Hitting WARN_ON_ONCE in i2c-designware-common.c

Hitting WARN_ON_ONCE in i2c-designware-common.c

[Jesper Dangaard Brouer]

Hi Maintainers,

I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
when booting the kernel on our Gen12 hardware.

I'm using devel kernel net-next at commit 1a9239bb425 (merge tag 
'net-next-6.15').

I assume you want this report.

Maybe it is not a critical error(?)
... looking the comment in the function:

  u32 i2c_dw_clk_rate(struct dw_i2c_dev *dev)
  {
	/*
	 * Clock is not necessary if we got LCNT/HCNT values directly from
	 * the platform code.
	 */
	if (WARN_ON_ONCE(!dev->get_clk_rate_khz))
		return 0;
	return dev->get_clk_rate_khz(dev);
  }

--Jesper

Boot error:

[   10.062651] i2c_designware AMDI0010:00: Unknown Synopsys component 
type: 0xffffffff
[   10.073312] pps_core: LinuxPPS API ver. 1 registered
[   10.073372] piix4_smbus 0000:00:14.0: SMBus Host Controller at 0xb00, 
revision 0
[   10.075433] i2c_designware AMDI0010:01: Unknown Synopsys component 
type: 0xffffffff
[   10.077727] ------------[ cut here ]------------
[   10.077728] WARNING: CPU: 15 PID: 1402 at 
drivers/i2c/busses/i2c-designware-common.c:575 i2c_dw_clk_rate+0x13/0x20 
[i2c_designware_core]
[   10.077733] Modules linked in: i2c_piix4(+) pps_core(+) i2c_smbus 
i2c_designware_platform(+) i2c_designware_core iosf_mbi i2c_core
[   10.077739] CPU: 15 UID: 0 PID: 1402 Comm: (udev-worker) Not tainted 
6.14.0-net-next2+ #24 PREEMPT(full)
[   10.077742] Hardware name: Lenovo HR355M-V3-G12/HR355M_V3_HPM, BIOS 
HR355M_V3.G.026 10/13/2023
[   10.077743] RIP: 0010:i2c_dw_clk_rate+0x13/0x20 [i2c_designware_core]
[   10.077745] Code: 5e 73 6f df 0f 1f 00 90 90 90 90 90 90 90 90 90 90 
90 90 90 90 90 90 0f 1f 44 00 00 48 8b 47 68 48 85 c0 74 05 ff e0 cc 66 
90 <0f> 0b 31 c0 e9 2f 73 6f df 0f 1f 40 00 90 90 90 90 90 90 90 90 90
[   10.077747] RSP: 0018:ffffc90021ee7ba0 EFLAGS: 00010246
[   10.077749] RAX: 0000000000000000 RBX: ffff88a056c00028 RCX: 
0000000000ffffae
[   10.077750] RDX: ffffc90021ee7bac RSI: ffffc900082490f4 RDI: 
ffff88a056c00028
[   10.077751] RBP: 000000000000012c R08: ffffffff825eb647 R09: 
0000000000000008
[   10.077752] R10: ffffc90021ee79b8 R11: 0000000000000001 R12: 
000000000000012c
[   10.077753] R13: ffff889852a61410 R14: ffff88a0532e6428 R15: 
ffff8881006db020
[   10.077754] FS:  00007f4d0f3d28c0(0000) GS:ffff88a8ac065000(0000) 
knlGS:0000000000000000
[   10.077755] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   10.077756] CR2: 00007f4d0fc75028 CR3: 0000002056f98003 CR4: 
0000000000770ef0
[   10.077757] PKRU: 55555554
[   10.077757] Call Trace:
[   10.077759]  <TASK>
[   10.077761]  ? __warn+0x85/0x150
[   10.077765]  ? i2c_dw_clk_rate+0x13/0x20 [i2c_designware_core]
[   10.077768]  ? report_bug+0x160/0x190
[   10.077772]  ? i2c_dw_clk_rate+0x13/0x20 [i2c_designware_core]
[   10.077777]  ? i2c_dw_clk_rate+0x15/0x20 [i2c_designware_core]
[   10.077779]  ? handle_bug+0x10d/0x160
[   10.077783]  ? exc_invalid_op+0x13/0x60
[   10.077785]  ? asm_exc_invalid_op+0x16/0x20
[   10.077789]  ? i2c_dw_clk_rate+0x13/0x20 [i2c_designware_core]
[   10.077791]  i2c_dw_set_timings_master+0x91/0x3c0 [i2c_designware_core]
[   10.077794]  i2c_dw_probe_master+0x5e/0x330 [i2c_designware_core]
[   10.077796]  ? pm_runtime_enable+0x59/0xe0
[   10.077800]  dw_i2c_plat_probe+0x305/0x380 [i2c_designware_platform]
[   10.077803]  platform_probe+0x3d/0x90
[   10.077807]  really_probe+0xc1/0x390
[   10.077811]  ? __pfx___driver_attach+0x10/0x10
[   10.077812]  __driver_probe_device+0x78/0x150
[   10.077814]  driver_probe_device+0x1f/0x90
[   10.077816]  __driver_attach+0xce/0x1c0
[   10.077818]  bus_for_each_dev+0x60/0xa0
[   10.077821]  bus_add_driver+0x10e/0x240
[   10.077824]  driver_register+0x55/0x100
[   10.077826]  ? __pfx_dw_i2c_init_driver+0x10/0x10 
[i2c_designware_platform]
[   10.077829]  do_one_initcall+0x40/0x210
[   10.077831]  ? security_kernel_post_read_file+0x2f/0x70
[   10.077836]  do_init_module+0x60/0x260
[   10.077840]  init_module_from_file+0x75/0xa0
[   10.077845]  idempotent_init_module+0xed/0x2c0
[   10.077848]  __x64_sys_finit_module+0x5d/0xc0
[   10.077851]  do_syscall_64+0x47/0x110
[   10.077853]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[   10.077855] RIP: 0033:0x7f4d0fae3799
[   10.077857] Code: 08 89 e8 5b 5d c3 66 2e 0f 1f 84 00 00 00 00 00 90 
48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 
05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 37 06 0d 00 f7 d8 64 89 01 48
[   10.077858] RSP: 002b:00007ffc1f421918 EFLAGS: 00000246 ORIG_RAX: 
0000000000000139
[   10.077860] RAX: ffffffffffffffda RBX: 00005606c614bcf0 RCX: 
00007f4d0fae3799
[   10.077861] RDX: 0000000000000000 RSI: 00007f4d0fc76efd RDI: 
000000000000000d
[   10.077862] RBP: 00007f4d0fc76efd R08: 0000000000000000 R09: 
00005606c6129bd0
[   10.077863] R10: 000000000000000d R11: 0000000000000246 R12: 
0000000000020000
[   10.077863] R13: 0000000000000000 R14: 00005606c614c1c0 R15: 
00007ffc1f421b50
[   10.077867]  </TASK>
[   10.077867] ---[ end trace 0000000000000000 ]---
[   10.078850] pps_core: Software ver. 5.3.6 - Copyright 2005-2007 
Rodolfo Giometti <giometti@linux.it>
[   10.087132] workqueue: work_for_cpu_fn hogged CPU for >10000us 5 
times, consider switching to WQ_UNBOUND
[   10.099417] PTP clock support registered
[   10.104491] piix4_smbus 0000:00:14.0: Using register 0x02 for SMBus 
port selection
[   10.711607] ice: Intel(R) Ethernet Connection E800 Series Linux Driver
[   10.712839] piix4_smbus 0000:00:14.0: Auxiliary SMBus Host Controller 
at 0xb20

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Andy Shevchenko]

On Tue, Apr 15, 2025 at 02:03:26PM +0200, Jesper Dangaard Brouer wrote:
> Hi Maintainers,
> 
> I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
> when booting the kernel on our Gen12 hardware.
> 
> I'm using devel kernel net-next at commit 1a9239bb425 (merge tag
> 'net-next-6.15').
> 
> I assume you want this report.
> 
> Maybe it is not a critical error(?)
> ... looking the comment in the function:

Have you forgotten to compile in the drivers/acpi/acpi_apd.c?

-- 
With Best Regards,
Andy Shevchenko

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Andy Shevchenko]

On Tue, Apr 15, 2025 at 03:17:37PM +0300, Andy Shevchenko wrote:
> On Tue, Apr 15, 2025 at 02:03:26PM +0200, Jesper Dangaard Brouer wrote:
> > Hi Maintainers,
> > 
> > I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
> > when booting the kernel on our Gen12 hardware.
> > 
> > I'm using devel kernel net-next at commit 1a9239bb425 (merge tag
> > 'net-next-6.15').
> > 
> > I assume you want this report.
> > 
> > Maybe it is not a critical error(?)
> > ... looking the comment in the function:
> 
> Have you forgotten to compile in the drivers/acpi/acpi_apd.c?

Also that driver has the missing error check in acpi_apd_setup() for
clk_register_clkdev().

-- 
With Best Regards,
Andy Shevchenko

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Jesper Dangaard Brouer]

On 15/04/2025 14.19, Andy Shevchenko wrote:
> On Tue, Apr 15, 2025 at 03:17:37PM +0300, Andy Shevchenko wrote:
>> On Tue, Apr 15, 2025 at 02:03:26PM +0200, Jesper Dangaard Brouer wrote:
>>> Hi Maintainers,
>>>
>>> I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
>>> when booting the kernel on our Gen12 hardware.
>>>
>>> I'm using devel kernel net-next at commit 1a9239bb425 (merge tag
>>> 'net-next-6.15').
>>>
>>> I assume you want this report.
>>>
>>> Maybe it is not a critical error(?)
>>> ... looking the comment in the function:
>>
>> Have you forgotten to compile in the drivers/acpi/acpi_apd.c?
> 

I have double checked that drivers/acpi/acpi_apd.o is compiled.

> Also that driver has the missing error check in acpi_apd_setup() for
> clk_register_clkdev().
> 

Are you saying I'm missing a .config option?

--Jesper

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Andy Shevchenko]

On Tue, Apr 15, 2025 at 02:30:10PM +0200, Jesper Dangaard Brouer wrote:
> On 15/04/2025 14.19, Andy Shevchenko wrote:
> > On Tue, Apr 15, 2025 at 03:17:37PM +0300, Andy Shevchenko wrote:
> > > On Tue, Apr 15, 2025 at 02:03:26PM +0200, Jesper Dangaard Brouer wrote:
> > > > 
> > > > I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
> > > > when booting the kernel on our Gen12 hardware.
> > > > 
> > > > I'm using devel kernel net-next at commit 1a9239bb425 (merge tag
> > > > 'net-next-6.15').
> > > > 
> > > > I assume you want this report.
> > > > 
> > > > Maybe it is not a critical error(?)
> > > > ... looking the comment in the function:
> > > 
> > > Have you forgotten to compile in the drivers/acpi/acpi_apd.c?
> 
> I have double checked that drivers/acpi/acpi_apd.o is compiled.

Hmm... But is the function that creates clock is compiled? And does it succeed
(see also my comment about missing error check which you probably want to
 address independently on the affection on this issue).

> > Also that driver has the missing error check in acpi_apd_setup() for
> > clk_register_clkdev().
> 
> Are you saying I'm missing a .config option?

I'm saying that please double check that the clock is provided and AFAICT it
has to be created by the above mentioned module.

-- 
With Best Regards,
Andy Shevchenko

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Jarkko Nikula]

Hi

On 4/15/25 3:03 PM, Jesper Dangaard Brouer wrote:
> Hi Maintainers,
> 
> I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
> when booting the kernel on our Gen12 hardware.
> 
> I'm using devel kernel net-next at commit 1a9239bb425 (merge tag 'net- 
> next-6.15').
> 
Is this a regression so was this working before?

Support for this platform was added back in 2016 by the commit 
e4e666ba74d4 ("i2c: designware: Add device HID for future AMD I2C 
controller").

> [   10.062651] i2c_designware AMDI0010:00: Unknown Synopsys component 
> type: 0xffffffff
> [   10.073312] pps_core: LinuxPPS API ver. 1 registered
> [   10.073372] piix4_smbus 0000:00:14.0: SMBus Host Controller at 0xb00, 
> revision 0
> [   10.075433] i2c_designware AMDI0010:01: Unknown Synopsys component 
> type: 0xffffffff

These "Unknown Synopsys component type: 0xffffffff" errors indicate IP 
is not alive. Perhaps linked to the missing clock or not powered.

Re: Hitting WARN_ON_ONCE in i2c-designware-common.c

[Jesper Dangaard Brouer]

On 15/04/2025 14.28, Jarkko Nikula wrote:
> Hi
> 
> On 4/15/25 3:03 PM, Jesper Dangaard Brouer wrote:
>> Hi Maintainers,
>>
>> I'm hitting a WARN_ON_ONCE in drivers/i2c/busses/i2c-designware-common.c
>> when booting the kernel on our Gen12 hardware.
>>
>> I'm using devel kernel net-next at commit 1a9239bb425 (merge tag 'net- 
>> next-6.15').
>>
> Is this a regression so was this working before?
> 

I'm seeing this when doing kernel development and booting net-next on
our Gen12 servers[1].  [1] https://blog.cloudflare.com/gen-12-servers/

I've not seen this WARN on our production kernels (stable 6.6 and 6.12).
This is likely due to different kernel configs.

> Support for this platform was added back in 2016 by the commit 
> e4e666ba74d4 ("i2c: designware: Add device HID for future AMD I2C 
> controller").
> 

Maybe we have just not enabled the module on our prod kernels.

>> [   10.062651] i2c_designware AMDI0010:00: Unknown Synopsys component type: 0xffffffff
>> [   10.073312] pps_core: LinuxPPS API ver. 1 registered
>> [   10.073372] piix4_smbus 0000:00:14.0: SMBus Host Controller at 0xb00, revision 0
>> [   10.075433] i2c_designware AMDI0010:01: Unknown Synopsys component type: 0xffffffff
> 
> These "Unknown Synopsys component type: 0xffffffff" errors indicate IP 
> is not alive. Perhaps linked to the missing clock or not powered.

It is possible that this hardware is special as I believe it is 
customized for us.

The dmesg output also contained:
  Hardware name: Lenovo HR355M-V3-G12/HR355M_V3_HPM
(see the extra -G12 to the SKU.)

--Jesper