On 12/8/22 03:27, marius@xxxxxxxxxxxxxx wrote: > December 6, 2022 8:08 AM, "Damien Le Moal" <damien.lemoal@xxxxxxxxxxxxxxxxxx> wrote: > >> By the way, what is the maximum speed of you pmp box ? Is it really 1.5 >> Gbps (sata gen 1) ? > > No. It's 3 Gbps. Hmmm... OK. Given that when it works (with the revert), the speed reduction is to 1.5 Gbps. So it seems that the revert changes the error return into checking the lowest speed because we do not already have a previous speed... See below. > >> Without reverting the above patch, could you please try the following patch ? > > I did more tests, with and without the patch, with and without the commit reverted. > All logs start when eSATA cable is connected. I didn't include boot logs. > [...] > [ 153.028797][ T30] resume, after do loop > [ 153.088796][ T30] debounce, SCR=0x113 > [ 153.148796][ T30] debounce, SCR=0x113 > [ 153.208796][ T30] debounce, SCR=0x113 > [ 153.268796][ T30] debounce, SCR=0x113 > [ 153.328796][ T30] debounce, SCR=0x113 > [ 153.388795][ T30] debounce, SCR=0x113 > [ 153.448796][ T30] debounce, SCR=0x113 > [ 153.508796][ T30] debounce, SCR=0x113 > [ 153.568796][ T30] debounce, SCR=0x113 > [ 153.572654][ T30] resume, return at end of function > [ 153.576511][ T30] hardreset, Offline=>Online > [ 153.586069][ T30] hardreset, sata_pmp_supported, before check_ready > [ 153.758800][ T30] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) > [ 153.772539][ T30] ata2.15: Port Multiplier 1.2, 0x197b:0x0325 r193, 8 ports, feat 0xf/0x1f > > So, is this a quirk of the mvebu controller or the Areca DAS? BTW, Areca uses some mvebu SOC too. > I'm going to try the Areca DAS with a few other controllers. I want to see what happens. Can you try with no patches and no revert adding the following options to your kernel command line ? libata.force=2:3Gbps and libata.force=2:1.5Gbps One at a time. Do not try them together :) This will force the initial link speed to be set for ata2 port. > > Marius Dinu > > -- Damien Le Moal Western Digital Research
