Re: Prestera driver fail to probe twice

["Köry Maincent" <kory.maincent@bootlin.com>]

Hello Andrew,

On Thu, 8 Feb 2024 00:32:43 +0100
Andrew Lunn <andrew@lunn.ch> wrote:

> On Tue, Feb 06, 2024 at 04:54:06PM +0100, Köry Maincent wrote:
> > Hello,
> > 
> > It seems the prestera driver has never been tested as a module or in a
> > probe defer case:  
> 
> Hi Köry
> 
> Could you hack a -EPROBE_DEFER failure? If you can show that does not
> work, the driver is clearly broken because phylink could return that.

That is how I have noticed the issue. I was trying to test my PSE pd692x0
driver as a module that prestera depends on (I will drop the PoE v3 soon).
The PSE core is returning -EPROBE_DEFER in case we can not find the PSE
provider.

Here is a boot dmesg for example:

[    1.898897] Prestera DX 0000:01:00.0: Loading mrvl/prestera/mvsw_prestera_fw-v4.1.img ...
[    1.907155] Prestera DX 0000:01:00.0: FW version '4.1.0'
[    5.535427] Prestera DX 0000:01:00.0: Prestera FW is ready
[   13.458823] Prestera DX 0000:01:00.0: using random base mac address
[   13.596594] prestera_port_sfp_bind : 403 -EPROBE_DEFER (The hack to get the PSE is in this function)
[   13.632517] ahci f2540000.sata: supply ahci not found, using dummy regulator
[   13.639759] ahci f2540000.sata: supply phy not found, using dummy regulator
[   13.646846] platform f2540000.sata:sata-port@0: supply target not found, using dummy regulator
[   13.655866] platform f2540000.sata:sata-port@1: supply target not found, using dummy regulator
[   13.666105] ahci f2540000.sata: masking port_map 0x3 -> 0x3
[   13.671960] ahci f2540000.sata: AHCI 0001.0000 32 slots 2 ports 6 Gbps 0x3 impl platform mode
[   13.680598] ahci f2540000.sata: flags: 64bit ncq sntf led only pmp fbs pio slum part sxs 
[   13.690393] scsi host1: ahci
[   13.693744] scsi host2: ahci
[   13.696789] ata1: SATA max UDMA/133 mmio [mem 0xf2540000-0xf256ffff] port 0x100 irq 40 lpm-pol 0
[   13.705640] ata2: SATA max UDMA/133 mmio [mem 0xf2540000-0xf256ffff] port 0x180 irq 40 lpm-pol 0
[   13.787389] mvpp2 f2000000.ethernet: using 8 per-cpu buffers
[   13.813450] mvpp2 f2000000.ethernet eth0: Using random mac address a2:00:a8:8c:7f:ca
[   13.862023] i2c i2c-2: Added multiplexed i2c bus 3
[   13.867101] i2c i2c-2: Added multiplexed i2c bus 4
[   13.872104] i2c i2c-2: Added multiplexed i2c bus 5
[   13.877181] i2c i2c-2: Added multiplexed i2c bus 6
[   13.882034] i2c-mux-gpio i2c-mux: 4 port mux on mv64xxx_i2c adapter adapter
[   14.030349] ata2: SATA link down (SStatus 0 SControl 300)
[   14.195606] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[   14.201982] ata1.00: ATA-9: M.2 (S42) 3TE7, S20730A, max UDMA/133
[   14.208142] ata1.00: 53742528 sectors, multi 16: LBA48 NCQ (depth 32)
[   14.214832] ata1.00: configured for UDMA/133
[   14.219477] scsi 1:0:0:0: Direct-Access     ATA      M.2 (S42) 3TE7   30A  PQ: 0 ANSI: 5
[   14.228723] sd 1:0:0:0: [sdb] 53742528 512-byte logical blocks: (27.5 GB/25.6 GiB)
[   14.236465] sd 1:0:0:0: [sdb] Write Protect is off
[   14.241341] sd 1:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
[   14.250491] sd 1:0:0:0: [sdb] Preferred minimum I/O size 512 bytes
[   14.259197]  sdb: sdb1 sdb2 sdb3 sdb4
[   14.263962] sd 1:0:0:0: [sdb] Attached SCSI removable disk
[   18.896377] Prestera DX 0000:01:00.0: waiting for FW loader is timed out
[   18.903242] Prestera DX: probe of 0000:01:00.0 failed with error -110


Regards,
-- 
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com