1. New controller algorithms
- Added new analysis algorithm for Netac NT2039 controller. Implemented autodetection of settings. Examples of different solutions are available in Global Solution Center.
Important: after using this method, in the case of unsatisfactory results it's necessary to use "Find the shift points" from FAT catalogs or "Data quality analysis" mode with the addition of all detected shift points.
- Enhanced Jmicron (JMF601/602) algorithm.
- Added new algorithm for ITE1168 controller. Implemented autodetection settings. Examples of different solutions are available in Global Solution Center.
Recommendation for data recovery of tasks with ITE1168 controller:
- It is recommended to use ready solution from Global Solution Center
- Transformation Graph is preferable mode to use during data preparation
- After the page transformation procedure (if it necessary) user should join all dumps and launch algorithm on it.
Important: there is no need to remove typical interleave.
In case of an incorrect order of the dump, message with correct order will be placed into log. In the case of the order correct initially image will be constructed successfully.
- Added new analysis algorithm for Stec 24850-00211(MACH4) controller. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for Stec 24850-01211 (MACH2,MACH2+) controller. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for SSS6690 controller. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for SSS6691 controller. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for ChipsBank CBM2092 controllers. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for ChipsBank CBM2093 controllers. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for AlcorMicro AU6990 controllers. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for Olympus xD Type 2 controllers. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for Silicon Motion (SM) controller family SM267EN. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for Silicon Motion (SM) controller family SM3254Q AE. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2251 50-F. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2232BD. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2136CC. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2134CE. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2136CF. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
- Added new analysis algorithm for PHISON controller family PS2134CH. Implemented auto-detect settings. Examples of different solutions are available in Global Solution Center.
2. Resources of PC-3000 Flash SSD Edition.
In PC-3000 Flash SSD Edition 5.0.4. now It is possible to add and edit resources, it is parameters separated by type, that used in different modes of the software.
Current version of software provides the following types of resources:
- power schemes.
- XOR combination which has been designed for a previous version of the data preparation method “XOR Transformation”. This method has been excluded from the tool, and now this type of resource is not available for changes and additions.
- XOR transformation for pages – different XOR pattern for the whole page.
- reading rules - special rules for reading chips.
The independent editing of the name and content (data) is available for each resource type.
3. Advanced support of TLC (Triple Level Cells) memory chips.
New feature for TLC memory reading has been added. Now it is possible to add own reading rules. TLC uses non-linear page and block addressing within memory that doesn't allow to read memory as continous sequence of pages.
TLC memory chips have following structure:
Chip → Planes → Blocks → Pages
For understanding of differences between TLC and other memory chips is neccessary to introduce concepts of "maximal address space" and "used address space". In SLC and MLC both values are similar, in TLC they always differ. This peculiarity of TLC depends on possibility to store 3 bits of information within one memory cell that breaks usual miltiplicity of power of 2.
Every TLC memory chip might consist of several planes which has maximal address space of N blocks. N - number that equals to power of 2, e.g. 1024, 2048, 4096, etc. The real address space for each plane is less because of oddness of information that is stored in memory cell. For example some memory chips have number of blocks equal to 1404.
The same situation is with blocks which make up the plane. For every block, some continous space of M pages is dedicated. M - number that equals to power of 2 as well, e.g. 64, 128, 256, 512. The real address space for each block is less, it is 3/4 of maximal space. For example some memory chips have number of pages equal to 192 when M=256.
Thus, for TLC memory reading there are additional parameters which should be selected in PC-3000 Flash SSD Edition software:
- number of planes
- maximal number of blocks per plane (in blocks)
- real number of blocks per plane (in blocks)
- maximal number of pages per block (in pages)
- real number of pages per block (in pages).
Addition and editing of reading rules is implemented via "Resources-->reading rules" menu of PC-3000 Flash SSD Edition. After creation of new reading rule it's possible to select own rule for certain memory chip. Value 0 means no particular reading rules.
Block size for TLC memory chips differs from usual ones (512,1024,2048...) and mostly multiple of 3/4 of usual size.
If the real number of blocks is entered wrongly, every plane will have excess number of blocks filled by FF. The same situation is with real size of block (number of pages per block). If entered value is wrong, every block will have excess number of pages filled by FF.
4. New XOR patterns.
Now it's possible to add own patterns into the database of resources of PC-3000 Flash SSD Edition software, apply it and use in autodetection of XOR elimination.
Addition and editing of XOR patterns is implemented via "Resources → XOR for page" menu of PC-3000 Flash SSD Edition.
To add new XOR it's necessary to enter following parameters:
- page size (in bytes)
- block size (in pages)
- path to binary file with XOR pattern (size of file should be equal to product of page size and block size)
To eliminate XOR in the task it's recommended to use Complex analysis with marked flag "Check necessity of XOR elimination" or data inspection method "Check necessity of XOR elimination"
5. New driver (version 2.0.).
Power management has been improved in the new driver of device, sleep mode bug is fixed.
Warning! It is recomended to always install the latest version of drivers!
6. Version autodetection.
When ambiguilty occurs in some block while image building, the most probable block is placed into the image and the rest of applicants (blocks) are placed into the list of versions.
In case with large number of versions (block-applicants) the task gets to be more complex.
New mode can be launched from any map, folder or file and allows to autoselect the correct version of blocks for entire area of object.
As a result of execution, method selects the flag "Use" for the versions which are correct. It is possible to add versions manually, it also can be used in autoselection.
7. Detailed LOG for Block Number Type 1 [0000].
Now it is possible to control the process of task analysis and image building in order to find the cause of incorrect work of algorithm. Use the maximal detalization of log to get access to this informarion.
Description of log:
Stage 1: Selection of number from service area of each block basing on specified parameters
Sector: 0 Marker_Bytes: 0x1003 Block : 0x0003 ( 3) Counter: 00
Sector: 1024 Marker_Bytes: 0x1108 Block : 0x0108 ( 264) Counter: 00
Sector: 2048 Marker_Bytes: 0x10F9 Block : 0x00F9 ( 249) Counter: 00
Sector: 3072 Marker_Bytes: 0x1014 Block : 0x0014 ( 20) Counter: 00
Sector: 4096 Marker_Bytes: 0x1006 Block : 0x0006 ( 6) Counter: 00
Sector: 5120 Marker_Bytes: 0x10FA Block : 0x00FA ( 250) Counter: 00
Sector: 6144 Marker_Bytes: 0x1011 Block : 0x0011 ( 17) Counter: 00
Sector: 7168 Marker_Bytes: 0x101A Block : 0x001A ( 26) Counter: 00
Sector: 8192 Marker_Bytes: 0x1016 Block : 0x0016 ( 22) Counter: 00
Sector – number of the first sector of block whose service information was used to identify number of block.
Marker_Bytes – initial bytes from which the number of block will be generated in accordance with conditions of algorithm.
Block – number of block calculated from Marker_Bytes using order of bytes, mask and shift.
Counter – value that shows how many times current number of block has been found. It allows to estimate number of banks, approximately.
Stage 2: Analysis of markers – Distributing by banks.
Analysis of markers. Distributing by banks.
Distribution of continuous chains in the bank before adjustment (Bank 0 )
From 0000 To 03F2 Size 1012
Distribution of continuous chains in the bank after adjustment
From 0000 To 03F2 Size 1012
Distribution of continuous chains in the bank before adjustment (Bank 1 )
From 0000 To 03F2 Size 1012
Distribution of continuous chains in the bank after adjustment
From 0000 To 03F2 Size 1012
Distribution of continuous chains in the bank before adjustment (Bank 2 )
From 0000 To 0009 Size 11
From 0010 To 037D Size 879
From 038F To 038F Size 2
Software estimates continuity of numbers of blocks for each bank and removes wrong groups.
From and To - describe certain continuous chain of sequential numbers of blocks within bank.
Stage 3: Conditions of banks building.
Image building is performed in accordance with the following parameters:
Bank 00 Block Number D: 001012 H: 03f4
Bank 01 Block Number D: 001012 H: 03f4
Bank 02 Block Number D: 000895 H: 037f
Bank 03 Block Number D: 000009 H: 0009
This log shows current size of banks which is found by algorithm of analysis. In the case of incorrect determination of bank size, data will be shifted, equally to the excess or missing blocks.
Stage 4: Image building.
Image = 0 Bank = 0 Block = 0x0000 ( 0) From = 846848( Block 0x033B (827) ...Several versions have been found for the current block.
...Block 0x03EC ( 1004) Sector 1028096
Image = 1024 Bank = 0 Block = 0x0001 ( 1) From = 1039360( Block 0x03F7 (1015)
Image = 2048 Bank = 0 Block = 0x0002 ( 2) From = 787456( Block 0x0301 (769)
Image = 3072 Bank = 0 Block = 0x0003 ( 3) From = 0( Block 0x0000 (0) )
Image = 4096 Bank = 0 Block = 0x0004 ( 4) From = 1038336( Block 0x03F6 (1014)
Image = 5120 Bank = 0 Block = 0x0005 ( 5) From = 1046528( Block 0x03FE (1022)
Image - number of the first sector of block in the final result that is being formed.
Bank - number of the bank that is being formed.
Block - number of the block that is being formed. (hex and dec views).
From - number of the first sector of source block that is being placed to the specified position of analysis result.
8. Innovations in the interface.1. Method "XOR transformation" has been changed to "XOR tranfromation for page". In the new concept it's necessary to perform XOR checking and elimination before page transformation but after joining by bytes (if required, e.g. Indilinx controller). All old solutions where XOR was used from previous version work correctly.
2. Drop-down list of controllers has been changed to the blank field. In this field the real marking of controller should be entered. It is very important to enter correct controller marking for further online solution search via software.
3. In the graph mode now it's possible to get real a ID of result in log by clicking on one of results.
4. New method "Synchronization of the channels" has been added to data preparation menu. This operation is necessary for data recovery from drives where multi-channel read/write operations were used (typical for SSD). This operation allows to eliminate asynchrony of allocation of symmetrical data within channels for successful joining to one result for further data analysis. In the current version of software this function is necessary only for JMicron controllers. It's recommended to apply it before any operations between memory chips/parts.
5. New flag has been added into Complex analysis parameters which helps to perform all operations within transformation graph.
6. New XOR patterns have been added into database, its automatical detection is implemented.
7. New additional parameters have been added in the Block Number Type 1. Now it is possible to enter additional parameters of block selection.
The specified conditions used for selection of blocks which form the final image of data. Every condition comprises 4 parameters:
- byte position – position of byte within sector which value is controlled,
- mask (heximal)
- condition – can assume 2 variants: equal or not equal
- value (heximal) – value after applying the mask is compared with this field
Sometimes additional conditions allow to improve the quality of result of analysis. For example to analyze only blocks which have byte 0x50 in the first byte of service area it is necessary to add following condition.
Position: 512
Mask: 0xFF
Condition: =
Value: 0x50
Under this condition all the rest blocks won't be analyzed.
8. Complex analysis has been optimized, its speed and efficiency is dramatically increased. If during "conversions within page" check the process is interrupted, and some conversions are confirmed, the pop-up message will appear that will offer to apply found conversion.
Results of operations will be placed into the transformation graph for the further work.
9. Now if the error occurs while image building it is possible to choose delete or leave result of analysis.
10. New function of automatic generation of Versions has been added into "Block Number Type 1".
If ambiguous situation occurs while image building and several applicants of block are compatible to the certain position, the most suitable block will be placed on the position inside image and other applicants will be added into list of “Versions” for this block. If the error occurs in image it is possible to change the certain block which was wrongly placed on its position.
All known bugs have been fixed.
|
