Re: Internal error: Oops: 0000000096000044 [#11] SMP

[Dave Jiang <dave.jiang@intel.com>]

On 5/21/25 1:38 PM, Itaru Kitayama wrote:
> Dave
> 
>> On May 22, 2025, at 0:31, Dave Jiang <dave.jiang@intel.com> wrote:
>>
>>
>>
>> On 5/21/25 1:39 AM, Itaru Kitayama wrote:
>>> Hi,
>>> On arm64/virt QEMU, the cxl/next (as of today) kernel prints out Internal errors:
>>>
>>> [   80.968299] [     T48] Internal error: Oops: 0000000096000044 [#11]
>>> SMP
>>> [   80.989250] [     T48] Modules linked in: cxl_mock_mem(O) cfg80211
>>> rfkill cxl_test(O) cxl_mem(O) cxl_pmem(O) cxl_acpi(O) cxl_port(O)
>>> cxl_mock(O) libnvdimm encrypted_keys trusted caam_jr caam asn1_encoder
>>> caamhash_desc caamalg_desc error crypto_engine authenc libdes fuse drm
>>> backlight ip_tables x_tables sm3_ce sm3 sha3_ce sha512_ce sha512_arm64
>>> cxl_core(O) fwctl btrfs blake2b_generic xor xor_neon raid6_pq
>>> zstd_compress ipv6
>>> [   80.992210] [     T48] CPU: 1 UID: 0 PID: 48 Comm: kworker/u8:2
>>> Tainted: G      D    O        6.15.0-rc4-00040-g128ad8fa385b #40 PREEMPT
>>> [   80.992791] [     T48] Tainted: [D]=DIE, [O]=OOT_MODULE
>>> [   80.993039] [     T48] Hardware name: QEMU QEMU Virtual Machine, BIOS
>>> 2025.02-3ubuntu2 04/04/2025
>>> [   80.993400] [     T48] Workqueue: async async_run_entry_fn
>>> [   80.994718] [     T48] pstate: 61402005 (nZCv daif +PAN -UAO -TCO
>>> +DIT -SSBS BTYPE=--)
>>> [   80.995329] [     T48] pc : cxl_mock_mbox_send+0xec/0x12c0
>>> [cxl_mock_mem]
>>
>> Can you do this in your kernel tree?
>> ./scripts/faddr2line tools/testing/cxl/test/cxl_mock_mem.ko cxl_mock_mbox_send+0xec/0x12c0
> 
> realm@machine-1:~/projects/cxl$ ./scripts/faddr2line tools/testing/cxl/test/cxl_mock_mem.ko cxl_mock_mbox_send+0xec/0x12c0
> cxl_mock_mbox_send+0xec/0x12c0:
> mock_get_event at /home/realm/projects/cxl/tools/testing/cxl/test/mem.c:277
> (inlined by) cxl_mock_mbox_send at /home/realm/projects/cxl/tools/testing/cxl/test/mem.c:1571
> 
>>
>> I've not see this issue on x86 running cxl/next. How consistently can you reproduce this? If it's every time, is it possible for you to do a git bisect on the kernel and see which commit causes this please? Thanks!
> 
> Fairly reliably (100% of the boot time, and cx/fixes did not change this BTW, which branch is seen as stable for you folks?), yes, I should try git bisect.

The current cxl/next is based on 6.15-rc4, which should have everything that was in cxl/fixes. And you do not see this with 6.14-final? git bisect would be very helpful. Thank you!

> 
> Itaru.
> 
>>
>> DJ
>>
>>> [   80.995691] [     T48] lr : cxl_internal_send_cmd+0x40/0x104
>>> [cxl_core]
>>> [   80.996189] [     T48] sp : ffff800080d0b9f0
>>> [   80.996380] [     T48] x29: ffff800080d0ba70 x28: fff0000008dd2410
>>> x27: fff00000088fb390
>>> [   80.996714] [     T48] x26: ffff800080d0bb07 x25: 0000000000000100
>>> x24: 0000000000000003
>>> [   80.997135] [     T48] x23: 0000000000000020 x22: fff0000008dd2410
>>> x21: 0000000000000002
>>> [   80.998119] [     T48] x20: fff00000088fb080 x19: ffff800080d0bb08
>>> x18: 00000000ffffffff
>>> [   80.998419] [     T48] x17: 0000000000000000 x16: ffffa8d169128748
>>> x15: ffff800080d0b5ad
>>> [   80.999243] [     T48] x14: ffff800080d0b400 x13: ffff800080d0b5b8
>>> x12: fff000006f7a0000
>>> [   81.000519] [     T48] x11: 0000000000000058 x10: 0000000000000018 x9
>>> : fff000006f7a0000
>>> [   81.001337] [     T48] x8 : ffff800080d0bb48 x7 : fff0000074fa0000 x6
>>> : fff0000074fa0000
>>> [   81.002497] [     T48] x5 : fff000007f937508 x4 : 0000000000000001 x3
>>> : 0000000000001000
>>> [   81.003993] [     T48] x2 : 0000000000001000 x1 : 0000000000000000 x0
>>> : 0000000000000088
>>> [   81.004223] [     T48] Call trace:
>>> [   81.004795] [     T48]  cxl_mock_mbox_send+0xec/0x12c0 [cxl_mock_mem]
>>> (P)
>>> [   81.005136] [     T48]  cxl_internal_send_cmd+0x40/0x104 [cxl_core]
>>> [   81.005520] [     T48]  cxl_mem_get_records_log+0xbc/0x198 [cxl_core]
>>> [   81.006042] [     T48]  cxl_mem_get_event_records+0xb0/0xc0
>>> [cxl_core]
>>> [   81.006246] [     T48]  cxl_mock_mem_probe+0x568/0x6f0 [cxl_mock_mem]
>>> [   81.006417] [     T48]  platform_probe+0x68/0xd8
>>> [   81.008340] [     T48]  really_probe+0xc0/0x39c
>>> [   81.008885] [     T48]  __driver_probe_device+0xd0/0x14c
>>> [   81.009539] [     T48]  driver_probe_device+0x3c/0x120
>>> [   81.010239] [     T48]  __driver_attach_async_helper+0x50/0xec
>>> [   81.011130] [     T48]  async_run_entry_fn+0x34/0x14c
>>> [   81.011276] [     T48]  process_one_work+0x148/0x284
>>> [   81.011420] [     T48]  worker_thread+0x2c4/0x3e0
>>> [   81.011552] [     T48]  kthread+0x12c/0x204
>>> [   81.011693] [     T48]  ret_from_fork+0x10/0x20
>>> [   81.011840] [     T48] Code: 54001b28 a90c6bf9 52801100 f9400a61
>>> (a9007c3f)
>>> [   81.013772] [     T48] ---[ end trace 0000000000000000 ]---
>>>
>>> How serious is this?
>>>
>>> Thanks,
>>> Itaru.
> 
>