ST1000LM035 - no ID after ROM swap

[andy17d]

I got ST1000LM035 with a burnt PCB.
I changed the ROM and the disk goes to DRD state. However, it does not report an ID.
Does this model have SED?

Terminal show:

Code:

Boot 0x80M
QB
Rst 0x80MSrv DETCR init 0x0000
FTU
PrgF
(P) SATA Reset

RAW OFF
PASS
(DOS Table) Worst Count: 00000000 At SU: FFFFFFFF NT: 00000000 OT: 00000000
MSP213
(MC POR Duration): 000000004F
Intel 6G->3G
Send Status: COMRESET seen
CSpd= 3Gbps
DSC
FAFH_TRIGGER_POWER_UP

[pepe]

i guess you get 0x74 in status reg, don't you?
pepe

[andy17d]

0x74 in status reg

Where is it?

[pepe]

pc3k status reg
i meant 0x71, sorry.
bsy drd dwf dsc drq crr idx err
0 1 1 1 0 0 0 1
after any ata command.

[pepe]

Code:

bsy drd dwf dsc drq crr idx err
  0  1   1   1   0   0   0   1

[andy17d]

After any operation my drive status is:

Code:

bsy drd dwf dsc drq crr idx err ... abr
  0  1   1   1   0   0   0   1 ...   1


[pepe]

yes, that's because device fault is set. And it is set because of donor pcb.
I usually clear it manually or patch the fw so that it does not set it. There is some way in pc3k to do it automatically, but i never used it and don't know how it works or if it works at all.
pepe

[andy17d]

Is the drive is recognizing that I installed another PCB (from the donor) to it despite I've swap ROM?

Do I understand correctly that this PCB has some extra register stored outside the ROM chip?

[pepe]

I did not research the cause of this problem, just reseted it in terminal and moved on (or patched the fw so that it does not set that flag whatever the problem is). Perhaps it is related to SED and cpu or the log in flash.

pepe

[andy17d]

Thank you Pepe for informations.

[northwind]

There is some way in pc3k to do it automatically, but i never used it and don't know how it works or if it works at all.

It usually works fairly good.
If it doesn't then of course there is manual way to bypass the problem.

[sin]

Hi!
3k has clear fault state in RAM. Pretty cool function.

I would love to understand what is the philosophy behind manually clearing it.

--

PS: I am in love with 3k now.

[pepe]

the philosophy?
it could be described as follows:
I need to know far more than pc3k lets users know so i researched a lot of stuff and i usually apply a whole bunch of patches to firmwares, so it does not take a lot to do this one too. I was doing these well before PC3k offered that option, so i just do it my way.
Perhaps i should attend to a training to get an idea of it's new features

pepe

[sin]

Heheh... Philosophy here would mean the concept.
You are one of the greatest personalities here who has indepth knowledge about drives.
I am just 6 years old in a cocoon and have so many things to learn/self learn.
--

[sin]

Seeking some education from you, when you work on developing your own patches, what do all do you have to deal with
example
1)obfuscation
2)compression/decompression etc
3)encryption with certain areas?
4)signature signing
5)checksums
..
Is it al lof it?
--

[pepe]

I don't think i can be of a huge help in your education, as obviously i can't just disclose the most interesting parts of it. The OS of the drives is quite complex even without additional obfuscation and encryption. But to provide some level of answer, yes, i need to deal with compression, signature validation, checksums, then i can start off to achieve my main goal: changing the behaviours i want to change.

pepe

[sin]

Dude you are God
Inspirational stuff... <3

--

[sin]

My last question before I stop bugging you further
How much time does it go in just finding the main() while rev engineering it
Also, once you rev eng it, how much time do you generally invest in RnD to locate and modify the behavior.


I have had a very strange view by some body....they say most of the tool mfgs bribe out the factories/employees of drive mfgs to extract the source codes and then make their way around.

Disassembly + Decompiling (with loads of other things that we discussed above ) if far difficult than just procuring the C version of the same..
is this true? (bribe to get C-source)

Just curious.
--

[pepe]

Dude you are God

please restrain yourself from such statements

Of course, if one can put his hands on source code, it can be very useful, provided he has comprehension about what he is looking at. I have source code for 1-2 pre-F3 families, well, i can tell you it is not a lot easier to read than reversing it coz it is split into hundreds of source files and a lot of conditional compilation which make the code hard to read. Of course it is helpful if one knows how to use it.
The problem is far more complex than finding main(). I've been doing it for more than 10 years, so i have some insight and find basic things quickly, however new drives come with a lot more complex feature set than older ones, so there's challenge for every day.

pepe

[sin]

Bravo.

Sure i would maintain the decorum and restrain it....however, you still are...
all of the embedded folks here...
Gods