ONFI 2.3a Spec

Open NAND Flash Interface Specification
Revision 2.3a
19-October-2011
Hynix Semiconductor
Intel Corporation
Micron Technology, Inc.
Phison Electronics Corp.
SanDisk Corporation
Sony Corporation
Spansion
This 2.3a revision of the Open NAND Flash Interface specification ("Final Specification") is
available for download at www.onfi.org.
SPECIFICATION DISCLAIMER
THIS SPECIFICATION IS PROVIDED TO YOU “AS IS” WITH NO WARRANTIES
WHATSOEVER, INCLUDING ANY WARRANTY OF MERCHANTABILITY, NONINFRINGEMENT, OR FITNESS FOR ANY PARTICULAR PURPOSE. THE AUTHORS OF THIS
SPECIFICATION DISCLAIM ALL LIABILITY, INCLUDING LIABILITY FOR INFRINGEMENT OF
ANY PROPRIETARY RIGHTS, RELATING TO USE OR IMPLEMENTATION OF INFORMATION
IN THIS SPECIFICATION. THE AUTHORS DO NOT WARRANT OR REPRESENT THAT SUCH
USE WILL NOT INFRINGE SUCH RIGHTS. THE PROVISION OF THIS SPECIFICATION TO
YOU DOES NOT PROVIDE YOU WITH ANY LICENSE, EXPRESS OR IMPLIED, BY
ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS.
Copyright 2005-2011, Hynix Semiconductor, Intel Corporation, Micron Technology, Inc., Phison
Electronics Corp., SanDisk Corporation, Sony Corporation, Spansion. All rights reserved.
For more information about ONFI, refer to the ONFI Workgroup website at www.onfi.org.
All product names are trademarks, registered trademarks, or servicemarks of their respective
owners.
ONFI Workgroup Technical Editor:
Amber Huffman
Intel Corporation
2111 NE 25th Ave M/S JF2-53
Hillsboro, OR 97124 USA
Tel: (503) 264-7929
Email: [email protected]
ii
Table of Contents
1.
Introduction ............................................................................................................................... 1
1.1. Goals and Objectives ........................................................................................................ 1
1.2. EZ NAND Overview .......................................................................................................... 1
1.3. References ........................................................................................................................ 1
1.4. Definitions, abbreviations, and conventions ...................................................................... 2
1.4.1.
Definitions and Abbreviations .................................................................................... 2
1.4.2.
Conventions ............................................................................................................... 4
2. Physical Interface ..................................................................................................................... 7
2.1. TSOP-48 and WSOP-48 Pin Assignments ....................................................................... 7
2.2. LGA-52 Pad Assignments ............................................................................................... 10
2.3. BGA-63 Ball Assignments ............................................................................................... 12
2.4. BGA-100 Ball Assignments ............................................................................................. 16
2.5. Signal Descriptions ......................................................................................................... 20
2.6. CE_n Signal Requirements ............................................................................................. 26
2.6.1.
Source Synchronous Data Interface Requirements ................................................ 26
2.7. Absolute Maximum DC Ratings ...................................................................................... 26
2.8. Recommended DC Operating Conditions ....................................................................... 27
2.8.1.
I/O Power (VccQ) and I/O Ground (VssQ)............................................................... 27
2.9. AC Overshoot/Undershoot Requirements ...................................................................... 28
2.10.
DC and Operating Characteristics ............................................................................... 28
2.11.
Calculating Pin Capacitance ....................................................................................... 31
2.12.
Staggered Power-up.................................................................................................... 32
2.13.
Power Cycle Requirements ......................................................................................... 32
2.14.
Independent Data Buses ............................................................................................. 32
2.15.
Bus Width Requirements ............................................................................................. 32
2.16.
Ready/Busy (R/B_n) Requirements ............................................................................ 33
2.16.1.
Power-On Requirements ...................................................................................... 33
2.16.2.
R/B_n and SR[6] Relationship ............................................................................. 34
2.17.
Write Protect ................................................................................................................ 34
3. Memory Organization ............................................................................................................. 35
3.1. Addressing ...................................................................................................................... 36
3.1.1.
Multi-plane Addressing ............................................................................................ 37
3.1.2.
Logical Unit Selection .............................................................................................. 38
3.1.3.
Multiple LUN Operation Restrictions ........................................................................ 38
3.2. Factory Defect Mapping .................................................................................................. 39
3.2.1.
Device Requirements............................................................................................... 39
3.2.2.
Host Requirements .................................................................................................. 40
3.3. Extended ECC Information Reporting ............................................................................. 40
3.3.1.
Byte 0: Number of bits ECC correctability ............................................................... 41
3.3.2.
Byte 1: Codeword size ............................................................................................. 41
3.3.3.
Byte 2-3: Bad blocks maximum per LUN ................................................................. 41
3.3.4.
Byte 4-5: Block endurance ....................................................................................... 41
3.4. Discovery and Initialization.............................................................................................. 42
3.4.1.
CE_n Discovery ....................................................................................................... 42
3.4.2.
Target Initialization ................................................................................................... 42
4. Data Interface and Timing ...................................................................................................... 44
4.1. Data Interface Types ....................................................................................................... 44
4.1.1.
Signal Function Reassignment ................................................................................ 44
4.1.2.
Bus State ................................................................................................................. 45
4.1.3.
Source Synchronous and Repeat Bytes .................................................................. 46
4.1.4.
Data Interface / Timing Mode Transitions ................................................................ 47
4.2. Timing Parameters .......................................................................................................... 48
4.2.1.
General Timings ....................................................................................................... 48
4.2.2.
Asynchronous .......................................................................................................... 49
iii
4.2.3.
Source Synchronous................................................................................................ 56
Number of LUNs per x8 data bus ........................................................................................... 64
4.3. Timing Diagrams ............................................................................................................. 69
4.3.1.
Asynchronous .......................................................................................................... 69
4.3.2.
Source Synchronous................................................................................................ 76
4.4. Command Examples ....................................................................................................... 88
4.4.1.
Asynchronous .......................................................................................................... 88
4.4.2.
Source Synchronous................................................................................................ 91
5. Command Definition ............................................................................................................... 96
5.1. Command Set ................................................................................................................. 96
5.2. Command Descriptions ................................................................................................... 99
5.3. Reset Definition ............................................................................................................. 102
5.4. Synchronous Reset Definition ....................................................................................... 102
5.5. Reset LUN Definition ..................................................................................................... 103
5.6. Read ID Definition ......................................................................................................... 104
5.7. Read Parameter Page Definition .................................................................................. 106
5.7.1.
Parameter Page Data Structure Definition ............................................................ 108
5.7.2.
Extended Parameter Page Data Structure Definition ............................................ 122
5.8. Read Unique ID Definition............................................................................................. 125
5.9. Block Erase Definition ................................................................................................... 127
5.10.
Read Status Definition ............................................................................................... 127
5.11.
Read Status Enhanced Definition ............................................................................. 130
5.12.
Read Status and Read Status Enhanced required usage ........................................ 131
5.13.
Status Field Definition................................................................................................ 131
5.14.
Read Definition .......................................................................................................... 132
5.15.
Read Cache Definition............................................................................................... 135
5.16.
Page Program Definition ........................................................................................... 139
5.17.
Page Cache Program Definition ................................................................................ 141
5.18.
Copyback Definition................................................................................................... 144
5.19.
Small Data Move ....................................................................................................... 149
5.20.
Change Read Column Definition ............................................................................... 152
5.21.
Change Read Column Enhanced Definition ............................................................. 152
5.22.
Change Write Column Definition ............................................................................... 154
5.23.
Change Row Address Definition ............................................................................... 154
5.24.
Set Features Definition .............................................................................................. 156
5.25.
Get Features Definition.............................................................................................. 158
5.26.
Feature Parameter Definitions .................................................................................. 158
5.26.1.
Timing Mode ....................................................................................................... 159
5.26.2.
I/O Drive Strength ............................................................................................... 160
5.26.3.
EZ NAND control ................................................................................................ 160
6. Multi-plane Operations ......................................................................................................... 161
6.1. Requirements ................................................................................................................ 161
6.2. Status Register Behavior .............................................................................................. 162
6.3. Multi-plane Page Program ............................................................................................ 162
6.4. Multi-plane Copyback Read and Program .................................................................... 165
6.5. Multi-plane Block Erase ................................................................................................ 168
6.6. Multi-plane Read ........................................................................................................... 169
7. Behavioral Flows .................................................................................................................. 174
7.1. Target behavioral flows ................................................................................................. 174
7.1.1.
Variables ................................................................................................................ 174
7.1.2.
Idle states ............................................................................................................... 174
7.1.3.
Idle Read states ..................................................................................................... 176
7.1.4.
Reset command states .......................................................................................... 178
7.1.5.
Read ID command states ...................................................................................... 180
7.1.6.
Read Parameter Page command states................................................................ 181
7.1.7.
Read Unique ID command states .......................................................................... 182
iv
7.1.8.
Page Program and Page Cache Program command states ................................. 183
7.1.9.
Block Erase command states ................................................................................ 186
7.1.10.
Read command states ....................................................................................... 188
7.1.11.
Set Features command states ........................................................................... 190
7.1.12.
Get Features command states ........................................................................... 191
7.1.13.
Read Status command states ............................................................................ 191
7.1.14.
Read Status Enhanced command states ........................................................... 192
7.2. LUN behavioral flows .................................................................................................... 193
7.2.1.
Variables ................................................................................................................ 193
7.2.2.
Idle command states .............................................................................................. 193
7.2.3.
Idle Read states ..................................................................................................... 195
7.2.4.
Status states .......................................................................................................... 196
7.2.5.
Reset states ........................................................................................................... 197
7.2.6.
Block Erase command states ................................................................................ 197
7.2.7.
Read command states ........................................................................................... 199
7.2.8.
Page Program and Page Cache Program command states ................................. 201
A.
Sample Code for CRC-16 (Informative) ........................................................................... 205
B.
Spare Size Recommendations (Informative) .................................................................... 207
C.
Device Self-Initialization with PSL (Informative) ............................................................... 208
D.
ICC Measurement Methodology ....................................................................................... 209
E.
Measuring Timing Parameters to/From Tri-State ............................................................. 217
F. EZ NAND: End to End Data Path Protection (INFORMATIVE) ........................................... 218
v
1. Introduction
1.1. Goals and Objectives
This specification defines a standardized NAND Flash device interface that provides the means
for a system to be designed that supports a range of NAND Flash devices without direct design
pre-association. The solution also provides the means for a system to seamlessly make use of
new NAND devices that may not have existed at the time that the system was designed.
Some of the goals and requirements for the specification include:
 Support range of device capabilities and new unforeseen innovation
 Consistent with existing NAND Flash designs providing orderly transition to ONFI
 Capabilities and features are self-described in a parameter page such that hard-coded
chip ID tables in the host are not necessary
 Flash devices are interoperable and do not require host changes to support a new Flash
device
 Define a higher speed NAND interface that is compatible with existing NAND Flash
interface
 Allow for separate core (Vcc) and I/O (VccQ) power rails
 Support for offloading NAND lithography specific functionality to a controller stacked in
the NAND package (EZ NAND)
1.2. EZ NAND Overview
EZ NAND includes the control logic packaged together with NAND to perform the NAND
management functionality that is lithography specific (e.g. ECC), while retaining the NAND
protocol infrastructure. EZ NAND delivers an ECC offloaded solution with minimal command
and/or protocol changes. The device parameter page will specify if EZ NAND is supported.
1.3. References
This specification is developed in part based on existing common NAND Flash device behaviors,
including the behaviors defined in the following datasheets:

Micron MT29F8G08ABC data sheet available at
http://www.micron.com/products/partdetail?part=MT29F8G08ABCBBH1-12

Numonyx NAND04GB2D data sheet available at
http://www.numonyx.com/Documents/Datasheets/NAND04G-B2D_NAND08G-BxC.pdf
The specification also makes reference to the following specifications and standards:

ONFI Block Abstracted NAND revision 1.1. Specification is available at
http://www.onfi.org.
1
1.4. Definitions, abbreviations, and conventions
1.4.1. Definitions and Abbreviations
The terminology used in this specification is intended to be self-sufficient and does not rely on
overloaded meanings defined in other specifications. Terms with specific meaning not directly
clear from the context are clarified in the following sections.
1.4.1.1. address
The address is comprised of a row address and a column address. The row address identifies
the page, block, and LUN to be accessed. The column address identifies the byte or word within
a page to access. The least significant bit of the column address shall always be zero in the
source synchronous data interface.
1.4.1.2. asynchronous
Asynchronous is when data is latched with the WE_n signal for writes and RE_n signal for reads.
1.4.1.3. block
Consists of multiple pages and is the smallest addressable unit for erase operations.
1.4.1.4. column
The byte (x8 devices) or word (x16 devices) location within the page register.
1.4.1.5. defect area
The defect area is where factory defects are marked by the manufacturer. Refer to section 3.2.
1.4.1.6. device
The packaged NAND unit. A device consists of one or more targets.
1.4.1.7. DDR
Acronym for double data rate.
1.4.1.8. Dword
A Dword is thirty-two (32) bits of data. A Dword may be represented as 32 bits, as two adjacent
words, or as four adjacent bytes. When shown as bits the least significant bit is bit 0 and most
significant bit is bit 31. The most significant bit is shown on the left. When shown as words the
least significant word (lower) is word 0 and the most significant (upper) word is word 1. When
shown as bytes the least significant byte is byte 0 and the most significant byte is byte 3. See
Figure 1 for a description of the relationship between bytes, words, and Dwords.
1.4.1.9. latching edge
The latching edge describes the edge of the CLK or the DQS signal that the contents of the data
bus are latched on for the source synchronous data interface. For data cycles the latching edge
is both the rising and falling edges of the DQS signal. For command and address cycles the
latching edge is the rising edge of the CLK signal.
2
1.4.1.10. LUN (logical unit number)
The minimum unit that can independently execute commands and report status. There are one
or more LUNs per target.
1.4.1.11. na
na stands for “not applicable”. Fields marked as “na” are not used.
1.4.1.12. O/M
O/M stands for Optional/Mandatory requirement. When the entry is set to “M”, the item is
mandatory. When the entry is set to “O”, the item is optional.
1.4.1.13. page
The smallest addressable unit for read and program operations.
1.4.1.14. page register
Register used to read data from that was transferred from the Flash array. For program
operations, the data is placed in this register prior to transferring the data to the Flash array. If EZ
NAND is supported a buffer exists in the EZ NAND controller that may be used to facilitate
Copyback operations. Refer to section 5.18 for information on EZ NAND Copyback operations.
1.4.1.15. partial page
A portion of the page, referred to as a partial page, may be programmed if the target supports
more than one program per page as indicated in the parameter page. The host may choose to
read only a portion of the data from the page register in a read operation; this portion may also be
referred to as a partial page.
1.4.1.16. read request
A read request is a data output cycle request from the host that results in a data transfer from the
device to the host. Refer to section 4.1.2 for information on data output cycles.
1.4.1.17. row
Refers to the block and page to be accessed.
1.4.1.18. source synchronous
Source synchronous is when the strobe (DQS) is forwarded with the data to indicate when the
data should be latched. The strobe signal, DQS, can be thought of as an additional data bus bit.
1.4.1.19. SR[ ]
SR refers to the status register contained within a particular LUN. SR[x] refers to bit x in the
status register for the associated LUN. Refer to section 5.13 for the definition of bit meanings
within the status register.
1.4.1.20. target
An independent Flash component with its own CE_n signal.
3
1.4.1.21. Uncorrectable Bit Error Rate, or ratio (UBER)
A metric for the rate of occurrence of data errors, equal to the number of data errors per bits read.
Mathematically, it may be represented as:
UBER = cumulative number of data errors / cumulative number of bits read
Note: The cumulative number of bits read is the sum of all bits of data read back from the device,
with multiple reads of the same memory bit as multiple bits read. For example, if a 100GB device
12
is read ten times then there would be about 1TB (8x10 bits) read. The cumulative number of
data errors is the count of the physical pages for which the device fails to return correct data.
Note: This metric only applies to devices that support EZ NAND. EZ NAND delivers an ECC
offloaded solution, and thus this metric applies. For raw NAND solutions where the host provides
the ECC solution, the UBER is dependent on the host controller capability and UBER for that
solution is not within the scope of this specification.
1.4.1.22. Vtt
Termination voltage.
1.4.1.23. word
A word is sixteen (16) bits of data. A word may be represented as 16 bits or as two adjacent
bytes. When shown as bits the least significant bit is bit 0 and most significant bit is bit 15. The
most significant bit is shown on the left. When shown as bytes the least significant byte (lower) is
byte 0 and the most significant byte (upper) is byte 1. See Figure 1 for a description of the
relationship between bytes, words and Dwords.
1.4.2. Conventions
The names of abbreviations and acronyms used as signal names are in all uppercase (e.g.,
CE_n). Fields containing only one bit are usually referred to as the "name" bit instead of the
"name" field. Numerical fields are unsigned unless otherwise indicated.
1.4.2.1. Precedence
If there is a conflict between text, figures, state machines, timing diagrams, and tables, the
precedence shall be state machines and timing diagrams, tables, figures, and then text.
1.4.2.2. Keywords
Several keywords are used to differentiate between different levels of requirements.
1.4.2.2.1. mandatory
A keyword indicating items to be implemented as defined by this specification.
1.4.2.2.2. may
A keyword that indicates flexibility of choice with no implied preference.
1.4.2.2.3. optional
A keyword that describes features that are not required by this specification. However, if any
optional feature defined by the specification is implemented, the feature shall be implemented in
the way defined by the specification.
4
1.4.2.2.4. reserved
A keyword indicating reserved bits, bytes, words, fields, and opcode values that are set-aside for
future standardization. Their use and interpretation may be specified by future extensions to this
or other specifications. A reserved bit, byte, word, or field shall be cleared to zero, or in
accordance with a future extension to this specification. The recipient shall not check reserved
bits, bytes, words, or fields.
1.4.2.2.5. shall
A keyword indicating a mandatory requirement. Designers are required to implement all such
mandatory requirements to ensure interoperability with other products that conform to the
specification.
1.4.2.2.6. should
A keyword indicating flexibility of choice with a strongly preferred alternative. Equivalent to the
phrase “it is recommended”.
1.4.2.3. Byte, word and Dword Relationships
Figure 1 illustrates the relationship between bytes, words and Dwords.
7
6
5
4
3
2
1
0
Byte
1
5
1
4
1
3
1
2
1
1
1
0
9
8
7
6
5
4
3
2
1
0
Word
Byte 1
3 3 2
1 0 9
2
8
2
7
2
6
2
5
2
4
2
3
2
2
2
1
2
0
1
9
1
8
1
7
1
6
1
5
1
4
1
3
1
2
1
1
Byte 0
1
0
9
8
7
6
5
4
3
2
1
0
Dword
Word 1
Byte 3
Word 0
Byte 2
Figure 1
Byte 1
Byte 0
Byte, word and Dword relationships
1.4.2.4. Behavioral Flow Diagrams
For each function to be completed a state machine approach is used to describe the sequence
and externally visible behavior requirements. Each function is composed of several states to
accomplish a set goal. Each state of the set is described by an individual state table. Table 1
5
below shows the general layout for each of the state tables that comprise the set of states for the
function.
State name
Action list
Transition condition 0

Next state 0
Transition condition 1

Next state 1
Table 1
State Table Cell Description
Each state is identified by a unique state name. The state name is a brief description of the
primary action taken during the state. Actions to take while in the state are described in the
action list.
Each transition is identified by a transition label and a transition condition. The transition label
consists of the state designator of the state from which the transition is being made followed by
the state designator of the state to which the transition is being made. The transition condition is a
brief description of the event or condition that causes the transition to occur and may include a
transition action that is taken when the transition occurs. This action is described fully in the
transition description text. Transition conditions are listed in priority order and are not required to
be mutually exclusive. The first transition condition that evaluates to be true shall be taken.
Upon entry to a state, all actions to be executed in that state are executed. If a state is re-entered
from itself, all actions to be executed in the state are executed again.
It is assumed that all actions are executed within a state and that transitions from state to state
are instantaneous.
6
2. Physical Interface
2.1. TSOP-48 and WSOP-48 Pin Assignments
Figure 2 defines the pin assignments for devices using 48-pin TSOP or 48-pin WSOP packaging for 8-bit data access. Figure 3 defines the pin
assignments for devices using 48-pin TSOP or 48-pin WSOP packaging for 16-bit data access. The package with 16-bit data access does not
support the source synchronous data interface. The physical dimensions of the TSOP package is defined in the JEDEC document MO-142
variation DD. The physical dimensions of the WSOP package is defined in the JEDEC document MO-259.
7
Ssync Async
R
R
R
R/B3_n
R/B2_n
R/B1_n
R/B0_n
W/R_n
CE0_n
CE1_n
R
Vcc
Vss
CE2_n
CE3_n
CLE
ALE
CLK
WP_n
VSP3
R
R
R
R
R
R
R
R/B3_n
R/B2_n
R/B1_n
R/B0_n
RE_n
CE0_n
CE1_n
R
Vcc
Vss
CE2_n
CE3_n
CLE
ALE
WE_n
WP_n
VSP3
R
R
R
R
Async Ssync
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
48-pin TSOP
and
48-pin WSOP
Figure 2
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
VssQ
R
R
R
IO7
IO6
IO5
IO4
R
VccQ
VSP1
Vcc
Vss
VSP2*
VccQ
R
IO3
IO2
IO1
IO0
R
R
R
VssQ
VssQ
R
R
R
DQ7
DQ6
DQ5
DQ4
R
VccQ
VSP1
Vcc
Vss
DQS
VccQ
R
DQ3
DQ2
DQ1
DQ0
R
R
R
VssQ
48-pin TSOP/WSOP pinout for 8-bit data access
NOTE: For a source synchronous capable part, pin 35 is not used when configured in the asynchronous data interface. Specifically,
VSP2 is present for asynchronous only parts.
8
R
R
R
R/B3_n
R/B2_n
R/B1_n
R/B0_n
RE_n
CE0_n
CE1_n
R
Vcc
Vss
CE2_n
CE3_n
CLE
ALE
WE_n
WP_n
VSP3
R
R
R
R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
48-pin TSOP
and
48-pin WSOP
Figure 3
48-pin TSOP/WSOP pinout for 16-bit data access
9
48
47
46
45
44
43
42
41
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
VssQ
IO15
IO14
IO13
IO7
IO6
IO5
IO4
IO12
VccQ
VSP1
Vcc
Vss
VSP2
VccQ
IO11
IO3
IO2
IO1
IO0
IO10
IO9
IO8
VssQ
2.2. LGA-52 Pad Assignments
Figure 4 defines the pad assignments for devices using 52-pad LGA packaging with 8-bit data
access. An option is specified for two independent 8-bit data buses. Figure 5 defines the pad
assignments for devices using 52-pad LGA packaging with 16-bit data access. The physical
dimensions of the package are 12mmx17mm or 14mmx18mm. Figure 6 defines the pad spacing
requirements for the 52-pad LGA package for both package dimensions. These LGA packages
do not support the source synchronous data interface.
A
8
7
R/B1_
1_n
B C D E
VDDi
R/B1_
0_n
6
H J
K L
R
RE_0
_n
M N
R
R/B0_
1_n
RE_1
_n
Vcc
5
F G
IO7
_1
IO6
_1
IO7
_0
Vss
R
VssQ
IO5
_1
IO5
_0
VccQ
Vcc
CE0_
0_n
CE0_
1_n
R/B0_
0_n
WP_1
_n
IO6
_0
IO4
_0
IO4
_1
CLE
_0
CLE
_1
WE_0
_n
IO0
_0
IO2
_0
Vss
IO3
_1
4
3
2
1
0
ALE
_1
Vss
CE1_
0_n
CE1_
1_n
OA
ALE
_0
R
OB
Figure 4
WP_0
_n
WE_1
_n
IO1
_0
IO0
_1
R
OC
IO3
_0
IO1
_1
VssQ
IO2
_1
R
R
R
OD
OE
OF
LGA pinout for 8-bit data access
10
VccQ
A
8
R/B3
_n
B C D E
VDDi
6
RE
_n
Vcc
5
H J
K L
R
R/B2
_n
7
F G
M N
R
R/B1
_n
R
IO15
Vss
IO14
IO7
R
VssQ
IO13
IO5
VccQ
Vcc
CE0
_n
CE1
_n
R/B0
_n
R
IO6
IO4
IO12
CLE
R
WE
_n
IO0
IO2
Vss
IO11
4
3
2
Vss
1
0
CE2
_n
CE3
_n
OA
WP
_n
R
ALE
R
OB
Figure 5
R
IO1
IO8
R
OC
IO3
IO9
VssQ
IO10
VccQ
R
R
R
OD
OE
OF
LGA pinout for 16-bit data access
11
14.00 +/- 0.15
12.00 +/- 0.10
10.00
2.00
1.00
1.00
1 mm
pad diameter
1.30
OA
A
B
G
1.00
H
J
18.00 +/- 0.15
F
17.00 +/- 0.10
OC
E
2.50
2.50
12.00
D
1.00
OB
C
OD
K
L
OE
M
N
OF
8
7
6
5
4
3
2
1
0
0.7 mm
pad diameter
Figure 6
LGA-52 pad spacing requirements (bottom view, dimensions in millimeters)
2.3. BGA-63 Ball Assignments
Figure 7 defines the ball assignments for devices using 63-ball BGA packaging with 8-bit data
access for the asynchronous data interface. Figure 8 defines the ball assignments for devices
using 63-ball BGA packaging with 8-bit data access for the source synchronous data interface.
Figure 9 defines the ball assignments for devices using 63-ball BGA packaging with 16-bit data
access for the asynchronous data interface. The 63-ball BGA package with 16-bit data access
does not support the source synchronous data interface. Figure 10 defines the ball spacing
requirements for the 63-ball BGA package. The solder ball diameter is 0.45 mm post reflow.
12
1
2
A
R
R
B
R
3
4
5
6
7
8
C
WP_n
ALE
VSS
CE0_n
WE_n
R/B0_n
D
VCC
RE_n
CLE
CE1_n
CE2_n
R/B1_n
E
R
R
R
R
CE3_n
R/B2_n
F
R
R
R
R
VSS
R/B3_n
G
VSP3
VCC
VSP1
R
R
VSP2
H
R
IO0
R
R
R
VCCQ
J
R
IO1
R
VCCQ
IO5
IO7
K
VSSQ
IO2
IO3
IO4
IO6
VSSQ
9
10
R
R
R
R
L
R
R
R
R
M
R
R
R
R
Figure 7
BGA-63 ball assignments for 8-bit data access, asynchronous only data
interface
13
Note that WE_n is located at ball H7 when a source synchronous capable part is used in
asynchronous mode.
1
2
A
R
R
B
R
3
4
5
6
7
8
C
WP_n
ALE
VSS
CE0_n
R
R/B0_n
D
VCC
W/R_n
CLE
CE1_n
CE2_n
R/B1_n
E
R
R
R
R
CE3_n
R/B2_n
F
R
R
VREFQ
R
VSS
R/B3_n
G
VSP3
VCC
VSP1
R
R
VSP2
H
R
DQ0
DQS_c
CLK_c
CLK_t
VCCQ
J
R
DQ1
DQS_t
VCCQ
DQ5
DQ7
K
VSSQ
DQ2
DQ3
DQ4
DQ6
VSSQ
9
10
R
R
R
R
L
R
R
R
R
M
R
R
R
R
Figure 8
BGA-63 ball assignments for 8-bit data access, source synchronous data
interface
14
1
2
A
R
R
B
R
3
4
5
6
7
8
C
WP_n
ALE
VSS
CE0_n
WE_n
R/B0_n
D
VCC
RE_n
CLE
CE1_n
CE2_n
R/B1_n
E
R
R
R
R
CE3_n
R/B2_n
F
R
R
R
R
VSS
R/B3_n
G
VSP3
VCC
VSP1
IO13
IO15
VSP2
H
IO8
IO0
IO10
IO12
IO14
VCCQ
J
IO9
IO1
IO11
VCCQ
IO5
IO7
K
VSSQ
IO2
IO3
IO4
IO6
VSSQ
9
10
R
R
R
R
L
R
R
R
R
M
R
R
R
R
Figure 9
BGA-63 ball assignments for 16-bit, asynchronous only data access
15
0.80 TYP
A10
A1
0.80 TYP
0.45 mm
ball diameter
post reflow
8.80
4.40
3.60
7.20
Figure 10
BGA-63 ball spacing requirements (top view, dimensions in millimeters)
2.4. BGA-100 Ball Assignments
Figure 11 defines the ball assignments for devices using 100-ball BGA packaging with dual 8-bit
data access for the asynchronous data interface. Figure 12 defines the ball assignments for
devices using 100-ball BGA packaging with dual 8-bit data access for the source synchronous
data interface. Figure 13 defines the ball spacing requirements for the 100-ball BGA package.
The solder ball diameter is 0.45 mm post reflow. The 100-ball BGA has two package sizes:
12mm x 18mm and 14mm x 18mm.
16
1
2
A
R
R
B
R
3
4
5
6
7
8
9
10
R
R
R
C
D
R
RFT
VSP3_1
WP_1_n
VSP2_1
VSP1_1
RFT
R
E
R
RFT
VSP3_0
WP_0_n
VSP2_0
VSP1_0
RFT
VDDi
F
VCC
VCC
VCC
VCC
VCC
VCC
VCC
VCC
G
VSS
VSS
VSS
VSS
VSS
VSS
VSS
VSS
H
VSSQ
VCCQ
R
R
R/B0_1_n
R/B1_1_n
VCCQ
VSSQ
J
IO0_1
IO2_1
ALE_1
CE1_1_n
R/B0_0_n
R/B1_0_n
IO5_1
IO7_1
K
IO0_0
IO2_0
ALE_0
CE1_0_n
CE0_1_n
CE0_0_n
IO5_0
IO7_0
L
VCCQ
VSSQ
VCCQ
CLE_1
RE_1_n
VCCQ
VSSQ
VCCQ
M
IO1_1
IO3_1
VSSQ
CLE_0
RE_0_n
VSSQ
IO4_1
IO6_1
N
IO1_0
IO3_0
NC
NC
NC
WE_1_n
IO4_0
IO6_0
P
VSSQ
VCCQ
NC
NC
NC
WE_0_n
VCCQ
VSSQ
R
T
R
U
R
Figure 11
R
R
R
R
BGA-100 ball assignments for dual 8-bit data access, asynchronous data
interface
17
1
2
A
R
R
B
R
3
4
5
6
7
8
9
10
R
R
R
C
D
R
RFT
VSP3_1
WP_1_n
VSP2_1
VSP1_1
RFT
R
E
R
RFT
VSP3_0
WP_0_n
VSP2_0
VSP1_0
RFT
VDDi
F
VCC
VCC
VCC
VCC
VCC
VCC
VCC
VCC
G
VSS
VSS
VSS
VSS
VSS
VSS
VSS
VSS
H
VSSQ
VCCQ
VREFQ_1
VREFQ_0
R/B0_1_n
R/B1_1_n
VCCQ
VSSQ
J
DQ0_1
DQ2_1
ALE_1
CE1_1_n
R/B0_0_n
R/B1_0_n
DQ5_1
DQ7_1
K
DQ0_0
DQ2_0
ALE_0
CE1_0_n
CE0_1_n
CE0_0_n
DQ5_0
DQ7_0
L
VCCQ
VSSQ
VCCQ
CLE_1
W/R_1_n
VCCQ
VSSQ
VCCQ
M
DQ1_1
DQ3_1
VSSQ
CLE_0
W/R_0_n
VSSQ
DQ4_1
DQ6_1
N
DQ1_0
DQ3_0
DQS_1_c
DQS_1_t
CLK_1_c
CLK_1_t
DQ4_0
DQ6_0
P
VSSQ
VCCQ
DQS_0_c
DQS_0_t
CLK_0_c
CLK_0_t
VCCQ
VSSQ
R
T
R
U
R
Figure 12
R
R
R
BGA-100 ball assignments for dual 8-bit data access, source synchronous data
interface
18
R
0.45 mm
ball diameter
post reflow
14mm wide package
14
A1
A10
9
8
7
16
18
5
1 TYP
1 TYP
4.5
9
12
Figure 13
12mm wide package
BGA-100 ball spacing requirements (top view, dimensions in millimeters)
19
2.5. Signal Descriptions
Table 2 provides the signal descriptions.
Where the pin function has an inverted logic sense, that is, the function is true or invoked for a
low signal, the overbar, trailing slash (\), #, or _n symbol is appended to the symbol. The _t and
_c symbols represent true and complementary logic states for differential pairs.
Signal
Name
Input /
Output
R/B_x_n
O
RE_x_n
I
W/R_x_n
I
CE_x_n
I
Vcc
I
VccQ
I
Vss
I
VssQ
I
VREFQ_x
I
VDDi
na
CLE_x
I
ALE_x
I
Description
Ready/Busy
The Ready/Busy signal indicates the target status. When low, the signal
indicates that one or more LUN operations are in progress. This signal is
an open drain output and requires an external pull-up. See section 2.16
for requirements.
Read Enable
The Read Enable signal enables serial data output. This signal shares the
same pin as W/R_x_n in the source synchronous data interface.
Write/Read Direction
The Write/Read Direction signal indicates the owner of the DQ bus and
DQS signal in the source synchronous data interface. This signal shares
the same pin as RE_x_n in the asynchronous data interface.
Chip Enable
The Chip Enable signal selects the target. When Chip Enable is high and
the target is in the ready state, the target goes into a low-power standby
state. When Chip Enable is low, the target is selected. See section 2.6 for
additional requirements.
Power
The Vcc signal is the power supply to the device.
I/O Power
The VccQ signal is the power supply for input and/or output signals. Refer
to section 2.8.1.
Ground
The Vss signal is the power supply ground.
I/O Ground
The VssQ signal is the ground for input and/or output signals. Refer to
section 2.8.1.
Voltage Reference
This signal is reserved for future use.
ASIC Voltage Control
This signal is used to assist in stabilizing the internal power supply to a
NAND controller ASIC (e.g. EZ NAND) by connecting to an external
capacitor.
Command Latch Enable
The Command Latch Enable signal is one of the signals used by the host
to indicate the type of bus cycle (command, address, data). Refer to
section 4.1.2.
Address Latch Enable
The Address Latch Enable signal is one of the signals used by the host to
indicate the type of bus cycle (command, address, data). Refer to section
4.1.2.
20
Signal
Name
Input /
Output
WE_x_n
I
CLK_x_t
I
CLK_x_c
I
WP_x_n
I
IO0_0 –
IO7_0
(DQ0_0 –
DQ7_0)
DQS_x_t
I/O
DQS_x_c
I/O
IO8 –
IO15
I/O
IO0_1 –
IO7_1
(DQ0_1 –
DQ7_1)
I/O
VSP_x
R
RFT
I/O
Description
Write Enable
The Write Enable signal controls the latching of input data in the
asynchronous data interface. Data, commands, and addresses are latched
on the rising edge of WE_x_n. This signal shares the same pin as CLK_x
in the source synchronous data interface.
Clock
The Clock signal is used as the clock in the source synchronous data
interface. This signal shares the same pin as WE_x_n in the
asynchronous data interface.
Clock Complement
This signal is reserved for future use.
Write Protect
The Write Protect signal disables Flash array program and erase
operations. See section 2.17 for requirements.
I/O Port 0, bits 0-7
The I/O port is an 8-bit wide bidirectional port for transferring address,
command, and data to and from the device. Also known as DQ0_0 –
DQ7_0 for the source synchronous data interface.
Data Strobe
The data strobe signal that indicates the data valid window for the source
synchronous data interface.
Data Strobe Complement
This signal is reserved for future use.
I/O Port 0, bits 8-15
These signals are used in a 16-bit wide target configuration. The signals
are the upper 8 bits for the 16-bit wide bidirectional port used to transfer
data to and from the device.
I/O Port 1, bits 0-7
The I/O port is an 8-bit wide bidirectional port for transferring address,
command, and data to and from the device. These pins may be used as
an additional 8-bit wide bidirectional port for devices that support two
independent data buses. Also known as DQ0_1 – DQ7_1 for the source
synchronous data interface.
Vendor Specific
The function of these signals is defined and specified by the NAND
vendor. Devices shall have an internal pull-up or pull-down resistor on
these signals to yield ONFI compliant behavior when a signal is not
connected by the host. Any VSP signal not used by the NAND vendor
shall not be connected internal to the device.
Reserved
These pins shall not be connected by the host.
Reserved for Test
These pins shall not be connected by the host.
Table 2
Signal descriptions
Table 3 provides the signal mapping to pin/pad/ball for each package type listed within the ONFI
specification. These signal mappings are required if the packages listed in this specification are
implemented. The “Async Only” signal mappings apply to packages where the device is not
source synchronous capable. When the device is source synchronous capable, the “Src Sync”
signal mappings shall be used. If a signal is marked as “na” then the corresponding package
21
does not implement that signal. Any signal that does not have an associated number is implicitly
numbered “0”. For example, WP_n is equivalent to WP0_n.
Devices may be implemented with other package types and be ONFI compliant if all other ONFI
requirements within this specification are satisfied.
22
TSOP /
WSOP
Async
only x8
TSOP /
WSOP
Src Sync
x8
TSOP /
WSOP
Async
only x16
LGA
Async
only x8
LGA
Async
only x16
BGA-63
Async
only x8
BGA-63
Src Sync
x8
BGA-63
Async
only x16
BGA-100
Async
only x8
BGA-100
Src Sync
x8
na
na
na
na
J6
H6
J7
H7
M6
L6
na
na
na
na
na
na
K7
K6
K5
J5
F2
F3
F4
F5
F6
F7
F8
F9
H3
H8
L2
L4
L7
L9
P3
na
na
na
na
J6
H6
J7
H7
M6
L6
M6
L6
na
na
na
na
K7
K6
K5
J5
F2
F3
F4
F5
F6
F7
F8
F9
H3
H8
L2
L4
L7
L9
P3
Signal Name
M/O/R
R/B0_n
R/B1_n
R/B2_n
R/B3_n
R/B0_0_n
R/B0_1_n
R/B1_0_n
R/B1_1_n
RE_0_n
RE_1_n
W/R_0_n
W/R_1_n
CE0_n
CE1_n
CE2_n
CE3_n
CE0_0_n
CE0_1_n
CE1_0_n
CE1_1_n
Vcc
M
O
O
O
M
O
O
O
M
O
M
O
M
O
O
O
M
O
O
O
M
7
6
5
4
na
na
na
na
8
na
na
na
9
10
14
15
na
na
na
na
12
37
7
6
5
4
na
na
na
na
8
na
8
na
9
10
14
15
na
na
na
na
12
37
7
6
5
4
na
na
na
na
8
na
na
na
9
10
14
15
na
na
na
na
12
37
na
na
na
na
E5
E7
A7
OA8
C7
D6
na
na
na
na
na
na
A5
C5
A1
OA0
B6
M6
E5
E7
A7
OA8
na
na
na
na
C7
na
na
na
A5
C5
A1
OA0
na
na
na
na
B6
M6
C8
D8
E8
F8
na
na
na
na
D4
na
na
na
C6
D6
D7
E7
na
na
na
na
D3
G4
C8
D8
E8
F8
na
na
na
na
D4
na
D4
na
C6
D6
D7
E7
na
na
na
na
D3
G4
C8
D8
E8
F8
na
na
na
na
D4
na
na
na
C6
D6
D7
E7
na
na
na
na
D3
G4
VccQ
M
34
39
34
39
34
39
N1
N7
N1
N7
H8
J6
H8
J6
H8
J6
23
TSOP /
WSOP
Async
only x8
TSOP /
WSOP
Src Sync
x8
TSOP /
WSOP
Async
only x16
LGA
Async
only x8
LGA
Async
only x16
BGA-63
Async
only x8
BGA-63
Src Sync
x8
BGA-63
Async
only x16
13
36
B2
F6
L3
B2
F6
L3
C5
F7
C5
F7
C5
F7
25
48
25
48
M2
OE8
M2
OE8
K8
K3
K8
K3
K8
K3
na
na
na
16
na
17
na
18
na
18
na
na
na
19
na
na
na
na
16
na
17
na
18
na
na
na
na
na
19
na
na
na
OB8
A3
C3
C1
D2
E3
E1
na
na
na
na
F2
G5
na
na
OB8
A3
na
C1
na
E3
na
na
na
na
na
F2
na
na
na
na
D5
na
C4
na
C7
na
na
na
na
na
C3
na
F5
na
na
D5
na
C4
na
H7
na
H7
na
H6
na
C3
na
na
na
na
D5
na
C4
na
C7
na
na
na
na
na
C3
na
Signal Name
M/O/R
Vss
M
13
36
13
36
VssQ
M
25
48
VREFQ_0
VREFQ_1
VDDi
CLE_0
CLE_1
ALE_0
ALE_1
WE_0_n
WE_1_n
CLK_0_t
CLK_1_t
CLK_0_c
CLK_1_c
WP_0_n
WP_1_n
R
R
O
M
O
M
O
M
O
M
O
R
R
M
O
na
na
na
16
na
17
na
18
na
na
na
na
na
19
na
24
BGA-100
Async
only x8
BGA-100
Src Sync
x8
P8
G2
G3
G4
G5
G6
G7
G8
G9
H2
H9
L3
L8
M4
M7
P2
P9
na
na
E9
M5
L5
K4
J4
P7
N7
na
na
na
na
E5
D5
P8
G2
G3
G4
G5
G6
G7
G8
G9
H2
H9
L3
L8
M4
M7
P2
P9
H5
H4
E9
M5
L5
K4
J4
P7
N7
P7
N7
P6
N6
E5
D5
Signal Name
M/O/R
TSOP /
WSOP
Async
only x8
IO0_0 / DQ0_0
IO1_0 / DQ1_0
IO2_0 / DQ2_0
IO3_0 / DQ3_0
IO4_0 / DQ4_0
IO5_0 / DQ5_0
IO6_0 / DQ6_0
IO7_0 / DQ7_0
IO8
IO9
IO10
IO11
IO12
IO13
IO14
IO15
IO0_1 / DQ0_1
IO1_1 / DQ1_1
IO2_1 / DQ2_1
IO3_1 / DQ3_1
IO4_1 / DQ4_1
IO5_1 / DQ5_1
IO6_1 / DQ6_1
IO7_1 / DQ7_1
DQS_0_t
DQS_1_t
DQS_0_c
DQS_1_c
VSP0_0
VSP1_0
VSP2_0
VSP0_1
VSP1_1
VSP2_1
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
O
O
O
O
O
O
O
O
M
O
R
R
O
O
O
O
O
O
29
30
31
32
41
42
43
44
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
38
35
20
na
na
na
TSOP /
WSOP
Src Sync
x8
TSOP /
WSOP
Async
only x16
LGA
Async
only x8
LGA
Async
only x16
BGA-63
Async
only x8
BGA-63
Src Sync
x8
BGA-63
Async
only x16
BGA-100
Async
only x8
BGA-100
Src Sync
x8
29
30
31
32
41
42
43
44
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
35
na
na
na
38
na
20
na
na
na
29
30
31
32
41
42
43
44
26
27
28
33
40
45
46
47
na
na
na
na
na
na
na
na
na
na
na
na
38
35
20
na
na
na
G3
H2
J3
K2
L5
K6
J5
H6
na
na
na
na
na
na
na
na
G1
J1
L1
N3
N5
L7
J7
G7
na
na
na
na
na
na
na
na
na
na
G3
H2
J3
K2
L5
K6
J5
H6
G1
J1
L1
N3
N5
L7
J7
G7
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
H4
J4
K4
K5
K6
J7
K7
J8
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
G5
G8
G3
na
na
na
H4
J4
K4
K5
K6
J7
K7
J8
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
J5
na
H5
na
G5
G8
G3
na
na
na
H4
J4
K4
K5
K6
J7
K7
J8
H3
J3
H5
J5
H6
G6
H7
G7
na
na
na
na
na
na
na
na
na
na
na
na
G5
G8
G3
na
na
na
K2
N2
K3
N3
N8
K8
N9
K9
na
na
na
na
na
na
na
na
J2
M2
J3
M3
M8
J8
M9
J9
na
na
na
na
E7
E6
E4
D7
D6
D4
K2
N2
K3
N3
N8
K8
N9
K9
na
na
na
na
na
na
na
na
J2
M2
J3
M3
M8
J8
M9
J9
P5
N5
P4
N4
E7
E6
E4
D7
D6
D4
Table 3
25
Signal mappings: TSOP, LGA, BGA packages
2.6. CE_n Signal Requirements
If one or more LUNs are active and the host sets CE_n to one, then those operations continue
executing until completion at which point the target enters standby. After the CE_n signal is
transitioned to one, the host may drive a different CE_n signal to zero and begin operations on
another target. Note that if using a dual x8 package (e.g. BGA-100), then operations may
execute in parallel on two different CE_n’s if they are connected to different 8-bit data buses.
When SR[6] for a particular LUN is cleared to zero and the CE_n signal for the corresponding
target is cleared to zero, the host may only issue the Reset, Synchronous Reset, Read Status, or
Read Status Enhanced commands to that LUN.
2.6.1. Source Synchronous Data Interface Requirements
When using the source synchronous data interface, the following requirements shall be met if the
device does not support CLK being stopped during data input:
1. CLK shall only stop or start when CE_n is high.
When using the source synchronous data interface, the following requirements shall be met if the
device supports CLK being stopped during data input:
1. CLK shall only stop or start when either:
a. CE_n is high, or
b. CE_n is low and the bus state is data input
When using the source synchronous data interface, the following requirements shall always be
met:
1. CLK shall only change frequency when CE_n is high.
2. When CE_n is low, CLK shall maintain the same frequency.
3. CE_n shall only transition from one to zero when the CLK is stable and has a valid period
based on the timing mode selected.
4. The interface shall be in an idle state (see section 4.1.2) when CE_n changes value.
CE_n shall only transition when the following are true:
a. ALE and CLE are both cleared to zero, and
b. There is no data transfer on the DQ/DQS signals during the current clock period.
2.7. Absolute Maximum DC Ratings
Stresses greater than those listed in Table 4 may cause permanent damage to the device. This is
a stress rating only. Operation beyond the recommended operating conditions specified in Table
5 and the DC and operating characteristics listed in Table 9 and Table 10 is not recommended.
Except as defined in section 2.9, extended exposure beyond these conditions may affect device
reliability.
Table 4 defines the voltage on any pin relative to Vss and/or VssQ for devices based on their Vcc
and VccQ typical voltages.
26
Parameter
Symbol
Rating
Vcc = 3.3V and VccQ = 3.3V nominal
Vcc Supply Voltage
VCC
-0.6 to +4.6
Voltage Input
VIN
-0.6 to +4.6
VccQ Supply Voltage
VCCQ
-0.6 to +4.6
Vcc = 3.3V and VccQ = 1.8V nominal
Vcc Supply Voltage
VCC
-0.6 to +4.6
Voltage Input
VIN
-0.2 to +2.4
VccQ Supply Voltage
VCCQ
-0.2 to +2.4
Vcc = 1.8V and VccQ = 1.8V nominal
Vcc Supply Voltage
VCC
-0.2 to +2.4
Voltage Input
VIN
-0.2 to +2.4
VccQ Supply Voltage
VCCQ
-0.2 to +2.4
Table 4
Units
V
V
V
Absolute maximum DC ratings
2.8. Recommended DC Operating Conditions
Parameter
Supply voltage for 3.3V
devices
Supply voltage for 1.8V
devices
Supply voltage for 3.3V
I/O signaling
Supply voltage for 1.8V
I/O signaling
Ground voltage supply
Ground voltage supply
for I/O signaling
Table 5
Symbol
Min
Typ
Max
Units
VCC
2.7
3.3
3.6
V
VCC
1.7
1.8
1.95
V
2.7
3.3
3.6
V
1.7
1.8
1.95
V
0
0
0
V
0
0
0
V
VCCQ
(VCCQH)
VCCQ
(VCCQL)
VSS
VSSQ
Recommended DC operating conditions
2.8.1. I/O Power (VccQ) and I/O Ground (VssQ)
VccQ and Vcc may be distinct and unique voltages. VccQ shall be less than or equal to Vcc,
including during power-on ramp. The device shall support one of the following VccQ/Vcc
combinations:
 Vcc = 3.3V, VccQ = 3.3V
 Vcc = 3.3V, VccQ = 1.8V
 Vcc = 1.8V, VccQ = 1.8V
All parameters, timing modes, and other characteristics are relative to the supported voltage
combination.
If a device has the same Vcc and VccQ voltage levels, then VccQ and VssQ are not required to
be connected internal to the device. Specifically, the device may use Vcc and Vss exclusively as
the I/O and core voltage supply.
27
2.9. AC Overshoot/Undershoot Requirements
The device may have AC overshoot or undershoot from VccQ and VssQ levels. Table 6 defines
the maximum values that the AC overshoot or undershoot may attain. These values apply for
both 3.3V and 1.8V VccQ levels.
Maximum Value
Parameter
Peak amplitude allowed
for overshoot area
Peak amplitude allowed
for undershoot area
Maximum Overshoot
area above VccQ
Maximum Undershoot
area below VssQ
Table 6
> 133 MT/s
and
<= 166 MT/s
> 166 MT/s
and
<= 200 MT/s
Unit
<= 100 MT/s
> 100 MT/s
and
<= 133 MT/s
1
1
1
1
V
1
1
1
1
V
3
2.25
1.8
1.5
V-ns
3
2.25
1.8
1.5
V-ns
AC Overshoot/Undershoot Maximum Values
Figure 14 displays pictorially the parameters described in Table 6.
Maximum
Amplitude
Volts (V)
Overshoot Area
VccQ
VssQ
Maximum
Amplitude
Figure 14
2.10.
Undershoot Area
Overshoot/Undershoot Diagram
DC and Operating Characteristics
All operating current ratings in this section are specified per active logical unit (LUN). A LUN is
active when there is a command outstanding to it. All other current ratings in this section are
specified per LUN (regardless of whether it is active).
For high performance applications it may be desirable to draw increased current for ICC1-ICC4.
For these applications, the device may draw up to 100 mA per active LUN in both 3.3V and 1.8V
devices. Increased current may be used to improve sustained write performance.
28
All ICC measurements are measured with each Vcc pin decoupled with a 0.1 µF capacitor. The
ICC definition assumes outputs change between one and zero every other data cycle (once per
CLK period, every other DQS transition) for data signals. The test conditions and measurement
methodology for the ICC values is defined in Appendix D.
Parameter
Symbol
Array read current
Array program current
Array erase current
ICC1
ICC2
ICC3
I/O burst read current
4
ICC4R
I/O burst write current
Bus idle current
Test
Conditions
Refer to
Appendix D
ICC4W
ICC5
Standby current,
CMOS
ISB
Staggered
power-up current
1
IST
CE_n=VccQ0.2V,
WP_n=0V/VccQ
CE_n=VccQ0.2V
tRise = 1 ms
cLine = 0.1 µF
Min
Typ
Max
Units
-
-
50
50
50
mA
mA
mA
-
-
50
mA
-
-
50
mA
-
-
10
mA
-
-
50
µA
-
-
10
mA
NOTE:
1. Refer to Appendix C for an exception to the IST current requirement.
2. ICC1, ICC2, and ICC3 as listed in this table are active current values. For details on how to calculate
the active current from the measured values, refer to Appendix D.
3. During cache operations, increased ICC current is allowed while data is being transferred on the bus
and an array operation is ongoing. For a cached read this value is ICC1 + ICC4R; for a cached write
this value is ICC2(active) + ICC4W.
4. For ICC4R the test conditions in Appendix D specify IOUT = 0 mA and requires static outputs with no
output switching. When outputs are not static, additional VccQ current will be drawn that is highly
dependent on system configuration. IccQ may be calculated for each output pin assuming 50% data
switching as (IccQ = 0.5 * CL * VccQ * frequency), where CL is the capacitive load.
Table 7
DC and Operating Conditions for raw NAND, measured on Vcc rail
29
Parameter
Symbol
Array read current
Array program current
Array erase current
I/O burst read current
I/O burst write current
Bus idle current
Standby current,
CMOS
Staggered
power-up current
Min
Typ
Max
Per LUN
Controller
Per LUN
Per LUN
Controller
Per LUN
Per LUN
Controller
Per LUN
-
-
85
200
15
85
75
15
85
75
15
LUN
-
-
50
Controller
-
-
50
ICC4W
LUN
-
-
50
ICCQ4W
ICC5
ICCQ5
ICCQ5
ISB
Controller
-
-
50
Per LUN
Controller
Per LUN
Per LUN
-
-
45
10
15
50
Controller
-
-
1000
Per LUN
-
-
10
Controller
-
-
20
ICC1
ICCQ1
ICCQ1
ICC2
ICCQ2
ICCQ2
ICC3
ICCQ3
ICCQ3
4
ICC4R
4
ICCQ4R
ISBQ
1
IST
ISTQ
Test
Conditions
Refer to
Appendix D
CE_n=VccQ0.2V,
WP_n=0V/VccQ
CE_n=VccQ0.2V
tRise = 1 ms
cLine = 0.1 µF
Units
mA
mA
mA
mA
mA
mA
µA
mA
NOTE:
1. Refer to Appendix C for an exception to the IST current requirement.
2. ICC1, ICC2, and ICC3 as listed in this table are active current values. For details on how to calculate the active
current from the measured values, refer to Appendix D.
3. During cache operations, increased ICC current is allowed while data is being transferred on the bus and an array
operation is ongoing. For a cached read this value is ICC1 + ICC4 R on Vcc and ICCQ1 on VccQ; for a cached write
this value is ICC2(active) + ICC4W on Vcc, and ICCQ2 on VccQ.
4. For ICC4R the test conditions in Appendix D specify IOUT = 0 mA and requires static outputs with no output
switching. When outputs are not static, additional VccQ current will be drawn that is highly dependent on system
configuration. IccQ may be calculated for each output pin assuming 50% data switching as (IccQ = 0.5 * C L * VccQ
* frequency), where CL is the capacitive load.
Table 8
DC and Operating Conditions for EZ NAND, measured on Vcc or VccQ rail
The maximum leakage current requirements (ILI and ILO) in Table 9 and Table 10 are tested
across the entire allowed VccQ range, specified in Table 5.
DC signal specifications apply to the following signals and only when using the source
synchronous data interface: CLK, DQ[7:0], DQS, ALE, CLE, and W/R_n. For all signals in
asynchronous and all other signals in source synchronous, the AC signal specification shall be
met. For signals where DC signal specifications apply, the transition times are measured between
VIL (DC) and VIH (AC) for rising input signals and between VIH (DC) and VIL (AC) for falling
input signals. The receiver will effectively switch as a result of the signal crossing the AC input
level and remain in that state as long as the signal does not ring back above (below) the DC input
LOW (HIGH) level.
The parameters in Table 9 and Table 10 apply to power-on default values in the device. If I/O
drive strength settings or other device settings are changed, these values may be modified. The
output characteristics for a device that supports driver strength settings (as indicated in the
parameter page) are specified in the impedance tables (Table 28 and Table 29).
30
Parameter
Symbol
Test Conditions
Min
Typ
Max
Units
Standby current,
CE_n=VccQ-0.2V,
ISBQ
25
µA
CMOS
WP_n=0V/VccQ
Input leakage current
ILI
VIN=0V to VccQ
+-10
µA
Output leakage
ILO
VOUT=0V to VccQ
+-10
µA
current
DC Input high
VIH (DC)
VccQ * 0.7
VccQ + 0.3
V
voltage
AC Input high
VIH (AC)
VccQ * 0.8
VccQ + 0.3
V
voltage
DC Input low voltage
VIL (DC)
-0.3
VccQ * 0.3
V
AC Input low voltage
VIL (AC)
-0.3
VccQ * 0.2
V
VccQ *
1
VOH
IOH=-400 µA
V
Output high voltage
0.67
1
VOL
IOL=2.1 mA
0.4
V
Output low voltage
Output low current
IOL(R/B_
VOL=0.4 V
8
10
mA
(R/B_n)
n)
NOTE:
1. VOH and VOL defined in this table shall only apply to devices in asynchronous mode that do not
support driver strength settings. If driver strength settings are supported then Table 28 shall be
used to derive the output driver impedance values.
Table 9
Parameter
DC and Operating Conditions for VccQ of 3.3V, measured on VccQ rail
Symbol
Test Conditions
Min
Typ
Max
Units
Standby current,
CE_n=VccQ-0.2V,
ISBQ
25
µA
CMOS
WP_n=0V/VccQ
Input leakage current
ILI
VIN=0V to VccQ
+-10
µA
Output leakage
ILO
VOUT=0V to VccQ
+-10
µA
current
DC Input high
VIH (DC)
VccQ * 0.7
VccQ+0.3
V
voltage
AC Input high
VIH (AC)
VccQ * 0.8
VccQ+0.3
V
voltage
DC Input low voltage
VIL (DC)
-0.3
VccQ * 0.3
V
AC Input low voltage
VIL (AC)
-0.3
VccQ * 0.2
V
1
VOH
IOH=-100
µA
VccQ
–
0.1
V
Output high voltage
1
VOL
IOL=100 µA
0.1
V
Output low voltage
Output low current
IOL(R/B_n)
VOL=0.2 V
3
4
mA
(R/B_n)
NOTE:
1. VOH and VOL defined in this table shall only apply to devices in asynchronous mode that do not
support driver strength settings. If driver strength settings are supported then Table 29 shall be
used to derive the output driver impedance values.
Table 10
2.11.
DC and Operating Conditions for VccQ of 1.8V, measured on VccQ
rail
Calculating Pin Capacitance
To calculate the pin capacitance for all loads on the I/O bus, the host should utilize the reported
pin capacitance per target in Read Parameter Page (refer to section 5.7). The maximum
31
capacitance may be used, or the typical capacitance if provided by the device may be used. The
algorithm to use is:
PinCapacitance = 0;
for (target = 0; target < TotalTargets; target++)
PinCapacitance += GetCapacitanceFromRPP(target);
This methodology will calculate an accurate maximum or typical pin capacitance, respectively,
accounting for all targets present.
2.12.
Staggered Power-up
Subsystems that support multiple Flash devices may experience power system design issues
related to the current load presented during the power-on condition. To limit the current load
presented to the host at power-on, all devices shall support power-up in a low-power condition.
Until a Reset (FFh) command is received by the target after power-on, the target shall not draw
more than IST of current per LUN and ISTQ (if present for devices that support EZ NAND). For
example, a target that contains 4 LUNs may draw up to 40 mA of current until a Reset (FFh)
command is received after power-on.
This value is measured with a nominal rise time (tRise) of 1 millisecond and a line capacitance
(cLine) of 0.1 µF. The measurement shall be taken with 1 millisecond averaging intervals and
shall begin after Vcc reaches Vcc_min and VccQ reaches VccQ_min.
2.13.
Power Cycle Requirements
As part of a power cycle, the host shall hold both the Vcc and VccQ voltage levels below 100 mV
for a minimum time of 100 ns. If these requirements are not met as part of a power cycle
operation, the device may enter an indeterminate state.
2.14.
Independent Data Buses
There may be two independent 8-bit data buses in some ONFI packages (i.e. the LGA and the
100-ball BGA package). For packages that support either two independent data buses or a
single data bus (e.g. LGA-52) then CE0_n and CE2_n shall use the same pins as the first data
bus CE_n pins (marked as CE0_0_n and CE1_0_n) and CE1_n and CE3_n shall use the same
pins as the second data bus CE_n pins (marked as CE0_1_n and CE1_1_n). Note that CE0_n,
CE1_n, CE2_n, and CE3_n may all use the first data bus and the first set of control signals
(RE0_n, CLE0_n, ALE0_n, WE0_n, and WP0_n) if the device does not support independent data
buses.
Implementations may tie the data lines and control signals (RE_n, CLE, ALE, WE_n, WP_n, and
DQS) together for the two independent 8-bit data buses externally to the device.
2.15.
Bus Width Requirements
All targets per device shall use the same data bus width. All targets shall either have an 8-bit bus
width or a 16-bit bus width. Note that devices that support the source synchronous interface shall
have an 8-bit bus width.
When the host supports a 16-bit bus width, only data is transferred at the 16-bit width. All
address and command line transfers shall use only the lower 8-bits of the data bus. During
32
command transfers, the host may place any value on the upper 8-bits of the data bus. During
address transfers, the host shall set the upper 8-bits of the data bus to 00h.
2.16.
2.16.1.
Ready/Busy (R/B_n) Requirements
Power-On Requirements
Once VCC and VccQ reach the VCC minimum and VccQ minimum values, respectively, listed in
Table 5 and power is stable, the R/B_n signal shall be valid after RB_valid_Vcc and shall be set
to one (Ready) within RB_device_ready, as listed in Table 11. R/B_n is undefined until 50 µs
has elapsed after VCC has started to ramp. The R/B_n signal is not valid until both of these
conditions are met.
Parameter
RB_valid_Vcc
Raw NAND
10 µs
EZ NAND
250 µs
RB_device_ready
1 ms
2 ms
Table 11
R/B_n Power-on Requirements
During power-on, VccQ shall be less than or equal to Vcc at all times. Figure 15 shows VccQ
ramping after Vcc, however, they may ramp at the same time.
50 µs
(min)
VccQ
RB_valid_Vcc
Vcc
>= 0 µs
(min)
R/B_n
Vcc ramp
starts
RB_device_ready
VccQ =
VccQ_min
Reset (FFh)
is issued
Vcc =
Vcc_min
Figure 15
Undefined
R/B_n Power-On Behavior
Ready/Busy is implemented as an open drain circuit, thus a pull-up resistor shall be used for
termination. The combination of the pull-up resistor and the capacitive loading of the R/B_n
circuit determines the rise time of R/B_n.
33
2.16.2.
R/B_n and SR[6] Relationship
R/B_n shall reflect the logical AND of the SR[6] (Status Register bit 6) values for all LUNs on the
corresponding target. For example, R/B3_n is the logical AND of the SR[6] values for all LUNs
on CE3_n. Thus, R/B_n reflects whether any LUN is busy on a particular target.
2.17.
Write Protect
When cleared to zero, the WP_n signal disables Flash array program and erase operations. This
signal shall only be transitioned while there are no commands executing on the device. After
modifying the value of WP_n, the host shall not issue a new command to the device for at least
tWW delay time.
Figure 16 describes the tWW timing requirement, shown with the start of a Program command.
The transition of the WP_n signal is asynchronous and unrelated to any CLK transition in the
source synchronous data interface. The bus shall be idle for tWW time after WP_n transitions
from zero to one before a new command is issued by the host, including Program. The bus shall
be idle for tWW time after WP_n transitions from one to zero before a new command is issued by
the host.
As defined for
Page Program
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
80h
C1
C2
R1
R2
R3
tWW
DQx
SR[6]
WP_n
Bus shall
be idle
Figure 16
Write Protect timing requirements, example
34
3. Memory Organization
Figure 17 shows an example of a Target memory organization. In this case, there are two logical
units where each logical unit has two planes.
Page 0
Page 0
Page 0
Page 1
Page 1
Page 1
Page 1
Page P
Page P
Page P
Page P
Block 0
Block 1
Block 0
Block 1
Page 0
Page 0
Page 0
Page 0
Page 1
Page 1
Page 1
Page 1
Page P
Page P
Page P
Page P
Block 2
Block 3
Block 2
Block 3
Page 0
Page 0
Page 0
Page 0
Page 1
Page 1
Page 1
Page 1
Page P
Page P
Page P
Page P
Logical Unit 1
Logical Unit 0
Page 0
Block B
Block B+1
Block B
Block B+1
Page Register
Page Register
Page Register
Page Register
Plane
Address 0
Plane
Address 1
Plane
Address 0
Plane
Address 1
Figure 17
Target memory organization
A device contains one or more targets. A target is controlled by one CE_n signal. A target is
organized into one or more logical units (LUNs).
A logical unit (LUN) is the minimum unit that can independently execute commands and report
status. Specifically, separate LUNs may operate on arbitrary command sequences in parallel.
For example, it is permissible to start a Page Program operation on LUN 0 and then prior to the
operation’s completion to start a Read command on LUN 1. See multiple LUN operation
restrictions in section 3.1.3. A LUN contains at least one page register and a Flash array. The
number of page registers is dependent on the number of multi-plane operations supported for that
LUN. The Flash array contains a number of blocks.
35
A block is the smallest erasable unit of data within the Flash array of a LUN. There is no
restriction on the number of blocks within the LUN. A block contains a number of pages.
A page is the smallest addressable unit for read and program operations. A page consists of a
number of bytes or words. The number of user data bytes per page, not including the spare data
area, shall be a power of two. The number of pages per block shall be a multiple of 32.
Each LUN shall have at least one page register. A page register is used for the temporary storage
of data before it is moved to a page within the Flash array or after it is moved from a page within
the Flash array. If EZ NAND is supported then a buffer exists in the EZ NAND controller that
provides for temporary storage of data that may then be transferred to or from the page register
within each LUN.
The byte or word location within the page register is referred to as the column.
There are two mechanisms to achieve parallelism within this architecture. There may be multiple
commands outstanding to different LUNs at the same time. To get further parallelism within a
LUN, multi-plane operations may be used to execute additional dependent operations in parallel.
3.1. Addressing
There are two address types used: the column address and the row address. The column
address is used to access bytes or words within a page, i.e. the column address is the byte/word
offset into the page. The least significant bit of the column address shall always be zero in the
source synchronous data interface, i.e. an even number of bytes is always transferred. The row
address is used to address pages, blocks, and LUNs.
When both the column and row addresses are required to be issued, the column address is
always issued first in one or more 8-bit address cycles. The row addresses follow in one or more
8-bit address cycles. There are some functions that may require only row addresses, like Block
Erase. In this case the column addresses are not issued.
For both column and row addresses the first address cycle always contains the least significant
address bits and the last address cycle always contains the most significant address bits. If there
are bits in the most significant cycles of the column and row addresses that are not used then
they are required to be cleared to zero.
The row address structure is shown in Figure 18 with the least significant row address bit to the
right and the most significant row address bit to the left.
MSB
LUN Address
LSB
Block Address
Figure 18
Page Address
Row Address Layout
The number of blocks and number of pages per block is not required to be a power of two. In the
case where one of these values is not a power of two, the corresponding address shall be
rounded to an integral number of bits such that it addresses a range up to the subsequent power
of two value. The host shall not access upper addresses in a range that is shown as not
supported. For example, if the number of pages per block is 96, then the page address shall be
rounded to 7 bits such that it can address pages in the range of 0 to 127. In this case, the host
shall not access pages in the range from 96 to 127 as these pages are not supported.
36
The page address always uses the least significant row address bits. The block address uses the
middle row address bits and the LUN address uses the most significant row address bit(s).
3.1.1. Multi-plane Addressing
The multi-plane address comprises the lowest order bits of the block address as shown in Figure
19. The following restrictions apply to the multi-plane address when executing a multi-plane
command sequence on a particular LUN:
 The plane address bit(s) shall be distinct from any other multi-plane operation in the
multi-plane command sequence.
 The page address shall be the same as any other multi-plane operations in the multiplane command sequence.
MSB
LSB
LUN Address
Block Address
Page Address
Plane
Address bit(s)
Figure 19
Plane Address Location
3.1.1.1. Multi-plane Block Address Restrictions
The device may indicate multi-plane block address restrictions. The specific cases are:
 No restriction: All block address bits may be different between two plane addresses.
 Full restriction: All block address bits (other than the plane address bits) shall be the
same between two plane addresses.
 Lower bit XNOR restriction: If the XNOR of the lowest plane address bits (bit 0) is
one between two plane addresses, then there is a full restriction between these two
plane addresses. If the XNOR of the lower plane address bits is zero between two
plane addresses, then there is no restriction between these two plane addresses.
Table 12 illustrates the three types of restrictions for a four plane operation.
Restriction Type
No restriction
Plane Address
0
Block A
Plane Address
1
Block B
Plane Address
2
Block C
Plane Address
3
Block D
XNOR restriction
Block A
Block B
Block A+2
Block B+2
Full restriction
Block A
Block A+1
Block A+2
Block A+3
Table 12
Four plane address restriction
Table 13 describes whether there is a lower bit XNOR restriction between two plane addresses A
and B, based on their plane address bits for a 4 plane implementation. If there is a lower bit
XNOR restriction, then the block addresses (other than the plane address bits) shall be the same
between multi-plane addresses A and B.
37
Multi-plane
Address bits A
00b
Multi-plane
Address bits B
01b
Lower Bit XNOR
0 XNOR 1 = 0
XNOR Restriction
Between A and B
No
00b
10b
0 XNOR 0 = 1
Yes
00b
11b
0 XNOR 1 = 0
No
01b
10b
1 XNOR 0 = 0
No
01b
11b
1 XNOR 1 = 1
Yes
10b
11b
0 XNOR 1 = 0
No
Table 13
4-way lower bit XNOR restriction
3.1.2. Logical Unit Selection
Logical units within one target share a single data bus with the host. The host shall ensure that
only one LUN is selected for data output to the host at any particular point in time to avoid bus
contention.
The host selects a LUN for future data output by issuing a Read Status Enhanced command to
that LUN. The Read Status Enhanced command shall deselect the output path for all LUNs that
are not addressed by the command. The page register selected for output within the LUN is
determined by the previous Read (Cache) commands issued, and is not impacted by Read
Status Enhanced.
3.1.3. Multiple LUN Operation Restrictions
LUNs are independent entities. A multiple LUN operation is one in which two or more LUNs are
simultaneously processing commands. This implies that R/B_n is cleared to zero when the
subsequent LUN operation is issued.
When a Page Program command (80h) is issued on any LUN that is not preceded by an 11h
command, all idle LUNs may clear their page registers if the program page register clear
enhancement is not supported or enabled. Thus, the host should not begin a Page Program
command on a LUN while a Read Page operation is either ongoing or has completed but the data
has not been read from another LUN, as the contents of the page register for the Read operation
are lost. A Read Page can be issued to one LUN while a Page Program is ongoing within a
second LUN without any restriction. If the program page register clear enhancement is enabled,
this restriction does not apply.
When issuing a Page Program command (80h), the host should not select another LUN until after
all data has been input and a 10h or 15h command has been issued. In the case of multi-plane
operations, all data input for all multi-plane addresses should be completed prior to selecting
another LUN.
When issuing Reads to multiple LUNs, the host shall take steps to avoid issues due to column
address corruption. Specifically, if the column addresses in Reads issued to multiple LUNs are
different, then the host shall issue a Change Read Column before starting to read out data from a
newly selected LUN. If the column addresses are the same, then no Change Read Column is
necessary.
If a multiple LUN operation has been issued, then the next status command issued shall be Read
Status Enhanced. Read Status Enhanced causes LUNs that are not selected to turn off their
38
output buffers. This ensures that only the LUN selected by the Read Status Enhanced command
responds to a subsequent data output cycle. After a Read Status Enhanced command has been
completed, the Read Status command may be used until the next multiple LUN operation is
issued.
When the host has issued Read Page commands to multiple LUNs at the same time, the host
shall issue Read Status Enhanced before reading data from either LUN. This ensures that only
the LUN selected by the Read Status Enhanced command responds to a data output cycle after
being put in data output mode with a 00h command, and thus avoiding bus contention. If the host
issues a Change Read Column (Enhanced) for any LUN that Read Page commands are
outstanding for, then the host shall issue a Change Read Column (Enhanced) prior to transferring
data from any subsequent LUN that is part of the multiple LUN read sequence. An example
sequence is shown below:
1) Read Page command issued to LUN 0
2) Read Page command issued to LUN 1
3) Read Status Enhanced selects LUN 0
4) Change Read Column (Enhanced) issued to LUN 0 (if needed by host)
5) Data transferred from LUN 0
6) Read Status Enhanced selects LUN 1
7) Change Read Column (Enhanced) issued to LUN 1 (required if action 4 taken)
8) Data transferred from LUN 1
The host may substitute Change Read Column Enhanced for the Read Status Enhanced /
Change Read Column sequence if all LUNs are not busy.
3.2. Factory Defect Mapping
The Flash array is not presumed to be pristine, and a number of defects may be present that
renders some blocks unusable. Block granularity is used for mapping factory defects since those
defects may compromise the block erase capability.
3.2.1. Device Requirements
If a block is defective and 8-bit data access is used, the manufacturer shall mark the block as
defective by setting the first byte in the defect area, as shown in Figure 20, of the first or last page
of the defective block to a value of 00h. If a block is defective and 16-bit data access is used, the
manufacturer shall mark the block as defective by setting the first word in the defect area of the
first or last page of the defective block to a value of 0000h.
# of data
bytes - 1
Byte
0 1 2
Area
Figure 20
# of data bytes +
# of spare bytes - 1
# of data
bytes
…
…
n/a
Defect Area
Area marked in factory defect mapping
39
3.2.2. Host Requirements
The host shall not erase or program blocks marked as defective by the manufacturer, and any
attempt to do so yields indeterminate results.
Figure 21 outlines the algorithm to scan for factory mapped defects. This algorithm should be
performed by the host to create the initial bad block table prior to performing any erase or
programming operations on the target. The initial state of all pages in non-defective blocks is FFh
(or FFFFh for 16-bit access) for all page addresses, although some bit errors may be present if
they are correctable via the required ECC reported to the host. A defective block is indicated by a
byte value equal to 00h for 8-bit access or a word value equal to 0000h for 16-bit access being
present at the first byte/word location in the defect area of either the first page or last page of the
block. The host shall check the first byte/word of the defect area of both the first and last past
page of each block to verify the block is valid prior to any erase or program operations on that
block.
NOTE: Over the lifetime use of a NAND device, the defect area of defective blocks may
encounter read disturbs that cause values to change. The manufacturer defect markings may
change value over the lifetime of the device, and are expected to be read by the host and used to
create a bad block table during initial use of the part.
for (i=0; i<NumLUNs; i++)
{
for (j=0; j<BlocksPerLUN; j++)
{
Defective=FALSE;
ReadPage(lun=i; block=j; page=0; DestBuff=Buff);
if (Buff[PageSize] == 00h) // Value checked for is 0000h for 16-bit access
Defective=TRUE;
ReadPage(lun=i; block=j; page=PagesPerBlock-1; DestBuff=Buff);
if (Buff[PageSize] == 00h) // Value checked for is 0000h for 16-bit access
Defective=TRUE;
if (Defective)
MarkBlockDefective(lun=i; block=j);
}
}
Figure 21
Factory defect scanning algorithm
3.3. Extended ECC Information Reporting
The device may report extended ECC information in the extended parameter page. The required
ECC correctability is closely related to other device parameters, like the number of valid blocks
and the number of program/erase cycles supported. Extended ECC information allows the
device to specify multiple valid methods for using the device.
40
Table 14 defines the extended ECC information block.
Byte
0
1
2-3
4-5
6-7
Table 14
Definition
Number of bits ECC correctability
Codeword size
Bad blocks maximum per LUN
Block endurance
Reserved
Extended ECC Information Block Definition
The definition of each field follows in the subsequent sections.
3.3.1. Byte 0: Number of bits ECC correctability
This field indicates the number of bits that the host should be able to correct per codeword. The
codeword size is reported in byte 1. With this specified amount of error correction by the host,
the target shall achieve the block endurance specified in bytes 4-5. When the specified amount
of error correction is applied by the host and the block endurance is followed, then the maximum
number of bad blocks specified in bytes 2-3 shall not be exceeded by the device. All used bytes
in the page shall be protected by host controller ECC including the spare bytes if the ECC
requirement reported in byte 0 has a value greater than zero.
When this value is cleared to zero, the target shall return valid data if the ECC Information Block
is valid (the Codeword size is non-zero).
3.3.2. Byte 1: Codeword size
The number of bits of ECC correctability specified in byte 0 is based on a particular ECC
codeword size. The ECC codeword size is specified in this field as a power of two. The
minimum value that shall be reported is 512 bytes (a value of 9).
If a value of 0 is reported then this ECC Information Block is invalid and should not be used.
3.3.3. Byte 2-3: Bad blocks maximum per LUN
This field contains the maximum number of blocks that may be defective at manufacture and over
the life of the device per LUN. The maximum rating assumes that the host is following the block
endurance requirements and the ECC requirements reported in this extended ECC information
block.
3.3.4. Byte 4-5: Block endurance
This field indicates the maximum number of program/erase cycles per addressable page/block.
This value assumes that the host is using the ECC correctability reported in byte 0.
The block endurance is reported in terms of a value and a multiplier according to the following
multiplier
equation: value x 10
. Byte 4 comprises the value. Byte 5 comprises the multiplier. For
example, a block endurance of 75,000 cycles would be reported as a value of 75 and a multiplier
3
of 3 (75 x 10 ). The value field shall be the smallest possible; for example 100,000 shall be
5
reported as a value of 1 and a multiplier of 5 (1 x 10 ).
41
3.4. Discovery and Initialization
3.4.1. CE_n Discovery
There may be up to four chip enable (CE_n) signals on a package, one for each separately
addressable target. To determine the targets that are connected, the procedure outlined in this
section shall be followed for each distinct CE_n signal. CE_n signals shall be used sequentially
on the device; CE0_n is always connected and CE_n signals shall be connected in a numerically
increasing order. The host shall attempt to enumerate targets connected to all host CE_n
signals.
The discovery process for a package that supports independent dual data buses includes
additional steps to determine which data bus the target is connected to. The LGA and 100-ball
BGA package with 8-bit data access are the packages within ONFI that have a dual data bus
option.
3.4.1.1. Single Data Bus Discovery
The CE_n to test is first pulled low by the host to enable the target if connected, while all other
CE_n signals are pulled high. The host shall then issue the Reset (FFh) command to the target.
Following the reset, the host should then issue a Read ID command to the target. If the ONFI
signature is returned by the Read ID command with address 20h, then the corresponding target is
connected. If the ONFI signature is not returned or any step in the process encountered an
error/timeout, then the CE_n is not connected and no further use of that CE_n signal shall be
done.
3.4.1.2. Dual Data Bus Discovery
The CE_n to test is first pulled low by the host to enable the target if connected, while all other
CE_n signals are pulled high. The host shall then issue the Reset (FFh) command to the target.
Following the reset, the host should then issue a Read ID command with address 20h to the
target. If the ONFI signature is returned by the Read ID command, then the corresponding target
is connected.
If the ONFI signature is not returned (or any step in the process encountered an error/timeout),
then the second 8-bit data bus should be probed. The host shall issue the Reset (FFh) command
to the target using the second 8-bit data bus. Following the reset, the host should then issue a
Read ID command with address 20h to the target on the second 8-bit data bus. If the ONFI
signature is returned by the Read ID command, then the corresponding target is connected and is
using the second 8-bit data bus. After discovering that the target is using the second 8-bit data
bus, all subsequent commands to that target shall use the second 8-bit data bus including Read
Parameter Page.
If after this point a valid ONFI signature is not discovered or further errors were encountered, then
the CE_n is not connected and no further use of that CE_n signal shall be done.
3.4.2. Target Initialization
To initialize a discovered target, the following steps shall be taken. The initialization process
should be followed for each connected CE_n signal, including performing the Read Parameter
Page (ECh) command for each target. Each chip enable corresponds to a unique target with its
own independent properties that the host shall observe and subsequently use.
The host should issue the Read Parameter Page (ECh) command. This command returns
information that includes the capabilities, features, and operating parameters of the device.
42
When the information is read from the device, the host shall check the CRC to ensure that the
data was received correctly and without error prior to taking action on that data.
If the CRC of the first parameter page read is not valid (refer to section 5.7.1.4), the host should
read redundant parameter page copies. The host can determine whether a redundant parameter
page is present or not by checking if the first four bytes contain at least two bytes of the
parameter page signature. If the parameter page signature is present, then the host should read
the entirety of that redundant parameter page. The host should then check the CRC of that
redundant parameter page. If the CRC is correct, the host may take action based on the contents
of that redundant parameter page. If the CRC is incorrect, then the host should attempt to read
the next redundant parameter page by the same procedure.
The host should continue reading redundant parameter pages until the host is able to accurately
reconstruct the parameter page contents. The host may use bit-wise majority or other ECC
techniques to recover the contents of the parameter page from the parameter page copies
present. When the host determines that a parameter page signature is not present (refer to
section 5.7.1.1), then all parameter pages have been read.
The Read ID and Read Parameter Page commands only use the lower 8-bits of the data bus.
The host shall not issue commands that use a word data width on x16 devices until the host
determines the device supports a 16-bit data bus width in the parameter page.
After successfully retrieving the parameter page, the host has all information necessary to
successfully communicate with that target. If the host has not previously mapped defective block
information for this target, the host should next map out all defective blocks in the target. The
host may then proceed to utilize the target, including erase and program operations.
43
4. Data Interface and Timing
4.1. Data Interface Types
ONFI supports two different data interface types: asynchronous and source synchronous. The
asynchronous data interface is the traditional NAND interface that uses RE_n to latch data read,
WE_n to latch data written, and does not include a clock. The source synchronous data interface
includes a clock that indicates where commands and addresses should be latched and a data
strobe that indicates where data should be latched.
On power-up, the device shall operate in asynchronous data interface timing mode 0. After the
host determines that the source synchronous data interface is supported in the parameter page,
the host may select a source synchronous timing mode by using Set Features with a Feature
Address of 01h. Refer to section 5.26.1.
The source synchronous data interface uses a DDR protocol. Thus, an even number of bytes is
always transferred. The least significant bit of the column address shall always be zero in the
source synchronous data interface. If the least significant bit of the column address is set to one
in the source synchronous data interface then the results are indeterminate.
4.1.1. Signal Function Reassignment
The function of some signals is different when using the asynchronous data interface versus
when using the source synchronous data interface. When source synchronous is selected, the
function of RE_n and WE_n is modified and DQS is enabled.
WE_n becomes the clock signal (CLK) when in source synchronous mode. CLK shall be enabled
with a valid clock period whenever a command cycle, address cycle, or data cycle is occurring.
CLK shall maintain the same frequency while CE_n is driven to zero. Refer to section 4.2.3.
RE_n becomes the write/read direction signal (W/R_n) when in source synchronous mode. This
signal indicates the owner of the DQ data bus and the DQS signal. The host shall only transition
W/R_n when ALE and CLE are latched to zero.
The I/O bus is renamed to the DQ bus in the source synchronous interface.
A strobe signal for the DQ data bus is used in source synchronous mode, called DQS (DQ
strobe). DQS is bi-directional and is used for all data transfers. DQS is not used for command or
address cycles. The latching edge of DQS is center aligned to the valid data window for data
transfers from the host to the device (writes). The latching edge of DQS is aligned to the
transition of the DQ bus for data transfers from the device to the host (reads). DQS should be
pulled high by the host and shall be ignored by the device when operating in the asynchronous
data interface.
When W/R_n changes from one to zero, the host shall tri-state the DQ bus and the DQS signal
and then the device shall drive DQS to zero. When W/R_n changes from zero to one, the device
shall tri-state the DQ bus and the DQS signal. DQS and the DQ bus should be driven high by the
host during idle when no data operations are outstanding and W/R_n is set to one. There is a
turn-around time whenever W/R_n changes its value where the DQS signal is tri-stated (as
neither the host nor the device is driving the signal), see section 4.3.2.6.
44
Symbol
Source
synchronous
ALE
CE_n
CLE
DQ[7:0]
DQS
W/R_n
CLK
WP_n
R/B_n
Asynchronous
ALE
CE_n
CLE
I/O[7:0]
—
RE_n
WE_n
WP_n
R/B_n
Table 15
Type
Input
Input
Input
I/O
I/O
Input
Input
Input
Output
Description
Address latch enable
Chip enable
Command latch enable
Data inputs/outputs
Data strobe
Read enable / (Write / Read_n direction)
Write enable / Clock
Write protect
Ready / Busy_n
Signal Reassignments between Data Interface Types
4.1.2. Bus State
ALE and CLE are used to determine the current bus state in asynchronous and source
synchronous data interfaces. Table 16 describes the bus state for asynchronous. Note that in
asynchronous the value 11b for ALE/CLE is undefined.
CE_n
ALE
CLE
WE_n
RE_n
1
0
0
0
0
0
0
X
0
0
1
0
0
1
X
0
1
0
0
0
1
X
1
0
0
0
1
X
X
1
1
1
1
0
X
Table 16
Asynchronous Bus
State
Standby
Idle
Command cycle
Address cycle
Data input cycle
Data output cycle
Undefined
ALE/CLE value and asynchronous bus state
Table 17 describes the bus state for source synchronous. In source synchronous the value 11b
for ALE/CLE is used for data transfers. The bus state lasts for an entire CLK period, starting with
the rising edge of CLK. Thus, for data cycles there are two data input cycles or two data output
cycles per bus state. The idle bus state is used to terminate activity on the DQ bus after a
command cycle, an address cycle, or a stream of data.
The value of CE_n shall only change when the source synchronous bus state is idle (i.e. ALE and
CLE are both cleared to zero) and no data is being transmitted during that clock period.
45
CE_n
ALE
CLE
W/R_n
CLK
1
X
X
X
0
0
0
1
0
0
0
0
0
0
1
1
0
1
0
1
0
1
1
1
0
1
1
0
0
0
1
0
0
1
0
0
X
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Rising edge to
rising edge
Source
Synchronous Bus
State
Standby
1
Idle
Bus Driving
1
Command cycle
Address cycle
Data input cycle
2
Data output cycle
2
Reserved
Reserved
NOTE:
1. When W/R_n is cleared to ‘0’, the device is driving the DQ bus and DQS
signal. When W/R_n is set to ‘1’ then the DQ and DQS signals are not driven
by the device.
2. There are two data input/output cycles from the rising edge of CLK to the next
rising edge of CLK.
Table 17
ALE/CLE value and source synchronous bus state
4.1.2.1. Pausing Data Input/Output
The host may pause data input or data output by inserting Idle cycles.
In the asynchronous data interface, pausing data input or data output is done by maintaining
WE_n or RE_n at a value of one, respectively.
In the source synchronous data interface, pausing data input or data output is done by clearing
ALE and CLE both to zero. The host may continue data transfer by setting ALE and CLE both to
one after the applicable tCAD time has passed.
4.1.3. Source Synchronous and Repeat Bytes
The source synchronous interface uses DDR to achieve a high data transfer rate. However,
certain configuration and settings commands are not often used and do not require a high data
transfer rate. Additionally, these commands typically are not serviced by the pipeline used for
data transfers.
To avoid adding unnecessary complexity and requirements to implementations for these
commands, the data is transferred using single data rate. Specifically, the same data byte is
repeated twice and shall conform to the timings required for the source synchronous data
interface. The data pattern in these cases is D0 D0 D1 D1 D2 D2 etc. The receiver (host or device)
shall only latch one copy of each data byte. CLK should not be stopped during data input for
these commands. The receiver is not required to wait for the repeated data byte before
beginning internal actions.
46
The commands that repeat each data byte twice in the source synchronous data interface are:
Set Features, Read ID, Get Features, Read Status, and Read Status Enhanced.
4.1.4. Data Interface / Timing Mode Transitions
4.1.4.1. Asynchronous to Source Synchronous
To transition from an asynchronous timing mode to a source synchronous timing mode, the
procedure described in this section shall be followed. The Set Features command is used to
change the data interface and timing mode. The Set Features command (EFh), Feature Address,
and the four parameters are entered using the previously selected timing mode in the
asynchronous data interface. When issuing the Set Features command, the host shall drive the
DQS signal high. After the fourth parameter, P4, is entered until the tITC time has passed the
host shall not issue any commands to the device.
Prior to issuing any new commands to the device, the host shall transition CE_n high. When
CE_n is high, the host selects the new CLK rate. After issuing the Set Features command and
prior to transitioning CE_n high, the host shall observe the following requirements:
 ALE and CLE shall be cleared to zero
 RE_n / W/R_n shall be set to one
 WE_n / CLK shall be set to one
 DQS shall be set to one
4.1.4.2. Source Synchronous to Source Synchronous
To transition from a source synchronous timing mode to another source synchronous timing
mode, the procedure described in this section shall be followed. The Set Features command is
used to change the timing mode. The Set Features command (EFh), Feature Address, and the
four parameters are entered using the previously selected timing mode in the source
synchronous data interface. After the fourth parameter, P4, is entered until the tITC time has
passed the host shall not issue any commands to the device.
Prior to issuing any new commands to the device, the host shall transition CE_n high. When
CE_n is high, the host selects the new CLK rate. After issuing the Set Features command and
prior to transitioning CE_n high, the host shall observe the following requirements:
 ALE and CLE shall be cleared to zero
 W/R_n shall be set to one
 CLK shall continue running at the previously selected speed grade
4.1.4.3. Source Synchronous to Asynchronous
To transition from a source synchronous timing mode to an asynchronous timing mode, the
procedure described in this section shall be followed. To transition from the source synchronous
data interface to the asynchronous data interface, the Reset (FFh) command shall be used. After
the Reset is issued, the host shall not issue any commands to the device until after the tITC time
has passed. Note that after the tITC time has passed, only status commands may be issued by
the host until the Reset completes.
The host shall transition to the asynchronous data interface. Then the host shall issue the Reset
(FFh) command described in the previous paragraph using asynchronous timing mode 0, thus the
host transitions to the asynchronous data interface prior to issuing the Reset (FFh). A device in
any timing mode is required to recognize a Reset (FFh) command issued in asynchronous timing
mode 0. After issuing the Reset (FFh) and prior to transitioning CE_n high, the host shall observe
the following requirements:
 ALE and CLE shall be cleared to zero
47


RE_n / W/R_n shall be set to one
WE_n / CLK shall be set to one
After CE_n has been pulled high and then transitioned low again, the host should issue a Set
Features to select the appropriate asynchronous timing mode.
4.2. Timing Parameters
All timing parameters are from a host perspective. For example, the “Minimum WE_n pulse width”
is the minimum allowed WE_n pulse width that the host is permitted to present to the device while
still assuring correct operation of the device. The behavior of the device when the required host
minimum and maximum times are not adhered to is undefined. Note that the host needs to
account for channel effects in meeting the specified timings with the device.
4.2.1. General Timings
This section describes timing parameters that apply regardless of the data interface type being
used.
For execution of the first Read Parameter Page command, prior to complete initialization, a tR
value of 200 microseconds and tCCS value of 500 ns shall be used. For page reads, including
execution of additional Read Parameter Page commands after initialization is complete, the value
for tR and tCCS contained in the parameter page shall be used.
There are three maximums listed for tRST in the asynchronous and source synchronous data
interfaces. The target is allowed a longer maximum reset time when a program or erase
operation is in progress. The maximums correspond to:
1. The target is not performing an erase or program operation.
2. The target is performing a program operation.
3. The target is performing an erase operation.
Table 18 defines the array timing parameters. The array timing parameter values are either
returned in the parameter page (tR, tPROG, tBERS, and tCCS) or they are statically defined in
Table 19.
.
Parameter
Description
1
tBERS
Block erase time
tCCS
Change Column setup time
1
tPLEBSY
1
tPLPBSY
1
Busy time for multi-plane erase operation
Busy time for multi-plane program operation
tPLRBSY
Busy time for multi-plane read operation
tPCBSY
Program cache busy time
1
tPROG
Page program time
1
tR
Page read time
1
tRCBSY
Read cache busy time
NOTE:
1. Measured from the falling edge of SR[6] to the rising edge of SR[6].
Table 18
Array Timing Parameter Descriptions
48
There are “short” busy times associated with cache operations (tRCBSY, tPCBSY) and multiplane operations (tPLEBSY, tPLPBSY, and tPLRBSY). Typical and maximum times for these
busy times are listed in Table 19.
Parameter
Typical
Maximum
tPLEBSY
500 ns
tBERS
tPLPBSY
500 ns
tPROG
tPLRBSY
500 ns
tR
tPCBSY
3 µs
tPROG
tRCBSY
3 µs
tR
NOTE:
1. Typical times for tPCBSY and tRCBSY are the recommended interval at which the host
should consider polling status. Device busy time may be longer than the typical value.
Table 19
Cache and Multi-plane Short Busy Times
4.2.2. Asynchronous
Table 20 defines the descriptions of all timing parameters. Table 23 and Table 24 define the
requirements for timing modes 0, 1, 2, 3, 4, and 5. Timing mode 0 shall always be supported and
the device operates in this mode at power-on. A host shall only begin use of a more advanced
timing mode after determining that the device supports that timing mode in Read Parameter
Page.
The host shall use EDO data output cycle timings, as defined in section 4.3.1.5, when running
with a tRC value less than 30 ns.
49
Parameter
tADL
Description
3
ALE to data loading time
tALH
ALE hold time
tALS
ALE setup time
tAR
ALE to RE_n delay
3
tCCS
WE_n high to RE_n low, value specified in parameter page
tCEA
CE_n access time
tCEH
CE_n high hold time
tCH
CE_n hold time
tCHZ
2
CE_n high to output hi-Z
tCLH
CLE hold time
tCLR
CLE to RE_n delay
tCLS
CLE setup time
tCOH
CE_n high to output hold
tCS
CE_n setup time
tDH
Data hold time
tDS
Data setup time
1
tFEAT
2
tIR
1
tITC
Busy time for Set Features and Get Features
Output hi-Z to RE_n low
Interface and Timing Mode Change time
tRC
RE_n cycle time
tREA
RE_n access time
tREH
RE_n high hold time
tRHOH
RE_n high to output hold
tRHW
2
tRHZ
RE_n high to WE_n low
RE_n high to output hi-Z
tRLOH
RE_n low to output hold
tRP
RE_n pulse width
tRR
Ready to RE_n low (data only)
Device reset time, measured from the falling edge of R/B_n to
the rising edge of R/B_n.
tRST
4
tWB
WE_n high to SR[6] low
tWC
WE_n cycle time
tWH
WE_n high hold time
3
tWHR
WE_n high to RE_n low
tWP
WE_n pulse width
tWW
WP_n transition to WE_n low
NOTE:
1. Measured from the falling edge of SR[6] to the rising edge of SR[6].
2. Refer to Appendix E for measurement technique.
3. tADL is used for Program operations. tWHR is used for Read ID, Read Status, and
Read Status Enhanced commands. tCCS is used for commands that modify the
column address and thus impact the data pipeline; these commands include Change
Read Column and Change Write Column.
4. Commands (including Read Status / Read Status Enhanced) shall not be issued until
50
after tWB is complete.
Table 20
Asynchronous Timing Parameter Descriptions
The testing conditions that shall be used to verify that a device complies with a particular
asynchronous timing mode are listed in Table 21 for devices that support the asynchronous data
interface only and do not support driver strength settings.
Parameter
Value
Input pulse levels
0.0 V to VccQ
Input rise and fall times
5 ns
Input and output timing levels
VccQ / 2
Output load for VccQ of 3.3V
CL = 50 pF
Output load for VccQ of 1.8V
CL = 30 pF
Table 21
Testing Conditions for Asynchronous Only Devices
The testing conditions that shall be used to verify compliance with a particular timing mode for
devices that support driver strength settings are listed in Table 22. This includes all devices that
support the source synchronous data interface. It also includes devices that only support the
asynchronous data interface that support driver strength settings.
Parameter
Value
Positive input transition
VIL (DC) to VIH (AC)
Negative input transition
VIH (DC) to VIL (AC)
Minimum input slew rate
tIS = 1.0 V/ns
Input timing levels
VccQ / 2
Output timing levels
VccQ / 2
Driver strength
Output capacitive load
Nominal
1
CL = 5 pF
EZ NAND output reference load
50 Ohm to Vtt
NOTE:
1. Assumes small propagation delay from output to CL.
Table 22
Testing Conditions for Devices that Support Driver Strength Settings
NAND
Package
Rtt = 50 Ohms
Output
Vtt = 0.5 x VccQ
51
Figure 22
EZ NAND Output Reference Load
52
Parameter
Mode 0
Mode 1
Mode 2
Unit
100
50
35
ns
Min
Max
Min
Max
Min
Max
tADL
200
—
100
—
100
—
ns
tALH
20
—
10
—
10
—
ns
tALS
50
—
25
—
15
—
ns
tAR
25
—
10
—
10
—
ns
tCEA
—
100
—
45
—
30
ns
tCEH
20
—
20
—
20
—
ns
tCH
20
—
10
—
10
—
ns
tCHZ
—
100
—
50
—
50
ns
tCLH
20
—
10
—
10
—
ns
tCLR
20
—
10
—
10
—
ns
tCLS
50
—
25
—
15
—
ns
tCOH
0
—
15
—
15
—
ns
tCS
70
—
35
—
25
—
ns
tDH
20
—
10
—
5
—
ns
tDS
40
—
20
—
15
—
ns
tFEAT
—
1
—
1
—
1
µs
tIR
10
—
0
—
0
—
ns
tITC
—
1
—
1
—
1
µs
tRC
100
—
50
—
35
—
ns
tREA
—
40
—
30
—
25
ns
tREH
30
—
15
—
15
—
ns
tRHOH
0
—
15
—
15
—
ns
tRHW
200
—
100
—
100
—
ns
tRHZ
—
200
—
100
—
100
ns
tRLOH
0
—
0
—
0
—
ns
tRP
50
—
25
—
17
—
ns
tRR
40
—
20
20
—
5000
—
2
tRST
—
250000
—
tWB
—
200
—
—
—
10/30/
500
150/
150/
500
100
ns
tRST (raw NAND)
—
10/30/
500
150/
150/
500
100
tWC
100
—
45
—
35
—
ns
tWH
30
—
15
—
15
—
ns
tWHR
120
—
80
—
80
—
ns
tWP
50
—
25
—
17
—
ns
tWW
100
—
100
—
100
—
(EZ NAND)
—
—
µs
µs
ns
ns
NOTE:
1. To easily support EDO capable devices, tCHZ and tRHZ maximums are higher in
modes 1, 2, and 3 than typically necessary for a non-EDO capable device.
2. If the reset is invoked using a Reset (FFh) command then the EZ NAND device has
250 ms to complete the reset operation regardless of the timing mode. If the reset is
invoked using LUN Reset (FAh) then the values are as shown.
53
Table 23
Asynchronous Timing Modes 0, 1, and 2
54
Parameter
Mode 4
(EDO capable)
Mode 3
30
Mode 5
(EDO capable)
Unit
20
ns
25
Min
Max
Min
Max
Min
Max
70
—
70
—
ns
100
—
tALH
5
—
5
—
5
—
ns
tALS
10
—
10
—
10
—
ns
tAR
10
—
10
—
10
—
ns
tCEA
—
25
—
25
—
25
ns
tCEH
20
—
20
—
20
—
ns
tCH
5
—
5
—
5
—
ns
tCHZ
—
50
—
30
—
30
ns
tCLH
5
—
5
—
5
—
ns
tCLR
10
—
10
—
10
—
ns
tCLS
10
—
10
—
10
—
ns
tCOH
15
—
15
—
15
—
ns
tCS
25
—
20
—
15
—
ns
tDH
5
—
5
—
5
—
ns
tDS
10
—
10
—
7
—
ns
tFEAT
—
1
—
1
—
1
µs
tIR
0
—
0
—
0
—
ns
tITC
—
1
—
1
—
1
µs
tRC
30
—
25
—
20
—
ns
tREA
—
20
—
20
—
16
ns
tREH
10
—
10
—
7
—
ns
tRHOH
15
—
15
—
15
—
ns
tRHW
100
—
100
—
100
—
ns
tRHZ
—
100
—
100
—
100
ns
tRLOH
0
—
5
—
5
—
ns
tRP
15
—
12
—
10
—
ns
tRR
20
20
20
—
2
tRST
—
150/15
0/500
—
—
—
10/30/
500
150/
150/
500
100
ns
tRST (raw NAND)
—
10/30/
500
20
—
ns
tADL
tWB
—
100
—
—
10/30/
500
150/
150/
500
100
tWC
30
—
25
—
tWH
10
—
10
—
7
—
ns
tWHR
60
—
60
—
60
—
ns
tWP
15
—
12
—
10
—
ns
tWW
100
—
100
—
100
—
(EZ NAND)
—
—
—
µs
µs
ns
ns
NOTE:
1. To easily support EDO capable devices, tCHZ and tRHZ maximums are higher in
modes 1, 2, and 3 than typically necessary for a non-EDO capable device.
2. If the reset is invoked using a Reset (FFh) command then the EZ NAND device has
250 ms to complete the reset operation regardless of the timing mode. If the reset is
invoked using LUN Reset (FAh) then the values are as shown.
55
Table 24
Asynchronous Timing Modes 3, 4, and 5
4.2.3. Source Synchronous
All source synchronous timing parameters are referenced to the rising edge of CLK or the
latching edge of DQS. Note that R/B_n and WP_n are always asynchronous signals.
If CLK is a different frequency than those described in the source synchronous timing modes,
then the host shall meet the setup and hold requirements for the next fastest timing mode.
For parameters measured in clocks (e.g. tDSH), the parameter is measured starting from a
latching edge of CLK or DQS, respectively.
56
Parameter
Description
tAC
Access window of DQ[7:0] from CLK
tADL
5
Address cycle to data loading time
tCAH
Command, Address, Data delay
(command to command, address to address,
command to address, address to command, command/address
to start of data)
Command/address DQ hold time
tCALH
W/R_n, CLE and ALE hold time
tCADf, tCADs
tCALS
W/R_n, CLE and ALE setup time
tCAS
5
tCCS
Command/address DQ setup time
Change Column setup time, value specified in parameter page
tCEH
CE_n high hold time
tCH
CE_n hold time
tCK(avg)
2
Absolute clock period, measured from rising edge to the next
consecutive rising edge
tCK(abs)
3
tCKH(abs)
tCKL(abs)
Average clock cycle time, also known as tCK
Clock cycle high
3
Clock cycle low
tCKWR
Data output end to W/R_n high
tCS
CE_n setup time
tDH
Data DQ hold time
tDPZ
Data input pause setup time
tDQSCK
4
tDQSD
Access window of DQS from CLK
W/R_n low to DQS/DQ driven by device
tDQSH
DQS input high pulse width
tDQSHZ
4
W/R_n high to DQS/DQ tri-state by device
tDQSL
DQS input low pulse width
tDQSQ
DQS-DQ skew, DQS to last DQ valid, per access
tDQSS
Data input to first DQS latching transition
tDS
Data DQ setup time
tDSC
DQS cycle time
tDSH
DQS falling edge to CLK rising – hold time
tDSS
DQS falling edge to CLK rising – setup time
tDVW
Output data valid window
1
tFEAT
Busy time for Set Features and Get Features
tHP
Half-clock period
57
1
tITC
Interface and Timing Mode Change time
tJIT(per)
The deviation of a given tCK(abs) from tCK(avg)
tQH
DQ-DQS hold, DQS to first DQ to go non-valid, per access
tQHS
Data hold skew factor
tRHW
Data output cycle to command, address, or data input cycle
tRR
Ready to data output cycle (data only)
Device reset time, measured from the falling edge of R/B_n to the
rising edge of R/B_n.
tRST
6
tWB
5
tWHR
CLK rising edge to SR[6] low
Command, address or data input cycle to data output cycle
tWPRE
DQS write preamble
tWPST
DQS write postamble
tWRCK
W/R_n low to data output cycle
tWW
WP_n transition to command cycle
NOTE:
1. Measured from the falling edge of SR[6] to the rising edge of SR[6].
2. tCK(avg) is the average clock period over any consecutive 200 cycle window.
3. tCKH(abs) and tCKL(abs) include static offset and duty cycle jitter.
4. Refer to Appendix E for measurement technique.
5. tADL is used for Program operations. tWHR is used for Read ID, Read Status, and
Read Status Enhanced commands. tCCS is used for commands that modify the
column address and thus impact the data pipeline; these commands include Change
Read Column and Change Write Column.
6. Commands (including Read Status / Read Status Enhanced) shall not be issued until
after tWB is complete.
Table 25
Source Synchronous Timing Parameter Descriptions
The device may be configured with multiple driver strengths with the Set Features command.
There is a 50 Ohm, 35 Ohm, 25 Ohm, and 18 Ohm setting that the device may support. Support
for all four driver strength settings is required for devices that support the source synchronous
data interface. A device that only supports the asynchronous data interface may support all or a
subset of driver strength settings. Devices that support driver strength settings shall comply with
the output driver requirements in this section.
Setting
Driver Strength
18 Ohms
2.0x = 18 Ohms
25 Ohms
1.4x = 25 Ohms
35 Ohms
1.0x = 35 Ohms
50 Ohms
0.7x = 50 Ohms
18 Ohms
2.0x = 18 Ohms
25 Ohms
1.4x = 25 Ohms
35 Ohms
1.0x = 35 Ohms
50 Ohms
0.7x = 50 Ohms
Table 26
VccQ
3.3V
1.8V
I/O Drive Strength Settings
The impedance values correspond to several different VccQ values are defined in Table 28 for
3.3V VccQ and Table 29 for 1.8V VccQ. The test conditions that shall be used to verify the
impedance values is specified in Table 27.
58
Condition
Temperature (TA)
VccQ
(3.3V)
VccQ
(1.8V)
Process
Minimum Impedance
TOPER (Min) degrees Celsius
3.6V
1.95V
Fast-fast
Nominal Impedance
25 degrees Celsius
3.3V
1.8V
Typical
Maximum impedance
TOPER (Max) degrees Celsius
2.7V
1.7V
Slow-slow
Table 27
Testing Conditions for Impedance Values
59
RON = 18 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
28.7
16.2
7.0
Ohms
0.5 x VccQ
36.0
18.0
9.0
Ohms
0.8 x VccQ
50.0
21.0
11.8
Ohms
0.2 x VccQ
50.0
21.0
11.8
Ohms
0.5 x VccQ
36.0
18.0
9.0
Ohms
0.8 x VccQ
28.7
14.0
7.0
Ohms
Minimum
Unit
RON = 25 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
0.2 x VccQ
40.0
22.3
9.3
Ohms
0.5 x VccQ
50.0
25.0
12.6
Ohms
0.8 x VccQ
68.0
29.0
16.3
Ohms
0.2 x VccQ
68.0
29.0
16.3
Ohms
0.5 x VccQ
50.0
25.0
12.6
Ohms
0.8 x VccQ
40.0
19.0
9.3
Ohms
RON = 35 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
58.0
32.0
12.8
Ohms
0.5 x VccQ
70.0
35.0
18.0
Ohms
0.8 x VccQ
95.0
40.0
23.0
Ohms
0.2 x VccQ
95.0
40.0
23.0
Ohms
0.5 x VccQ
70.0
35.0
18.0
Ohms
0.8 x VccQ
58.0
32.0
12.8
Ohms
RON = 50 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
80.0
45.0
18.4
Ohms
0.5 x VccQ
100.0
50.0
25.0
Ohms
0.8 x VccQ
136.0
57.0
32.0
Ohms
0.2 x VccQ
136.0
57.0
32.0
Ohms
0.5 x VccQ
100.0
50.0
25.0
Ohms
0.8 x VccQ
80.0
45.0
18.4
Ohms
Table 28
Impedance Values for 3.3V VccQ
60
RON = 18 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
34.0
13.5
7.5
Ohms
0.5 x VccQ
31.0
18.0
9.0
Ohms
0.8 x VccQ
44.0
23.5
11.0
Ohms
0.2 x VccQ
44.0
23.5
11.0
Ohms
0.5 x VccQ
31.0
18.0
9.0
Ohms
0.8 x VccQ
34.0
13.5
7.5
Ohms
RON = 25 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
47.0
19.0
10.5
Ohms
0.5 x VccQ
44.0
25.0
13.0
Ohms
0.8 x VccQ
61.5
32.5
16.0
Ohms
0.2 x VccQ
61.5
32.5
16.0
Ohms
0.5 x VccQ
44.0
25.0
13.0
Ohms
0.8 x VccQ
47.0
19.0
10.5
Ohms
RON = 35 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
66.5
27.0
15.0
Ohms
0.5 x VccQ
62.5
35.0
18.0
Ohms
0.8 x VccQ
88.0
52.0
22.0
Ohms
0.2 x VccQ
88.0
52.0
22.0
Ohms
0.5 x VccQ
62.5
35.0
18.0
Ohms
0.8 x VccQ
66.5
27.0
15.0
Ohms
RON = 50 Ohms
Description
R_pulldown
R_pullup
VOUT to
VssQ
Maximum
Nominal
Minimum
Unit
0.2 x VccQ
95.0
39.0
21.5
Ohms
0.5 x VccQ
90.0
50.0
26.0
Ohms
0.8 x VccQ
126.5
66.5
31.5
Ohms
0.2 x VccQ
126.5
66.5
31.5
Ohms
0.5 x VccQ
90.0
50.0
26.0
Ohms
0.8 x VccQ
95.0
39.0
21.5
Ohms
Table 29
Impedance Values for 1.8V VccQ
The pull-up and pull-down impedance mismatch is defined in Table 30. Impedance mismatch is
the absolute value between pull-up and pull-down impedances. Both are measured at the same
temperature and voltage. The testing conditions that shall be used to verify the impedance
mismatch requirements are: VccQ = VccQ(min), VOUT = VccQ x 0.5, and T A is across the full
operating range.
61
Maximum
Minimum
Unit
18 Ohms
6.3
0.0
Ohms
25 Ohms
8.8
0.0
Ohms
35 Ohms
12.3
0.0
Ohms
50 Ohms
17.5
0.0
Ohms
Output Impedance
Table 30
Pull-up and Pull-down Impedance Mismatch
The input slew rate requirements that the device shall comply with are defined in Table 31. The
testing conditions that shall be used to verify the input slew rate are listed in Table 32.
Description
Timing Mode
2
3
0
1
Input slew rate (min)
0.5
0.5
0.5
Input slew rate (max)
4.5
4.5
4.5
Derating factor for setup
times, address and command
TBD
TBD
Derating factor for hold times,
address and command
TBD
Derating factor for setup
times, data input and output
Derating factor for hold times,
data input and output
5
0.5
0.5
0.5
V/ns
4.5
4.5
4.5
V/ns
TBD
TBD
TBD
TBD
ns per 100 mV
TBD
TBD
TBD
TBD
TBD
ns per 100 mV
TBD
TBD
TBD
TBD
TBD
TBD
ns per 100 mV
TBD
TBD
TBD
TBD
TBD
TBD
ns per 100 mV
Table 31
Input Slew Rate Requirements
Parameter
Value
Positive input transition
VIL (DC) to VIH (AC)
Negative input transition
VIH (DC) to VIL (AC)
Table 32
Unit
4
Testing Conditions for Input Slew Rate
The output slew rate requirements that the device shall comply with are defined in Table 33 and
Table 34 for a single LUN per 8-bit data bus. The testing conditions that shall be used to verify
the output slew rate are listed in Table 35.
Output Slew Rate
Min
Max
Unit
Normative or
Recommended
18 Ohms
1.5
10.0
V/ns
Normative
25 Ohms
1.5
9.0
V/ns
Normative
35 Ohms
1.2
7.0
V/ns
Normative
50 Ohms
1.0
5.5
V/ns
Recommended
Description
Table 33
Output Slew Rate Requirements for 3.3V VccQ
62
Output Slew Rate
Min
Max
Unit
Normative or
Recommended
18 Ohms
1.0
5.5
V/ns
Normative
25 Ohms
0.85
5.0
V/ns
Normative
35 Ohms
0.75
4.0
V/ns
Normative
50 Ohms
0.60
4.0
V/ns
Recommended
Description
Table 34
Output Slew Rate Requirements for 1.8V VccQ
Parameter
Value
VOL(DC)
0.3 * VccQ
VOH(DC)
0.7 * VccQ
VOL(AC)
0.2 * VccQ
VOH(AC)
0.8 * VccQ
Positive output transition
VOL (DC) to VOH (AC)
Negative output transition
1
tRISE
1
tFALL
VOH (DC) to VOL (AC)
Time during falling edge from VOH(DC) to VOL(AC)
Output slew rate rising edge
[VOH(AC) – VOL(DC)] / tRISE
Output slew rate falling edge
[VOH(DC) – VOL(AC)] / tFALL
Time during rising edge from VOL(DC) to VOH(AC)
Output load
Refer to Table 22
NOTE:
1. Refer to Figure 23.
2. Output slew rate is verified by design and characterization; it may not be subject to production test.
3. The minimum slew rate is the minimum of the rising edge and the falling edge slew rate. The
maximum slew rate is the maximum of the rising edge and the falling edge slew rate.
Table 35
Figure 23
Testing Conditions for Output Slew Rate
tRISE and tFALL Definition for Output Slew Rate
63
The testing conditions that shall be used to verify that a device complies with a particular source
synchronous timing mode are listed in Table 36. The test conditions are the same regardless of
the number of LUNs per Target.
Parameter
Value
Positive input transition
VIL (DC) to VIH (AC)
Negative input transition
VIH (DC) to VIL (AC)
Minimum input slew rate
tIS = 1.0 V/ns
Input timing levels
VccQ / 2
Output timing levels
VccQ / 2
Driver strength
Nominal
Output capacitive load
1
CL = 5 pF
NOTE:
1. Assumes small propagation delay from output to CL.
Table 36
Testing Conditions for Source Synchronous Timing Modes
The input capacitance requirements are defined in Table 37. The testing conditions that shall be
used to verify the input capacitance requirements are: temperature of 25 degrees Celsius, VIN =
0V, and a CLK frequency of 100 MHz. The capacitance delta values measure the pin-to-pin
capacitance for all LUNs within a package, including across data buses. The capacitance delta
values change based on the number of LUNs per x8 data bus. All targets that share an I/O
should report an equivalent Typical capacitance in order to meet the capacitance delta
requirements. If targets with different Typical capacitance values share an I/O bus, then the
values in these tables are not applicable and detailed topology and signal integrity analysis needs
to be completed by the implementer to determine the bus speed that is achievable.
EZ NAND has a controller stacked in the NAND package and the LUNs are physically connected
to the controller and not to the external data bus. EZ NAND values are defined separately from
the values specified in Table 37. For the Typical ranges in Table 38 and Table 39, EZ NAND is
specified separately since ASIC capacitance tends to be higher.
Parameter
Number of LUNs per x8 data bus
1
2
4
8
Typical Variance
±1.0
±1.5
±2.0
±4.0
Delta
1.4
1.7
2.0
4.0
Unit
pF
pF
NOTE:
1. Typical capacitance values for pin groups CCK, CIN, and CIO are specified in the
parameter page. The allowable range for Typical capacitance values is specified
in Table 38 for CLK and input pins and Table 39 for I/O pins.
2. Input capacitance for CE_n and WP_n shall not exceed the maximum capacitance
value for CLK and input pins. However, CE_n and WP_n do not have to meet
delta or variance requirements.
Table 37
Input Capacitance, Minimum and Maximums
The Typical capacitance values shall be constrained to the ranges defined in Table 38 for CLK
and input pins and Table 39 for I/O pins for devices in a BGA package. Capacitance is shared for
64
LUNs that share the same 8-bit data bus in the same package, thus the ranges are specific to the
number of LUNs per data bus.
Parameter
Min
Source Synchronous
Timing Modes 0-5
Typ
Typ
Low
High
Max
Typ
Variance
Unit
1 LUN per x8
data bus
2.0
3.0
5.4
6.4
±0.5
pF
2 LUNs per x8
data bus
3.5
5.0
7.7
9.2
±1.0
pF
4 LUNs per x8
data bus
5.0
7.0
12.3
14.3
±2.0
pF
8 LUNs per x8
data bus
6.0
10.0
23.3
27.3
±4.0
pF
EZ NAND
3.0
5.0
6.5
8.5
±2.0
pF
Table 38
Input Capacitance for CLK and input pins, Typical Ranges
Parameter
Min
Source Synchronous
Timing Modes 0-5
Typ
Typ
Low
High
Max
Typ
Variance
Unit
1 LUN per x8
data bus
3.0
4.0
5.7
6.7
±0.5
pF
2 LUNs per x8
data bus
5.0
6.5
8.5
10.0
±1.0
pF
4 LUNs per x8
data bus
8.9
10.9
14.9
16.9
±2.0
pF
8 LUNs per x8
data bus
16.7
20.7
28.7
32.7
±4.0
pF
EZ NAND
3.0
5.0
6.5
8.5
±2.0
pF
Table 39
Input Capacitance for I/O pins, Typical Ranges
NOTE: Capacitance ranges are not defined for the TSOP package due to the varying TSOP
package construction techniques and bond pad locations. For the TSOP package compared to
the BGA package, the input capacitance delta values do not apply, input capacitance values for
I/O pins is similar, and input capacitance values for input pins could be significantly higher. For
higher speed applications, the BGA-63 or BGA-100 packages are recommended due to their
lower and more consistent input capacitance and input capacitance delta values.
Table 40 describes the standard source synchronous timing modes. The host is not required to
have a clock period that exactly matches any of the clock periods listed for the standard timing
modes. The host shall meet the setup and hold times for the timing mode selected. If the host
selects timing mode n using Set Features, then its clock period shall be faster than the clock
period of timing mode n-1 and slower than or equal to the clock period of timing mode n. For
example, if the host selects timing mode 2, then the following equation shall hold:
30 ns > host clock period >= 20 ns
65
If timing mode 0 is selected, then the clock period shall be no slower than 100 ns. The only
exception to this requirement is when the host is issuing a Reset (FFh) in asynchronous timing
mode 0 (see section 4.1.4.3).
Timing parameters that indicate a latency requirement before a data input, data output, address
or command cycle shall be satisfied to the rising clock edge after the latency in nanoseconds has
elapsed. To calculate the first edge where the associated transition may be made, it is calculated
as follows:
= RoundUp{[tParam + tCK] / tCK}
66
Parameter
Mode 0
50
Mode 1
30
~20
tAC
tADL
tCADf
tCADs
tCAH
tCALH
tCALS
tCAS
tCEH
tCH
tCK(avg) or
tCK
Mode 2
20
~33
Mode 3
15
~50
~66
Min
3
100
25
45
10
10
10
10
20
10
Max
25
—
—
—
—
—
—
—
—
—
Min
3
100
25
45
5
5
5
5
20
5
Max
25
—
—
—
—
—
—
—
—
—
Min
3
70
25
45
4
4
4
4
20
4
Max
25
—
—
—
—
—
—
—
—
—
Min
3
70
25
45
3
3
3
3
20
3
Max
25
—
—
—
—
—
—
—
—
—
50
—
30
—
20
—
15
—
Mode 4
12
~83
Min
Max
3
70
25
45
2.5
2.5
2.5
2.5
20
2.5
12
tCS
tDH
tDPZ
tDQSCK
tDQSD
35
5
1.5
3
0
—
—
—
25
18
25
2.5
1.5
3
0
Minimum: tCK(avg) + tJIT(per) min
Maximum: tCK(avg) + tJIT(per) max
0.57
0.43
0.57
0.43
0.57
0.43
0.57
0.43
0.57
0.43
0.57
0.43
Minimum: RoundUp{[tDQSCK(max) + tCK] / tCK}
Maximum: —
15
—
15
—
15
—
1.1
—
1.7
—
1.3
—
—
1.5
—
1.5
—
1.5
25
3
25
3
25
3
18
0
18
0
18
0
tDQSH
0.4
0.6
0.4
0.6
0.4
0.6
0.4
0.6
tDQSHZ
—
20
—
20
—
20
—
tDQSL
0.4
0.6
0.4
0.6
0.4
0.6
tDQSQ
tDQSS
tDS
tDSC
tDSH
—
0.75
5
50
0.2
5
1.25
—
—
—
—
0.75
3
30
0.2
2.5
1.25
—
—
—
—
0.75
2
20
0.2
1.7
1.25
—
—
—
tCK(abs)
tCKH(abs)
tCKL(abs)
0.43
0.43
0.57
0.57
0.43
0.43
tCKWR
67
25
—
—
—
—
—
—
—
—
—
—
Mode 5
10
~100
Min
Max
3
70
25
45
2
2
2
2
20
2
10
Unit
ns
MHz
25
—
—
—
—
—
—
—
—
—
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
—
ns
ns
0.57
0.57
0.43
0.43
0.57
0.57
tCK
tCK
tCK
—
—
—
25
18
15
0.9
1.5
3
0
—
—
—
25
18
0.4
0.6
0.4
0.6
20
—
20
—
20
0.4
0.6
0.4
0.6
0.4
0.6
—
0.75
1.5
15
0.2
1.3
1.25
—
—
—
—
0.75
1.1
12
0.2
1.0
1.25
—
—
—
—
0.75
0.9
10
0.2
0.85
1.25
—
—
—
ns
ns
tDSC
ns
ns
tCK or
4
tDSC
ns
tCK or
4
tDSC
ns
tCK
ns
ns
tCK
tDSS
tDVW
tFEAT
tHP
tITC
tJIT(per)
tQH
tQHS
tRHW
tRR
tRST (raw
NAND)
2
tRST
(EZ NAND)
0.2
—
0.2
—
0.2
—
1
—
1
—
—
-0.7
1
0.7
—
-0.7
1
0.7
—
-0.7
—
100
20
6
—
—
10/30/
500
150/
150/
500
100
—
—
—
—
—
—
100
20
3
—
—
10/30/
500
150/
150/
500
100
—
—
—
—
—
—
100
20
—
—
—
—
—
—
—
0.2
tDVW = tQH – tDQSQ
1
—
tHP = min(tCKL, tCKH)
1
—
0.7
-0.6
tQH = tHP – tQHS
2
—
—
100
—
20
10/30/
—
500
150/
150/
—
500
100
—
—
80
—
1.5
—
1.5
—
20
—
100
—
0.2
—
0.2
—
1
—
1
—
1
1
0.6
—
-0.6
1
0.6
—
-0.5
1
0.5
1.5
—
—
10/30/
500
150/
150/
500
100
—
—
—
—
—
—
100
20
1.2
—
—
10/30/
500
150/
150/
500
100
—
—
—
—
—
—
100
20
1.0
—
—
10/30/
500
150/
150/
500
100
—
—
—
—
—
—
—
—
—
tCK
ns
µs
ns
µs
ns
ns
ns
ns
ns
µs
µs
tWB
—
—
—
—
—
ns
tWHR
80
80
80
80
80
ns
tWPRE
1.5
1.5
1.5
1.5
1.5
tCK
tWPST
1.5
1.5
1.5
1.5
1.5
tCK
tWRCK
20
20
20
20
20
ns
tWW
100
100
100
100
100
ns
NOTE:
1. tDQSHZ is not referenced to a specific voltage level, but specifies when the device output is no longer driving.
2. tCK(avg) is the average clock period over any consecutive 200 cycle window.
3. tCKH(abs) and tCKL(abs) include static offset and duty cycle jitter.
4. tDQSL and tDQSH are relative to tCK when CLK is running. If CLK is stopped during data input, then tDQSL and tDQSH are relative to tDSC.
5. If the reset is invoked using a Reset (FFh) command then the EZ NAND device has 250 ms to complete the reset operation regardless of the timing mode. If
the reset is invoked using Synchronous Reset (FCh) or LUN Reset (FAh) then the values are as shown.
Table 40
68
Source Synchronous Timing Modes
4.3. Timing Diagrams
4.3.1. Asynchronous
4.3.1.1. Command Latch Timings
The requirements for the R/B_n signal only apply to commands where R/B_n is cleared to zero after the command is issued, as specified in the
command definitions.
CLE
tCLS
tCLH
tCS
tCH
CE_n
tWP
WE_n
tALS
tALH
ALE
tDS
IO0-7
tDH
Command
R/B_n
tWB
Figure 24
69
Command latch timings
4.3.1.2. Address Latch Timings
tCLS
CLE
tCS
CE_n
tWC
tWP
WE_n
tWH
ALE
tALS
tDS
IO0-7
t ALH
tDH
Address
Figure 25
70
Address latch timings
4.3.1.3. Data Input Cycle Timings
Data input may be used with CE_n don’t care. However, if CE_n don’t care is used tCS and tCH timing requirements shall be met by the host,
and WE_n falling edge shall always occur after the CE_n falling edge (i.e. tCS > tWP).
tCLH
CLE
tCH
CE_n
tWP
tWP
WE_n
tWP
tWH
t WC
tALS
ALE
tDS
IOx
t DH
DIN 0
Figure 26
71
tDS
tDH
DIN 1
Data input cycle timings
tDS
tDH
DIN n
4.3.1.4. Data Output Cycle Timings
Data output may be used with CE_n don’t care. However, if CE_n don’t care is used tCEA and tCOH timing requirements shall be met by the host
and RE_n shall either remain low or RE_n falling edge shall occur after the CE_n falling edge.
tCHZ
tCEA
CE_n
t COH
tRP
RE_n
t RP
tRP
t RHOH
tREH
tRR
tRHZ
tRC
R/B_n
tRHZ
tREA
IOx
DOUT 0
Figure 27
72
tREA
tRHZ
DOUT 1
Data output cycle timings
tREA
tRHZ
DOUT n
4.3.1.5. Data Output Cycle Timings (EDO)
EDO data output cycle timings shall be used if the host drives tRC less than 30 ns. Data output may be used with CE_n don’t care. However, if
CE_n don’t care is used tCEA and tCOH timing requirements shall be met by the host.
tCHZ
CE_n
tCOH
tRP
RE_n
tRR
R/B_n
t REH
tRC
tREA
tRHZ
tREA
tRHOH
tRLOH
IOx
DOUT 0
DOUT 1
tCEA
Figure 28
73
EDO data output cycle timings
DOUT n
4.3.1.6. Read Status Timings
tCLR
CLE
tCLH
tCLS
tCS
tCH
tCEA
CE_n
tWP
tCHZ
WE_n
t WHR
tCOH
RE_n
tRHZ
tDS
IO7-0
t DH
tRHOH
70h
Status
t IR
Figure 29
74
Read Status timings
tREA
4.3.1.7. Read Status Enhanced Timings
CLE
tCLH
tCLS
tCEA
tCH
tCS
CE_n
tWC
tWP
tWP
WE_n
tWH
t ALH
tCHZ
tALH
tALS
tAR
tCOH
ALE
RE_n
tRHZ
tDS
IO7-0
tDH
78h
tWHR
R1
Figure 30
75
R2
R3
Read Status Enhanced timings
t REA
tRHOH
Status
4.3.2. Source Synchronous
For the command, address, data input, and data output diagrams, the tCS timing parameter may
consume multiple clock cycles. The host is required to satisfy tCS by the rising edge of CLK
shown in the diagrams, and thus needs to pull CE_n low far enough in advance to meet this
requirement (which could span multiple clock cycles).
4.3.2.1. Command Cycle Timings
tCH
CE_n
tCS
tCALS
CLE
tCALS tCALH
tCALS
tCAD
ALE
tCKL tCKH
CLK
tCAD starts for next
non-idle cycle
tCK
W/R_n
tCALS
tCALH
tDQSHZ
DQS
tCAS
DQ[7:0]
tCAH
Command
Figure 31
Command cycle timings
NOTE:
1. The cycle that tCAD is measured from may be an idle cycle (as shown), another command
cycle, an address cycle, or a data cycle. The idle cycle is shown in this diagram for simplicity.
2. ALE and CLE shall be in a valid state when CE_n transitions from one to zero. In the
diagram, it appears that tCS and tCALS are equivalent times. However, tCS and tCALS
values are not the same, the timing parameter values should be consulted in Table 40.
76
4.3.2.2. Address Cycle Timings
tCH
CE_n
tCS
tCALS
CLE
tCAD
tCALS
ALE
tCALS tCALH
tCKL tCKH
CLK
tCAD starts for next
non-idle cycle
tCK
W/R_n
tCALS
tCALH
tDQSHZ
DQS
tCAS
DQ[7:0]
tCAH
Address
Figure 32
Address cycle timings
NOTE:
1. ALE and CLE shall be in a valid state when CE_n transitions from one to zero. In the
diagram, it appears that tCS and tCALS are equivalent times. However, tCS and tCALS
values are not the same, the timing parameter values should be consulted in Table 40.
77
4.3.2.3. Data Input Cycle Timings
Data input cycle timing describes timing for data transfers from the host to the device (i.e. data writes).
For the Set Features command, the same data byte is repeated twice. The data pattern in this case is D0 D0 D1 D1 D2 D2 etc. The device shall
only latch one copy of each data byte. CLK should not be stopped during data input for the Set Features command. The device is not required to
wait for the repeated data byte before beginning internal actions.
78
tCH
CE_n
tCS
CLE
tCALS
tCALH
tCALS
tCALH
tCAD
ALE
tCKL tCKH
CLK
tCAD starts for next
non-idle cycle
tCK
W/R_n
tDSH
tDQSS
tDSS tDSH
tDSH
tDSS tDSH
tDSS
DQS
tWPRE
DQ[7:0]
tDQSH tDQSL tDQSH
D0
D2
D1
D3
tDQSL tDQSH
DN-2
tDH
79
DN
tDS
tDS
Figure 33
DN-1
tWPST
Data input cycle timing
tDH
4.3.2.4. Data Input Cycle Timings, CLK stopped
The host may save power during the data input cycles by holding the CLK signal high (i.e. stopping the CLK). The host may only stop the CLK
during data input if the device supports this feature as indicated in the parameter page. Data input cycle timing describes timing for data transfers
from the host to the device (i.e. data writes). Figure 34 describes data input cycling with the CLK signal stopped. The values of the ALE, CLE,
and W/R_n signals are latched on the rising edge of CLK and thus while CLK is held high these signals are don’t care.
Figure 35 shows data input cycling with the CLK signal stopped where the host has optionally paused data input. The host may pause data input
if it observes the tDPZ timing parameter for re-starting data input to the device. When re-starting the CLK, the host shall observe the indicated
timing parameters in Figure 34 and Figure 35, which include tDSS and tDSH.
80
A
B
tCH
CE_n
tCS
tCALS
CLE
tCALS tCALH
tCAD
tCALS
ALE
tCALS
tCALH
tCKL tCKH
CLK
tCAD starts for next
non-idle cycle
tCK
W/R_n
tCALS
tCALH
tDSS tDSH
tDSC
tDQSS
tDSS
DQS
tWPRE
DQ[7:0]
tDQSH tDQSL tDQSH
Dn
D1
D2
D3
tDQSL tDQSH
DN-2
DN-1
tDH
81
DN
tDS
tDS
Figure 34
tWPST
Data input cycle timing, CLK stopped
tDH
A
B
CE_n
tCALS
CLE
tCALH
tCALS
ALE
tCALH
CLK
W/R_n
tCALS
tCALH
tDQSS
tDSC
tDSC
tDPZ
tDSS tDSH
DQS
tDQSH tDQSL tDQSH
DQ[7:0]
DM
DM+1 DM+2
DM+3
tDQSH tDQSL tDQSH
DM+4
DM+5 DM+6 DM+7
tDQSL tDQSH
DN-2
DN-1
DN
tDS
tDS
tDH
tDH
Figure 35
Data input cycle timing, CLK stopped with data pause
82
4.3.2.5. Data Output Cycle Timings
Data output cycle timing describes timing for data transfers from the device to the host (i.e. data reads). The host shall not start data output (i.e.
transition ALE/CLE to 11b) until the tDQSD time has elapsed.
For the Read ID, Get Features, Read Status, and Read Status Enhanced commands, the same data byte is repeated twice. The data pattern in
this case is D0 D0 D1 D1 D2 D2 etc. The host shall only latch one copy of each data byte.
A calculated parameter, tCKWR, indicates when W/R_n may be transitioned from a zero to one. This parameter is calculated as:
 tCKWR(min) = RoundUp{[tDQSCK(max) + tCK] / tCK}
83
tCH
CE_n
tCS
tCAD starts for next
non-idle cycle
CLE
tCALS
tCALH
tCALS
tCALH
tCAD
ALE
tCKH
tCKL
tHP
CLK
tHP
tHP
tHP
tCK
tHP
tHP
tCKWR
tDQSCK
tDQSCK
tDQSCK
tCALS
tWRCK
tDQSCK
W/R_n
tDQSCK
tDQSCK
tCALS tDQSHZ
tDQSD
DQS
tAC
DQ[7:0]
tDVW tDVW tDVW
D0
D1
tQH
tQH
tDQSQ
Don’t Care
Data Transitioning
Figure 36
84
tDVW tDVW
DN-2
D2
tDQSQ
DN-1
DN
tDQSQ
Device Driving
Data output cycle timing
tDQSQ
tQH
tQH
4.3.2.6. W/R_n Behavior Timings
Figure 37 describes the ownership transition of the DQ bus and DQS signal. The host owns the DQ bus and DQS signal when W/R_n is one.
The device owns the DQ bus and DQS signal when W/R_n is zero. The host shall tri-state the DQ bus and DQS signal whenever W/R_n is zero.
When W/R_n transitions from one to zero, the bus ownership is assumed by the device. The device shall start driving the DQS signal low within
tDQSD after the transition of W/R_n to zero. When W/R_n transitions from zero to one, the bus ownership is assumed by the host. The device
shall tri-state the DQ bus and DQS signal within tDQSHZ after the transition of W/R_n to one.
tCALS
tCALS
CLK (WE_n)
W/R_n (RE_n)
DQ[7:0]
tDQSD
tDQSHZ
DQS
Tri-state
Device Driving
Figure 37
85
W/R_n timing
4.3.2.7. Satisfying Timing Requirements
In some cases there are multiple timing parameters that shall be satisfied prior to the next phase
of a command operation. For example, both tDQSD and tCCS shall be satisfied prior to data
output commencing for the Change Write Column command. The host and device shall ensure
all timing parameters are satisfied. In cases where tADL, tCCS, tRHW, or tWHR are required,
then these are the governing parameters (i.e. these parameters are the longest times).
Figure 38 and Figure 39 show an example of a Read Status command that includes all the timing
parameters for both the command and data phase of the operation. It may be observed that
tWHR is the governing parameter prior to the data transfer starting. Also note that the same data
byte is transmitted twice (D0, D0) for the Read Status command.
A
CE_n
tCS
tCALS
CLE
tCALS tCALH
tWHR
tCALS
tCAD
tCAD
ALE
tCKL tCKH
CLK
tCK
W/R_n
tCALS
tCALH
tDQSHZ
DQS
tCAS
DQ[7:0]
tCAH
70h
Figure 38
Read Status including tWHR and tCAD timing requirements
NOTE:
1. ALE and CLE shall be in a valid state when CE_n transitions from one to zero. In the
diagram, it appears that tCS and tCALS are equivalent times. However, tCS and tCALS
values are not the same, the timing parameter values should be consulted in Table 40.
86
A
tCH
CE_n
CLE
tCALS
tWHR
ALE
tCALS
tHP
CLK
tHP
tCALS
tDQSCK
tDQSD
W/R_n
tDQSCK
tCALS tDQSHZ
DQS
tAC
DQ[7:0]
tDVW tDVW
D0
D0
tQH
tQH
tDQSQ
Figure 39
tDQSQ
Read Status including tWHR and tCAD timing requirements, continued
87
4.4. Command Examples
4.4.1. Asynchronous
This section shows examples of commands using the asynchronous data interface. Figure 40
and Figure 41 show an example of Change Read Column.
A
CE_n
CLE
ALE
WE_n
RE_n
DQx
Figure 40
05h
C1
C2
E0h
Asynchronous Data Interface: Change Read Column example
88
A
CE_n
CLE
ALE
WE_n
tCCS
RE_n
DQx
E0h
Figure 41
D0
D1
D2
D3
Asynchronous Data Interface: Change Read Column example, continued
89
Figure 42 shows an example of Change Write Column.
CE_n
CLE
ALE
WE_n
tCCS
RE_n
DQx
85h
C1
Figure 42
C2
D0
D1
Asynchronous Data Interface: Change Write Column example
90
D2
D3
4.4.2. Source Synchronous
This section shows examples of commands using the source synchronous data interface. Figure 43 through Figure 45 show an example of
Change Read Column. Figure 45 shows a continuation of data transfer, after the host stops the transfer for a period of time.
91
A
CLK
CE_n
CLE
ALE
W/R_n
DQ[7:0]
tCAD
05h
tCAD
C1
tCAD
C2
E0h
DQS
DQ, DQS driven by host
Figure 43
Source Synchronous Data Interface: Change Read Column example, command issue
92
A
B
tCCS
CLK
CE_n
CLE
ALE
W/R_n
DQ[7:0]
E0h
D0
D1
D2
D3
DQS
DQ, DQS
driven by host
Figure 44
tDQSD turnaround on
DQ, DQS
DQ, DQS driven by device
tDQSHZ
turn-around
on DQ, DQS
Source Synchronous Data Interface: Change Read Column example, data transfer
93
DQ, DQS
driven by
host
B
CLK
CE_n
CLE
ALE
W/R_n
DQ[7:0]
D0
D1
D2
D3
D4
D5
DQS
DQ, DQS driven by device
Figure 45
tDQSHZ
turn-around
on DQ, DQS
DQ, DQS
driven by
host
Source Synchronous Data Interface: Change Read Column example, data transfer continue
94
Figure 46 shows an example of Change Write Column.
tCCS
CLK
CE_n
CLE
ALE
W/R_n
tCAD
DQ[7:0]
85h
tCAD
C1
C2
D0
D1
D2
D3
DQS
DQ, DQS driven by host
Figure 46
Source Synchronous Data Interface: Change Write Column example
95
5. Command Definition
5.1. Command Set
Table 41 outlines the ONFI command set.
The value specified in the first command cycle identifies the command to be performed. Some
commands have a second command cycle as specified in Table 41. Typically, commands that
have a second command cycle include an address.
96
Command
O/M
Read
Multi-plane
Copyback Read
Change Read Column
Change Read Column
Enhanced
Read Cache Random
Read Cache
Sequential
Read Cache End
Block Erase
Multi-plane
Read Status
Read Status Enhanced
Page Program
Multi-plane
Page Cache Program
Copyback Program
Multi-plane
Small Data Move
2
Change Write Column
1
st
nd
1 Cycle
2
Cycle
M
O
O
M
O
00h
00h
00h
05h
06h
30h
32h
35h
E0h
E0h
O
O
00h
31h
31h
O
M
O
M
O
M
O
O
O
O
O
3Fh
60h
60h
70h
78h
80h
80h
80h
85h
85h
85h
M
85h
Acceptable
while
Accessed
LUN is
Busy
Acceptable
while
Other
LUNs are
Busy
Y
Y
Y
Y
Y
Target
level
commands
Y
Y
D0h
D1h
Y
Y
10h
11h
15h
10h
11h
11h
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
1
O
85h
Y
Change Row Address
Read ID
M
90h
Y
Read Parameter Page
M
ECh
Y
Read Unique ID
O
EDh
Y
Get Features
O
EEh
Y
Set Features
O
EFh
Y
Reset LUN
O
FAh
Y
Y
Synchronous Reset
O
FCh
Y
Y
Y
Reset
M
FFh
Y
Y
Y
NOTE:
1. Change Write Column specifies the column address only. Change Row Address
specifies the row address and the column address. Refer to the specific command
definitions.
2. Small Data Move’s first opcode may be 80h if the operation is a program only with no
data output. For the last second cycle of a Small Data Move, it is a 10h command to
confirm the Program or Copyback operation.
Table 41
Command set
Reserved opcodes shall not be used by the device, as the ONFI specification may define the use
of these opcodes in a future revision. Vendor specific opcodes may be used at the discretion of
the vendor and shall never be defined for standard use by ONFI. Future Standardization
opcodes are those opcodes already being used commonly in the industry and may be defined for
standard use by ONFI for those same purposes. Future Standardization opcodes may be used
by compliant ONFI implementations with the common industry usage. Block abstracted NAND
opcodes are opcodes used in a BA NAND implementation.
97
Type
Vendor Specific
Future Standardization
Block Abstracted NAND
Reserved
Opcode
02h – 04h, 08h, 16h – 17h, 19h, 1Dh, 20h – 22h, 25h – 29h, 2Bh,
2Dh – 2Fh, 33h, 36h – 39h, 3B – 3Eh, 40h – 41h, 48h, 4Ch, 53h
– 55h, 68h, 72h – 75h, 84h, 87h – 89h, 91h – BFh, CFh, F1-F4h
23h – 24h, 2Ah, 2Ch, 34h, 3Ah, 65h, 71h, 79h – 7Bh, 81h, 8Ch
C0h - CEh
01h, 07h, 09h – 0Fh, 12h-14h, 18h, 1Ah – 1Ch, 1Eh – 1Fh, 42h –
47h, 49h – 4Bh, 4Dh – 52h, 56h – 5Fh, 62h – 64h, 66h – 67h,
69h – 6Fh, 76h – 77h, 7Ch – 7Fh, 82h – 83h, 86h, 8Ah – 8Bh,
8Dh – 8Fh, D2h –DFh, E1h – EBh, F0h, F5h – F9h, FBh, FDh –
FEh
Table 42
Opcode Reservations
98
5.2. Command Descriptions
The command descriptions in section 5 are shown in a format that is agnostic to the data interface being used (when the command may be used
in either data interface). An example of the agnostic command description for Change Write Column is shown in Figure 47. The agnostic
command examples shall be translated to a command description for the particular data interface selected. The command description for Change
Write Column in the asynchronous data interface is shown in Figure 48. The command description for Change Write Column in the source
synchronous data interface is shown in Figure 49. Note that the timing parameters defined in section 4 shall be observed for each command (e.g.
the tCAD timing parameter for the source synchronous data interface).
Cycle Type
CMD
ADDR
ADDR
DIN
DIN
Dk
Dk+1
tCCS
DQx
85h
C1
C2
SR[6]
Figure 47
99
Agnostic command description
CE_n
CLE
ALE
WE_n
tCCS
RE_n
IOx
85h
C1
Figure 48
100
C2
D0
Asynchronous data interface command description
D1
tCCS
CLK
CE_n
CLE
ALE
W/R_n
tCAD
DQ[7:0]
85h
tCAD
C1
C2
DQS
Figure 49
Source synchronous data interface command description
101
D0
D1
5.3. Reset Definition
The Reset function puts the target in its default power-up state and places the target in the
asynchronous data interface. This command shall only be issued when the host is configured to
the asynchronous data interface. The device shall also recognize and execute the Reset
command when it is configured to the source synchronous data interface. The R/B_n value is
unknown when Reset is issued; R/B_n is guaranteed to be low tWB after the Reset is issued.
Note that some feature settings are retained across Reset commands (as specified in section
5.26). As part of the Reset command, all LUNs are also reset. The command may be executed
with the target in any state, except during power-on when Reset shall not be issued until R/B_n is
set to one. Figure 50 defines the Reset behavior and timings.
CLE
ALE
WE_n
RE_n
IOx
FFh
DQS
tWB
tRST
R/B_n
Figure 50
Reset timing diagram
5.4. Synchronous Reset Definition
The Synchronous Reset command resets the target and all LUNs. The command may be
executed with the target in any state. Figure 51 defines the Synchronous Reset behavior and
102
timings. The R/B_n value is unknown when Synchronous Reset is issued; R/B_n is guaranteed
to be low tWB after the Synchronous Reset is issued.
This command shall be supported by devices that support the source synchronous data interface.
This command is only accepted in source synchronous operation. The host should not issue this
command when the device is configured to the asynchronous data interface. The target shall
remain in the source synchronous data interface following this command.
CLK
CLE
ALE
W/R_n
DQ[7:0]
FCh
DQS
tWB
tRST
R/B_n
Figure 51
Synchronous Reset timing diagram
5.5. Reset LUN Definition
The Reset LUN command is used to reset a particular LUN. This command is accepted by only
the LUN addressed as part of the command. The command may be executed with the LUN in
any state. Figure 52 defines the Reset LUN behavior and timings. The SR[6] value is unknown
when Reset LUN is issued; SR[6] is guaranteed to be low tWB after the Reset LUN command is
issued. This command does not affect the data interface configuration for the target.
Reset LUN should be used to cancel ongoing command operations, if desired. When there are
issues with the target, e.g. a hang condition, the Reset (FFh) or Synchronous Reset (FCh)
commands should be used.
103
Cycle Type
CMD
ADDR
ADDR
ADDR
DQ[7:0]
FAh
R1
R2
R3
tWB
tRST
SR[6]
Figure 52
Reset LUN timing diagram
5.6. Read ID Definition
The Read ID function identifies that the target supports the ONFI specification. If the target
supports the ONFI specification, then the ONFI signature shall be returned. The ONFI signature
is the ASCII encoding of ‘ONFI’ where ‘O’ = 4Fh, ‘N’ = 4Eh, ‘F’ = 46h, and ‘I’ = 49h. Reading
beyond four bytes yields indeterminate values. Figure 53 defines the Read ID behavior and
timings.
When issuing Read ID in the source synchronous data interface, each data byte is received
twice. The host shall only latch one copy of each data byte. See section 4.3.2.5.
For the Read ID command, only addresses of 00h and 20h are valid. To retrieve the ONFI
signature an address of 20h shall be entered (i.e. it is not valid to enter an address of 00h and
read 36 bytes to get the ONFI signature).
Cycle Type
CMD
ADDR
DOUT
DOUT
DOUT
DOUT
4Fh
4Eh
46h
49h
tWHR
DQ[7:0]
90h
20h
R/B_n
Figure 53
Read ID timing diagram for ONFI signature
The Read ID function can also be used to determine the JEDEC manufacturer ID and the device
ID for the particular NAND part by specifying an address of 00h. Figure 54 defines the Read ID
behavior and timings for retrieving the JEDEC manufacturer ID and device ID. Reading beyond
the first two bytes yields values as specified by the manufacturer.
104
Cycle Type
CMD
ADDR
DOUT
DOUT
MID
DID
tWHR
DQ[7:0]
90h
00h
R/B_n
Figure 54
Read ID timing diagram for manufacturer ID
MID
Manufacturer ID for manufacturer of the part, assigned by JEDEC.
DID
Device ID for the part, assigned by the manufacturer.
The Read ID command may be issued using either the asynchronous or source synchronous
data interfaces. The timing parameters for each data interface are shown in Figure 55 and Figure
56.
Figure 55
Read ID command using asynchronous data interface
105
Figure 56
Read ID command using source synchronous data interface
NOTE: The data bytes in Figure 56 are repeated twice (on the rising and falling edge of CLK).
5.7. Read Parameter Page Definition
The Read Parameter Page function retrieves the data structure that describes the target’s
organization, features, timings and other behavioral parameters. There may also be additional
information provided in an extended parameter page. Figure 57 defines the Read Parameter
Page behavior.
Values in the parameter page are static and shall not change. The host is not required to read
the parameter page after power management events.
The first time the host executes the Read Parameter Page command after power-on, timing mode
0 shall be used. If the host determines that the target supports more advanced timing modes,
those supported timing modes may be used for subsequent execution of the Read Parameter
Page command.
The Change Read Column command may be issued following execution of the Read Parameter
Page to read specific portions of the parameter page.
Read Status may be used to check the status of Read Parameter Page during execution. After
completion of the Read Status command, 00h shall be issued by the host on the command line to
continue with the data output flow for the Read Parameter Page command.
Read Status Enhanced and Change Read Column Enhanced shall not be used during execution
of the Read Parameter Page command.
106
A
Cycle Type
CMD
ADDR
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
DQ[7:0]
ECh
00h
P00
P10
...
P01
P11
...
tRR
tWB
tR
R/B_n
A
Cycle Type
DQ[7:0]
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
P0n
P1n
...
E00
E10
...
E01
E11
...
R/B_n
Figure 57
Read Parameter Page command timing
P0k-Pnk
The kth copy of the parameter page data structure. See section 5.7.1. Reading bytes beyond the end of the final
parameter page copy (or beyond the final extended parameter page copy if supported) returns indeterminate values.
E0k-Enk
The kth copy of the extended parameter page data structure. See section 5.7.2. Reading bytes beyond the end of the
final extended parameter page copy returns indeterminate values. This field is only present when the extended parameter
page is supported, as indicated in the Features supported field of the parameter page.
107
5.7.1. Parameter Page Data Structure Definition
Table 43 defines the parameter page data structure. For parameters that span multiple bytes, the
least significant byte of the parameter corresponds to the first byte. See section 1.4.2.3 for more
information on the representation of word and Dword values.
Values are reported in the parameter page in units of bytes when referring to items related to the
size of data access (as in an 8-bit data access device). For example, the target will return how
many data bytes are in a page. For a device that supports 16-bit data access, the host is
required to convert byte values to word values for its use.
Unused fields should be cleared to 0h by the target.
Byte
0-3
4-5
6-7
8-9
O/M
Description
Revision information and features block
M
Parameter page signature
Byte 0: 4Fh, “O”
Byte 1: 4Eh, “N”
Byte 2: 46h, “F”
Byte 3: 49h, “I”
M
Revision number
6-15
Reserved (0)
5
1 = supports ONFI version 2.3
4
1 = supports ONFI version 2.2
3
1 = supports ONFI version 2.1
2
1 = supports ONFI version 2.0
1
1 = supports ONFI version 1.0
0
Reserved (0)
M
Features supported
10-15 Reserved (0)
9
1 = supports EZ NAND
8
1 = supports program page register clear
enhancement
7
1 = supports extended parameter page
6
1 = supports multi-plane read operations
5
1 = supports source synchronous
4
1 = supports odd to even page Copyback
3
1 = supports multi-plane program and erase
operations
2
1 = supports non-sequential page programming
1
1 = supports multiple LUN operations
0
1 = supports 16-bit data bus width
M
Optional commands supported
10-15 Reserved (0)
9
1 = supports Reset LUN
8
1 = supports Small Data Move
7
1 = supports Change Row Address
6
1 = supports Change Read Column Enhanced
5
1 = supports Read Unique ID
4
1 = supports Copyback
3
1 = supports Read Status Enhanced
2
1 = supports Get Features and Set Features
1
1 = supports Read Cache commands
108
Byte
10-11
12-13
14
15-31
32-43
44-63
64
65-66
67-79
O/M
O
O
Description
0
1 = supports Page Cache Program command
Reserved (0)
Extended parameter page length
Number of parameter pages
Reserved (0)
Manufacturer information block
M
Device manufacturer (12 ASCII characters)
M
Device model (20 ASCII characters)
M
JEDEC manufacturer ID
O
Date code
Reserved (0)
116-127
Memory organization block
M
Number of data bytes per page
M
Number of spare bytes per page
Obsolete – Number of data bytes per partial page
Obsolete – Number of spare bytes per partial page
M
Number of pages per block
M
Number of blocks per logical unit (LUN)
M
Number of logical units (LUNs)
M
Number of address cycles
4-7
Column address cycles
0-3
Row address cycles
M
Number of bits per cell
M
Bad blocks maximum per LUN
M
Block endurance
M
Guaranteed valid blocks at beginning of target
M
Block endurance for guaranteed valid blocks
M
Number of programs per page
Obsolete – Partial programming attributes
M
Number of bits ECC correctability
M
Number of plane address bits
4-7
Reserved (0)
0-3
Number of plane address bits
O
Multi-plane operation attributes
6-7
Reserved (0)
5
1 = lower bit XNOR block address restriction
4
1 = read cache supported
3
Address restrictions for cache operations
2
1 = program cache supported
1
1 = no block address restrictions
0
Overlapped / concurrent multi-plane support
O
EZ NAND support
3-7
Reserved (0)
2
1 = Requires Copyback Adjacency
1
1 = supports Copyback for other planes & LUNs
0
1 = supports enable/disable of automatic retries
Reserved (0)
128
Electrical parameters block
M
I/O pin capacitance, maximum
80-83
84-85
86-89
90-91
92-95
96-99
100
101
102
103-104
105-106
107
108-109
110
111
112
113
114
115
109
Byte
129-130
O/M
Description
M
Asynchronous timing mode support
6-15
Reserved (0)
5
1 = supports timing mode 5
4
1 = supports timing mode 4
3
1 = supports timing mode 3
2
1 = supports timing mode 2
1
1 = supports timing mode 1
0
1 = supports timing mode 0, shall be 1
Obsolete – Asynchronous program cache timing mode support
133-134
M
135-136
M
137-138
M
139-140
M
tPROG Maximum page program time (µs)
tBERS Maximum block erase time (µs)
tR Maximum page read time (µs)
tCCS Minimum change column setup time (ns)
141-142
O
143
O
144-145
O
Source synchronous timing mode support
6-15
Reserved (0)
5
1 = supports timing mode 5
4
1 = supports timing mode 4
3
1 = supports timing mode 3
2
1 = supports timing mode 2
1
1 = supports timing mode 1
0
1 = supports timing mode 0
Source synchronous features
3-7
Reserved (0)
2
1 = device supports CLK stopped for data input
1
1 = typical capacitance values present
0
tCAD value to use
CLK input pin capacitance, typical
146-147
O
I/O pin capacitance, typical
148-149
O
Input pin capacitance, typical
150
M
Input pin capacitance, maximum
151
M
152-153
O
154-155
O
Driver strength support
3-7
Reserved (0)
2
1 = supports 18 Ohm drive strength
1
1 = supports 25 Ohm drive strength
0
1 = supports driver strength settings
tR Maximum multi-plane page read time (µs)
tADL Program page register clear enhancement tADL value (ns)
156-157
O
tR Typical page read time for EZ NAND (µs)
131-132
158-163
Reserved (0)
164-165
166-253
254-255
Vendor block
M
Vendor specific Revision number
Vendor specific
M
Integrity CRC
256-511
512-767
768+
Redundant Parameter Pages
M
Value of bytes 0-255
M
Value of bytes 0-255
O
Additional redundant parameter pages
110
Table 43
Parameter page definitions
5.7.1.1. Byte 0-3: Parameter page signature
This field contains the parameter page signature. When two or more bytes of the signature are
valid, then it denotes that a valid copy of the parameter page is present.
Byte 0 shall be set to 4Fh.
Byte 1 shall be set to 4Eh.
Byte 2 shall be set to 46h.
Byte 3 shall be set to 49h.
5.7.1.2. Byte 4-5: Revision number
This field indicates the revisions of the ONFI specification that the target complies to. The target
may support multiple revisions of the ONFI specification. This is a bit field where each defined bit
corresponds to a particular specification revision that the target may support.
Bit 0 shall be cleared to zero.
Bit 1 when set to one indicates that the target supports the ONFI revision 1.0 specification.
Bit 2 when set to one indicates that the target supports the ONFI revision 2.0 specification.
Bit 3 when set to one indicates that the target supports the ONFI revision 2.1 specification.
Bit 4 when set to one indicates that the target supports the ONFI revision 2.2 specification.
Bit 5 when set to one indicates that the target supports the ONFI revision 2.3 specification.
Bits 6-15 are reserved and shall be cleared to zero.
5.7.1.3. Byte 6-7: Features supported
This field indicates the optional features that the target supports.
Bit 0 when set to one indicates that the target’s data bus width is 16-bits. Bit 0 when cleared to
zero indicates that the target’s data bus width is 8-bits. The host shall use the indicated data bus
width for all ONFI commands that are defined to be transferred at the bus width (x8 or x16). Note
that some commands, like Read ID, always transfer data as 8-bit only. If the source synchronous
data interface is supported, then the data bus width shall be 8-bits.
Bit 1 when set to one indicates that the target supports multiple LUN operations (see section
3.1.3). If bit 1 is cleared to zero, then the host shall not issue commands to a LUN unless all
other LUNs on the Target are idle (i.e. R/B_n is set to one).
Bit 2 when set to one indicates that the target supports non-sequential page programming
operations, such that the host may program pages within a block in arbitrary order. Bit 2 when
cleared to zero indicates that the target does not support non-sequential page programming
operations. If bit 2 is cleared to zero, the host shall program all pages within a block in order
starting with page 0.
111
Bit 3 when set to one indicates that the target supports multi-plane program and erase
operations. Refer to section 5.7.1.28.
Bit 4 when set to one indicates that there are no even / odd page restrictions for Copyback
operations. Specifically, a read operation may access an odd page and then program the
contents to an even page using Copyback. Alternatively, a read operation may access an even
page and then program the contents to an odd page using Copyback. Bit 4 when cleared to zero
indicates that the host shall ensure that Copyback reads and programs from odd page to odd
page or alternatively from even page to even page.
Bit 5 when set to one indicates that the source synchronous data interface is supported by the
target. If bit 5 is set to one, then the target shall indicate the source synchronous timing modes
supported in the source synchronous timing mode support field. Bit 5 when cleared to zero
indicates that the source synchronous data interface is not supported by the target.
Bit 6 when set to one indicates that the target supports multi-plane read operations. Refer to
section 5.7.1.28.
Bit 7 when set to one indicates the target includes an extended parameter page that is stored in
the data bytes following the last copy of the parameter page. If bit 7 is cleared to zero, then an
extended parameter page is not supported. Refer to section 5.7.2. This bit shall be cleared to
zero for devices that support EZ NAND. Note: This bit was inadvertently specified in the BA
NAND specification to show support for BA NAND. If the device supports BA NAND, then the
number of bits of ECC correctability should be cleared to 0h in byte 112 of the parameter page.
Bit 8 when set to one indicates that the target supports clearing only the page register for the LUN
addressed with the Program (80h) command. If bit 8 is cleared to zero, then a Program (80h)
command clears the page register for each LUN that is part of the target. At power-on, the device
clears the page register for each LUN that is part of the target. Refer to section 5.26.1 for how to
enable this feature.
Bit 9 when set to one indicates that the target supports EZ NAND. If bit 9 is cleared to zero, then
the target does not support EZ NAND and is configured as raw NAND.
Bits 10-15 are reserved and shall be cleared to zero.
5.7.1.4. Byte 8-9: Optional commands supported
This field indicates the optional commands that the target supports.
Bit 0 when set to one indicates that the target supports the Page Cache Program command. If bit
0 is cleared to zero, the host shall not issue the Page Cache Program command to the target.
For targets that support EZ NAND, this field shall be cleared to zero indicating that the target
does not support the Page Cache Program command.
Bit 1 when set to one indicates that the target supports the Read Cache Random, Read Cache
Sequential, and Read Cache End commands. If bit 1 is cleared to zero, the host shall not issue
the Read Cache Sequential, Read Cache Random, or Read Cache End commands to the target.
For targets that support EZ NAND, this field shall be cleared to zero indicating that the target
does not support Read Cache Random, Read Cache Sequential, or Read Cache End
commands.
Bit 2 when set to one indicates that the target supports the Get Features and Set Features
commands. If bit 2 is cleared to zero, the host shall not issue the Get Features or Set Features
commands to the target.
112
Bit 3 when set to one indicates that the target supports the Read Status Enhanced command. If
bit 3 is cleared to zero, the host shall not issue the Read Status Enhanced command to the
target. Read Status Enhanced shall be supported if the target has multiple LUNs or supports
multi-plane operations.
Bit 4 when set to one indicates that the target supports the Copyback Program and Copyback
Read commands. If bit 4 is cleared to zero, the host shall not issue the Copyback Program or
Copyback Read commands to the target. If multi-plane operations are supported and this bit is
set to one, then multi-plane copyback operations shall be supported.
Bit 5 when set to one indicates that the target supports the Read Unique ID command. If bit 5 is
cleared to zero, the host shall not issue the Read Unique ID command to the target.
Bit 6 when set to one indicates that the target supports the Change Read Column Enhanced
command. If bit 6 is cleared to zero, the host shall not issue the Change Read Column Enhanced
command to the target.
Bit 7 when set to one indicates that the target supports the Change Row Address command. If bit
7 is cleared to zero, the host shall not issue the Change Row Address command to the target.
Bit 8 when set to one indicates that the target supports the Small Data Move command for both
Program and Copyback operations. If bit 8 is cleared to zero, the target does not support the
Small Data Move command for Program or Copyback operations. The Small Data Move
command is mutually exclusive with overlapped multi-plane support. Refer to section 5.19.
When bit 8 is set to one, the device shall support the 11h command to flush any internal data
pipeline regardless of whether multi-plane operations are supported.
Bit 9 when set to one indicates that the target supports the Reset LUN command. If bit 9 is
cleared to zero, the host shall not issue the Reset LUN command.
Bits 10-15 are reserved and shall be cleared to zero.
5.7.1.5. Byte 12-13: Extended parameter page length
If the target supports an extended parameter page as indicated in the Features supported field,
then this field specifies the length of the extended parameter page in multiples of 16 bytes. Thus,
a value of 2 corresponds to 32 bytes and a value of 3 corresponds to 48 bytes. The minimum
size is 3, corresponding to 48 bytes.
5.7.1.6. Byte 14: Number of parameter pages
If the target supports an extended parameter page as indicated in the Features supported field,
then this field specifies the number of parameter pages present, including the original and the
subsequent redundant versions. As an example, a value of 3 means that there are three
parameter pages present and thus the extended parameter page starts at byte 768. The number
of extended parameter pages should match the number of parameter pages.
5.7.1.7. Byte 32-43: Device manufacturer
This field contains the manufacturer of the device. The content of this field is an ASCII character
string of twelve bytes. The device shall pad the character string with spaces (20h), if necessary,
to ensure that the string is the proper length.
There is no standard for how the manufacturer represents their name in the ASCII string. If the
host requires use of a standard manufacturer ID, it should use the JEDEC manufacturer ID (refer
to section 5.7.1.9).
113
5.7.1.8. Byte 44-63: Device model
This field contains the model number of the device. The content of this field is an ASCII character
string of twenty bytes. The device shall pad the character string with spaces (20h), if necessary,
to ensure that the string is the proper length.
5.7.1.9. Byte 64: JEDEC manufacturer ID
This field contains the JEDEC manufacturer ID for the manufacturer of the device.
5.7.1.10. Byte 65-66: Date code
This field contains a date code for the time of manufacture of the device. Byte 65 shall contain
the two least significant digits of the year (e.g. a value of 05h to represent the year 2005). Byte
66 shall contain the workweek, where a value of 00h indicates the first week of January.
If the date code functionality is not implemented, the value in this field shall be 0000h.
5.7.1.11. Byte 80-83: Number of data bytes per page
This field contains the number of data bytes per page. The value reported in this field shall be a
power of two. The minimum value that shall be reported is 512 bytes.
5.7.1.12. Byte 84-85: Number of spare bytes per page
This field contains the number of spare bytes per page. There are no restrictions on the value.
Appendix B lists recommendations for the number of bytes per page based on the page size and
the number of bits of ECC correctability for the device.
5.7.1.13. Byte 86-89: Obsolete – Number of data bytes per partial page
This field is obsolete. It previously contained the number of data bytes per partial page.
5.7.1.14. Byte 90-91: Obsolete – Number of spare bytes per partial page
This field is obsolete. It previously contained the number of spare bytes per partial page.
5.7.1.15. Byte 92-95: Number of pages per block
This field contains the number of pages per block. This value shall be a multiple of 32. Refer to
section 3.1 for addressing requirements.
5.7.1.16. Byte 96-99: Number of blocks per logical unit
This field contains the number of blocks per logical unit. There are no restrictions on this value.
Refer to section 3.1 for addressing requirements.
5.7.1.17. Byte 100: Number of logical units (LUNs)
This field indicates the number of logical units the target supports. Logical unit numbers are
sequential, beginning with a LUN address of 0. This field shall be greater than zero.
114
5.7.1.18. Byte 101: Number of Address Cycles
This field indicates the number of address cycles used for row and column addresses. The
reported number of address cycles shall be used by the host in operations that require row and/or
column addresses (e.g. Page Program).
Bits 0-3 indicate the number of address cycles used for the row address. This field shall be
greater than zero.
Bits 4-7 indicate the number of address cycles used for the column address. This field shall be
greater than zero.
NOTE: Throughout this specification examples are shown with 2-byte column addresses and 3byte row addresses. However, the host is responsible for providing the number of column and
row address cycles in each of these sequences based on the values in this field.
5.7.1.19. Byte 102: Number of bits per cell
This field indicates the number of bits per cell in the Flash array. This field shall be greater than
zero.
For some devices, including implementations supporting EZ NAND, the device may be
constructed of Flash arrays of different types. A value of FFh indicates that the number of bits
per cell is not specified.
5.7.1.20. Byte 103-104: Bad blocks maximum per LUN
This field contains the maximum number of blocks that may be defective at manufacture and over
the life of the device per LUN. The maximum rating assumes that the host is following the block
endurance requirements and the ECC requirements reported in the parameter page.
5.7.1.21. Byte 105-106: Block endurance
This field indicates the maximum number of program/erase cycles per addressable page/block.
This value assumes that the host is using at least the minimum ECC correctability reported in the
parameter page.
A page may be programmed in partial operations subject to the value reported in the Number of
programs per page field. However, programming different locations within the same page does
not count against this value more than once per full page.
The block endurance is reported in terms of a value and a multiplier according to the following
multiplier
equation: value x 10
. Byte 105 comprises the value. Byte 106 comprises the multiplier.
For example, a target with an endurance of 75,000 cycles would report this as a value of 75 and a
3
multiplier of 3 (75 x 10 ). For a write once device, the target shall report a value of 1 and a
multiplier of 0. For a read-only device, the target shall report a value of 0 and a multiplier of 0.
The value field shall be the smallest possible; for example 100,000 shall be reported as a value of
5
1 and a multiplier of 5 (1 x 10 ).
5.7.1.22. Byte 107: Guaranteed valid blocks at beginning of target
This field indicates the number of guaranteed valid blocks starting at block address 0 of the
target. The minimum value for this field is 1h. The blocks are guaranteed to be valid for the
endurance specified for this area (see section 5.7.1.23) when the host follows the specified
number of bits to correct.
115
5.7.1.23. Byte 108-109: Block endurance for guaranteed valid blocks
This field indicates the minimum number of program/erase cycles per addressable page/block in
the guaranteed valid block area (see section 5.7.1.22). This value requires that the host is using
at least the minimum ECC correctability reported in the parameter page. This value is not
encoded. If the value is 0000h, then no minimum number of cycles is specified, though the
block(s) are guaranteed valid from the factory.
5.7.1.24. Byte 110: Number of programs per page
This field indicates the maximum number of times a portion of a page may be programmed
without an erase operation. After the number of programming operations specified have been
performed, the host shall issue an erase operation to that block before further program operations
to the affected page. This field shall be greater than zero. Programming the same portion of a
page without an erase operation results in indeterminate page contents.
5.7.1.25. Byte 111: Obsolete – Partial programming attributes
This field is obsolete. It previously indicated the attributes for partial page programming that the
target supports.
5.7.1.26. Byte 112: Number of bits ECC correctability
This field indicates the number of bits that the host should be able to correct per 512 bytes of
data. With this specified amount of error correction by the host, the target shall achieve the block
endurance specified in the parameter page. When the specified amount of error correction is
applied by the host and the block endurance is followed, then the maximum number of bad blocks
shall not be exceeded by the device. All used bytes in the page shall be protected by host
controller ECC including the spare bytes if the minimum ECC requirement has a value greater
than zero.
If the recommended ECC codeword size is not 512 bytes, then this field shall be set to FFh. The
host should then read the Extended ECC Information that is part of the extended parameter page
to retrieve the ECC requirements for this device.
When this value is cleared to zero, the target shall return valid data. For targets that support EZ
NAND, this field shall be cleared to zero indicating that the target returns corrected data to the
host.
5.7.1.27. Byte 113: Multi-plane addressing
This field describes parameters for multi-plane addressing.
Bits 0-3 indicate the number of bits that are used for multi-plane addressing. This value shall be
greater than 0h when multi-plane operations are supported. For information on the plane address
location, refer to section 3.1.1.
Bits 4-7 are reserved.
5.7.1.28. Byte 114: Multi-plane operation attributes
This field describes attributes for multi-plane operations. This byte is mandatory when multiplane operations are supported as indicated in the Features supported field.
Bit 0 indicates whether overlapped multi-plane operations are supported. If bit 0 is set to one,
then overlapped multi-plane operations are supported. If bit 0 is cleared to zero, then concurrent
multi-plane operations are supported.
116
Bit 1 indicates that there are no block address restrictions for the multi-plane operation. If set to
one all block address bits may be different between multi-plane operations. If cleared to zero,
there are block address restrictions. Refer to bit 5 for the specific block address restrictions
required.
Bit 2 indicates whether program cache is supported with multi-plane programs. If set to one then
program cache is supported for multi-plane program operations. If cleared to zero then program
cache is not supported for multi-plane program operations. Note that program cache shall not be
used with multi-plane copyback program operations. See bit 3 for restrictions on the multi-plane
addresses that may be used.
Bit 3 indicates whether the block address bits other than the multi-plane address bits of multiplane addresses may change during either: a) a program cache sequence between 15h
commands, or b) a read cache sequence between 31h commands. If set to one and bit 2 is set
to one, then the host may change the number of multi-plane addresses and the value of the block
address bits (other than the multi-plane address bits) in the program cache sequence. If set to
one and bit 4 is set to one, then the host may change the number of multi-plane addresses and
the value of the block address bits (other than the multi-plane address bits) in the read cache
sequence. If cleared to zero and bit 2 is set to one, then for each program cache operation the
block address bits (other than the plane address bits) and number of multi-plane addresses
issued to the LUN shall be the same. If cleared to zero and bit 4 is set to one, then for each read
cache operation the block address bits (other than the multi-plane address bits) and number of
multi-plane addresses issued to the LUN shall be the same.
Bit 4 indicates whether read cache is supported with multi-plane reads. If set to one then read
cache is supported for multi-plane read operations. If cleared to zero then read cache is not
supported for multi-plane read operations. Note that read cache shall not be used with multiplane copyback read operations.
Bit 5 indicates the type of block address restrictions required for the multi-plane operation. If set
to one then all block address bits (other than the multi-plane address bits) shall be the same if the
XNOR of the lower multi-plane address bits between two multi-plane addresses is one. If cleared
to zero, all block address bits (other than the multi-plane address bits) shall be the same
regardless of the multi-plane address bits between two plane addresses. See section 3.1.1.1 for
a detailed definition of interleaved block address restrictions. These restrictions apply to all multiplane operations (Read, Program, Erase, and Copyback Program).
Bits 6-7 are reserved.
5.7.1.1.
Byte 115: EZ NAND support
This field describes the EZ NAND attributes that the target supports. This field is only used if EZ
NAND is supported by the target.
Bit 0 indicates whether the target supports automatic retries being enabled and disabled explicitly
by the host using Set Features. If this bit is set to one, then the host may enable or disable
automatic retries using Set Features. If this bit is cleared to zero, then the EZ NAND controller
determines whether to perform a retry without host intervention. If the EZ NAND controller
executes an automatic retry, the typical page read time (tR) may be exceeded.
Bit 1 indicates whether Copyback is supported with destination planes or LUNs that are different
from the source. If this bit is set to one, then the plane or LUN of the destination for the
Copyback may be different than the source. If this bit is cleared to zero, then the plane or LUN of
the destination for a Copyback shall be the same as the source.
117
Bit 2 indicates whether the target requires Copyback Read and Copyback Program command
adjacency. If this bit is set to one, then each Copyback Read command issued shall be followed
by a Copyback Program command prior to any additional Copyback Read command being issued
to the target. I.e., A Copyback Read command shall be explicitly paired with aCopyback Program
command). If this bit is cleared to zero, then there may be multiple Copyback Read commands
prior to issuing a Copyback Program command. For ONFI 2.3, Copyback Adjacency is required
and this bit shall be set to one.
Bits 3-7 are reserved.
5.7.1.2. Byte 128: I/O pin capacitance, maximum
This field indicates the maximum I/O pin capacitance for the target in pF. This may be used by
the host to calculate the load for the data bus. Refer to section 2.11.
5.7.1.3. Byte 129-130: Asynchronous timing mode support
This field indicates the asynchronous timing modes supported. The target shall always support
asynchronous timing mode 0.
Bit 0 shall be set to one. It indicates that the target supports asynchronous timing mode 0.
Bit 1 when set to one indicates that the target supports asynchronous timing mode 1.
Bit 2 when set to one indicates that the target supports asynchronous timing mode 2.
Bit 3 when set to one indicates that the target supports asynchronous timing mode 3.
Bit 4 when set to one indicates that the target supports asynchronous timing mode 4.
Bit 5 when set to one indicates that the target supports asynchronous timing mode 5.
Bits 6-15 are reserved and shall be cleared to zero.
5.7.1.4. Byte 131-132: Obsolete – Asynchronous program cache timing
mode support
This field is obsolete. It previously indicated the asynchronous timing modes supported for Page
Cache Program operations.
5.7.1.5. Byte 133-134: Maximum page program time
This field indicates the maximum page program time (tPROG) in microseconds.
5.7.1.6. Byte 135-136: Maximum block erase time
This field indicates the maximum block erase time (tBERS) in microseconds.
5.7.1.7. Byte 137-138: Maximum page read time
This field indicates the maximum page read time (tR) in microseconds. For devices that support
EZ NAND, this value is the tR maximum at the end of life and is related to the uncorrectable bit
error rate (UBER) specified for the device.
118
5.7.1.8. Byte 139-140: Minimum change column setup time
This field indicates the minimum change column setup time (tCCS) in nanoseconds. After issuing
a Change Read Column command, the host shall not read data until a minimum of tCCS time has
elapsed. After issuing a Change Write Column command including all column address cycles,
the host shall not write data until a minimum of tCCS time has elapsed. The value of tCCS shall
always be longer than or equal to tWHR and tADL when the source synchronous data interface is
supported.
5.7.1.9. Byte 141-142: Source synchronous timing mode support
This field indicates the source synchronous timing modes supported. If the source synchronous
data interface is supported by the target, at least one source synchronous timing mode shall be
supported. The target shall support an inclusive range of source synchronous timing modes (i.e.
if timing mode n-1 and n+1 are supported, then the target shall also support timing mode n).
Bit 0 when set to one indicates that the target supports source synchronous timing mode 0.
Bit 1 when set to one indicates that the target supports source synchronous timing mode 1.
Bit 2 when set to one indicates that the target supports source synchronous timing mode 2.
Bit 3 when set to one indicates that the target supports source synchronous timing mode 3.
Bit 4 when set to one indicates that the target supports source synchronous timing mode 4.
Bit 5 when set to one indicates that the target supports source synchronous timing mode 5.
Bits 6-15 are reserved and shall be cleared to zero.
5.7.1.10. Byte 143: Source synchronous features
This field describes features and attributes for source synchronous operation. This byte is
mandatory when the source synchronous data interface is supported.
Bit 0 indicates the tCAD value that shall be used by the host. If bit 0 is set to one, then the host
shall use the tCADs (slow) value in source synchronous command, address and data transfers. If
bit 0 is cleared to zero, then the host shall use the tCADf (fast) value in source synchronous
command, address and data transfers.
Bit 1 indicates if the typical CLK, I/O and input pin capacitance values are reported in the
parameter page. If bit 1 is set to one, then the typical CLK, I/O and input pin capacitance values
are reported in the parameter page. If bit 1 is cleared to zero, then the typical capacitance fields
are not used.
Bit 2 indicates that the device supports the CLK being stopped during data input, as described in
Figure 34. If bit 2 is set to one, then the host may optionally stop the CLK during data input for
power savings. If bit 2 is set to one, the host may pause data while the CLK is stopped. If bit 2 is
cleared to zero, then the host shall leave CLK running during data input.
Bits 3-7 are reserved.
5.7.1.11. Byte 144-145: CLK input pin capacitance, typical
This field indicates the typical CLK input pin capacitance for the target. This value applies to the
CLK_t and CLK_c signals. This field is specified in 0.1 pF units. For example, a value of 31
corresponds to 3.1 pF. The variance from this value is less than +/- 0.5 pF per LUN per x8 data
119
bus. As an example, if two LUNs are present per x8 data bus than the total variance is less than
+/- 1.0 pF. This value is only valid if the typical capacitance values are supported as indicated in
the source synchronous features field. Additional constraints on the CLK input pin capacitance
are specified in section 4.2.3.
5.7.1.12. Byte 146-147: I/O pin capacitance, typical
This field indicates the typical I/O pin capacitance for the target. This field is specified in 0.1 pF
units. For example, a value of 31 corresponds to 3.1 pF. The variance from this value is less
than +/- 0.5 pF per LUN per x8 data bus. As an example, if two LUNs are present than the total
variance is less than +/- 1 pF. This value is only valid if the typical capacitance values are
supported as indicated in the source synchronous features field. Additional constraints on the I/O
pin capacitance are specified in section 4.2.3.
5.7.1.13. Byte 148-149: Input pin capacitance, typical
This field indicates the typical input pin capacitance for the target. This value applies to all inputs
except the following: CLK_t, CLK_c, CE_n and WP_n signals. This field is specified in 0.1 pF
units. For example, a value of 31 corresponds to 3.1 pF. The variance from this value is less
than +/- 0.5 pF per LUN per x8 data bus. As an example, if two LUNs are present than the total
variance is less than +/- 1 pF. This value is only valid if the typical capacitance values are
supported as indicated in the source synchronous features field. Additional constraints on the
input pin capacitance are specified in section 4.2.3.
5.7.1.14. Byte 150: Input pin capacitance, maximum
This field indicates the maximum input pin capacitance for the target in pF. This value applies to
all inputs, including CLK_t, CLK_c, CE_n, and WP_n. This may be used by the host to calculate
the load for the data bus. Refer to section 2.11.
5.7.1.15. Byte 151: Driver strength support
This field describes if the target supports configurable driver strengths and its associated
features.
Bit 0 when set to one indicates that the target supports configurable driver strength settings as
defined in Table 26. If this bit is set to one, then the device shall support both the 35 Ohm and 50
Ohm settings. If this bit is set to one, then the device shall power-on with a driver strength at the
35 Ohm value defined in Table 26. If this bit is cleared to zero, then the driver strength at poweron is undefined. This bit shall be set to one for devices that support the source synchronous data
interface.
Bit 1 when set to one indicates that the target supports the 25 Ohm setting in Table 26 for use in
the I/O Drive Strength setting. This bit shall be set to one for devices that support the source
synchronous data interface.
Bit 2 when set to one indicates that the target supports the 18 Ohm setting in Table 26 for use in
the I/O Drive Strength setting. This bit shall be set to one for devices that support the source
synchronous data interface.
Bits 3-7 are reserved.
5.7.1.16. Byte 152-153: Maximum multi-plane page read time
This field indicates the maximum page read time (tR) for multi-plane page reads in microseconds.
Multi-plane page read times may be longer than non-multi-plane pages read times. This field
120
shall be supported if the target supports multi-plane reads as indicated in the Features supported
field.
5.7.1.17. Byte 154-155: Program page register clear enhancement tADL value
This field indicates the ALE to data loading time (tADL) in nanoseconds when the program page
register clear enhancement is enabled. If the program page register clear enhancement is
disabled, then the tADL value is as defined for the selected timing mode. This increased tADL
value only applies to Program (80h) command sequences; it does not apply for Set Features,
Copyback, or other commands.
5.7.1.1.
Byte 156-157: Typical page read time for EZ NAND
This field indicates the typical page read time (tR) in microseconds for devices that support EZ
NAND. This field is not used for devices that do not support EZ NAND. For devices that include
multiple bits per cell in the NAND used, this value is an average of the tR typical values for the
pages (e.g. lower and upper pages).
5.7.1.2. Byte 164-165: Vendor specific Revision number
This field indicates a vendor specific revision number. This field should be used by vendors to
indicate the supported layout for the vendor specific parameter page area and the vendor specific
feature addresses. The format of this field is vendor specific.
5.7.1.3. Byte 166-253: Vendor specific
This field is reserved for vendor specific use.
5.7.1.4. Byte 254-255: Integrity CRC
The Integrity CRC (Cyclic Redundancy Check) field is used to verify that the contents of the
parameter page were transferred correctly to the host. The CRC of the parameter page is a word
(16-bit) field. The CRC calculation covers all of data between byte 0 and byte 253 of the
parameter page inclusive.
The CRC shall be calculated on byte (8-bit) quantities starting with byte 0 in the parameter page.
The bits in the 8-bit quantity are processed from the most significant bit (bit 7) to the least
significant bit (bit 0).
The CRC shall be calculated using the following 16-bit generator polynomial:
G(X) = X16 + X15 + X2 + 1
This polynomial in hex may be represented as 8005h.
The CRC value shall be initialized with a value of 4F4Eh before the calculation begins. There is
no XOR applied to the final CRC value after it is calculated. There is no reversal of the data
bytes or the CRC calculated value.
5.7.1.5. Byte 256-511: Redundant Parameter Page 1
This field shall contain the values of bytes 0-255 of the parameter page. Byte 256 is the value of
byte 0.
The redundant parameter page is used when the integrity CRC indicates that there was an error
in bytes 0-255. The redundant parameter page shall be stored in non-volatile media; the target
shall not create these bytes by retransmitting the first 256 bytes.
121
5.7.1.6. Byte 512-767: Redundant Parameter Page 2
This field shall contain the values of bytes 0-255 of the parameter page. Byte 512 is the value of
byte 0.
The redundant parameter page is used when the integrity CRC indicates that there was an error
in bytes 0-255 and in the first redundant parameter page. The redundant parameter page shall
be stored in non-volatile media; the target shall not create these bytes by retransmitting the first
256 bytes.
5.7.1.7. Byte 768+: Additional Redundant Parameter Pages
Bytes at offset 768 and above may contain additional redundant copies of the parameter page.
There is no limit to the number of redundant parameter pages that the target may provide. The
target may provide additional copies to guard against the case where all three mandatory copies
have invalid CRC checks.
The host should determine whether an additional parameter page is present by checking the first
Dword. If at least two out of four bytes match the parameter page signature, then an additional
parameter page is present.
5.7.2. Extended Parameter Page Data Structure Definition
The extended parameter page, if present, provides additional information to the host that there
was insufficient space to include in the parameter page. The extended parameter page is
organized in sections. Each section is a multiple of 16 bytes in length. The section types are
specified in Table 44.
Section Type
0
1
2
3-255
Section Definition
Unused section marker. No section present.
Section type and length specifiers.
Extended ECC information.
Reserved
Table 44
Section Type Definitions
Section types shall be specified in the extended parameter page in order (other than section type
value 0). For example, if section type 12 and section type 15 were both present in the extended
parameter page then section type 12 shall precede section type 15. There shall only be one
instantiation of each section type. All unused sections shall be marked with a section type value
of 0. When software encounters a section type value of 0, this marks the end of the valid
sections.
Table 45 defines the layout of section type 1. Section type 1 specifies additional sections when
more than eight sections are present in the extended parameter page. The length of section type
1 shall be a multiple of 16 bytes.
122
Byte
O/M
0
1
2
3
4
5
6 – (end)
Description
M
M
O
O
O
O
O
Table 45
Section 8 type
Section 8 length
Section 9 type
Section 9 length
Section 10 type
Section 10 length
Section 11 – n type & lengths
Section Type 1: Additional Section Type and Length Specifiers
Table 46 defines the layout of section type 2. Section type 2 specifies extended ECC information.
Each extended ECC information block is eight bytes in length. If an extended ECC information
block is not specified, then all values in that block shall be cleared to 0h. The length of section
type 2 shall be a multiple of 16 bytes.
Byte
0-7
8-15
16 – (end)
O/M
M
O
O
Table 46
Description
Extended ECC information block 0
Extended ECC information block 1
Extended ECC information block 2 – n (if present)
Section Type 2: Extended ECC Information
The definition of the extended ECC information block is specified in section 3.3.
Table 47 defines the extended parameter page data structure. For parameters that span multiple
bytes, the least significant byte of the parameter corresponds to the first byte. See section
1.4.2.3 for more information on the representation of word and Dword values.
Values are reported in the extended parameter page in units of bytes when referring to items
related to the size of data access (as in an 8-bit data access device). For example, the target will
return how many data bytes are in a page. For a device that supports 16-bit data access, the
host is required to convert byte values to word values for its use.
Unused fields should be cleared to 0h by the target.
123
Byte
O/M
0-1
2-5
6-15
16
17
18
19
20
21
22-31
32 – (end)
Description
Revision information and features block
M
Integrity CRC
M
Extended parameter page signature
Byte 0: 45h, “E”
Byte 1: 50h, “P”
Byte 2: 50h, “P”
Byte 3: 53h, “S”
Reserved (0)
M
Section 0 type
M
Section 0 length
M
Section 1 type
M
Section 1 length
O
Section 2 type
O
Section 2 length
O
Section 3 – 7 types & lengths
M
Section information
Table 47
Extended Parameter Page definition
5.7.2.1. Byte 0-1: Integrity CRC
The Integrity CRC (Cyclic Redundancy Check) field is used to verify that the contents of the
extended parameter page were transferred correctly to the host. The CRC of the extended
parameter page is a word (16-bit) field. The CRC calculation covers all of data between byte 2
and the end of the extended parameter page inclusive.
The CRC shall be calculated on byte (8-bit) quantities starting with byte 2 in the extended
parameter page to the end of the extended parameter page. The bits in the 8-bit quantity are
processed from the most significant bit (bit 7) to the least significant bit (bit 0).
The CRC shall be calculated using the following 16-bit generator polynomial:
G(X) = X16 + X15 + X2 + 1
This polynomial in hex may be represented as 8005h.
The CRC value shall be initialized with a value of 4F4Eh before the calculation begins. There is
no XOR applied to the final CRC value after it is calculated. There is no reversal of the data
bytes or the CRC calculated value.
5.7.2.2. Byte 2-5: Extended parameter page signature
This field contains the extended parameter page signature. When two or more bytes of the
signature are valid, then it denotes that a valid copy of the extended parameter page is present.
Byte 2 shall be set to 45h.
Byte 3 shall be set to 50h.
Byte 4 shall be set to 50h.
Byte 5 shall be set to 53h.
124
5.7.2.3. Byte 16: Section 0 type
Section 0 is the first section in the extended parameter page and begins at byte offset 32. This
field specifies the type of section 0. Section types are defined in Table 44.
5.7.2.4. Byte 17: Section 0 length
Section 0 is the first section in the extended parameter page and begins at byte offset 32. This
field specifies the length of section 0. The length is specified in multiples of 16 bytes. Thus, a
value of 1 corresponds to 16 bytes and a value of 2 corresponds to 32 bytes.
5.7.2.5. Byte 18: Section 1 type
Section 1 is the second section in the extended parameter page and starts immediately following
section 0. This field specifies the type of section 1. Section types are defined in Table 44. If
section 1 is not present, then the type field shall be cleared to 0.
5.7.2.6. Byte 19: Section 1 length
Section 1 is the second section in the extended parameter page and starts immediately following
section 0. This field specifies the length of section 1. The length is specified in multiples of 16
bytes. Thus, a value of 1 corresponds to 16 bytes and a value of 2 corresponds to 32 bytes. If
section 1 is not present, then the length field shall be cleared to 0.
5.7.2.7. Byte 20: Section 2 type
Section 2 is the third section in the extended parameter page and starts immediately following
section 1. This field specifies the type of section 2. Section types are defined in Table 44. If
section 2 is not present, then the type field shall be cleared to 0.
5.7.2.8. Byte 21: Section 2 length
Section 2 is the third section in the extended parameter page and starts immediately following
section 1. This field specifies the length of section 2. The length is specified in multiples of 16
bytes. Thus, a value of 1 corresponds to 16 bytes and a value of 2 corresponds to 32 bytes. If
section 2 is not present, then the length field shall be cleared to 0.
5.7.2.9. Byte 22-31: Section 3 – 7 types and lengths
Bytes 22-31 define the type and lengths for sections 3 – 7 in order, following the same definition
and layout as section 0 and 1 type and length definitions. If a section is not present, then the type
and length fields for that section shall be cleared to 0.
5.7.2.10. Byte 32 – (end): Section information
Section 0 begins at byte offset 32 and is a multiple of 16 bytes. If there are additional sections
(section 1, 2, 3, etc), each section starts immediately following the previous section and is a
multiple of 16 bytes.
5.8. Read Unique ID Definition
The Read Unique ID function is used to retrieve the 16 byte unique ID (UID) for the device. The
unique ID when combined with the device manufacturer shall be unique.
The UID data may be stored within the Flash array. To allow the host to determine if the UID is
without bit errors, the UID is returned with its complement, as shown in Table 48. If the XOR of
the UID and its bit-wise complement is all ones, then the UID is valid.
125
Bytes
0-15
16-31
Table 48
Value
UID
UID complement (bit-wise)
UID and Complement
To accommodate robust retrieval of the UID in the case of bit errors, sixteen copies of the UID
and the corresponding complement shall be stored by the target. For example, reading bytes 3263 returns to the host another copy of the UID and its complement.
Read Status Enhanced shall not be used during execution of the Read Unique ID command.
Figure 58 defines the Read Unique ID behavior. The host may use any timing mode supported
by the target in order to retrieve the UID data.
126
Cycle Type
CMD
ADDR
DOUT
DOUT
DOUT
DOUT
DOUT
DOUT
DQ[7:0]
EDh
00h
U00
U10
...
U01
U11
...
tRR
tWB
tR
R/B_n
Figure 58
U0k-Unk
Read Unique ID command timing
The kth copy of the UID and its complement. Sixteen copies are stored.
Reading beyond 512 bytes returns indeterminate values.
5.9. Block Erase Definition
The Block Erase function erases the block of data identified by the block address parameter on
the LUN specified. After a successful Block Erase, all bits shall be set to one in the block. SR[0]
is valid for this command after SR[6] transitions from zero to one until the next transition of SR[6]
to zero. Figure 59 defines the Block Erase behavior and timings.
Cycle Type
DQ[7:0]
CMD
ADDR
ADDR
ADDR
CMD
60h
R1
R2
R3
D0h
tWB
tBERS
SR[6]
Figure 59
Block Erase timing
R1-R3 The row address of the block to be erased. R1 is the least significant byte in the
row address.
5.10.
Read Status Definition
In the case of non-multi-plane operations, the Read Status function retrieves a status value for
the last operation issued. If multiple multi-plane operations are in progress on a single LUN, then
Read Status returns the composite status value for status register bits that are independent per
plane address. Specifically, Read Status shall return the combined status value of the
independent status register bits according to Table 49. See section 5.13 for status register bit
definitions.
127
Status Register bit
Bit 0, FAIL
Bit 1, FAILC
Bit 3, CSP
Table 49
Composite status value
OR
OR
OR
Composite Status Value
When issuing Read Status in the source synchronous data interface, each data byte is received
twice. The host shall only latch one copy of each data byte. See section 4.3.2.5.
Figure 60 defines the Read Status behavior and timings.
Cycle Type
DOUT
CMD
tWHR
DQ[7:0]
Figure 60
SR
70h
SR
Read Status timing
Status value as defined in section 5.13.
128
The Read Status command may be issued using either the asynchronous or source synchronous
data interfaces. The timing parameters for each data interface are shown in Figure 61 and Figure
62.
Figure 61
Read Status command using asynchronous data interface
129
Figure 62
Read Status command using source synchronous data interface
Note (1): For the asynchronous data interface, status may be continually read by pulsing RE_n or
leaving RE_n low. For the source synchronous interface, status may continually be read by
leaving ALE/CLE at a value of 11b.
5.11.
Read Status Enhanced Definition
The Read Status Enhanced function retrieves the status value for a previous operation on the
particular LUN and plane address specified. Figure 63 defines the Read Status Enhanced
behavior and timings. If the row address entered is invalid, the Status value returned has an
indeterminate value. The host uses Read Status Enhanced for LUN selection (refer to section
3.1.2). Note that Read Status Enhanced has no effect on which page register is selected for data
output within the LUN.
When issuing Read Status Enhanced in the source synchronous data interface, each data byte is
received twice. The host shall only latch one copy of each data byte. See section 4.3.2.5.
130
Cycle Type
CMD
ADDR
ADDR
DOUT
ADDR
tWHR
DQ[7:0]
78h
Figure 63
R1
R2
SR
R3
Read Status Enhanced timing
R1-R3 Row address of the previous operation to retrieve status for. R1 is the least
significant byte. The row address contains both the LUN and plane address to
retrieve status for.
SR
5.12.
Status value as defined in section 5.13.
Read Status and Read Status Enhanced required usage
In certain sequences only one status command shall be used by the host. This section outlines
situations in which a particular status command is required to be used.
If a command is issued to a LUN while R/B_n is cleared to zero, then the next status command
shall be Read Status Enhanced. Read Status Enhanced causes LUNs that are not selected to
turn off their output buffers. This ensures that only the LUN selected by the Read Status
Enhanced commands responds to a subsequent toggle of the RE_n input signal.
When the host has issued Read Page commands to multiple LUNs at the same time, the host
shall issue Read Status Enhanced before reading data from either LUN. Read Status Enhanced
causes LUNs that are not selected to turn off their output buffers. This ensures that only the LUN
selected by the Read Status Enhanced commands responds to a subsequent toggle of the RE_n
input signal after data output is selected with the 00h command. Refer to section 3.1.3 for
additional requirements if a Change Read Column (Enhanced) command is used as part of a
multiple LUN read sequence.
During and after Target level commands, the host shall not issue the Read Status Enhanced
command. In these sequences, the host uses Read Status to check for the status value. The
only exception to this requirement is if commands were outstanding to multiple LUNs when a
Reset was issued. In this case, the Read Status Enhanced command shall be used to determine
when each active LUN has completed Reset.
5.13.
Status Field Definition
The returned status register byte value (SR) for Read Status and Read Status Enhanced has the
format described below. If the RDY bit is cleared to zero, all other bits in the status byte (except
WP_n) are invalid and shall be ignored by the host.
Value
Status Register
FAIL
7
6
5
4
3
2
1
0
WP_n
RDY
ARDY
VSP
CSP
R
FAILC
FAIL
If set to one, then the last command failed. If cleared to zero, then the last
command was successful. For raw NAND operation, this bit is only valid for
131
program and erase operations. For EZ NAND operation, this bit is valid for read,
program, and erase operations. During program cache operations, this bit is only
valid when ARDY is set to one.
FAILC If set to one, then the command issued prior to the last command failed. If
cleared to zero, then the command issued prior to the last command was
successful. This bit is only valid for program cache operations. This bit is not
valid until after the second 15h command or the 10h command has been
transferred in a Page Cache Program sequence. When program cache is not
supported, this bit is not used and shall be cleared to zero. For EZ NAND
operation, this bit is not used (cache commands are not supported with EZ
NAND).
CSP
Command Specific: This bit has command specific meaning.
For EZ NAND read operations, if CSP (Threshold) bit is set to one then the last
read operation exceeded the ECC threshold and the host should take
appropriate action (e.g. rewrite the data to a new location). When FAIL is set to
one, the CSP (Threshold) bit is don’t care.
For all other operations, this bit is reserved.
ARDY If set to one, then there is no array operation in progress. If cleared to zero, then
there is a command being processed (RDY is cleared to zero) or an array
operation in progress. When overlapped multi-plane operations or cache
commands are not supported, this bit is not used.
RDY
If set to one, then the LUN or plane address is ready for another command and
all other bits in the status value are valid. If cleared to zero, then the last
command issued is not yet complete and SR bits 5:0 are invalid and shall be
ignored by the host. This bit impacts the value of R/B_n, refer to section 2.16.2.
When caching operations are in use, then this bit indicates whether another
command can be accepted, and ARDY indicates whether the last operation is
complete.
WP_n If set to one, then the device is not write protected. If cleared to zero, then the
device is write protected. This bit shall always be valid regardless of the state of
the RDY bit.
R
Reserved (0)
VSP
Vendor Specific
5.14.
Read Definition
The Read function reads a page of data identified by a row address for the LUN specified. The
page of data is made available to be read from the page register starting at the column address
specified. Figure 64 defines the Read behavior and timings. Reading beyond the end of a page
results in indeterminate values being returned to the host.
While monitoring the read status to determine when the tR (transfer from Flash array to page
register) is complete, the host shall re-issue a command value of 00h to start reading data.
132
Issuing a command value of 00h will cause data to be returned starting at the selected column
address.
133
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
DOUT
DOUT
DOUT
00h
C1
C2
R1
R2
R3
30h
D0
D1
D2
tRR
tWB
tR
SR[6]
Figure 64
Read timing
C1-C2 Column address of the page to retrieve. C1 is the least significant byte.
R1-R3 Row address of the page to retrieve. R1 is the least significant byte.
Dn
Data bytes read from the addressed page.
134
5.15.
Read Cache Definition
This command is not supported for EZ NAND.
The Read Cache Sequential and Read Cache Random functions permit a page to be read from
the page register while another page is simultaneously read from the Flash array for the selected
LUN. A Read Page command, as defined in section 5.14, shall be issued prior to the initial Read
Cache Sequential or Read Cache Random command in a read cache sequence. A Read Cache
Sequential or Read Cache Random command shall be issued prior to a Read Cache End (3Fh)
command being issued.
The Read Cache (Sequential or Random) function may be issued after the Read function is
complete (SR[6] is set to one). The host may enter the address of the next page to be read from
the Flash array. Data output always begins at column address 00h. If the host does not enter an
address to retrieve, the next sequential page is read. When the Read Cache (Sequential or
Random) function is issued, SR[6] is cleared to zero (busy). After the operation is begun SR[6] is
set to one (ready) and the host may begin to read the data from the previous Read or Read
Cache (Sequential or Random) function. Issuing an additional Read Cache (Sequential or
Random) function copies the data most recently read from the array into the page register. When
no more pages are to be read, the final page is copied into the page register by issuing the 3Fh
command. The host may begin to read data from the page register when SR[6]is set to one
(ready). When the 31h and 3Fh commands are issued, SR[6] shall be cleared to zero (busy) until
the page has finished being copied from the Flash array.
The host shall not issue a Read Cache Sequential (31h) command after the last page of a block
is read. If commands are issued to multiple LUNs at the same time, the host shall execute a
Read Status Enhanced (78h) command to select the LUN prior to issuing a Read Cache
Sequential (31h) or Read Cache End (3Fh) command for that LUN.
Figure 65 defines the Read Cache Sequential behavior and timings for the beginning of the cache
operations subsequent to a Read command being issued to the target. Figure 66 defines the
Read Cache Random behavior and timings for the beginning of the cache operations subsequent
to a Read command being issued to the target. In each case, SR[6] conveys whether the next
selected page can be read from the page register.
135
As defined for
Read
Cycle Type
DQx
CMD
CMD
DOUT
DOUT
DOUT
CMD
DOUT
30h
31h
D0
...
Dn
31h
D0
tWB
tRR
tWB
tR
tRCBSY
tRR
tWB
tRCBSY
SR[6]
Figure 65
Read Cache Sequential timing, start of cache operations
D0-Dn Data bytes/words read from page requested by the original Read or the previous cache operation.
136
As defined for
Read
Cycle Type
DQx
A
CMD
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
DOUT
DOUT
DOUT
30h
00h
C1
C2
R1
R2
R3
31h
D0
...
Dn
tRR
tWB
tRR
tWB
tR
tRCBSY
SR[6]
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
DOUT
00h
C1
C2
R1
R2
R3
31h
D0
tRR
tWB
tRCBSY
SR[6]
Figure 66
Read Cache Random timing, start of cache operations
C1-C2 Column address of the page to retrieve. C1 is the least significant byte. The column address is ignored.
R1-R3 Row address of the page to retrieve. R1 is the least significant byte.
D0-Dn Data bytes/words read from page requested by the original Read or the previous cache operation
137
Figure 67 defines the Read Cache (Sequential or Random) behavior and timings for the end of cache operations. This applies for both Read
Cache Sequential and Read Cache Random. A command code of 3Fh indicates to the target to transfer the final selected page into the page
register, without beginning another background read operation.
As defined for
Read Cache
(Sequential or Random)
Cycle Type
DQx
CMD
DOUT
DOUT
DOUT
CMD
DOUT
DOUT
DOUT
31h
D0
...
Dn
3Fh
D0
...
Dn
tRR
tWB
tRR
tWB
tRCBSY
tRCBSY
SR[6]
Figure 67
Read Cache timing, end of cache operations
D0-Dn Data bytes/words read from page requested by the previous cache operation.
138
5.16.
Page Program Definition
The Page Program command transfers a page or portion of a page of data identified by a column
address to the page register. The contents of the page register are then programmed into the
Flash array at the row address indicated. SR[0] is valid for this command after SR[6] transitions
from zero to one until the next transition of SR[6] to zero. Figure 68 defines the Page Program
behavior and timings. Writing beyond the end of the page register is undefined.
139
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0
D1
...
Dn
10h
tADL
DQx
80h
C1
C2
R1
R2
R3
tWB
tPROG
SR[6]
Figure 68
Page Program timing
C1-C2 Column address of the starting buffer location to write data to. C1 is the least significant byte.
R1-R3 Row address of the page being programmed. R1 is the least significant byte.
D0-Dn Data bytes/words to be written to the addressed page.
140
5.17.
Page Cache Program Definition
This command is not supported for EZ NAND.
The Page Cache Program function permits a page or portion of a page of data to be written to the
Flash array for the specified LUN in the background while the next page to program is transferred
by the host to the page register. After the 10h command is issued, all data is written to the Flash
array prior to SR[6] being set to one (ready). SR[0] is valid for this command after SR[5]
transitions from zero to one until the next transition. SR[1] is valid for this command after SR[6]
transitions from zero to one, and this is not the first operation.
Figure 69 and Figure 70 define the Page Cache Program behavior and timings. Note that tPROG
at the end of the caching operation may be longer than typical as this time also accounts for
completing the programming operation for the previous page. Writing beyond the end of the page
register is undefined.
If the program page register clear enhancement is supported, then the host may choose to only
clear the page register for the selected LUN and plane address when a Program (80h) command
is received. In this case, the tADL time may be longer than defined for the selected timing mode,
refer to section 5.7.1.17. Refer to section 5.26.1 for details on how to enable this feature.
141
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0
D1
...
Dn
15h
tADL
DQx
80h
C1
C2
R1
R2
R3
tWB
tPCBSY
SR[6]
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0
D1
...
Dn
15h
tADL
DQx
80h
C1
C2
R1
R2
R3
tWB
tPCBSY
SR[6]
Figure 69
142
Page Cache Program timing, start of operations
As defined for Page
Cache Program
Cycle Type
CMD
A
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0
D1
...
Dn
15h
tADL
DQx
80h
C1
C2
R1
R2
R3
tWB
tPCBSY
SR[6]
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0
D1
...
Dn
10h
tADL
DQx
80h
C1
C2
R1
R2
R3
tWB
tPROG
SR[6]
Figure 70
143
Page Cache Program timing, end of operations
C1-C2 Column address of the starting buffer location to write data to. C1 is the least significant byte.
R1-R3 Row address of the page being programmed. R1 is the least significant byte.
D0-Dn Data bytes/words to be written to the addressed page.
5.18.
Copyback Definition
The Copyback function reads a page of data from one location and then moves that data to a second location on the same LUN. If the target
supports EZ NAND then the data may be moved to a second location on a different plane or LUN. Refer to the parameter page to determine the
destinations support for Copyback. The data read from the first location may be read by the host, including use of Change Read Column. After
completing any data read out and issuing Copyback Program, the host may perform data modification using Change Write Column as needed.
Figure 71 defines the Copyback behavior and timings.
Copyback uses a single page register for the read and program operation. If the target supports EZ NAND, the buffer in the EZ NAND controller is
used for read and program operations and the page register within each LUN is not explicitly accessible.
When multi-plane addressing is supported, the multi-plane address for Copyback Read and Copyback Program for a non-multi-plane Copyback
operation shall be the same. If EZ NAND is supported, this restriction may not apply; refer to the parameter page.
Copyback may also have odd/even page restrictions. Specifically, when reading from an odd page, the contents may need to be written to an odd
page. Alternatively, when reading from an even page, the contents may need to be written to an even page. Refer to section 5.7.1.3.
This revision of the ONFI specification requires Copyback Adjacency for all implementations. In future ONFI revisions, for EZ NAND
implementations it is possible that Copyback Read and Copyback Program are not required to be adjacent. I.e., there may be multiple Copyback
Read commands prior to a Copyback Program being issued. This requirement also applies to multi-plane Copyback operations.
144
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1A
C2A
R1A
R2A
R3A
35h
tWB
tR
SR[6]
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
85h
C1B
C2B
R1B
R2B
R3B
10h
tWB
tPROG
SR[6]
Figure 71
Copyback timing
C1-C2A
Column address of the page to retrieve. C1A is the least significant byte.
R1-R3 A
Row address of the page to retrieve. R1A is the least significant byte.
C1-C2B
Column address of the page to program. C1B is the least significant byte.
145
R1-R3 B
Row address of the page to program. R1B is the least significant byte.
Figure 72 and Figure 73 define Copyback support for data output and data modification.
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
DOUT
DOUT
DOUT
00h
C1A
C2A
R1A
R2A
R3A
35h
Dj
...
Dj+n
tRR
tWB
tR
SR[6]
A
Cycle Type
CMD
ADDR
ADDR
CMD
DOUT
DOUT
DOUT
Dk
Dk+1
Dk+2
tCCS
DQx
05h
C1C
C2C
E0h
SR[6]
Figure 72
C1-C2A
Copyback Read with data output
Column address of the page to retrieve. C1A is the least significant byte.
146
R1-R3 A
Row address of the page to retrieve. R1A is the least significant byte.
Dj-(Dj+n)
Data bytes read starting at column address specified in C1-C2A.
C1-C2C
Column address of new location (k) to read out from the page register. C1C is the least significant byte.
Dk-Dk+n
Data bytes read starting at column address specified in C1-C2C.
147
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
Di
Di+1
tADL
DQx
85h
C1B
C2B
CMD
ADDR
ADDR
R1B
R2B
R3B
SR[6]
A
Cycle Type
DIN
DIN
DIN
CMD
Dj
Dj+1
Dj+2
10h
tCCS
DQx
85h
C1D
C2D
tWB
tPROG
SR[6]
Figure 73
Copyback Program with data modification
C1-C2B
Column address of the page to program. C1B is the least significant byte.
R1-R3 B
Row address of the page to program. R1B is the least significant byte.
Di-Di+n
Data bytes overwritten in page register starting at column address specified in C1-C2B.
C1-C2D
Column address of new location (j) to overwrite data at in the page register. C1D is the least significant byte.
148
Dj-Dj+n
5.19.
Data bytes overwritten starting at column address specified in C1-C2D.
Small Data Move
If the Small Data Move command is supported, as indicated in the parameter page, then the host may transfer data to the page register in
increments that are less than the page size of the device for both Program and Copyback operations (including multi-plane Program and
Copyback operations). The host may also read data out as part of the operation. If the Small Data Move is a program operation with no data
output, then the 80h opcode may be used for the first cycles. For Copyback and program operations that include data output, the 85h opcode
shall be used for the first cycles.
Figure 74 defines the data modification portion of a Program or Copyback Program with small data moves; this sequence may be repeated as
necessary to complete the data transfer. Figure 75 defines the final program operation that is used to complete the Program or Copyback
Program with small data move operation. The row address (R1B – R3B) shall be the same for all program portions of the sequence destined for
the same plane address. The function of the 11h command in a small data move operation is to flush any internal data pipeline in the device prior
to resuming data output.
149
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
CMD
Dj
...
Dj+n
11h
tCCS
DQx
PRG
C1B
C2B
R1B
R2B
R3B
tWB
tIPBSY
SR[6]
A
Cycle Type
B
CMD
ADDR
ADDR
CMD
DOUT
DOUT
DOUT
Dk
...
Dk+p
tCCS
DQx
05h
C1C
C2C
E0h
SR[6]
Figure 74
Small data moves, data modification
PRG
Program command, either 80h or 85h. Following any data output, the command shall be 85h.
C1-C2B
Column address to write to in the page register. C1B is the least significant byte.
R1-R3B
Row address of the page to program. R1B is the least significant byte.
150
Dj-(Dj+n)
Data bytes to update in the page register starting at column address specified in C1-C2B.
C1-C2C
Column address of the byte/word in the page register to retrieve. C1C is the least significant byte.
Dk-(Dk+p)
Data bytes read starting at column address specified in C1-C2C.
NOTE: If Change Read Column Enhanced is supported, this command may be substituted for Change Read Column in Figure 74. Use of the
Change Read Column (Enhanced) command and data output in this flow is optional; this flow may be used to incrementally transfer data for a
Program or Copyback Program.
B
Cycle Type
DOUT
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
tRHW
DQx
Dk+p
DIN
DIN
DIN
CMD
Dl
...
Dl+q
10h
tCCS
PRG
C1D
C2D
R1B
R2B
R3B
tWB
tPROG
SR[6]
Figure 75
Small data moves, end
PRG
Program command, either 80h or 85h. 85h shall be used if there is any data output as part of the command.
C1-C2D
Column address to write to in the page register. C1D is the least significant byte.
R1-R3B
Row address of the page to program. R1B is the least significant byte.
Dl-(Dl+q)
Data bytes to update in the page register starting at column address specified in C1-C2D.
151
5.20.
Change Read Column Definition
The Change Read Column function changes the column address from which data is being read in
the page register for the selected LUN. Change Read Column shall only be issued when the
LUN is in a read idle condition. Figure 76 defines the Change Read Column behavior and timings.
The host shall not read data from the LUN until tCCS ns after the E0h command is written to the
LUN. Refer to Figure 76.
Cycle Type
DOUT
DOUT
CMD
ADDR
ADDR
CMD
tRHW
DQx
Dn
Dn+1
DOUT
DOUT
DOUT
Dk
Dk+1
Dk+2
tCCS
05h
C1
C2
E0h
SR[6]
Figure 76
Dn
Change Read Column timing
Data bytes read prior to the column address change.
C1-C2 New column address to be set for subsequent data transfers. C1 is the least
significant byte.
Dk
5.21.
Data bytes being read starting with the new addressed column.
Change Read Column Enhanced Definition
The Change Read Column Enhanced function changes the LUN address, plane address and
column address from which data is being read in a page previously retrieved with the Read
command. This command is used when independent LUN operations or multi-plane operations
are being performed such that the entire address for the new column needs to be given. Figure
77 defines the Change Read Column Enhanced behavior and timings.
The Change Read Column Enhanced command shall not be issued by the host unless it is
supported as indicated in the parameter page. Change Read Column Enhanced shall not be
issued while Target level data output commands (Read ID, Read Parameter Page, Read Unique
ID, Get Features) are executing or immediately following Target level commands.
Change Read Column Enhanced causes idle LUNs (SR[6] is one) that are not selected to turn off
their output buffers. This ensures that only the LUN selected by the Change Read Column
Enhanced command responds to subsequent data output. If unselected LUNs are active (SR[6]
is zero) when Change Read Column Enhanced is issued, then the host shall issue a Read Status
Enhanced (78h) command prior to subsequent data output to ensure all LUNs that are not
selected turn off their output buffers.
152
Cycle Type
DOUT
DOUT
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
tRHW
DQx
Dn
Dn+1
06h
C1
Figure 77
C2
R1
R2
R3
E0h
Change Read Column Enhanced timing
Data bytes read prior to the row and column address change.
C1-C2 New column address to be set for subsequent data transfers. C1 is the least significant byte.
R1-R3 New row address to be set for subsequent data transfers. R1 is the least significant byte.
Dk
DOUT
DOUT
Dk
Dk+1
Dk+2
tCCS
SR[6]
Dn
DOUT
Data bytes being read starting with the new addressed row and column.
153
5.22.
Change Write Column Definition
The Change Write Column function changes the column address being written to in the page
register for the selected LUN. Figure 78 defines the Change Write Column behavior and timings.
The host shall not write data to the LUN until tCCS ns after the last column address is written to
the LUN. Refer to Figure 76.
As defined for Page
(Cache) Program
Cycle Type
DIN
As defined for Page
(Cache) Program
DIN
CMD
ADDR
ADDR
DIN
DIN
DIN
Dk
Dk+1
Dk+2
tCCS
DQx
Dn
Dn+1
85h
C1
C2
SR[6]
Figure 78
Change Write Column timing
C1-C2 New column address to be set for subsequent data transfers. C1 is the least
significant byte.
Dn
Data bytes being written to previous addressed column
Dk
Data bytes being written starting with the new addressed column
5.23.
Change Row Address Definition
The Change Row Address function changes the row and column address being written to for the
selected LUN. This mechanism may be used to adjust the block address, page address, and
column address for a Program that is in execution. The LUN and plane address shall be the
same as the Program that is in execution. Figure 79 defines the Change Row Address behavior
and timings.
The host shall not write data to the LUN until tCCS ns after the last row address is written to the
LUN. Refer to Figure 79.
154
As defined for Page
(Cache) Program
Cycle Type
DIN
As defined for Page
(Cache) Program
DIN
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
Dk
Dk+1
Dk+2
tCCS
DQx
Dn
Dn+1
85h
C1
C2
R1
R2
R3
SR[6]
Figure 79
Change Row Address timing
C1-C2 New column address to be set for subsequent data transfers. C1 is the least significant byte.
R1-R3 Row address of the page being programmed. The LUN address and plane address shall be the same as the Program in
execution. R1 is the least significant byte.
Dn
Data bytes being written prior to row address change; will be written to new row address
Dk
Data bytes being written to the new block and page, starting with the newly addressed column
155
5.24.
Set Features Definition
The Set Features function modifies the settings of a particular feature. For example, this function
can be used to enable a feature that is disabled at power-on. Parameters are always transferred
on the lower 8-bits of the data bus. Figure 80 defines the Set Features behavior and timings.
When issuing Set Features in the source synchronous data interface, each data byte is
transmitted twice. The device shall only latch one copy of each data byte. See section 4.3.2.3.
Set Features is used to change the timing mode and data interface type. When changing the
timing mode, the device is busy for tITC, not tFEAT. During the tITC time the host shall not poll
for status.
156
Cycle Type
CMD
ADDR
DIN
DIN
DIN
DIN
P1
P2
P3
P4
tADL
DQx
EFh
FA
tWB
tFEAT
R/B_n
Figure 80
Set Features timing
* NOTE: Busy time is tITC when setting the timing mode.
FA
Feature address identifying feature to modify settings for.
P1-P4 Parameters identifying new settings for the feature specified.
P1
P2
P3
P4
Sub feature parameter 1
Sub feature parameter 2
Sub feature parameter 3
Sub feature parameter 4
Refer to section 5.26 for the definition of features and sub feature parameters.
157
5.25.
Get Features Definition
The Get Features function is the mechanism the host uses to determine the current settings for a
particular feature. This function shall return the current settings for the feature (including
modifications that may have been previously made with the Set Features function). Parameters
are always transferred on the lower 8-bits of the data bus. After reading the first byte of data, the
host shall complete reading all desired data before issuing another command (including Read
Status or Read Status Enhanced). Figure 81 defines the Get Features behavior and timings.
When issuing Get Features in the source synchronous data interface, each data byte is received
twice. The host shall only latch one copy of each data byte. See section 4.3.2.5.
If Read Status is used to monitor when the tFEAT time is complete, the host shall issue a
command value of 00h to begin transfer of the feature data starting with parameter P1.
Cycle Type
CMD
ADDR
DOUT
DOUT
DOUT
DOUT
DQx
EEh
FA
P1
P2
P3
P4
tRR
tWB
tFEAT
R/B_n
Figure 81
FA
Get Features timing
Feature address identifying feature to return parameters for.
P1-P4 Current settings/parameters for the feature identified by argument P1
P1
P2
P3
P4
Sub feature parameter 1 setting
Sub feature parameter 2 setting
Sub feature parameter 3 setting
Sub feature parameter 4 setting
Refer to section 5.26 for the definition of features and sub feature parameters.
5.26.
Feature Parameter Definitions
If the Set Features and Get Features commands are not supported by the Target, then no feature
parameters are supported. Additionally, the Target only supports feature parameters defined in
ONFI specification revisions that the Target complies with.
Feature settings are volatile across power cycles. For each feature setting, whether the value
across resets is retained is explicitly stated.
Feature Address
00h
01h
02h-0Fh
Description
Reserved
Timing Mode
Reserved
158
10h
11h-1Fh
20h-4Fh
50h
51h-5Fh
60h-7Fh
80h-FFh
5.26.1.
I/O Drive Strength
Reserved for programmable I/O settings
Reserved
EZ NAND control
Reserved
Reserved for Block Abstracted NAND
Vendor specific
Timing Mode
This setting shall be supported if the Target complies with ONFI specification revision 1.0.
The Data Interface setting is not retained across Reset (FFh); after a Reset (FFh) the Data
Interface shall be asynchronous. All other settings for the timing mode are retained across Reset
(FFh) and Synchronous Reset (FCh) commands. If the Reset (FFh) command is issued when
the Data Interface is configured as source synchronous, then the host shall use the asynchronous
data interface with Timing Mode 0 until a new data interface and/or timing mode is selected with
Set Features. Hosts shall only set a timing mode that is explicitly shown as supported in the Read
Parameter Page.
The results of the host using Set Features to transition from the source synchronous data
interface to the asynchronous data interface is indeterminate. To transition to the asynchronous
data interface, the host should use the Reset (FFh) command.
Sub Feature
Parameter
7
6
P1
R
PC
P2
P3
P4
5
4
3
2
1
0
Data
Timing Mode Number
Interface
Reserved (0)
Reserved (0)
Reserved (0)
Timing Mode Number
Set to the numerical value of the maximum timing mode in use
by the host. Default power-on value is 0h.
Data Interface
00b = asynchronous (default power-on value)
01b = source synchronous
10-11b = Reserved
PC
The Program Clear bit controls the program page register clear
enhancement which defines the behavior of clearing the page
register when a Program (80h) command is received. If cleared
to zero, then the page register(s) for each LUN that is part of the
target is cleared when the Program (80h) command is received.
If set to one, then only the page register for the LUN and
interleave address selected with the Program (80h) command is
cleared and the tADL time for Program commands is as reported
in the parameter page.
Reserved / R
Reserved values shall be cleared to zero by the host. Targets
shall not be sensitive to the value of reserved fields.
159
5.26.2.
I/O Drive Strength
This setting shall be supported if the Target supports the source synchronous data interface. The
I/O drive strength setting shall be retained across Reset (FFh) and Synchronous Reset (FCh)
commands. The power-on default drive strength value is the Nominal (10b) setting.
Sub Feature
Parameter
P1
P2
P3
P4
7
6
5
4
3
2
Reserved (0)
Reserved (0)
Reserved (0)
Reserved (0)
1
0
Drive Strength
Drive strength
00b = 18 Ohm
01b = 25 Ohm
10b = 35 Ohm (power-on default)
11b = 50 Ohm
Reserved
Reserved values shall be cleared to zero by the host. Targets
shall not be sensitive to the value of reserved fields.
5.26.3.
EZ NAND control
This setting shall be supported if the device supports EZ NAND. This feature is used to control
settings for the EZ NAND device.
Sub Feature
Parameter
P1
P2
P3
P4
7
6
5
4
3
Reserved (0)
Reserved (0)
Reserved (0)
Reserved (0)
2
1
0
RD
Retry Disable (RD)
If set to one, then the EZ NAND device shall not automatically
perform retries. If cleared to zero, then the EZ NAND device
may automatically perform retries during error conditions at its
discretion. If automatic retries are disabled, the device may
exceed the UBER specified. Automatic retries shall only be
disabled if the device supports this capability as indicated in the
parameter page. If an EZ NAND controller executes an
automatic retry, the typical page read time (tR) may be
exceeded.
Reserved / R
Reserved values shall be cleared to zero by the host. Targets
shall not be sensitive to the value of reserved fields.
160
6. Multi-plane Operations
A LUN may support multi-plane read, program and erase operations. Multi-plane operations are
when multiple commands of the same type are issued to different blocks on the same LUN.
Refer to section 5.7.1.28 for addressing restrictions with multi-plane operations. There are two
methods for multi-plane operations: concurrent and overlapped.
When performing multi-plane operations, the operations/functions shall be the same type. The
functions that may be used in multi-plane operations are:
 Page Program
 Copyback Read and Program
 Block Erase
 Read
6.1. Requirements
When supported, the plane address comprises the lowest order bits of the block address as
shown in Figure 19. The LUN and page addresses are required to be the same. The block
address (other than the plane address bits) may be required to be the same, refer to section
5.7.1.28.
For copyback program operations, the restrictions are the same as for a multi-planeprogram
operation. However, copyback reads shall be previously issued to the same plane addresses as
those in the multi-plane copyback program operations. The reads for copyback may be issued
non-multi-plane or multi-plane. If the reads are non-multi-plane then the reads may have different
page addresses. If the reads are multi-plane then the reads shall have the same page
addresses.
Multi-plane operations enable operations of the same type to be issued to other blocks on the
same LUN. There are two methods for multi-plane operations: concurrent and overlapped. The
concurrent multi-plane operation waits until all command, address, and data are entered for all
plane addresses before accessing the Flash array. The overlapped multi-plane operation begins
its operation immediately after the command, address and data are entered and performs it in the
background while the next multi-plane command, address, and data are entered.
The plane address component of each address shall be distinct. A single multi-plane (cached)
program operation is shown in Figure 82. Between “Multi-plane Op 1” and “Multi-plane Op n”, all
plane addresses shall be different from each other. After the 10h or 15h (cached) command
cycle is issued, previously issued plane addresses can be used in future multi-plane operations.
80h
11h
<ADDR1>
CMD
CMD
80h
11h
<ADDR2>
CMD
CMD
Multi-plane Op 1
Multi-plane Op 2
Figure 82
…
10h
80h
<ADDRn > or 15h
CMD
CMD
Multi-plane Op n
Multi-plane Program (Cache)
For multi-plane erase operations, the plane address component of each address shall be distinct.
A single multi-plane erase operation is shown in Figure 83. Between “Multi-plane Op 1” and
“Multi-plane Op n”, all plane addresses shall be different from each other. After the D0h
command cycle is issued, previously issued plane addresses can be used in future multi-plane
operations.
161
60h
D1h
<ADDR1>
CMD
CMD
60h
D1h
<ADDR2>
CMD
CMD
Multi-plane Op 1
Multi-plane Op 2
Figure 83
…
60h
D0h
<ADDRn >
CMD
CMD
Multi-plane Op n
Multi-plane Erase
The plane address component of each address shall be distinct. A single multi-plane read
(cache) operation is shown in Figure 84. Between “Multi-plane Op 1” and “Multi-plane Op n”, all
plane addresses shall be different from each other. After the 30h or 31h (cached) command
cycle is issued, previously issued plane addresses can be used in future multi-plane operations.
00h
32h
<ADDR1>
CMD
CMD
00h
32h
<ADDR2>
CMD
CMD
Multi-plane Op 1
Multi-plane Op 2
Figure 84
30h
00h
<ADDRn > or 31h
CMD
CMD
…
Multi-plane Op n
Multi-plane Read (Cache)
6.2. Status Register Behavior
Some status register bits are independent per plane address. Other status register bits are
shared across the entire LUN. This section defines when status register bits are independent per
plane address. This is the same for concurrent and overlapped operations.
For multi-plane program and erase operations, the FAIL/FAILC bits are independent per plane
address. Table 50 lists whether a bit is independent per plane address or shared across the
entire LUN for multi-plane operations.
Value
Status Register
Independent
7
WP_n
N
Table 50
6
RDY
N
5
ARDY
N
4
VSP
N
3
R
N
2
R
N
1
FAILC
Y
0
FAIL
Y
Independent Status Register bits
6.3. Multi-plane Page Program
The Page Program command transfers a page or portion of a page of data identified by a column
address to the page register. The contents of the page register are then programmed into the
Flash array at the row address indicated. With a multi-plane operation, multiple programs can be
issued back to back to the LUN, with a shorter busy time between issuance of the next program
operation. Figure 85 defines the behavior and timings for two multi-plane page program
commands.
Cache operations may be used when doing multi-plane page program operations, as shown, if
supported by the target as indicated in the parameter page. Refer to section 5.7.1.27.
162
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0A
D1A
...
DnA
11h
tADL
DQx
80h
C1A
C2A
R1A
R2A
R3A
tWB
tPLPBSY
SR[6]
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
DIN
DIN
DIN
DIN
CMD
D0B
D1B
...
DnB
10h
tADL
DQx
80h
C1B
C2B
R1B
R2B
R3B
tWB
tPROG
SR[6]
Figure 85
Multi-plane Page Program timing
C1A-C2A
Column address for page A. C1A is the least significant byte.
R1A-R3A
Row address for page A. R1A is the least significant byte.
D0A-DnA
Data to program for page A.
C1B-C2B
Column address for page B. C1B is the least significant byte.
163
R1B-R3B
Row address for page B. R1B is the least significant byte.
D0B-DnB
Data to program for page B.
The row addresses for page A and B shall differ in the plane address bits.
Finishing a multi-plane program with a command cycle of 15h rather than 10h indicates that this is a cache operation. The host shall only issue a
command cycle of 15h to complete an multi-plane program operation if program cache is supported with multi-plane program operations, as
described in section 5.7.1.27.
164
6.4. Multi-plane Copyback Read and Program
The Copyback function reads a page of data from one location and then moves that data to a
second location. With a multi-plane operation, the Copyback Program function can be issued
back to back to the target, with a shorter busy time between issuance of the next Copyback
Program. Figure 86, Figure 87, and Figure 88 define the behavior and timings for two Copyback
Program operations. The reads for the Copyback Program may or may not be multi-plane.
Figure 86 defines the non-multi-plane read sequence and Figure 87 defines the multi-plane read
sequence.
The plane addresses used for the Copyback Read operations (regardless of multi-plane) shall be
the same as the plane addresses used in the subsequent multi-plane Copyback Program
operations. If EZ NAND is supported, this restricton may not apply; refer to the parameter page.
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1A
C2A
R1A
R2A
R3A
35h
DQx
tWB
tR
SR[6]
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1B
C2B
R1B
R2B
R3B
35h
tWB
tR
SR[6]
Figure 86
Non-multi-plane Copyback Read timing for multi-plane Copyback Program
C1A-C2A
Column address for source page A. C1A is the least significant byte.
R1A-R3A
Row address for source page A. R1A is the least significant byte.
C1B-C2B
Column address for source page B. C1B is the least significant byte.
R1B-R3B
Row address for source page B. R1B is the least significant byte.
165
The row addresses for all source pages shall differ in their plane address bits.
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1A
C2A
R1A
R2A
R3A
32h
DQx
tWB
tPLRBSY
SR[6]
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1B
C2B
R1B
R2B
R3B
35h
tWB
tR
SR[6]
Figure 87
Multi-plane Copyback Read timing for Multi-plane Copyback Program
C1A-C2A
Column address for source page A. C1A is the least significant byte.
R1A-R3A
Row address for source page A. R1A is the least significant byte.
C1B-C2B
Column address for source page B. C1B is the least significant byte.
R1B-R3B
Row address for source page B. R1B is the least significant byte.
The row addresses for all source pages shall differ in their plane address bits. The source page
addresses shall be the same for multi-plane reads.
166
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
85h
C1C
C2C
R1C
R2C
R3C
11h
DQx
tWB
tPLPBSY
SR[6]
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
85h
C1D
C2D
R1D
R2D
R3D
10h
tWB
tPROG
SR[6]
Figure 88
Multi-plane Copyback Program
C1C-C2C
Column address for destination page C. C1C is the least significant byte.
R1C-R3C
Row address for destination page C. R1C is the least significant byte.
C1D-C2D
Column address for destination page D. C1D is the least significant byte.
R1D-R3D
Row address for destination page D. R1D is the least significant byte.
The row addresses for all destination pages shall differ in their plane address bits. The page
address for all destination addresses for multi-plane copyback operations shall be identical.
167
6.5. Multi-plane Block Erase
Figure 89 defines the behavior and timings for a multi-plane block erase operation. Only two operations are shown, however additional erase
operations may be issued with a 60h/D1h sequence prior to the final 60h/D0h sequence depending on how many multi-plane operations the LUN
supports.
Cycle Type
DQ[7:0]
CMD
ADDR
ADDR
ADDR
CMD
CMD
ADDR
ADDR
ADDR
CMD
60h
R1A
R2A
R3A
D1h
60h
R1B
R2B
R3B
D0h
tWB
tWB
tPLEBSY
SR[6]
Figure 89
Multi-plane Block Erase timing
R1A-R3A
Row address for erase block A. R1A is the least significant byte.
R1B-R3B
Row address for erase block B. R1B is the least significant byte.
168
tBERS
6.6. Multi-plane Read
The Read command reads a page of data identified by a row address for the LUN specified. The
page of data is made available to be read from the page register starting at the column address
specified. With a multi-plane operation, multiple reads can be issued back to back to the LUN,
with a shorter busy time between issuance of the next read operation. Figure 90 defines the
behavior and timings for issuing two multi-plane read commands. Figure 91 defines the behavior
and timings for reading data after the multi-plane read commands are ready to return data.
Cache operations may be used when doing multi-plane read operations, as shown, if supported
by the target as indicated in the parameter page. Refer to section 5.7.1.27.
Change Read Column Enhanced shall be issued prior to reading data from a LUN. If data is read
without issuing a Change Read Column Enhanced, the output received is undefined.
A
Cycle Type
DQx
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1A
C2A
R1A
R2A
R3A
32h
tWB
tPLRBSY
SR[6]
A
Cycle Type
DQx
B
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1B
C2B
R1B
R2B
R3B
30h
tRR
tWB
tR
SR[6]
Figure 90
Multi-plane Read command issue timing
C1A-C2A
Column address for page A. C1A is the least significant byte.
R1A-R3A
Row address for page A. R1A is the least significant byte.
C1B-C2B
Column address for page B. C1B is the least significant byte.
R1B-R3B
Row address for page B. R1B is the least significant byte.
The row addresses for page A and B shall differ in the plane address bits.
169
B
Cycle Type
CMD
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
DOUT
DOUT
Dn
Dn+1
tCCS
DQx
06h
30h
C1
C2
R1
R2
R3
E0h
tRR
tWB
tR
SR[6]
Figure 91
Multi-plane Read data output timing, continued from command issue
C1-C2
Column address to read from. C1 is the least significant byte.
R1-R3
Row address to read from (specifies LUN and plane address). R1 is the least significant byte.
Dn
Data bytes read starting with addressed row and column.
The row address provided shall specify a LUN and plane address that has valid read data.
For Multi-plane Read Cache Sequential operations, the initial Multi-plane Read command issue is followed by a Read Cache confirmation opcode
31h, as shown in Figure 92.
170
As defined for
Multi-plane Read
D
B
Cycle Type
DQx
CMD
CMD
30h
31h
tRR
tWB
tR
tWB
tRCBSY
SR[6]
Figure 92
Multi-plane Read Cache Sequential command issue timing
For Multi-plane Read Cache Random operations, the initial multi-plane Read command issue is followed by another Read Multi-plane command
sequence where the last confirmation opcode is 31h, as shown in Figure 93.
171
B
C
Cycle Type
DQx
CMD
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
30h
00h
C1C
C2C
R1C
R2C
R3C
32h
tRR
tWB
tWB
tPLRBSY
tR
SR[6]
C
Cycle Type
DQx
D
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
00h
C1D
C2D
R1D
R2D
R3D
31h
tRR
tWB
tRCBSY
SR[6]
Figure 93
Multi-plane Read Cache Random command issue timing
C1C-C2C
Column address for page C. C1C is the least significant byte.
R1C-R3C
Row address for page C. R1C is the least significant byte.
C1D-C2D
Column address for page D. C1D is the least significant byte.
R1D-R3D
Row address for page D. R1D is the least significant byte.
172
The row addresses for page C and D shall differ in the plane address bits.
For Multi-plane Read Cache operations, two data output operations follow each Multi-plane Read Cache operation. The individual data output
sequences are described in Figure 91. Prior to the last set (i.e. two) data output operations, a Read Cache End command (3Fh) should be issued
by the host.
173
7. Behavioral Flows
7.1. Target behavioral flows
The Target state machine describes the allowed sequences when operating with the target. If
none of the arcs are true, then the target remains in the current state.
7.1.1. Variables
This section describes variables used within the Target state machine.
tbStatusOut
This variable is set to TRUE when a data read cycle should return the status value. The
power-on value for this variable is FALSE.
tbChgCol
This variable is set to TRUE when changing the column using Change Read Column is
allowed. The power-on value for this variable is FALSE.
tbChgColEnh
This variable is set to TRUE when changing the column using Change Read Column
Enhanced is allowed. The power-on value for this variable is FALSE.
tCopyback
This variable is set to TRUE if the Target is issuing a copyback command. The power-on
value for this variable is FALSE.
tLunSelected
This variable contains the LUN that is currently selected by the host. The power-on value for
this variable is 0.
tLastCmd
This variable contains the first cycle of the last command (other than 70h/78h) received by
the Target.
tReturnState
tbStatus78hReq
This variable contains the state to return to after status operations.
This variable is set to TRUE when the next status operation shall be a 78h command (and
not a 70h command). The power-on value for this variable is FALSE.
7.1.2. Idle states
1
The target performs the following actions:
1. R/B_n is cleared to zero.
2. Each LUN shall draw less than 10 mA of power per staggered
power-up requirement.
2
 T_PowerOnReady
1. Target is ready to accept FFh (Reset) command
NOTE:
1. This state is entered as a result of a power-on event when Vcc reaches Vcc_min.
2. This arc shall be taken within 1 millisecond of Vcc reaching Vcc_min.
T_PowerOn
T_PowerOnReady
The target performs the following actions:
1. R/B_n is set to one.
2. Each LUN shall draw less than 10mA of power per staggered
power-up requirement.
1. Command cycle FFh (Reset) received
 T_RST_PowerOn
174
T_Idle
tCopyback set to FALSE. tReturnState set to T_Idle.
1. WP_n signal transitioned
 T_Idle_WP_Transition
2. LUN indicates its SR[6] value transitioned
 T_Idle_RB_Transition
3. Command cycle received
 T_Cmd_Decode
1
Decode command received. tbStatusOut is set to FALSE. If R/B_n is
set to one and command received is not 70h (Read Status), then
tbStatus78hReq is set to FALSE.
1. (Command 80h (Page Program) or command 60h  T_Idle
(Block Erase) decoded) and WP_n is low
2. Command FFh (Reset) decoded
 T_RST_Execute
3. Command FCh (Synchronous Reset) decoded
 T_RST_Execute_Sync
4. Command FAh (Reset LUN) decoded
 T_RST_Execute_LUN
5. Command 90h (Read ID) decoded
 T_RID_Execute
6. Command ECh (Read Parameter Page) decoded
 T_RPP_Execute
7. Command EDh (Read Unique ID) decoded
 T_RU_Execute
8. Command 80h (Page Program) decoded and WP_n is
 T_PP_Execute
high
9. Command 60h (Block Erase) decoded and WP_n is
 T_BE_Execute
high
10. Command 00h (Read) decoded
 T_RD_Execute
11. Command EFh (Set Features) decoded
 T_SF_Execute
12. Command EEh (Get Features) decoded
 T_GF_Execute
13. Command 70h (Read Status) decoded
 T_RS_Execute
14. Command 78h (Read Status Enhanced) decoded
 T_RSE_Execute
NOTE:
1. The host shall ensure R/B_n is set to one before issuing Target level commands (Reset,
Read ID, Read Parameter Page, Read Unique ID, Set Features, Get Features).
T_Cmd_Decode
T_Idle_WP_Transition
Indicate WP_n value to all LUN state machines.
1. State entered from T_Idle_Rd
 T_Idle_Rd
2. Else
 T_Idle
1
T_Idle_RB_Transition
R/B_n is set to the AND of all LUN status register SR[6] values.
1. Unconditional
 tReturnState
NOTE:
1. R/B_n may transition to a new value prior to the Target re-entering an idle condition
when LUN level commands are in the process of being issued.
175
7.1.3. Idle Read states
T_Idle_Rd
1.
2.
3.
4.
5.
6.
Wait for read request (data or status) or other action. tReturnState
set to T_Idle_Rd.
WP_n signal transitioned
 T_Idle_WP_Transition
LUN indicates its SR[6] value transitioned
 T_Idle_RB_Transition
Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
Read request received and (tLastCmd set to 90h or
 T_Idle_Rd_XferByte
EEh)
Read request received and (tLastCmd set to ECh or
 T_Idle_Rd_LunByte
EDh)
Read request received and tbStatus78hReq set to
 T_Idle_Rd_LunData
1
FALSE
7. Command cycle 05h (Change Read Column) received  T_CR_Execute2
and tbChgCol set to TRUE
8. Command cycle 06h (Change Read Column Enhanced)  T_CRE_Execute2
received and tbChgColEnh set to TRUE
9. Command cycle of 31h received and tbStatus78hReq  T_Idle_Rd_CacheCmd
set to FALSE
10. Command cycle of 3Fh received and tLastCmd set to  T_Idle_Rd_CacheCmd
31h and tbStatus78hReq set to FALSE
11. Command cycle received
 T_Cmd_Decode
NOTE:
1. When tbStatus78hReq is set to TRUE, a Read Status Enhanced (78h) command
followed by a 00h command shall be issued by the host prior to reading data from a
particular LUN.
2. If there are reads outstanding on other LUNs for this target, a Change Read Column
(Enhanced) shall be issued before transferring data. Refer to section 3.1.3 that
describes multiple LUN operation restrictions.
T_Idle_Rd_CacheCmd
1. Unconditional
T_Idle_Rd_XferByte
1. Unconditional
Set tLastCmd to the command received. Pass command received to
LUN tLunSelected
 T_Idle_Rd
Return next byte of data.

T_Idle_Rd
T_Idle_Rd_LunByte
Request byte of data from page register of LUN tLunSelected.
1. Byte received from LUN tLunSelected
 T_Idle_Rd_XferHost
T_Idle_Rd_LunData
Request byte (x8) or word (x16) of data from page register of LUN
tLunSelected.
1. Byte or word received from LUN tLunSelected
 T_Idle_Rd_XferHost
176
T_Idle_Rd_XferHost
Transfer data byte or word received from LUN tLunSelected to host.
1. tReturnState set to T_RD_StatusOff and tCopyback set  T_RD_Copyback
to TRUE
2. tReturnState set to T_RD_StatusOff
 T_Idle_Rd
3. Else
 tReturnState
T_Idle_Rd_Status
Request status from LUN tLunSelected.
1. Status from LUN tLunSelected received

T_Idle_Rd_StatusEnd
1. Unconditional
T_Idle_Rd_StatusEnd
Transfer status byte received from LUN tLunSelected to host.
 tReturnState
T_CR_Execute
Wait for a column address cycle.
1. Column address cycle received

T_CR_Addr
T_CR_Addr
Store the column address cycle received.
1. More column address cycles required

2. All column address cycles received

T_CR_Execute
T_CR_WaitForCmd
T_CR_WaitForCmd
Wait for a command cycle.
1. Command cycle E0h received
T_CR_ReturnToData

T_CR_ReturnToData
Request that LUN tLunSelected select the column in the page
register based on the column address received.
1. tReturnState set to T_PP_MplWait
 T_PP_WaitForDataOut
2. tReturnState set to T_RD_Status_Off
 T_Idle_Rd
3. Else
 tReturnState
T_CRE_Execute
Wait for a column address cycle.
1. Column address cycle received

T_CRE_ColAddr
Store the column address cycle received.
1. More column address cycles required

2. All column address cycles received

177
T_CRE_ColAddr
T_CRE_Execute
T_CRE_RowAddrWait
T_CRE_RowAddrWait
Wait for a row address cycle.
1. Row address cycle received

T_CRE_RowAddr
T_CRE_RowAddr
Store the row address cycle received.
1. More row address cycles required

2. All row address cycles received

T_CRE_RowAddrWait
T_CRE_WaitForCmd
T_CRE_WaitForCmd
Wait for a command cycle.
1. Command cycle E0h received
T_CRE_ReturnToData
T_CRE_ReturnToData

The target performs the following actions:
1. Set tLunSelected to LUN selected by row address received.
2. Request that LUN tLunSelected select the column in the page
register based on the column address received.
3. Indicate plane address received to tLunSelected for use in
data output.
4. Request all idle LUNs not selected turn off their output
1
buffers.
1. tReturnState set to T_PP_MplWait
 T_PP_WaitForDataOut
2. tReturnState set to T_RD_Status_Off
 T_Idle_Rd
3. Else
 tReturnState
NOTE:
1. LUNs not selected only turn off their output buffers if they are in an idle condition (SR[6]
is one) when Change Read Column Enhanced is received. If LUNs are active (SR[6] is
zero) when Change Read Column Enhanced is issued, then the host shall issue a Read
Status Enhanced (78h) command prior to subsequent data output to ensure all LUNs
that are not selected turn off their output buffers.
7.1.4. Reset command states
T_RST_PowerOn
1. Unconditional
T_RST_PowerOn_Exec
1. Unconditional
The target performs the following actions:
1. tLastCmd set to FFh.
2. tbStatusOut is set to FALSE.
3. The target sends a Reset request to each LUN.
 T_RST_PowerOn_Exec
The target performs the following actions:
1. Target level reset actions are performed.
2. R/B_n is set to zero.
 T_RST_Perform
178
1
The target performs the following actions:
1. tLastCmd set to FFh.
2. The target selects the asynchronous data interface.
3. The target sends a Reset request to each LUN.
4. Set tbChgCol to FALSE.
5. Set tbChgColEnh to FALSE.
6. Request all LUNs invalidate page register(s).
 T_RST_Perform
T_RST_Execute
1. Unconditional
NOTE:
1. This state is entered as a result of receiving a Reset (FFh) command in any other state,
except if this is the first Reset after power-on.
T_RST_Execute_Sync
1
The target performs the following actions:
1. tLastCmd set to FCh.
2. tbStatusOut is set to FALSE.
3. The target sends a Reset request to each LUN.
4. Set tbChgCol to FALSE.
5. Set tbChgColEnh to FALSE.
6. Request all LUNs invalidate page register(s).
 T_RST_Perform
1. Unconditional
NOTE:
1. This state is entered as a result of receiving a Synchronous Reset (FCh) command in
any other state.
T_RST_Execute_LUN
1
1. Unconditional
The target performs the following actions:
1. tLastCmd set to FAh.
2. tbStatusOut is set to FALSE.
3. Set tbChgCol to FALSE.
4. Set tbChgColEnh to FALSE.
5. Wait for an address cycle.

179
T_RST_LUN_Addr
T_RST_LUN_AddrWait
Wait for an address cycle.
1. Address cycle received

T_RST_LUN_Addr
T_RST_LUN_Addr
Store the address cycle received.
1. More address cycles required
2. All address cycles received


T_RST_LUN_AddrWait
T_RST_LUN_Perform
T_RST_LUN_Perform
The target performs the following actions:
1. The target sends a Reset request to the addressed LUN.
2. R/B_n is cleared to zero.
3. Request the addressed LUN invalidate its page register.
1. Addressed LUN reset actions are complete and
 T_Idle
tbStatusOut is set to FALSE
2. Addressed LUN reset actions are complete and
 T_Idle_Rd
tbStatusOut is set to TRUE
T_RST_Perform
The target performs the following actions:
1. Target level reset actions are performed.
2. R/B_n is set to zero.
3. tReturnState set to T_RST_Perform.
1. Target and LUN reset actions are complete
 T_RST_End
2. Command cycle 70h (Read Status) received
 T_RS_Execute
3. Read request received and tbStatusOut is set to TRUE
 T_Idle_Rd_Status
T_RST_End
The target performs the following actions:
1. R/B_n is set to one.
1. tbStatusOut is set to FALSE

2. tbStatusOut is set to TRUE

T_Idle
T_Idle_Rd
7.1.5. Read ID command states
T_RID_Execute
The target performs the following actions:
1. tLastCmd set to 90h.
2. Wait for an address cycle.
3. Set tbChgCol to FALSE.
4. Set tbChgColEnh to FALSE.
5. Request all LUNs invalidate page register(s).
1. Address cycle of 00h received
 T_RID_Addr_00h
2. Address cycle of 20h received
 T_RID_Addr_20h
180
T_RID_Addr_00h
Wait for the read request.
1. Read byte request received
2. Command cycle received


T_RID_ManufacturerID
T_Cmd_Decode
T_RID_ManufacturerID
Return the JEDEC manufacturer ID.
1. Read byte request received
2. Command cycle received


T_RID_DeviceID
T_Cmd_Decode
T_RID_DeviceID
Return the device ID.
1
1. Unconditional
 T_Idle_Rd
NOTE:
1. Reading bytes beyond the device ID returns vendor specific values.
T_RID_Addr_20h
Wait for the read request.
1. Read byte request received
2. Command cycle received


T_RID_Signature
T_Cmd_Decode
1
T_RID_Signature
Return next ONFI signature byte.
1. Last ONFI signature byte returned

2. Else

NOTE:
1. Reading beyond the fourth byte returns indeterminate values.
T_Idle_Rd
T_RID_Addr_20h
7.1.6. Read Parameter Page command states
T_RPP_Execute
The target performs the following actions:
1. tLastCmd set to ECh.
2. Set tbChgCol to TRUE.
3. Set tbChgColEnh to FALSE.
4. Wait for an address cycle.
5. Request all LUNs invalidate page register(s).
6. Target selects LUN to execute parameter page read, sets
tLunSelected to the address of this LUN.
1. Address cycle of 00h received
 T_RPP_ReadParams
181
T_RPP_ReadParams
The target performs the following actions:
1. Request LUN tLunSelected clear SR[6] to zero.
2. R/B_n is cleared to zero.
3. Request LUN tLunSelected make parameter page data
available in page register.
4. tReturnState set to T_RPP_ReadParams_Cont.
1. Read of page complete
 T_RPP_Complete
2. Command cycle 70h (Read Status) received
 T_RS_Execute
3. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
T_RPP_ReadParams_Cont
1. Read of page complete
2. Command cycle 70h (Read Status) received
3. Read request received and tbStatusOut set to TRUE
T_RPP_Complete
1. Unconditional



T_RPP_Complete
T_RS_Execute
T_Idle_Rd_Status
Request LUN tLunSelected set SR[6] to one. R/B_n is set to one.
 T_Idle_Rd
7.1.7. Read Unique ID command states
T_RU_Execute
The target performs the following actions:
1. tLastCmd set to EDh.
2. Set tbChgCol to TRUE.
3. Set tbChgColEnh to FALSE.
4. Request all LUNs invalidate page register(s).
5. Wait for an address cycle.
6. Target selects LUN to execute unique ID read, sets
tLunSelected to the address of this LUN.
1. Address cycle of 00h received
 T_RU_ReadUid
T_RU_ReadUid
The target performs the following actions:
1. Request LUN tLunSelected clear SR[6] to zero.
2. R/B_n is cleared to zero.
3. Request LUN tLunSelected make Unique ID data available in
page register.
4. tReturnState set to T_RU_ReadUid.
1. LUN tLunSelected indicates data available in page
 T_RU_Complete
register
2. Command cycle 70h (Read Status) received
 T_RS_Execute
3. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
T_RU_Complete
1. Unconditional
Request LUN tLunSelected set SR[6] to one. R/B_n is set to one.
 T_Idle_Rd
182
7.1.8. Page Program and Page Cache Program command states
T_PP_Execute
The target performs the following actions:
1. tLastCmd set to 80h.
2. If R/B_n is cleared to zero, then tbStatus78hReq is set to
TRUE.
3. If the program page register clear enhancement is not
supported or disabled, request all LUNs clear their page
1
register(s).
1. Unconditional
 T_PP_AddrWait
NOTE:
1. Idle LUNs may choose to not clear their page register if the Program is not addressed to
that LUN.
T_PP_Copyback
1. Unconditional
If R/B_n is cleared to zero, then tbStatus78hReq is set to TRUE.
 T_PP_AddrWait
T_PP_AddrWait
Wait for an address cycle.
1. Address cycle received

T_PP_Addr
T_PP_Addr
Store the address cycle received.
1. More address cycles required
2. All address cycles received


T_PP_AddrWait
T_PP_LUN_Execute
183
T_PP_LUN_Execute
1. Unconditional
The target performs the following actions:
1. tLunSelected is set to the LUN indicated by the row address
received.
2. If the program page register clear enhancement is enabled,
request LUN tLunSelected clear the page register for the
plane address specified.
3. Target issues the Program with associated address to the
LUN tLunSelected.
 T_PP_LUN_DataWait
T_PP_LUN_DataWait
1.
2.
3.
4.
Wait for data byte/word or command cycle to be received from the
host.
Data byte/word received from the host
 T_PP_LUN_DataPass
Command cycle of 15h received and tCopyback set to  T_PP_Cmd_Pass
FALSE
Command cycle of 10h or 11h received
 T_PP_Cmd_Pass
Command cycle of 85h received
 T_PP_ColChg
T_PP_LUN_DataPass
1. Unconditional
Pass data byte/word received from host to LUN tLunSelected
 T_PP_LUN_DataWait
T_PP_Cmd_Pass
Pass command received to LUN tLunSelected
1. Command passed was 11h
 T_PP_MplWait
2. Command passed was 10h or 15h
 T_Idle
184
T_PP_MplWait
Wait for next Program to be issued. tReturnState set to
T_PP_MplWait.
1
 T_PP_AddrWait
1. Command cycle of 85h received
2
 T_PP_AddrWait
2. Command cycle of 80h received and tCopyback set to
FALSE
3. Command cycle of 05h received
 T_CR_Execute
4. Command cycle of 06h received
 T_CRE_Execute
5. Command cycle of 70h received
 T_RS_Execute
6. Command cycle of 78h received
 T_RSE_Execute
7. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
NOTE:
1. If the 85h is part of a Copyback, Change Row Address, or Small Data Move operation,
then the LUN address and plane address shall be the same as the preceding Program
operation. If the 85h is part of a Small Data Move operation, then the page address shall
also be the same as the preceding Program operation.
2. Address cycles for the Program operation being issued shall have the same LUN
address and page address as the preceding Program operation. The plane address
shall be different than the one issued in the preceding Program operation.
T_PP_ColChg
Wait for column address cycle.
1. Address cycle received

T_PP_ColChg_Addr
T_PP_ColChg_Addr
Store the address cycle received.
1. More column address cycles required
2. All address cycles received


T_PP_ColChg
T_PP_ColChg_LUN
T_PP_ColChg_LUN
1. Unconditional
Request that LUN tLunSelected change column address to column
address received.
 T_PP_ColChg_Wait
T_PP_ColChg_Wait
1.
2.
3.
4.
5.
Wait for an address cycle, data byte/word, or command cycle to be
received from the host
Address cycle received
 T_PP_RowChg_Addr
Data byte/word received from the host
 T_PP_LUN_DataPass
Command cycle of 15h received and tCopyback set to  T_PP_Cmd_Pass
FALSE
Command cycle of 10h or 11h received
 T_PP_Cmd_Pass
Command cycle of 85h received
 T_PP_ColChg
T_PP_RowChg
Wait for row address cycle.
1. Address cycle received
185

T_PP_RowChg_Addr
T_PP_RowChg_Addr
Store the address cycle received.
1. More row address cycles required
2. All address cycles received
T_PP_RowChg_LUN


T_PP_RowChg
T_PP_RowChg_LUN
Request that LUN tLunSelected change row address to row address
received.
1
1. Unconditional
 T_PP_LUN_DataWait
NOTE:
1. The LUN address and plane address shall be the same as previously specified for the
Program operation executing.
T_PP_WaitForDataOut
Wait for read request (data or status) or other action. tReturnState
set to T_PP_WaitForDataOut.
1. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
2. Read request received and tbStatus78hReq set to  T_Idle_Rd_LunData
1
FALSE
3. Command cycle of 70h received
 T_RS_Execute
4. Command cycle of 78h received
 T_RSE_Execute
5. Command cycle of 00h received
 T_RD_Execute
6. Command cycle received
 T_PP_MplWait
NOTE:
1. When tbStatus78hReq is set to TRUE, a Read Status Enhanced (78h) command
followed by a 00h command shall be issued by the host prior to reading data from a
particular LUN.
7.1.9. Block Erase command states
T_BE_Execute
The target performs the following actions:
1. tLastCmd set to 60h.
2. If R/B_n is cleared to zero, then tbStatus78hReq is set to
TRUE.
3. Wait for a row address cycle.
1. Address cycle received
 T_BE_Addr
T_BE_Addr
Store the row address cycle received.
1. More address cycles required

2. All address cycles received

T_BE_LUN_Execute
1. Unconditional
T_BE_Execute
T_BE_LUN_Execute
tLunSelected is set to the LUN indicated by the row address received.
Target issues the Erase with associated row address to the LUN
tLunSelected.
 T_BE_LUN_Confirm
186
T_BE_LUN_Confirm
Wait for D0h or D1h command cycle.
1. Command cycle of D0h or D1h received

T_BE_Cmd_Pass
T_BE_Cmd_Pass
Pass command received to LUN tLunSelected
1. Command passed was D1h
 T_BE_MplWait
2. Command passed was D0h
 T_Idle
T_BE_MplWait
Wait for next Erase to be issued. tReturnState set to T_BE_MplWait.
1. Command cycle of 60h received
 T_BE_Execute
2. Command cycle of 70h received
 T_RS_Execute
3. Command cycle of 78h received
 T_RSE_Execute
4. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
187
7.1.10.
Read command states
T_RD_Execute
1. tbStatusOut set to TRUE
2. Else


T_RD_StatusOff
T_RD_AddrWait
T_RD_StatusOff
tbStatusOut set to FALSE. tReturnState set to T_RD_StatusOff.
1. Address cycle received
 T_RD_Addr
2. Read request received and tLastCmd set to 80h
 T_PP_WaitForDataOut
3. Read request received and tLastCmd set to EEh
 T_Idle_Rd_XferHost
4. Read request received
 T_Idle_Rd_LunData
5. Command cycle of 05h received
 T_CR_Execute
6. Command cycle of 06h received
 T_CRE_Execute
T_RD_AddrWait
tLastCmd set to 00h. Set tbChgCol to TRUE. Set tbChgColEnh to
TRUE. If R/B_n is cleared to zero, then tbStatus78hReq is set to
TRUE. Wait for an address cycle.
1. Address cycle received
 T_RD_Addr
T_RD_Addr
Store the address cycle received.
3. More address cycles required
4. All address cycles received
T_RD_LUN_Execute


T_RD_AddrWait
T_RD_LUN_Execute
The target performs the following actions:
1. tLunSelected is set to the LUN indicated by the row address
received.
2. Issues the Read Page with address to LUN tLunSelected.
3. Requests all idle LUNs not selected to turn off their output
buffers.
1
1. Unconditional
 T_RD_LUN_Confirm
NOTE:
1. LUNs not selected will only turn off their output buffers if they are in an Idle state. If other
LUNs are active, the host shall issue a Read Status Enhanced (78h) command to ensure
all LUNs that are not selected turn off their output buffers prior to issuing the Read (00h)
command.
T_RD_LUN_Confirm
Wait for 30h, 31h, 32h, or 35h to be received.
1. Command cycle of 30h, 31h, 32h, or 35h received
 T_RD_Cmd_Pass
T_RD_Cmd_Pass
Pass command received to LUN tLunSelected
1. Command passed was 35h
 T_RD_Copyback
2. Command passed was 30h, 31h, or 32h
 T_Idle_Rd
188
T_RD_Copyback
tCopyback set to TRUE. tReturnState set to T_RD_Copyback.
1. Command cycle of 00h received
 T_RD_Execute
2. Command cycle of 05h received
 T_CR_Execute
3. Command cycle of 06h received
 T_CRE_Execute
4. Command cycle of 85h received
 T_PP_Copyback
5. Command cycle of 70h received
 T_RS_Execute
6. Command cycle of 78h received
 T_RSE_Execute
7. LUN indicates its SR[6] value transitions
 T_Idle_RB_Transition
8. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
9. Read request received
 T_Idle_Rd_LunData
189
7.1.11.
Set Features command states
T_SF_Execute
The target performs the following actions:
1. tLastCmd set to EFh.
2. Request all LUNs invalidate page register(s).
3. Wait for an address cycle.
1. Address cycle received
 T_SF_Addr
T_SF_Addr
1. Unconditional
Store the feature address received.

T_SF_WaitForParams
T_SF_WaitForParams
Wait for data byte to be received.
1. Data byte written to target

T_SF_StoreParam
T_SF_StoreParam
Store parameter received.
1. More parameters required
2. All parameters received


T_SF_WaitForParams
T_SF_Complete
T_SF_Complete
The target performs the following actions:
1. Request LUN tLunSelected clear SR[6] to zero.
2. R/B_n is cleared to zero.
3. Finish Set Features command.
4. tReturnState set to T_SF_Complete.
1. Set Features command complete
 T_SF_UpdateStatus
2. Command cycle 70h (Read Status) received
 T_RS_Execute
3. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
T_SF_UpdateStatus
The target performs the following actions:
1. Request LUN tLunSelected set SR[6] to one.
2. R/B_n is set to one.
1. tbStatusOut is set to FALSE
 T_Idle
2. tbStatusOut is set to TRUE
 T_Idle_Rd
190
7.1.12.
Get Features command states
T_GF_Execute
The target performs the following actions:
1. tLastCmd set to EEh.
2. Request all LUNs invalidate page register(s).
3. Set tbChgCol to FALSE.
4. Set tbChgColEnh to FALSE.
5. Wait for an address cycle.
1. Address cycle received
 T_GF_Addr
T_GF_Addr
1. Unconditional
Store the feature address received.

T_GF_RetrieveParams
T_GF_RetrieveParams
The target performs the following actions:
1. Request LUN tLunSelected clear SR[6] to zero.
2. R/B_n is cleared to zero.
3. Retrieve parameters.
4. tReturnState set to T_GF_RetrieveParams.
1. Parameters are ready to be transferred to the host
 T_GF_Ready
2. Command cycle 70h (Read Status) received
 T_RS_Execute
3. Read request received and tbStatusOut set to TRUE
 T_Idle_Rd_Status
T_GF_Ready
1. Unconditional
7.1.13.
Request LUN tLunSelected set SR[6] to one. R/B_n is set to one.
 T_Idle_Rd
Read Status command states
T_RS_Execute
1
 T_RS_Perform
1. tbStatus78hReq is set to FALSE
NOTE:
1. When tbStatus78hReq is set to TRUE, issuing a Read Status (70h) command is illegal.
T_RS_Perform
The target performs the following actions:
1. tbStatusOut is set to TRUE.
2. Indicate 70h command received to LUN tLunSelected.
1. tReturnState set to T_Idle
 T_Idle_Rd
2. Else
 tReturnState
191
7.1.14.
Read Status Enhanced command states
1
tbStatus78hReq is set to FALSE. tbStatusOut is set to TRUE. Wait
for a row address cycle.
1. Row address cycle received
 T_RSE_Addr
NOTE:
1. The host should not issue Read Status Enhanced following a Target level command
(Reset, Read ID, Read Parameter Page, Read Unique ID, Set Features, Get Features).
The status value read from the LUN selected with Read Status Enhanced may not
correspond with the LUN selected during the Target level command.
T_RSE_Execute
T_RSE_Addr
Store the row address cycle received.
1. More row address cycles required

2. All row address cycles received

T_RSE_Select
T_RSE_Execute
T_RSE_Select
The target performs the following actions:
1. Set tLunSelected to LUN selected by row address received.
2. Indicate 78h command and row address received to all LUNs.
1. tReturnState set to T_Idle
 T_Idle_Rd
2. Else
 tReturnState
192
7.2. LUN behavioral flows
The LUN state machine describes the allowed sequences when operating with the LUN. If none
of the arcs are true, then the LUN remains in the current state.
7.2.1. Variables
This section describes variables used within the LUN state machine.
lunStatus
This variable contains the current LUN status register value contents.
for this variable is 00h.
The power on value
lunFail[]
This array contains the FAIL and FAILC bits for each interleave address. For example,
lunFail[3][1] contains the FAILC bit for plane address 3. The power on value for each
variable in this array is 00b.
lunLastConfirm
This variable contains the last confirm command cycle (30h, 31h, 32h, 35h, 10h, 15h, 11h,
D0h, D1h). The power on value for this variable is FFh.
lunOutputMpl
This variable contains the plane address requested for data output. The power on value for
this variable is 0h.
lunReturnState
This variable contains the state to return to after status operations. The power on value for
this variable is L_Idle.
lunStatusCmd
This variable contains the last status command received.
variable is 70h.
The power on value for this
lunStatusMpl
This variable contains the plane address indicated in a previous 78h command. The power
on value for this variable is 0h.
lunbInterleave
This variable is set to one when the LUN is performing a multi-plane operation. The power
on value for this variable is FALSE.
lunbMplNextCmd
This variable is set to TRUE when the LUN is ready to receive the next multi-plane
command.
lunEraseAddr[]
This variable contains the block addresses of erases that have been suspended.
7.2.2. Idle command states
1
L_Idle
lunReturnState is set to L_Idle.
1. Target request received
 L_Idle_TargetRequest
NOTE:
1. This state is entered as a result of a power-on event when Vcc reaches Vcc_min.
193
L_Idle_TargetRequest
If Target indicates an address, the address is stored by the LUN.
1. Target requests LUN perform a Reset
 L_RST_Execute
2. Target indicates WP_n value
 L_WP_Update
3. Target requests SR register update
 L_SR_Update
4. Target requests status or status command received
 L_Status_Execute
5. Target indicates plane address for use in data output
 L_Idle_Mpl_DataOutAdd
r
6. Target indicates output buffer should be turned off
 L_Idle
7. Target requests page register clear
 L_Idle_ClearPageReg
8. Target requests page register invalidate
 L_Idle_InvalidPageReg
9. Target indicates Program request for this LUN
 L_PP_Execute
10. Target indicates Erase request for this LUN
 L_BE_Execute
11. Target indicates Erase Resume request for this LUN
 L_ER_Execute
12. Target indicates Read Page request for this LUN
 L_RD_Addr
13. Target indicates Read Parameter Page request
 L_Idle_RdPp
14. Target indicates Read Unique ID request
 L_Idle_RdUid
L_WP_Update
1. Unconditional
Set lunStatus[7] to the WP_n value indicated by the target.
 lunReturnState
L_SR_Update
1. Unconditional
Update lunStatus as indicated by the target.
L_Idle_Mpl_DataOutAddr
1. Unconditional
Set lunOutputMpl to plane address indicated by the target.
 lunReturnState
L_Idle_ClearPageReg
1. Unconditional
Set page register to all ones value.
L_Idle_InvalidPageReg
1. Unconditional
Invalidate page register.

L_Idle_RdPp
lunReturnState

lunReturnState

lunReturnState
The LUN performs the following actions:
1. LUN reads parameter page data into the page register.
2. lunReturnState set to L_Idle_RdPp_Cont.
1. Parameter page data transferred to page register
 L_Idle_RdPp_End
2. Target requests status or status command received
 L_Status_Execute
194
L_Idle_RdPp_Cont
1. Parameter page data transferred to page register
2. Target requests status or status command received
L_Idle_RdPp_End
1. Unconditional


L_Idle_RdPp_End
L_Status_Execute
LUN indicates to Target that parameter page data is in page register.
 L_Idle_Rd
L_Idle_RdUid
The LUN performs the following actions:
1. LUN reads Unique ID data into the page register.
2. lunReturnState set to L_Idle_RdUid.
1. Unique ID data transferred to page register
 L_Idle_RdUid_End
2. Target requests status or status command received
 L_Status_Execute
L_Idle_RdUid_End
1. Unconditional
LUN indicates to Target that Unique ID data is in page register.
 L_Idle_Rd
7.2.3. Idle Read states
L_Idle_Rd
1.
2.
3.
4.
5.
lunReturnState is set to L_Idle_Rd.
Background read operation complete
Target requests column address be selected
Read request received from Target
Command cycle 31h (Read Cache Sequential) received
Command cycle 3Fh (Read Cache End) received and
lunLastConfirm is 31h
6. Target request received
L_Idle_Rd_Finish
1. Unconditional





L_Idle_Rd_Finish
L_Idle_Rd_ColSelect
L_Idle_Rd_Xfer
L_RD_Cache_Next
L_RD_Cache_Xfer_End

L_Idle_TargetRequest

L_Idle_Rd
Set lunStatus[5] to one.
L_Idle_Rd_Xfer
Return to the target the next byte (x8) or word (x16) of data from page
register based on Target requested. Increments column address.
1. lunReturnState set to L_PP_Mpl_Wait
 L_PP_Mpl_Wait
2. Unconditional
 L_Idle_Rd
L_Idle_Rd_ColSelect
Select the column in the page register based on the column address
received from the target.
1. lunReturnState set to L_PP_Mpl_Wait
 L_PP_Mpl_Wait
2. Unconditional
 L_Idle_Rd
195
7.2.4. Status states
L_Status_Execute
1. Target requests status value
2. Target indicates 78h was received
3. Target indicates 70h was received
L_Status_Value
1. lunbInterleave set to TRUE and lunStatusCmd set to
70h
2. lunbInterleave set to TRUE and lunStatusCmd set to
78h
3. lunbInterleave set to FALSE
L_Status_Enhanced
1. LUN in row address indicated matches this LUN
2. Else



L_Status_Value
L_Status_Enhanced
L_Status_Legacy

L_Status_Mpl_Comp

L_Status_Mpl_Addr

L_Status_Lun


L_Status_Record_78h
L_Status_Output_Off
L_Status_Record_78h
lunStatusCmd is set to 78h and lunStatusMpl is set to plane address
indicated by Target. The LUN turns on its output buffer.
1. lunReturnState set to L_Idle and (lunLastConfirm set to
 L_Idle_Rd
30h, 31h, 32h, or 35h)
2. Else
 lunReturnState
L_Status_Output_Off
LUN turns off its output buffer.
1. lunReturnState set to L_Idle_Rd
2. Else


L_Idle
lunReturnState
L_Status_Legacy
1. Unconditional
lunStatusCmd is set to 70h.

lunReturnState
L_Status_Mpl_Comp
The LUN composes the status value to return as shown:
 status[7:2] = lunStatus[7:2]
 status[1] = for all x, OR of lunFail[x][1]
 status[0] = for all x, OR of lunFail[x][0]
Return status to the Target.

1. Unconditional
L_Status_Mpl_Addr
lunReturnState
The LUN composes the status value to return as shown:
 status[7:2] = lunStatus[7:2]
 status[1:0] = lunFail[lunStatusMpl][1:0]
196
Return status to the Target.
1. Unconditional
L_Status_Lun
1. Unconditional

lunReturnState

lunReturnState
Return lunStatus to the Target.
7.2.5. Reset states
1
The LUN performs the following actions:
1. lunStatus[6] is cleared to zero.
2. lunStatus[6] value is indicated to the Target.
3. Perform reset of the LUN.
4. lunbInterleave is set to FALSE.
5. lunReturnState is set to L_RST_Execute.
1. Reset of the LUN is complete
 L_RST_Complete
2. Target requests status or status command received
 L_Status_Execute
NOTE:
1. This state is entered as a result of receiving an indication from the Target state machine
to perform a Reset in any other state.
L_RST_Execute
L_RST_Complete
1. Unconditional
The LUN performs the following actions:
1. lunStatus[1:0] are cleared to 00b.
2. For all plane addresses x, clear lunFail[x][1:0] to 00b.
3. lunStatus[6] is set to one.
4. lunStatus[6] value is indicated to the Target.
5. Indicate to the Target state machine that Reset for this LUN is
complete.
 L_Idle
7.2.6. Block Erase command states
L_BE_Execute
1. Unconditional
lunbInterleave set to FALSE.
L_BE_WaitForCmd
Wait for a command cycle.
1. Command cycle D0h received
2. Command cycle D1h received
L_BE_Erase

L_BE_WaitForCmd


L_BE_Erase
L_BE_Mpl
The LUN performs the following actions:
1. lunStatus[6] is cleared to zero.
2. If lunbInterleave is TRUE, lunStatus[5] is cleared to zero.
3. lunStatus[6] value is indicated to the Target.
4. lunLastConfirm set to D0h.
5. Erase the requested block and any previously requested
blocks if lunbInterleave is set to TRUE and concurrent
197
interleaving is supported.

1. Unconditional
L_BE_Erase_Wait
lunReturnState set to L_BE_Erase_Wait.
1. Erase of requested block(s) complete and

lunbInterleave set to TRUE
2. Erase of requested block complete

3. Target requests page register clear

4. Target requests status or status command received

L_BE_Mpl
1. Unconditional
1. Unconditional
L_BE_Mpl_Overlap
1. Unconditional
L_BE_Mpl_Sts
L_BE_Sts
L_Idle_ClearPageReg
L_Status_Execute
The LUN performs the following actions in the order specified:
1. lunbInterleave set to TRUE.
2. lunLastConfirm set to D1h.
3. lunStatus[6:5] is cleared to 00b. lunStatus[6] value is
indicated to the Target.
4. LUN begins erasing block specified if overlapped is
supported.
5. lunbMplNextCmd is set to FALSE.
6. LUN prepares to receive the next block to erase.
 L_BE_Mpl_Wait
L_BE_Mpl_Wait
lunReturnState set to L_BE_Mpl_Wait.
1. An overlapped multi-plane Erase completed
2. Ready to receive the next Erase command and
lunbMplNextCmd is set to FALSE
3. Target indicates Erase request for this LUN and
lunbMplNextCmd is set to TRUE
4. Target requests status or status command received
L_BE_Mpl_NextCmd
L_BE_Erase_Wait


L_BE_Mpl_Overlap
L_BE_Mpl_NextCmd

L_BE_WaitForCmd

L_Status_Execute
The LUN performs the following actions in the order specified:
1. lunbMplNextCmd is set to TRUE.
2. If no array operations are in progress, lunStatus[5] is set to
one.
3. lunStatus[6] is set to one. lunStatus[6] value is indicated to
the Target.
 L_BE_Mpl_Wait
The LUN performs the following actions in the order specified for the
overlapped multi-plane operation that completed:
1. mplComplete set to plane address of completed operation
2. lunFail[mplComplete][0] is set to program status of operation.
If all array operations are complete, lunStatus[5] is set to one.
 lunReturnState
198
L_BE_Sts
1. Unconditional
L_BE_Mpl_Sts
1. Unconditional
The LUN performs the following actions in the order specified:
1. lunStatus[0] is set to erase status.
2. lunStatus[6] is set to one. lunStatus[6] value is indicated to
the Target.
 L_Idle
The LUN performs the following actions in the order specified for each
multi-plane operation that completed:
1. mplComplete set to interleave address of completed
operation.
2. lunFail[mplComplete][0] is set to erase status value.
lunStatus[6:5] is set to 11b and lunStatus[6] value is indicated to the
Target.
 L_Idle
7.2.7. Read command states
If caching is not supported, then all actions for status bit 5 are ignored.
L_RD_Addr
1. Unconditional
The LUN performs the following actions in the order specified:
1. Records address received from the target.
2. If multi-plane addressing is supported, selects the correct
page register based on the plane address.
3. Selects the column in the page register based on the column
address received.
 L_RD_WaitForCmd
L_RD_WaitForCmd
lunbInterleave set to FALSE. Wait for a command cycle.
1. Command cycle 30h or 35h received
 L_RD_ArrayRead
2. Command cycle 31h received and lunLastConfirm equal  L_RD_Cache_Xfer
to 30h or 31h
3. Command cycle 32h received
 L_RD_Mpl_Xfer
L_RD_ArrayRead
The LUN performs the following actions:
1. lunStatus[6:5] is cleared to 00b.
2. lunStatus[6] value is indicated to the target.
3. lunLastConfirm set to last command cycle (30h or 35h).
4. Read the requested page from the array. If concurrent multiplane operation, read all pages requested from the array.
5. lunReturnState set to L_RD_ArrayRead_Cont.
1. Read of requested page(s) complete
 L_RD_Complete
2. Target requests status or status command received
 L_Status_Execute
199
L_RD_ArrayRead_Cont
1. Read of requested page(s) complete
2. Target requests status or status command received
L_RD_Complete
1. Unconditional
L_RD_Cache_Next
1. Unconditional


L_RD_Complete
L_Status_Execute
lunStatus[6:5] is set to 11b. lunStatus[6] value is indicated to the
target.
 L_Idle_Rd
Select the next row address as the sequential increasing row address
to the last page read.
 L_RD_Cache_Xfer
L_RD_Cache_Xfer
The LUN performs the following actions:
1. lunStatus[6:5] is cleared to 00b. lunStatus[6] value is
indicated to the Target.
2. lunLastConfirm set to 31h.
3. Begin background read operation for selected address.
4. lunReturnState set to L_RD_Cache_Xfer.
1. Data available in page register for previous read
 L_RD_Cache_Sts
operation
2. Target requests status or status command received
 L_Status_Execute
L_RD_Cache_Xfer_End
The LUN performs the following actions:
1. lunStatus[6] is cleared to zero.
2. lunStatus[6] value is indicated to the target.
3. lunLastConfirm set to 3Fh.
4. lunReturnState set to L_RD_Cache_Xfer_End.
1. Data available in page register for previous read
 L_RD_Cache_Sts_End
operation
2. Target requests status or status command received
 L_Status_Execute
L_RD_Cache_Sts
1. Unconditional
L_RD_Cache_Sts_End
1. Unconditional
lunStatus[6] is set to one. lunStatus[6] value is indicated to the
Target.
 L_Idle_Rd
lunStatus[6:5] is set to 11b. lunStatus[6] value is indicated to the
Target.
 L_Idle_Rd
200
L_RD_Mpl_Xfer
The LUN performs the following actions:
1. lunStatus[6:5] is cleared to 00b.
2. lunStatus[6] value is indicated to the target.
3. lunLastConfirm set to 32h.
4. lunbMplNextCmd is set to FALSE.
5. LUN begins reading page specified if overlapped interleaving
is supported.
6. Prepare to receive the next page to read.
7. lunReturnState set to L_RD_Mpl_Xfer.
1. Target ready to receive next page to read
 L_RD_Mpl_Wait
2. Target requests status or status command received
 L_Status_Execute
L_RD_Mpl_Wait
lunStatus[6] is set to one. lunStatus[6] value is indicated to the
Target. lunReturnState set to L_RD_Mpl_Wait.
1. An overlapped multi-plane Read completed
 L_RD_Mpl_Overlap
2. Target indicates Read Page request for this LUN
 L_RD_Addr
3. Target requests status or status command received
 L_Status_Execute
L_RD_Mpl_Overlap
1. Unconditional
The LUN performs the following actions in the order specified for the
overlapped multi-plane operation that completed:
1. mplComplete set to plane address of completed operation.
If all array operations are complete, lunStatus[5] is set to one.
 lunReturnState
7.2.8. Page Program and Page Cache Program command states
If caching or overlapped interleaving is not supported, then all actions for status bit 5 are ignored.
If caching is not supported, then all actions for status bit 1 are ignored.
L_PP_Execute
1. Unconditional
lunbInterleave set to FALSE.
L_PP_Addr
The LUN performs the following actions in the order specified:
1. Records address received from the Target.
2. If multi-plane addressing is supported, selects the correct
page register based on the plane address.
3. Selects the column in the page register based on the column
address received.
 L_PP_WaitForData
1. Unconditional

L_PP_WaitForData
L_PP_Addr
Wait for data to be received. lunReturnState is set to
L_PP_WaitForData.
1. Target passes data byte or word to LUN
 L_PP_AcceptData
2. Command cycle 10h (program execute) received
 L_PP_Prog
201
3.
4.
5.
6.
Command cycle 15h (cache program) received
Command cycle 11h (interleave) received
Target requests column address be selected
Target requests row address be selected
L_PP_AcceptData
1. Unconditional
L_PP_Prog
1. Unconditional




Write the byte (x8) or word (x16) of data into the selected column
address in the page register. Increments column address.
 L_PP_WaitForData
The LUN performs the following actions in the order specified:
1. lunStatus[6:5] is cleared to 00h. lunStatus[6] value is
indicated to the Target.
2. lunLastConfirm set to 10h.
3. If only one page is specified to be programmed, clear
lunbInterleave to FALSE.
4. LUN begins programming page specified and any previous
pages specified if lunbInterleave is TRUE and concurrent
interleaving is supported.
 L_PP_Prog_Wait
L_PP_Prog_Wait
lunReturnState set to L_PP_Prog_Wait.
1. Write of all requested pages are complete and

lunbInterleave is set to TRUE
2. Write of requested page is complete and lunbInterleave 
is cleared to FALSE
3. Target requests status or status command received

L_PP_Cache
1. Unconditional
L_PP_Cache
L_PP_Mpl
L_PP_ColSelect
L_PP_RowSelect
L_PP_Mpl_Sts
L_PP_Sts
L_Status_Execute
The LUN performs the following actions in the order specified:
1. lunStatus[6:5] is cleared to 00b. lunStatus[6] value is
indicated to the Target.
2. lunLastConfirm set to 15h.
3. Wait for the page register to become available for data input.
4. Start background program operation.
 L_PP_Cache_Wait
L_PP_Cache_Wait
lunReturnState is set to L_PP_Cache_Wait.
1. Page register available for data input
 L_PP_CacheRdy
2. Target requests status or status command received
 L_Status_Execute
L_PP_CacheRdy
The LUN performs the following actions:
1. If lunbInterleave is set to FALSE, then lunStatus[1] is set to
the value of lunStatus[0].
2. If lunbInterleave is set to TRUE, then for all multi-plane
addresses, x, lunFail[x][1] is set to the value of lunFail[x][0].
3. lunStatus[6] is set to one. lunStatus[6] value is indicated to
the Target.
202

1. Unconditional
L_PP_CacheRdy_Wait
L_PP_CacheRdy_Wait
lunReturnState set to L_PP_CacheRdy_Wait.
1. Previous cache operation complete and lunbInterleave  L_PP_Mpl_Cache_Sts
set to TRUE
2. Previous cache operation complete
 L_PP_Cache_Sts
3. Target indicates Program request for this LUN
 L_PP_Addr
4. Target requests page register clear
 L_Idle_ClearPageReg
5. Target requests status or status command received
 L_Status_Execute
L_PP_Mpl
1. Unconditional
The LUN performs the following actions in the order specified:
1. lunbInterleave set to TRUE.
2. lunStatus[6:5] is cleared to 00b. lunStatus[6] value is
indicated to the Target.
3. lunLastConfirm set to 11h.
4. lunbMplNextCmd is set to FALSE.
5. LUN begins programming page specified if overlapped
interleaving is supported.
 L_PP_Mpl_Wait
L_PP_Mpl_Wait
lunReturnState set to L_PP_Mpl_Wait.
1. An overlapped multi-plane Program completed
2. A previous cache Program completed
3. LUN is ready to receive the next Program command and
lunbMplNextCmd is set to FALSE
4. Target indicates Program request for this LUN and
lunbMplNextCmd is set to TRUE
5. Target requests column address be selected
6. Target indicates plane address for use in data output
7. Target requests status or status command received
8. Read request received from Target
L_PP_Mpl_NextCmd
1. Unconditional
L_PP_Sts



L_PP_Mpl_Overlap
L_PP_Mpl_Cache_Sts
L_PP_Mpl_NextCmd

L_PP_Addr


L_Idle_Rd_ColSelect
L_Idle_Mpl_DataOutAdd
r
L_Status_Execute
L_Idle_Rd_Xfer


The LUN performs the following actions in the order specified:
1. lunbMplNextCmd is set to TRUE.
2. If no array operations are in progress, lunStatus[5] is set to
one.
3. lunStatus[6] is set to one. lunStatus[6] value is indicated to
the Target.
 L_PP_Mpl_Wait
The LUN performs the following actions in the order specified:
1. lunStatus[1] is set to program status of previous operation
2. lunStatus[0] is set to program status of final operation
3. lunStatus[6:5] is set to 11b.
4. lunStatus[6] value is indicated to the Target.
203

1. Unconditional
L_PP_Cache_Sts
1. Unconditional
L_PP_Mpl_Cache_Sts
1. Unconditional
L_PP_Mpl_Overlap
1. Unconditional
L_PP_Mpl_Sts
1. Unconditional
L_PP_ColSelect
1. Unconditional
L_PP_RowSelect
1. Unconditional
L_Idle
The LUN performs the following actions in the order specified:
1. lunStatus[0] is set to program status.
2. lunStatus[5] is set to one.
 lunReturnState
The LUN performs the following actions in the order specified for all
completed cache operations:
1. mplAddr set to interleave address of cache operation.
2. lunFail[mplAddr][0] is set to program status.
If all array operations are complete, lunStatus[5] is set to one.
 lunReturnState
The LUN performs the following actions in the order specified for the
overlapped multi-plane operation that completed:
1. mplComplete set to interleave address of completed
operation
2. lunFail[mplComplete][0] is set to program status of operation.
If all array operations are complete, lunStatus[5] is set to one.
 lunReturnState
The LUN performs the following actions in the order specified for each
multi-plane operation that completed:
1. mplComplete set to plane address of completed operation
2. lunFail[mplComplete][1] is set to program status of previous
operation.
3. lunFail[mplComplete][0] is set to program status of final
operation.
lunStatus[6:5] is set to 11b and lunStatus[6] value is indicated to the
Target.
 L_Idle
Select the column in the page register based on the column address
received that the target requested.
 L_PP_WaitForData
Select the block and page to program based on the row address
received from the target.
 L_PP_WaitForData
204
A. SAMPLE CODE FOR CRC-16 (INFORMATIVE)
This section provides an informative implementation of the CRC-16 polynomial. The example is intended as an aid in verifying an implementation
of the algorithm.
int main(int argc, char* argv[])
{
// Bit by bit algorithm without augmented zero bytes
const unsigned long crcinit = 0x4F4E; // Initial CRC value in the shift register
const int order = 16;
// Order of the CRC-16
const unsigned long polynom = 0x8005; // Polynomial
unsigned long i, j, c, bit;
unsigned long crc = crcinit;
// Initialize the shift register with 0x4F4E
unsigned long data_in;
int dataByteCount = 0;
unsigned long crcmask, crchighbit;
crcmask = ((((unsigned long)1<<(order-1))-1)<<1)|1;
crchighbit = (unsigned long)1<<(order-1);
// Input byte stream, one byte at a time, bits processed from MSB to LSB
printf("Input byte value in hex(eg. 0x30):");
printf("\n");
205
while(scanf("%x", &data_in) == 1)
{
c = (unsigned long)data_in;
dataByteCount++;
for (j=0x80; j; j>>=1) {
bit = crc & crchighbit;
crc<<= 1;
if (c & j) bit^= crchighbit;
if (bit) crc^= polynom;
}
crc&= crcmask;
printf("CRC-16 value: 0x%x\n", crc);
}
printf("Final CRC-16 value: 0x%x, total data bytes: %d\n", crc, dataByteCount);
return 0;
}
206
B. SPARE SIZE RECOMMENDATIONS (INFORMATIVE)
This appendix describes recommendations for the spare bytes per page based on the ECC
requirements reported in the parameter page. These recommendations are for raw NAND
implementations and do not apply to devices that support EZ NAND. Table 51 lists
recommendations for 2KB, 4KB, and 8KB page size devices.
Page Size
Number of bits ECC
correctability
Spare Bytes Per Page
Recommendation
2048 bytes
<= 8 bits
64 bytes
2048 bytes
> 8 bits
112 bytes
4096 bytes
<= 8 bits
128 bytes
4096 bytes
> 8 bits
218 or 224 bytes
8192 bytes
2
8192 bytes
<= 8 bits
256 bytes
> 8 bits
448 bytes
NOTE:
1. The number of bits ECC correctability is based on a 512 byte codeword size.
2. If more correction is required than spare area size allows for with a 512 byte
codeword size, it is recommended that the host use a larger ECC codeword size
(e.g. 1KB, 2KB, etc). The device manufacturer may provide guidance on the ECC
codeword size to use in the extended parameter page.
Table 51
Spare Area Size Recommendations for raw NAND
The host transfers bytes from the page register in discrete units that include data, metadata, and
the ECC check bytes. This discrete unit is recommended to be an even number of bytes for
devices that support the source synchronous data interface.
As an example, assume the page size is 8192 bytes and the ECC codeword size used is 1KB.
Then 1024 bytes of data will be transferred in each discrete unit, resulting in eight discrete units
of data being transferred for this page. The spare bytes for this page should be allocated to allow
enough storage for the metadata and check bytes, and should also be an even number when
divided by eight (i.e. the number of discrete units contained in that page).
207
C. DEVICE SELF-INITIALIZATION WITH PSL (INFORMATIVE)
Some devices store configuration information for the Flash array within the Flash array itself. The
device loads this information either at power-on or during the first Reset after power-on.
Vendors may choose to support PSL as one of the vendor specific pins. If PSL is supported, then
it shall have the following behavior:
 PSL = 0 V: Configuration information is loaded at power-on. The IST current may be up
to 15 mA and the time for R/B_n to become one is up to 5 ms.
 PSL = Vcc or not connected: Configuration information if supported is loaded during the
first Reset after power-on. There is no change to the IST current requirement. This
corresponds to the normally expected ONFI device operation.
If PSL is not supported by the device, then the IST requirement shall be met.
Refer to the device vendor’s datasheet to determine if self-initialization at power-on is supported.
208
D. ICC MEASUREMENT METHODOLOGY
This section defines the technique used to measure the ICC parameters defined in section 2.10.
The common testing conditions that shall be used to measure the DC and Operating Conditions
are defined in Table 52. The testing conditions that shall be used to measure the DC and
Operating Conditions that are data interface specific are defined in Table 53.
Parameter
Testing Condition
1.
2.
3.
4.
5.
6.
7.
Vcc = Vcc(min) to Vcc(max)
VccQ = VccQ(min) to VccQ(max)
CE_n = 0 V
WP_n = VccQ
IOUT = 0 mA
Measured across operating temperature range
N data input or data output cycles, where N is the number of bytes or
words in the page
8. No multi-plane operations
9. Sample sufficient number of times to remove measurement variability
10. Sample an equal ratio of page types that exist in a block. A page type
is a group of page addresses and is commonly referred to as upper or
lower page (or middle page for 3 bits per cell devices).
11. Choose the first good even/odd block pair beginning at blocks 2-3
General conditions
Array preconditioning for
ICC1 and ICC3
The array is preconditioned to match the data input pattern for ICC2.
Fixed wait time
(no R/B_n polling)
ICC1: tR = tR(max)
ICC2: tPROG = tPROG(max)
ICC3: tBERS = tBERS(max)
Table 52
Parameter
AC Timing Parameters
Bus idle data pattern
Repeated data pattern
(Used for ICC2 and ICC4W )
Array preconditioning for
ICC4R
Common Testing Conditions for ICC
Asynchronous
tWC = tWC(min)
tRC = tRC(min)
tADL = ~tADL(min)
tCCS = ~tCCS(min)
tRHW = ~tRHW(min)
IO[7:0] = FFh
IO[15:0] = FFFFh
Source Synchronous
tCK = tCK(avg)
tADL = ~tADL(min)
tCCS = ~tCCS(min)
tRHW = ~tRHW(min)
DQ[7:0] = FFh
IO[7:0] = A5h, AAh, 5Ah, 55h
IO[15:0] = A5A5h, AAAAh, 5A5Ah, 5555h
DQ[7:0] = A5h, AAh, 5Ah, 55h
The array is preconditioned to match the
following repeating data pattern:
IO[7:0] = A5h
IO[15:8] = A5A5h
The array is preconditioned to match the
following repeating data pattern:
DQ[7:0] = A5h
NOTE:
1. The value of tCK(avg) used should be the minimum tCK(avg) of the timing modes supported for the device. The
source synchronous timing modes supported by the device are indicated in the parameter page.
Table 53
Data Interface Specific Testing Conditions for ICC
209
The following figures detail the testing procedure for ICC1, ICC2, ICC3, ICC4R, ICC4W, and
ICC5.
210
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
IDLE
30h
FFh
Minimum cycle time between command, address cycles
DQx
00h
00h
00h
R2
R1
R3
tWB
tR(max)
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 94
ICC1 measurement procedure
To calculate the active current for ICC1, the following equations may be used.
(
(
(
(
)
)
211
)
)
(
(
)
(
(
)
(
)
)
(
)
(
)
(
(
)
)
)
Cycle Type
ADDR
CMD
ADDR
ADDR
ADDR
IDLE
ADDR
Minimum cycle time between command, address cycles
DQx
00h
80h
00h
R2
R1
DIN
DIN
DIN
Cycle(min)
tADL
FFh
R3
IDLE
CMD
D0
Dn
...
FFh
10h
tWB
tPROG(max)
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 95
ICC2 measurement procedure
To calculate the active current for ICC2, the following equations may be used.
(
(
)
(
(
)
(
)
)
(
(
(
(
(
)
(
For the source synchronous data interface, the tIO value is calculated as:
)
(
(
(
(
(
)
)]
)
For the asynchronous interface, the tIO value is calculated as:
212
)
)
(
)
))
(
(
)
)
)
(
)
)
Cycle Type
ADDR
CMD
ADDR
ADDR
CMD
IDLE
D0h
FFh
Min time between command, address cycles
DQ[7:0]
R1
60h
R2
R3
tWB
tBERS(max)
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 96
ICC3 measurement procedure
To calculate the active current for ICC3, the following equations may be used.
(
(
(
)
(
)
213
(
)
)
(
)
(
(
(
)
)
)
(
(
)
)
(
(
)
)
)
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
CMD
IDLE
30h
FFh
Minimum cycle time between command, address cycles
DQx
00h
00h
00h
R1
R2
R3
tWB
tR(max)
R/B_n
A
Cycle Type
CMD
ADDR
CMD
ADDR
Min time between
command, address cycles
DQx
05h
00h
00h
IDLE
DOUT
DOUT
DOUT
tCCS
E0h
FFh
tRHW
D0
...
Dn
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 97
214
IDLE
ICC4R measurement procedure
FFh
A
Cycle Type
CMD
ADDR
ADDR
ADDR
ADDR
ADDR
IDLE
Minimum cycle time between command, address cycles
DQx
80h
00h
00h
CMD
ADDR
ADDR
R1
R2
FFh
R3
R/B_n
A
Cycle Type
Min time between
cmd, addr cycles
DQx
85h
00h
IDLE
DIN
DIN
Cycle(min)
tCCS
00h
FFh
DIN
D0
...
Dn
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 98
215
ICC4W measurement procedure
Cycle Type
IDLE
DQ[7:0]
FFh
R/B_n
Repeat to gather sufficient ICC samples to remove measurement variability
Figure 99
216
ICC5 measurement procedure
E. MEASURING TIMING PARAMETERS TO/FROM TRI-STATE
There are several timing parameters that are measured to or from when:
 The device is no longer driving the NAND bus or a tri-state (hi-Z) condition
 The device begins driving from a tri-state (hi-Z) condition
These timing parameters include: tDQSD, tDQSHZ, tCHZ, tRHZ, and tIR. See section 4.2.
This appendix defines a two point method for measuring timing parameters that involve a tri-state
condition. Figure 100 defines a method to calculate the point when the device is no longer driving
the NAND bus or begins driving by measuring the signal at two different voltages. The voltage
measurement points are acceptable across a wide range (x = 20 mV up to x < 1/4 of VccQ). The
figure uses tDQSHZ and tDQSD as examples. However, the method should be used for any
timing parameter (asynchronous or source synchronous) that specifies that the device output is
no longer driving the NAND bus or specifies that the device begins driving the NAND bus from a
tri-state condition.
Figure 100
Two point method for measuring timing parameters with tri-state condition
217
F. EZ NAND: END TO END DATA PATH PROTECTION
(INFORMATIVE)
An EZ NAND implementation may choose to support end to end data protection. In certain
topologies with longer trace lengths or discontinuities, it may be helpful to include this capability.
However, it is not expected that most EZ NAND implementations need this type of protection at
speeds up to 200 MT/s.
One potential implementation is to include a two to four byte CRC for each data unit in the page.
The data unit is called the CRC sub-page, which is the individual unit that is protected. The
capability of end to end data protection should be reported in a vendor specific parameter page
bit. The feature may then be configured utilizing a vendor specific Set Features. If there is an
error detected as part of end to end data path protection, the error should be reported in the FAIL
bit as part of the associated Read or Program operation.
Included below is a recommended method for configuring the CRC capabilities as part Set
Features. The CRC polynomial to use is not specified.
Sub Feature
Parameter
7
P1
6
5
4
Reserved (0)
P2
P3
P4
3
2
CRC Mode
1
0
CRC Poly
Length
# of CRC sub-pages
Offset of CRC bytes (bits 31:00)
Offset of CRC bytes (bits 63:32)
CRC Poly Length
CRC polynomial length in bytes. Valid values are 2, 3 or 4.
CRC Mode
00b = no CRC check or generation
01b = CRC check on Program/store to Flash; CRC check on
read from Flash
10b = CRC check and discard on Program/store to Flash; regenerate CRC on read from Flash
11b = Reserved
# of CRC sub-pages
Specifies the number of CRC sub-pages within a page. The
CRC sub-page size is the page size divided by the number of
CRC sub-pages.
Offset of CRC bytes
Specifies the offset of the CRC bytes within each CRC sub-page.
E.g., a value of 0h indicates that the CRC is stored at the
beginning of the CRC sub-page.
218