts.infradead.org" <linux-nvme@...ts.infradead.org>,
 "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
 linux-mm@...ck.org, Harry Yoo <harry.yoo@...cle.com>,
 "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: Kernel oops with 6.14 when enabling TLS

+Cc NETWORKING [TLS] maintainers and netdev for input, thanks.

The full error is here:
https://lore.kernel.org/all/fcfa11c6-2738-4a2e-baa8-09fa8f79cbf3@suse.de/

On 3/4/25 11:20, Hannes Reinecke wrote:
> On 3/4/25 09:18, Vlastimil Babka wrote:
>> On 3/4/25 08:58, Hannes Reinecke wrote:
>>> On 3/3/25 23:02, Vlastimil Babka wrote:
>>>> On 3/3/25 17:15, Vlastimil Babka wrote:
>>>>> On 3/3/25 16:48, Matthew Wilcox wrote:
>>>>>> You need to turn on the debugging options Vlastimil mentioned and try to
>>>>>> figure out what nvme is doing wrong.
>>>>>
>>>>> Agree, looks like some error path going wrong?
>>>>> Since there seems to be actual non-large kmalloc usage involved, another
>>>>> debug parameter that could help: CONFIG_SLUB_DEBUG=y, and boot with
>>>>> "slab_debug=FZPU,kmalloc-*"
>>>>
>>>> Also make sure you have CONFIG_DEBUG_VM please.
>>>>
>>> Here you go:
>>>
>>> [  134.506802] page: refcount:0 mapcount:0 mapping:0000000000000000
>>> index:0x0 pfn:0x101ef8
>>> [  134.509253] head: order:3 mapcount:0 entire_mapcount:0
>>> nr_pages_mapped:0 pincount:0
>>> [  134.511594] flags:
>>> 0x17ffffc0000040(head|node=0|zone=2|lastcpupid=0x1fffff)
>>> [  134.513556] page_type: f5(slab)
>>> [  134.513563] raw: 0017ffffc0000040 ffff888100041b00 ffffea0004a90810
>>> ffff8881000402f0
>>> [  134.513568] raw: 0000000000000000 00000000000a000a 00000000f5000000
>>> 0000000000000000
>>> [  134.513572] head: 0017ffffc0000040 ffff888100041b00 ffffea0004a90810
>>> ffff8881000402f0
>>> [  134.513575] head: 0000000000000000 00000000000a000a 00000000f5000000
>>> 0000000000000000
>>> [  134.513579] head: 0017ffffc0000003 ffffea000407be01 ffffffffffffffff
>>> 0000000000000000
>>> [  134.513583] head: 0000000000000008 0000000000000000 00000000ffffffff
>>> 0000000000000000
>>> [  134.513585] page dumped because: VM_BUG_ON_FOLIO(((unsigned int)
>>> folio_ref_count(folio) + 127u <= 127u))
>>> [  134.513615] ------------[ cut here ]------------
>>> [  134.529822] kernel BUG at ./include/linux/mm.h:1455!
>> 
>> Yeah, just as I suspected, folio_get() says the refcount is 0.
>> 
>>> [  134.529835] Oops: invalid opcode: 0000 [#1] PREEMPT SMP
>>> DEBUG_PAGEALLOC NOPTI
>>> [  134.529843] CPU: 0 UID: 0 PID: 274 Comm: kworker/0:1H Kdump: loaded
>>> Tainted: G            E      6.14.0-rc4-default+ #309
>>> 03b131f1ef70944969b40df9d90a283ed638556f
>>> [  134.536577] Tainted: [E]=UNSIGNED_MODULE
>>> [  134.536580] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
>>> 0.0.0 02/06/2015
>>> [  134.536583] Workqueue: nvme_tcp_wq nvme_tcp_io_work [nvme_tcp]
>>> [  134.536595] RIP: 0010:__iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.542810] Code: e8 4c 39 e0 49 0f 47 c4 48 01 45 08 48 29 45 18 e9
>>> 90 fa ff ff 48 83 ef 01 e9 7f fe ff ff 48 c7 c6 40 57 4f 82 e8 6a e2 ce
>>> ff <0f> 0b e8 43 b8 b1 ff eb c5 f7 c1 ff 0f 00 00 48 89 cf 0f 85 4f ff
>>> [  134.542816] RSP: 0018:ffffc900004579d8 EFLAGS: 00010282
>>> [  134.542821] RAX: 000000000000005c RBX: ffffc90000457a90 RCX:
>>> 0000000000000027
>>> [  134.542825] RDX: 0000000000000000 RSI: 0000000000000002 RDI:
>>> ffff88817f423748
>>> [  134.542828] RBP: ffffc90000457d60 R08: 0000000000000000 R09:
>>> 0000000000000001
>>> [  134.554485] R10: ffffc900004579c0 R11: ffffc90000457720 R12:
>>> 0000000000000000
>>> [  134.554488] R13: ffffea000407be40 R14: ffffc90000457a70 R15:
>>> ffffc90000457d60
>>> [  134.554495] FS:  0000000000000000(0000) GS:ffff88817f400000(0000)
>>> knlGS:0000000000000000
>>> [  134.554499] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>> [  134.554502] CR2: 0000556b0675b600 CR3: 0000000106bd8000 CR4:
>>> 0000000000350ef0
>>> [  134.554509] Call Trace:
>>> [  134.554512]  <TASK>
>>> [  134.554516]  ? __die_body+0x1a/0x60
>>> [  134.554525]  ? die+0x38/0x60
>>> [  134.554531]  ? do_trap+0x10f/0x120
>>> [  134.554538]  ? __iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.568839]  ? do_error_trap+0x64/0xa0
>>> [  134.568847]  ? __iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.568855]  ? exc_invalid_op+0x53/0x60
>>> [  134.572489]  ? __iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.572496]  ? asm_exc_invalid_op+0x16/0x20
>>> [  134.572512]  ? __iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.576726]  ? __iov_iter_get_pages_alloc+0x676/0x710
>>> [  134.576733]  ? srso_return_thunk+0x5/0x5f
>>> [  134.576740]  ? ___slab_alloc+0x924/0xb60
>>> [  134.580253]  ? mempool_alloc_noprof+0x41/0x190
>>> [  134.580262]  ? tls_get_rec+0x3d/0x1b0 [tls
>>> 47f199c97f69357468c91efdbba24395e9dbfa77]
>>> [  134.580282]  iov_iter_get_pages2+0x19/0x30
>> 
>> Presumably that's __iov_iter_get_pages_alloc() doing get_page() either in
>> the " if (iov_iter_is_bvec(i)) " branch or via iter_folioq_get_pages()?
>> 
> Looks like it.
> 
>> Which doesn't work for a sub-size kmalloc() from a slab folio, which after
>> the frozen refcount conversion no longer supports get_page().
>> 
>> The question is if this is a mistake specific for this path that's easy to
>> fix or there are more paths that do this. At the very least the pinning of
>> page through a kmalloc() allocation from it is useless - the object itself
>> has to be kfree()'d and that would never happen through a put_page()
>> reaching zero.
>> 
> Looks like a specific mistake.
> tls_sw is the only user of sk_msg_zerocopy_from_iter()
> (which is calling into __iov_iter_get_pages_alloc()).
> 
> And, more to the point, tls_sw messes up iov pacing coming in from
> the upper layers.
> So even if the upper layers send individual iovs (where each iov might
> contain different allocation types), tls_sw is packing them together 
> into full records. So it might end up with iovs having _different_ 
> allocations.
> Which would explain why we only see it with TLS, but not with normal
> connections.
> 
> Or so my reasoning goes. Not sure if that's correct.
> 
> So I'd be happy with an 'easy' fix for now. Obviously :-)
> 
> Cheers,
> 
> Hannes

Powered by blists - more mailing lists