DEV9: various fixes for ATA Identify

[AKuHAK]

UDMA mode now set correctly
MDMA0 mode now set correctly
fixed user addressable sectors limit

Description of Changes

In continuation for #10427 as hdd size still wasnt set properly everywhere.

Rationale behind Changes

Fix critical values in ATA Identify response.

Suggested Testing Steps

Can be tested with https://github.com/ps2homebrew/PS2HDDTester/releases . After changes it can properly run UDMA tests and reports correct disk size for large disks:

Below attached ATA Identify response from one of the Sony original PS2 drives for the research purposes.
HDD_IDentify.bin.zip

[AKuHAK]

TBH whole HDDInfo should be rewritten in accordance to the provided actual hdd response and in accordance to the ATA-7 specs:
d1532v1r4b-ATA-ATAPI-7-1.pdf.zip
My vision is that all Obsolete and Retired values does not need any calculation or explanation and can be just filled with fixed values without attempts to calculate them. This will simplify the code readability a lot too. I can do that, but I am not sure if team will accept that (@TheLastRar ? )

[TheLastRar]

TBH whole HDDInfo should be rewritten in accordance to the provided actual hdd response and in accordance to the ATA-7 specs: d1532v1r4b-ATA-ATAPI-7-1.pdf.zip My vision is that all Obsolete and Retired values does not need any calculation or explanation and can be just filled with fixed values without attempts to calculate them. This will simplify the code readability a lot too. I can do that, but I am not sure if team will accept that (@TheLastRar ? )

Looking at your provided HDD Identity dump, the original HDD reports more obsolete/retired fields then we do.

As for providing fixed values. you can where it makes sense to (i.e. fixed values for default CHS fields as we are always >8GB)

As far as I can see, we aren't missing anything required for ATAPI-7, And comparing our Identity to the dump, I don't think we aren't too far off that, The most significant thing to me seems to be feature support.

[AKuHAK]

As far as I can see, we aren't missing anything required for ATAPI-7

Yes, currently pcsx2 correctly emulates everything (just MDMA modes doesnt work properly, cant figure out why). What I would like to say that the codebase is overcomplicated. CHS and sdma modes are retired for example.

Looking at your provided HDD Identity dump, the original HDD reports more obsolete/retired fields then we do.

Yes, because HDD itself is manufactured with ATAPI-6 standard (bit 6 in word 80), while ps2 itself supports ATAPI-7.

[TheLastRar]

Yes, currently pcsx2 correctly emulates everything (just MDMA modes doesnt work properly, cant figure out why).

Found the issue, I can address that in a PR once this has been merged. I'll also remove SDMA support as original hardware lacks it

What I would like to say that the codebase is overcomplicated. CHS and sdma modes are retired for example.

I somewhat doubt that having CHS and SDMA support contribute much to code complexity here

I am hesitant to remove things that original hardware support, however, we don't implement read/writes in CHS mode and I've yet to see an issue report about anything needing it.

If you are dead set on removing CHS support then you can propose that as a separate standalone pr

[AKuHAK]

Found the issue, I can address that in a PR once this has been merged. I'll also remove SDMA support as original hardware lacks it

Nice

[TheLastRar]

Noted Codacy issues are pre-existing

[refractionpcsx2]

Nice work folks