OCZ TRION100 480GB

[HaQue]

I have an OCZ TRION100 here. Anyone worked on these yet? Controller is Toshiba TC58NC1000GSB-00 but It has been reported the controller seems to be a Phison PS3110-10 (S10) under the hood. http://www.anandtech.com/show/9408/ocz-trion-100-240gb-480gb-960gb-review

I have a brand new drive to create a dump with patterns to try and crack this controller.

NANDs are 4 physical chips... Toshiba's A19nm 128Gbit TLC toggle NAND TH58TET0UDKBAEF (reported as 8 layers/NANDs each). BGA 272-ball chips.

Dumping them is a dang lot of wiring to adapter

wonder if anyone has any insight, if the xor is known and if mix has been worked out, or it is a total reverse engineer job from scratch..

thanks

[datahdd116]

you see this post https://forum.hddguru.com/viewtopic.php ... t=vertex+2

[HaQue]

you see this post https://forum.hddguru.com/viewtopic.php ... t=vertex+2

totally different controller, and pretty much everything

[solver_001]

Hello. I also have a Toshiba Q300 that unfortunately is dead. Tried power cycling but without any success. Should i try a re-flow? Thanks.

[HaQue]

honestly don't know. Haven't seen any evidence that anyone has worked out the actual issue with these yet, only guesswork. I suppose you could try, do one chip, try it, do the next chip, try it etc... Just don't overheat the chips or blast them in one spot for too long. This is a pretty risky procedure

[Flashguys]

I have a PS3110-10 SSD DUMP but have many BAD BITE Because ECC map is 40% red maybe the nand parameter is Incorrect .

[solver_001]

I would like to identify if the problem is a soldering one or something is wrong with the fw. Just do not know how. I have checked with mhdd and got the response that Drive is not ready. What do you think?

[HaQue]

Lots of DR companies would also like to know how.

Problem is, we don't get "Workshop manuals" from the manufacturers, so physical inspection and repair is hard. We don't the many years of experience in firmware development like the manufacturers, so reverse engineering firmware issues is hard. We don't have 10's to 100's of hours to figure out all the different configurations of flash controllers+nand chips, different data+service data layouts, encryption schemes, Whitening schemes, data de-duplication/write amplification/ JuJu such as in SandForce chips/ "next big thing in SSD improvements"

The complexity and different possible problems an SSD might have is SEVERELY underestimated. Just because there is a small controller and a stack of dump NAND chips, I think make people believe it is just a tiny thing that has gone wrong. maybe it is but finding it when you know nothing about electronics, filesystems and firmware design makes it quite a challenge.

All this takes some SERIOUS R&D efforts

[jermy]

I have a PS3110-10 SSD DUMP but have many BAD BITE Because ECC map is 40% red maybe the nand parameter is Incorrect .

could be, or maybe you overheat the chip.
if parameters IS correct, try reading with lower voltage.

Whooppss, just noticed that you are from FE
Subject: rusolut reader or PC-3000 Flash

We analyze it and release updates of our software weekly.

[HaQue]

I think he was quoting the website, I don't think he is from FE. http://www.flash-extractor.com/

[jermy]

I apologize if I misunderstood

[RDane]

Hi
OCZ 240 TRN100-25SAT3 240G

NAND ID 983AA89276 TLC DDR

[DR-Kiev]

you see this post https://forum.hddguru.com/viewtopic.php ... t=vertex+2

totally different controller, and pretty much everything

I disagree here with you. There is an opinion that Phison 3110 the same core (Xtensa) based as Sandforce . I would wish it is not. Very soon it will be clarified .

[HaQue]

Are you talking about the Xtensa DPU from Cadence? As I understand it, Phison licences a Data Plane Unit from Cadence, which would make up a part of their processor architecture, but not be a whole core. Sandforce also licence stuff from Cadence, but I can't see any link between Sandforce and Phison in this instance. the S10 looks to me to be wholly developed by Phison. With all the buzzwords around how it works though, I don't think an easy controller to recover from.

Not clear if Xtensa is used in 3110, but I think problems lie elsewhere.

https://ip.cadence.com/about/customer-profiles
https://ip.cadence.com/news/409/330/Phison-Licenses-Tensilica-s-Dataplane-Processor-DPU-for-NAND-Flash-Memory-Controllers-and-SSD-Applications
https://ip.cadence.com/news/257/330/SandForce-Licenses-Tensilica-s-Diamond-Standard-108Mini-for-Storage-Controller-Chipsets.htm

http://www.phison.com/English/newProductView.asp?ID=246&SortID=63