Crucial V4
October 21st, 2014, 5:15
Hello,
I have a broken crusial V4. The controller is a PHISON PS3105-S5-1. I unsoldered and dumped one chip to see if I could do something.
So it doesn't seem encrypted but i'm having trouble getting the xor.
Can someone confirm (or not) that the xor is [ID=40: BLK=256 PG=8640 (1074-1070x7)] in PC3k Flash. It seems to work on the first two sectors of the page but not after...
For info, pictures of the drive :
I have a broken crusial V4. The controller is a PHISON PS3105-S5-1. I unsoldered and dumped one chip to see if I could do something.
So it doesn't seem encrypted but i'm having trouble getting the xor.
Can someone confirm (or not) that the xor is [ID=40: BLK=256 PG=8640 (1074-1070x7)] in PC3k Flash. It seems to work on the first two sectors of the page but not after...
For info, pictures of the drive :
Re: Crucial V4
October 21st, 2014, 6:28
what ECC did you detect ?
Re: Crucial V4
October 21st, 2014, 7:13
No ECC detected... that's also a problem...
I'm starting to think I'm playing out of my league.
I'm starting to think I'm playing out of my league.
Re: Crucial V4
October 21st, 2014, 8:34
Just find a XOR key fragments and you'll see what's the problem. The best block for this is LBN0000/LBN0001.
As I remember there's nothing special in this key, here it's appearance (do not remember if it was "S" version.)
As I remember there's nothing special in this key, here it's appearance (do not remember if it was "S" version.)
Re: Crucial V4
October 21st, 2014, 10:45
Sasha Sheremetov wrote:Just find a XOR key fragments and you'll see what's the problem. The best block for this is LBN0000/LBN0001.
As I remember there's nothing special in this key, here it's appearance (do not remember if it was "S" version.)
I don't think that PC3k can do that...http://rusolut.com/xor-key-library/ wrote:Controller generates one page of pattern for first page in block, then cyclically shift it -1 byte (-8 bits) for next pages in block. Pattern xored wih Data area of page only.
I'm still dumping the other chips in the mean time...
By the way, the https version of rusolut doesn't work !
Re: Crucial V4
October 21st, 2014, 13:25
Can you upload to my FTP (PM) one dump? I can try find XOR for this.
Re: Crucial V4
October 23rd, 2014, 4:31
Just detect page structure.
mostly only first page bigger then other but this controller can have few bigger pages.
After detect all sector lenght in page add this for ECC.
if ecc detected run fix and post structure, page size and block size here.
I will make xor for use in PC3000.
mostly only first page bigger then other but this controller can have few bigger pages.
After detect all sector lenght in page add this for ECC.
if ecc detected run fix and post structure, page size and block size here.
I will make xor for use in PC3000.
Re: Crucial V4
October 23rd, 2014, 6:56
Structure for this case is: DATA2048_SA8_ECC90_(DATA2048_SA2_ECC90)x3.
Re: Crucial V4
October 23rd, 2014, 10:05
Arvika is right.
Considering this chip is MLC chip, block size must be 128 (256) pages = 1105920 (2211840) bytes.
Considering this chip is MLC chip, block size must be 128 (256) pages = 1105920 (2211840) bytes.
Re: Crucial V4
October 24th, 2014, 2:40
Dear Gurus, thank you for your comments.
Structure given by arvika seems ok but I still can't find and correct ECC... I'll work on that today.
The chip IDed 0x2CA805CB so block size seems to be 256 pages.
In the mean time i'll also try to work out something else.
Again thanks for your comments.
Structure given by arvika seems ok but I still can't find and correct ECC... I'll work on that today.
The chip IDed 0x2CA805CB so block size seems to be 256 pages.
In the mean time i'll also try to work out something else.
Again thanks for your comments.
Re: Crucial V4
October 24th, 2014, 3:20
Block is 512 pages.
Re: Crucial V4
October 24th, 2014, 4:03
This ECC not has in PC3000Flash.
Re: Crucial V4
November 7th, 2014, 6:35
XOR was found for this case remotely 
Powered by phpBB © phpBB Group.