Re: [PATCH v3 3/3] serial: sb1250-duart: Convert to use a platform device

[Florian Fainelli <f.fainelli@gmail.com>]

On 11/3/26 13:42, Florian Fainelli wrote:
> On 8/6/26 03:54, Maciej W. Rozycki wrote:
>> Prevent a crash from happening as the first serial port is initialised:
>>
>>    pata-swarm: PATA interface at GenBus slot 4
>>    workingset: timestamp_bits=62 max_order=18 bucket_order=0
>>    Block layer SCSI generic (bsg) driver version 0.4 loaded (major 253)
>>    CPU 1 Unable to handle kernel paging request at virtual address 
>> 0000000000000208, epc == ffffffff8067f8f8, ra == ffffffff80666330
>>    Oops[#1]:
>>    CPU: 1 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.19.0-dirty #27 NONE
>>    $ 0   : 0000000000000000 0000000014001fe0 0000000000000020 
>> ffffffff80666130
>>    $ 4   : 0000000000000000 a800000100e6f118 ffffffff8112cbc0 
>> 0000000000000000
>>    $ 8   : 0000000000000002 0000000000000000 0000000000000000 
>> 0000000001a80000
>>    $12   : 0000000000000000 ffffffff809fd488 000000004ddf14dd 
>> ffffffff00000000
>>    $16   : a800000100e6f000 0000000000000000 ffffffff8112c1d0 
>> a800000100e6f000
>>    $20   : 0000000000000000 00000000000004d0 0000000000000004 
>> ffffffff8112c1d0
>>    $24   : 0000000000000001 0000000000000003
>>    $28   : a80000010007c000 a80000010007fcb0 00000000000000ef 
>> ffffffff80666330
>>    Hi    : fffffffffffffdb9
>>    Lo    : 0000000000000035
>>    epc   : ffffffff8067f8f8 __dev_fwnode+0x0/0x8
>>    ra    : ffffffff80666330 serial_base_ctrl_add+0xb8/0x180
>>    Status: 14001fe3    KX SX UX KERNEL EXL IE
>>    Cause : 80800008 (ExcCode 02)
>>    BadVA : 0000000000000208
>>    PrId  : 03040102 (SiByte SB1)
>>    Process swapper/0 (pid: 1, threadinfo=(____ptrval____), 
>> task=(____ptrval____), tls=0000000000000000)
>>    Stack : 0000000000000000 ffffffff80cd5178 ffffffff80cd0000 
>> ffffffff8112c1c8
>>            0000000000000260 ffffffff806655c4 a800000100275bc0 
>> ffffffff8064ac88
>>            004000408112c000 0000000000000002 0000000000000000 
>> ffffffff801965d0
>>            a800000100786ba0 ffffffff80cd5178 a800000100786ba0 
>> 0000000000000004
>>            a800000100275bc0 0000000000000000 0000000000000000 
>> ffffffff80cd5178
>>            0000000000000000 ffffffff8112c1c8 0000000000000260 
>> 00000000000004d0
>>            0000000000000004 ffffffff80bf0000 00000000000000ef 
>> ffffffff80d171dc
>>            ffffffff80d17120 ffffffff80d25658 0000000000000000 
>> ffffffff80d50000
>>            ffffffff80d2f928 ffffffff80d50000 ffffffff80d25698 
>> ffffffff80cfcecc
>>            00ffffff80b84428 0000000000000000 0000000000000006 
>> 0000000000000006
>>            ...
>>    Call Trace:
>>    [<ffffffff8067f8f8>] __dev_fwnode+0x0/0x8
>>    [<ffffffff80666330>] serial_base_ctrl_add+0xb8/0x180
>>    [<ffffffff806655c4>] serial_core_register_port+0x174/0x8e0
>>    [<ffffffff80d171dc>] sbd_init+0xbc/0xf4
>>    [<ffffffff80cfcecc>] do_one_initcall+0x64/0x150
>>    [<ffffffff80cfd284>] kernel_init_freeable+0x25c/0x30c
>>    [<ffffffff809ff6e4>] kernel_init+0x24/0x118
>>    [<ffffffff80112d20>] ret_from_kernel_thread+0x14/0x1c
>>    Code: 67bd0010  03e00008  2402ffea <03e00008> dc820208  67bdffa0  
>> ffbe0050  ffb70048  ffb60040
>>    ---[ end trace 0000000000000000 ]---
>>
>> -- where a pointer is dereferenced that has been derived from a null
>> pointer to the port's parent device.
>>
>> Since no device is available with legacy probing and it's not anymore a
>> preferable way to discover devices anyway, switch the driver to using a
>> platform device and use it as the port's parent device.
>>
>> Use platform_driver_probe() because SB1250 DUART devices are embedded
>> onchip the SoC and therefore not that straightforward to remove.
>>
>> An unfortunate consequence of the switch to a platform device is we now
>> hand the console over from the bootconsole much later in the bootstrap.
>> The CFE console handler appears good enough though to work so late and
>> in particular with interrupts enabled.
>>
>> Conversely, only starting the console device so late lets the port
>> initialisation code always use fsleep() for transmitter draining, so
>> drop the other variant since it is now dead code.
>>
>> Since there is one way only remaining to reach sbd_init_port() now, drop
>> the port initialisation marker as no longer needed and go through the
>> channel resets unconditionally.
>>
>> Fixes: 84a9582fd203 ("serial: core: Start managing serial controllers 
>> to enable runtime PM")
>> Signed-off-by: Maciej W. Rozycki <macro@orcam.me.uk>
>> Cc: stable@vger.kernel.org # needs to use .remove_new for <= 6.10
> 
> On BCM91250E (Sentosa):
> 
> Tested-by: Florian Fainelli <f.fainelli@gmail.com>

Could we apply Maciej's sb1250-duart patches because this is preventing 
the use of those systems while everything else works just fine.

Thanks!
-- 
Florian