[ale] Cloning laptop drive didn't work

[Alex Carver]

Finally had a chance to try the kernel options but no luck. I get 
repeating errors of the form:
I/O error, dev sda, sector 0 op 0x0:(READ) flags 0x80700 phys_seg 1 prio 
class 2
Then these repeat several times before the I/O error above repeats:
ata1.00: status { DRDY DRQ }
ata1.00: exception Emask 0x0 SAct 0x0 Serr 0x0 action 0x6 frozen
ata1.00: failed command: READ MULTIPLE
ata1.00 cmd c4/00:08:00:00:00/00:00:00:00:00/e0 tag 0 pio 4096 in
         res 58/00:01:00:00:00/00:00:00:00:00/e0 Emask 0x2 (HSM violation)

(everything scrolled off screen too fast so I couldn't get the initial logs)

Eventually it gives up.

Dropping back to the original kernel settings the errors are different:

This one goes slower so these are the very first messages:
ata1.00: Read log 0x00 page 0x00 failed, Emask 0x400
ata1.00: revalidation failed (errno=-5)
ata1: SRST failed (errno=-16)
ata1: SRST failed (errno=-16)
ata1: SRST failed (errno=-16)
ata1: SRST failed (errno=-16)
ata1: reset failed, giving up


On 2024-04-28 21:26, Alex Carver via Ale wrote:
> This is in a laptop so there's only one IDE drive on the port, single 
> drive jumper same as the existing HDD.
> 
> The very first error encountered with the SDD in place is ata1.00 
> revalidation failed. This happens when the system is switching from 
> initramfs to mounting the root filesystem so we haven't even gotten to 
> anything more interesting yet where I can even start with loading things.
> 
> According to the boot logs it discovers that there's only a 40 wire 
> connection so the system is limited to UDMA/33. The chipset according to 
> lspci is a ULi Electronics M5229 rev 03.
> 
> I removed the drive at the moment so I'll put the SSD back in and try 
> the boot flags.
> 
> On 2024-04-28 17:32, Solomon Peachy via Ale wrote:
>> On Sun, Apr 28, 2024 at 11:11:01AM -0700, Alex Carver via Ale wrote:
>>> When I boot using the adapter in the IDE bay I get a lot of ata1.00
>>> revalidation failed messages plus many other ata errors. All I did 
>>> was move
>>> the drive out of the IDE bay and onto the USB cable and it worked.
>>>
>>> Now I'm completely lost why this worked.
>>
>> The short version is that the USB adapter is hiding the messy details of
>> ATA behind a USB Mass Storage API (and a much slower interface speed)
>> whereas when it's directly attached something is not as compatible as it
>> claims to be.
>>
>> A few questions:
>>
>>   0) What is the type of ATA controller / chipset?  Intel PIIX or
>>      something else?  (Being PATA implies it's something relatively 
>> ancient..)
>>   1) Do you know exactly what the first error is?
>>   2) How soon do you see these errors; during early bootup or only after
>>      the rootfs is remounted read/write and more interesting stuff starts
>>      happening?  (ie does going into single-user mode make a difference?)
>>   3) Is the adapter jumpered correctly for master/slave and plugged into
>>      the appropriate place on an 80-pin cable?  (These SSDs transfer data
>>      a _lot_ faster than any PATA drive ever dream of...)
>>
>> I think it would be informative to tell Linux to dial back what it's
>> trying to do.. Try adding 'libata.force=noncq libata.dma=0' to the
>> kernel cmdline and see what happens?
>>
>> (I've been trying to hunt down a sorta-similar problem -- An mSATA SSD
>>   plugged into a PATA adapter... on an 6th-gen iPod.  Same setup works
>>   fine on a 5th-gen, which is usually far more finicky..  Ah, the joys of
>>   baremetal hackery and starting at the ATA specs until I go 
>> crosseyed...)
>>
>>   - Solomon
>>
>>
>> _______________________________________________
>> Ale mailing list
>> Ale at ale.org
>> https://mail.ale.org/mailman/listinfo/ale
>> See JOBS, ANNOUNCE and SCHOOLS lists at
>> http://mail.ale.org/mailman/listinfo
> 
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> https://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo