APPARATUSES AND METHODS FOR SUBROW ADDRESSING
Systems, apparatuses, and methods related to subrow addressing for electronic memory and/or storage are described. Independent subrow addressing may enable energy consumed by performance of an operation on a particular subset of data values stored by a row to more closely correspond to the size of the particular subset of data values relative to energy consumed by addressing and activating the complete row. For instance, one such apparatus includes a plurality of subrows within a row of memory cells and a controller configured to selectably address and manage an activation state of each subrow of the plurality of subrows. The apparatus further includes subrow driver circuitry coupled to the controller. The subrow driver circuitry is configured to maintain one or more subrows of the plurality in the activation state based at least in part on signaling from the controller.
The present disclosure relates generally to semiconductor memory and methods, and more particularly, to apparatuses and methods for subrow addressing.
BACKGROUNDMemory devices are typically provided as internal, semiconductor, integrated circuits in computers or other electronic systems. There are many different types of memory including volatile and non-volatile memory. Volatile memory can require power to maintain its data (e.g., host data, error data, etc.) and includes random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), synchronous dynamic random access memory (SDRAM), and thyristor random access memory (TRAM), among others. Non -volatile memory can provide persistent data by retaining stored data when not powered and can include NAND flash memory, NOR flash memory, and resistance variable memory such as phase change random access memory (PCRAM), resistive random access memory (RRAM), and magnetoresistive random access memory (MRAM), such as spin torque transfer random access memory (STT RAM), among others.
Electronic systems often include a number of processing resources (e.g., one or more processors), which may retrieve and execute instructions and store the results of the executed instructions to a suitable location. A processor can include a number of functional units such as arithmetic logic unit (ALU) circuitry, floating point unit (FPU) circuitry, and a combinatorial logic block, for example, which can be used to execute instructions by performing logical operations such as AND, OR, NOT, NAND, NOR, and XOR, and invert (e.g., inversion) logical operations on data (e.g., one or more operands). For example, functional unit circuitry may be used to perform arithmetic operations such as addition, subtraction, multiplication, and division on operands via a number of operations. In many instances, addressing protocols may activate a complete row of memory cells and the data values from all of the memory cells in the row may be accessed regardless of how much of the data is intended to be used in performing an operation by a processing resource.
The present disclosure includes systems, apparatuses, and methods related to subrow addressing in a row of a memory array. Independent subrow addressing in accordance with embodiments described herein may provide various benefits, such as reducing energy consumption compared to previous approaches to array operation. For example, a number of embodiments may enable energy consumed by performance of an operation on a particular subset of data values stored by the row to more closely correspond to the size of the particular subset of data values relative to energy consumed by addressing and activating the complete row. For instance, one such apparatus includes a plurality of subrows within a row of memory cells and a controller configured to selectably address and manage an activation state of each subrow of the plurality of subrows. The apparatus further includes subrow driver circuitry coupled to the controller. The subrow driver circuitry is configured to maintain one or more subrows of the plurality in the activation state based at least in part on signaling from the controller.
Implementations of address circuitry (e.g., for a DRAM configuration conforming to standards and protocols such as JEDEC, DDR3, DDR4, etc.) for a memory device may be configured to activate a complete row of memory cells at a time (e.g., per write/read cycle), for instance, within a subarray including a plurality of rows of memory cells. Such an implementation may further access the data values from all of the memory cells in the row via sensing circuity (e.g., read/latch circuitry, as described herein). A host and/or a controller for the memory device may then access the data values stored by the sensing circuitry to enable performance of an operation by a processing resource on the stored data values. However, the operation may be performed, for instance, on only a portion of the data values stored by the sensing circuitry (e.g., 128 data values from among 16,384 data values).
Accessing all the memory cells and storing all of the data values from the complete row, regardless of the amount of data to which the operation is directed, may result in various drawbacks. For instance, at least half of the energy (e.g., electrical power) used by a memory device may result from performance of accessing all the memory cells and storing all of the data values from complete rows to enable performance of operations thereon.
In contrast, one or more subrows may be specifically addressed at positions within the row that include the data values upon which the operation is intended to be performed. The subrows described herein are intended to mean separate portions, which may be sequential, that each includes different memory cells of the complete row. These subrows may be specifically addressed by inhibiting activation of a number of other subrows at particular positions in the row at which data values are stored that are not to be utilized in performance of the operation. Inhibiting activation of a number of subrows may notably reduce energy consumption by the memory device.
For example, in an embodiment in which a row includes 16,384 (16K) memory cells addressed as sixteen subrows each including 1024 memory (1K) cells, addressing and activating only one 1K subrow to access the data values stored therein (e.g., the data values intended for use in performance of an operation) may reduce the energy consumption to one sixteenth ( 1/16) of that potentially used for addressing and activation of all 16 of the 1K subrows in combination with storage by the sensing circuitry of data values accessed therefrom. Hence, a number of embodiments of the present disclosure may provide various benefits (e.g., a technical advantage) by, for example, independent subrow addressing enabling energy consumed by performance of an operation on a particular subset of data values stored by a row to more closely correspond to the size of the particular subset of data values relative to energy consumed by addressing and activating the complete row.
The embodiments of 16K memory cells per row and/or being divided into sixteen 1K subrows are presented by way of example and not limitation. For example, a row may include 1024, 2048, 4,096, 8192, or 16,384 memory cells, among other possibilities, and each row may include a plurality of portions that divide the memory cells into 2, 4, 8, 16, or 32 subrows, among other possibilities.
The figures herein follow a numbering convention in which the first digit or digits of a reference number correspond to the figure number and the remaining digits identify an element or component in the figure. Similar elements or components between different figures may be identified by the use of similar digits. For example, 104 may reference element “04” in
The mask register 104 may selectably enable inactivation of a subrow at a particular position (e.g., subrow 108-0) in the row 107 by inhibiting (e.g., overriding, cancelling, etc.) a signal from the global row driver 101 to activate subrow driver circuitry (e.g., subrow drivers shown at 105-0, 105-1, . . . , 105-N-1 in
A plurality of subrows in a particular row may, in a number of embodiments, each have their activation inhibited by entries corresponding to the particular subrows being latched (e.g., stored) by the mask register 104. Such entries may be selected by a host (e.g., as shown at 211 and described in connection with
The boxes illustrated at 108-0, 108-1, . . . , 108-N-1 in
A read/latch stripe 124 shown in
The connections between the various elements of the circuitry 100 in
The circuitry 100 for subrow addressing illustrated in
In some embodiments, the subrow driver circuitry described herein (e.g., as shown at 205) may be formed as a multiplexer configured to selectably control activation of the different subrows in a row and/or a subarray. As such, in a number of embodiments, the row and/or the subarray may be formed without subrow drivers and subrow decoders positioned between the subrows.
As described herein, various embodiments may allow a computing system 210 to allocate a number of locations (e.g., subarrays) in a bank to store (e.g., hold) data. An embodiment of a bank and/or a section of a bank that may include a plurality of subarrays is shown at 221 and 323 and described in connection with
In a number of embodiments, a row (e.g., as shown at 107 in
A number of embodiments of the present disclosure include read/latch circuitry (e.g., sense amplifiers 506 and associated circuitry), which may be formed on pitch with sense lines of an array of memory cells. The read/latch circuitry and other data storage components described herein are capable of performing data sensing and/or storage (e.g., caching, latching, buffering etc.) of data local to the array of memory cells.
In order to appreciate the improved subrow addressing techniques described herein, a discussion of an apparatus for implementing such techniques (e.g., a memory device 220 having these capabilities and an associated host 211) follows. According to various embodiments, program instructions (e.g., commands) involving a memory device having the subrow addressing capabilities described herein may distribute implementation of the commands (e.g., signals) and data over multiple read/latch and subrow addressing circuitries that may implement operations and may move and store the commands and data within the memory array (e.g., without having to transfer such back and forth over a bus between a host and the memory device). Thus, data for a memory device having the subrow addressing capabilities described herein may be accessed and used in less time, along with using less power. For example, a time and power advantage may be realized by increasing the speed, rate, and/or efficiency of data being accessed only from a number subrows and stored in read/latch circuitry (e.g., sensing circuitry) in order to enable data processing for requested memory operations (e.g., reads, writes, logical operations, etc.) to be performed only on data values from subrows in which data values intended for use in the operations are stored.
The system 210 may include host 211 coupled to memory device 220, which includes the memory array 230 and the controller 240, among the various circuitry for subrow addressing, as shown and described herein. Host 211 may be responsible for execution of an operating system (OS) and/or various applications that may be loaded thereto (e.g., from memory device 220 via controller 240). Host 211 may include a system motherboard and backplane and may include a number of processing resources (e.g., one or more processors 272, microprocessors, or some other type of controlling circuitry) capable of accessing the memory device 220 (e.g., via controller 240) to perform operations on data values moved from the memory device 220 (e.g., using subrow addressing signals provided via controller 240). Controller 240 also may, in a number of embodiments, include a number of processing resources for performance of processing operations. The system 210 that may include separate integrated circuits or both the host 211 and the memory device 220 may be on the same integrated circuit. The system 210 may, for instance, be a server system and a high performance computing (HPC) system or a portion thereof. Although the example shown in
The controller 240 (e.g., bank control logic and sequencer) may include control circuitry, in the form of hardware, firmware, or software, or combinations thereof. As an example, the controller 240 may include a state machine, a sequencer, and/or some other types of control circuitry, which may be implemented in the form of an application specific integrated circuit (ASIC) coupled to a printed circuit board. In a number of embodiments, the controller 240 may be co-located with the host 211 (e.g., in a system-on-chip (SOC) configuration).
For clarity, description of the system 210 has been simplified to focus on features with particular relevance to the present disclosure. For example, the array 230 may be a DRAM array, SRAM array, STT RAM array, PCRAM array, TRAM array, RRAM array, FeRAM array, phase-change memory array, 3D XPoint™ array, NAND flash array, and/or NOR flash array, for instance. The array 230 may include memory cells arranged in rows (e.g., in a plurality of subarrays) coupled by access lines (which may be referred to herein as word lines or select lines) and columns coupled by sense lines (which may be referred to herein as data lines or digit lines). Although a single bank 221 and a single memory array 230 are shown in
The memory device 220 may include address circuitry 242 to latch address signals provided over a data bus 256 (e.g., an I/O bus from host 211) by I/O circuitry 244 (e.g., provided to external ALU circuitry and to DRAM DQs via local I/O lines and global I/O lines). Status and exception information may be provided from the controller 240 of the memory device 220 to a channel controller 243, for example, through a control bus 254, which in turn may be provided from the channel controller 243 to host 211. Address signals may be received (e.g., from channel controller 243 or another host component) through address circuitry 242 and may be decoded (e.g., via a subarray decoder and/or a row decoder in the address circuitry 242) and/or a column decoder 249 coupled to the read latch circuitry 250 to access the memory array 130. A global row driver 201 configured to activate a complete row (e.g., as shown at 107 in
Data may be sensed (read) from memory array 230 by sensing voltage and/or current changes on sense lines (digit lines) using sensing circuitry (e.g., shown as read/latch circuitry 250 in
The controller 240 may decode signals (e.g., commands) provided by control bus 254 from host 211. The controller 240 may control operations by issuing signals determined from the decoded commands from host 211. These signals may include chip enable signals, write enable signals, address signals (e.g., subarray address signals, row address signals, and/or subrow address signals), and/or mode signals that may be used to control operations performed on the memory array 230, including data sense, data store, subarray addressing, row addressing, subrow addressing, data move, data write, and data erase operations, among other operations. In various embodiments, the controller 240 may be responsible for executing instructions from host 211 and accessing the memory array 230.
The controller 240 may, in various embodiments, include a mode register 238 configured to enable selection between all of the plurality of memory cells being activated in a row being addressed (e.g., responsive to signals from the global row driver 201) and activation of selected subrows of the plurality of memory cells in the row being addressed (e.g., responsive to latched subrow entries in the mask register 204). In a number of embodiments, all of the memory cells being activated in the row being addressed may be a default mode selection. Selection between the modes in the mode register 238 may be performed responsive to mode signals provided by host 211 though control bus 254.
In the default mode, the controller 240 may direct that signals be sent via the global row driver 201 to all the subrow drivers of the row (e.g., subrow drivers 105-0, 105-2, . . . , 105-N-1) to activate all of the corresponding subrows (e.g., subrows 108-1, 108-2, . . . , 108-N-1 corresponding to row 107 in
In a number of embodiments, a memory device (e.g., as shown at 220 and described in connection with
As described herein, the activation state of a subrow may include the subrow being in an active state or in an inactive state. The active state is intended to at least mean the subrow, and the memory cells therein, being accessible by the sensing circuitry (e.g., shown as read/latch circuitry 250 in
For example, in a number of embodiments, the controller 240 may be configured to selectably address and to direct that a subrow at a particular position in the row (e.g., at the position of any one or more of subrows 108-0, 108-1, . . . , 108-N-1 in row 107) remains inactive. To selectably address may, in a number of embodiments, be to use a row address corresponding to a particular row of a plurality of rows and the subrow at the particular position in the row may be directed to remain inactive via an indicator of the subrow (e.g., included in and/or accompanying the row address) including a signal to direct that the subrow remain inactive (e.g., as opposed to another signal to direct that the subrow be activated). The controller may be coupled to the subrow driver circuitry 105, 205 configured to inhibit activation of the subrow at the particular position.
As described herein, a mask register (e.g., as shown at 104 and 204 and described in connection with
As described herein, a global row driver (e.g., as shown at 101 and 201 and described in connection with
The mask register 204 may be configured to selectably latch a number of selections (e.g., provided by host 211 and/or a user) from among the plurality of subrows to remain in a second activation state (e.g., inactive). The latched second activation state may inhibit the first activation state via the first and second subrow drivers being deselected (e.g., inactivated) responsive to the latched number of selections.
The subrow driver circuitry 205 may include a subrow driver coupled to the one or more subrows at a particular position. For example, in a number of embodiments, subrow driver 105-1 may only be coupled to subrow 108-1. In a number of other embodiments, subrow driver 105-1 may be coupled both to subrow 108-1 and subrow 108-0 and/or subrow 108-2, among other possible combinations and/or numbers of subrows being coupled to a given subrow driver. Each of the subrow drivers may be configured to control activation of the one or more subrows. The subrow driver circuitry 205 may further include a subrow decoder (e.g., as shown at 427 and described in connection with
The subrow driver circuitry 205 may include a first subrow driver (e.g., subrow driver 105-0) coupled to one of the subrows (e.g., 108-0) at a particular position within the row 107 and a second subrow driver (e.g., any of subrow drivers 105-1, 105-2, . . . , 105-N-1) coupled to another one of the corresponding subrows (e.g., any of subrows 108-1, 108-2, . . . , 108-N-1) at a different position within the row 107. The first and second subrow drivers may each be configured to inhibit activation of the corresponding subrow responsive to signaling from the controller 240.
The controller 240 may be configured to selectably address and direct that the subrow at the particular position (e.g., subrow 108-0) remains inactive responsive to signaling that inhibits activation of the first subrow driver (e.g., subrow driver 105-0). In combination, the controller 240 may be configured to selectably address and direct that the subrow at the different position (e.g., any of subrows 108-1, 108-2, . . . , 108-N-1) remains inactive responsive to signaling that inhibits activation of the second subrow driver (e.g., any of subrow drivers 105-1, 105-2, . . . , 105-N-1). Accordingly, the first and second subrow drivers (e.g., all of subrow drivers 105-0, 105-1, . . . , 105-N-1) are each configured to be inactivated responsive to signaling from the controller.
As illustrated in
The bank 221 or bank section 323 may include 64 subarrays, 128 subarrays, 256 subarrays, 512 subarrays, among various other possible numbers of subarrays. However, embodiments are not so limited, such that some embodiments of a bank may have a different number of subarrays than just presented. In a number of embodiments, the subarrays may have the same number of rows in each subarray (e.g., 256 rows, 512 rows, 1024 rows, 2048 rows, among various other possible numbers of rows). However, embodiments are not so limited, such that at least some of a plurality of subarrays within the bank or bank section may have different numbers of rows.
Each column 322 is configured to be coupled to read/latch circuitry 250 (e.g., as further described in connection with
Each of the of the subarrays 325-0, 325-1, . . . , 325-N-1 may include a respective plurality of rows (e.g., a respective group of “Y” rows 319) within a memory array (e.g., as shown at 230 and 530 and described in connection with
A number of the plurality of subrow drivers 305 may be positioned between subrows 308 of each of a plurality of rows 319. Each subrow driver 305 may be coupled to one subrow 308 of the plurality of subrows in each of the plurality of rows 319. A predetermined number of memory cells (e.g., 128, 256, 512, 1024, or 2048 memory cells) may separate the number of the subrow drivers for each of the plurality of subrows.
Controller 340 shown in
The controller 340 may be configured to provide a row activation signal to the row (e.g., via global row driver 201) and inhibit activation of a selected one of the plurality subrows 308 responsive to the row activation signal via a signal received by a subrow driver 305 corresponding to the selected one of the plurality of subrows. The controller may address a subrow driver coupled to a subrow at a particular position using a signal for selective activation (e.g., activation or inactivation) of the subrow driver. A subrow decoder (e.g., coupled to the addressed subrow driver as part of the subrow driver circuitry 205) may be configured to decode the signal to select or deselect a gate for activation of the subrow driver. For example, decoding the signal and selecting the gate may result in activation of the coupled subrow driver, whereas decoding the signal and deselecting the gate may result in inactivation of the coupled subrow driver (e.g., as described further in connection with
A mask register (e.g., as shown at 104, 204, and 304 and described in connection with
As described herein, a memory device (e.g., as shown at 220 and described in connection with
Access to a plurality of memory cells of the first subrow by sense amplifiers coupled to the plurality of memory cells (e.g., as described in connection with
Accordingly, responsive to inhibiting activation of a number of subrows, access to the number of subrows, and the plurality of memory cells therein, may be prevented and storage of a number of data values corresponding to the number of subrows, and the plurality of memory cells therein, also may be prevented. By inhibiting the activation of the subrows, and consequently preventing access to the subrows and storage of the data values, an amount of energy used by (e.g., electrical power to enable operation of) the memory device may be reduced. The reduced amount of energy that is used may correspond to the number of subrows being inhibited from activation relative to energy used for activation of all of the plurality of subrows within the row in a different operating mode of the memory device (e.g., the default mode, as described herein).
All of the subrows being activated within a row that is addressed is consistent with implementations conforming to standards and protocols (e.g., JEDEC, DDR3, DDR4, etc.) that activate a complete row of memory cells at a time. However, activating all of the subrows, accessing all of the subrows, and storing all of the data values accessed therefrom may use an amount of energy that may be reduced (e.g., is unnecessary) for access to, and performance of operations on, data values stored only in a subset of the subrows in the row. Hence, the independent subrow addressing described herein (e.g., utilizing the mask register, mode register, subrow addressing circuitry, and signaling pathways, etc.) may provide a technical advantage by enabling energy consumed by performance of an operation on a particular subset of data values stored by a number of subrows of a row to more closely correspond to the size of the particular subset of data values.
Whereas each subrow driver 105-0, 105-1, . . . , 105-N-1 shown in
The controller 340 may be configured, as shown at 341, to provide data to (e.g., as received from host 211) the bank 221, section of the bank 323, rows 319, and/or subrows 308 and/or to retrieve and/or access data therefrom. The section of the bank 323 illustrated in
The global row driver 401 may, in some embodiments, be coupled to a number of global row lines 426 that are individually coupled to every row/subrow to activate one row/subrow at a time. Each global row line 426 may be configured to carry an activation signal (e.g., having a value of 1 in binary) sent from the global row driver 401 to a selected row and the subrow driver (e.g., as shown at 405-0, 405-1, etc.) of the selected subrow. The activation signal may be sent, for example, to a number of gates 428 (e.g., logic gates configured to perform a Boolean AND function based upon input of two binary values) in the subrow drivers 405-0, 405-1 corresponding (e.g., coupled to) each of the respective subrows 408-0, 408-1.
In a number of embodiments, as shown in
For example, each subrow decoder 427 for each subrow driver 405 may be configured to allow selection of any of the subrows that would be potentially activated via the signal from the global row driver 401. The selection may be made consistent with the entries stored in the mask register 304. Using appropriately configured circuitry coupled to, for example, the gate 428 (e.g., the AND logic gate), a signal may be sent to the subrow driver 405 to selectably disable activation of (e.g., deselect) a particular subrow that would be activated via the activation signal from the global row driver 401. For example, an activation signal with a binary value of 1 may be sent from the global row driver 401 to the AND logic gate and a signal (e.g., a first signal with a binary value of 1) may be sent from the subrow decoder 427 as a second input to the AND logic gate to deselect activation of a particular subrow 408 of the row potentially activated via the activation signal from the global row driver 401.
In a number of embodiments, when there is no entry stored in the mask register 304 for the particular subrow, a signal (e.g., a second signal with a binary value of 0) may be sent from the subrow decoder 427 as a second input to the AND logic gate in order to select activation of the particular subrow 408 of the row in combination with the activation signal from the global row driver 401. In some embodiments, when there is no entry stored in the mask register 304 for the particular subrow, a signal may not be sent from the subrow decoder 427 as a second input to the AND logic gate. In such an instance, a default signal with a binary value of 0 may be input to the AND logic gate to select for activation of the particular subrow 408 of the row in combination with the activation signal from the global row driver 401.
In some embodiments, each global row line 426 from the global row driver 401 may be coupled to a plurality (e.g., 2, 4, 6, 16, etc.) of rows, along with the corresponding subrows (e.g., to overcome potential crowding of and/or limited area for the global row lines 426). For example, each global row line 426 may be coupled to a sequence of four subrows among a plurality of subrows (e.g., 512 subrows above and/or below the other) in each of 16 mats. An activation signal from the global row driver 401 may then potentially activate all four subrows coupled to a particular global row line 426, via the appropriate subrow driver 405, in each of the 16 mats. In such an embodiment, circuitry of the subrow decoder 427 may be configured to determine and direct the appropriately disabled activation, via the subrow driver 405, of one or more of the potentially activated subrows (e.g., based on the entries stored in the mask register 304).
In the embodiment illustrated in
Memory cells may be coupled to different sense lines and/or access lines. For example, a first source/drain region of a transistor 502-1 may be coupled to sense line 509-1, a second source/drain region of transistor 502-1 may be coupled to capacitor 503-1, and a gate of a transistor 502-1 may be coupled to access line 512-Y. A first source/drain region of transistor 502-2 may be coupled to sense line 509-2, a second source/drain region of transistor 502-2 may be coupled to capacitor 503-2, and a gate of a transistor 502-2 may be coupled to access line 512-X. The cell plate, as shown in
As described herein, the transistors 502 and capacitors 503 may contribute to formation of the pairs of complementary memory cells in a single row of the memory array 530 that are coupled to the complementary sense lines (e.g., sense lines 509-1 and 509-2). The number of data values (e.g., voltages) sensed from the memory cells (e.g., in read operations) may correspond to the number of columns of memory cells and/or pairs of sense lines (e.g., 4,096, 8,192, 16,384, etc.) that intersect a row, for example, of a subarray 325 shown in and described in connection with
The memory array 530 illustrated in
A plurality of sense amplifiers (e.g., as shown at 506) may be coupled to the respective plurality of columns (e.g., as shown at 322 and described in connection with
At block 661, the method 660 may include receiving a first signal by a first subrow driver coupled to a first subrow of a plurality of subrows within a row of memory cells. The operations of block 661 may, in a number of embodiments, be performed by a subrow driver 105, 305, or 405 (e.g., as described in connection with
At block 662, the method 660 may include deselecting activation of the first subrow driver to inhibit activation of the first subrow in response to receiving the first signal. The operations of block 662 may, in a number of embodiments, be performed (e.g., as described in connection with
At block 663, the method 660 may include receiving a second signal by a second subrow driver (e.g., subrow driver 405-1) coupled to a second subrow (e.g., subrow 408-1) of the plurality of subrows. The second signal may be sent by the global row driver 401 to the second subrow driver.
At block 664, the method 660 may include selecting activation of the second subrow driver 405-1 to activate the second subrow 408-1 in response to receiving the second signal. In a number of embodiments, (e.g., in the absence of the first signal to inhibit activation of the second subrow 408-1), the second signal to a gate 428 of the second subrow driver 405-1 may select activation of the second subrow 408-1. Activation of the second subrow driver 405-1 may thus be selected to activate the second subrow 408-1 in response to receiving the second signal.
Ordinal positioning, as used herein, is used to distinguish between relative positions of elements within respective groups of elements. For example, rows of memory cells may each include a sequence of 16 subrows (e.g., subrow 0 through subrow 15). In this example, subrow 0 from a particular row (e.g., a first subrow of the particular row) has a different ordinal position than any of subrows 1 through 15 (e.g., a last subrow) of the row. However, use herein of ordinal numbers such as “first” and “second” is not intended to indicate a particular ordinal position of an element, unless the context clearly dictates otherwise. For example, consider a subrow having an ordinal position of subrow 0 within a particular row and a different subrow having an ordinal position of subrow 4. In this example, subrow 0 might be referred to as a “first” subrow and subrow 4 might be referred to as a “second” subrow, despite not having an ordinal position of subrow 2. Alternatively, subrow 4 might be referred to as a “first” subrow and subrow 0 might be referred to as a “second” subrow.
In the above detailed description of the present disclosure, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration how one or more embodiments of the disclosure may be practiced. These embodiments are described in sufficient detail to enable those of ordinary skill in the art to practice the embodiments of this disclosure, and it is to be understood that other embodiments may be utilized and that process, electrical, and structural changes may be made without departing from the scope of the present disclosure.
As used herein, designators such as “X”, “Y”, “N”, “M”, etc., particularly with respect to reference numerals in the drawings, indicate that a number of the particular feature so designated may be included. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting. As used herein, the singular forms “a”, “an”, and “the” include singular and plural referents, unless the context clearly dictates otherwise, as do “a number of”, “at least one”, and “one or more” (e.g., a number of memory arrays may refer to one or more memory arrays), whereas a “plurality of” is intended to refer to more than one of such things. Furthermore, the words “can” and “may” are used throughout this application in a permissive sense (i.e., having the potential to, being able to), not in a mandatory sense (i.e., must). The term “include,” and derivations thereof, means “including, but not limited to”. The terms “coupled” and “coupling” mean to be directly or indirectly connected physically for access to and/or for movement (transmission) of instructions (e.g., control signals, address signals, etc.) and data, as appropriate to the context. The terms “data” and “data values” are used interchangeably herein and may have the same meaning, as appropriate to the context (e.g., one or more data units or “bits”).
While example embodiments including various combinations and configurations of read/latch circuitry, sense amplifiers, read/latch stripes, subrow driver circuitry, subrow drivers, subrow decoders, mask registers, mode registers, and/or multiplexers, among other circuitry for subrow addressing shown and described herein, have been illustrated and described herein, embodiments of the present disclosure are not limited to those combinations explicitly recited herein. Other combinations and configurations of the read/latch circuitry, sense amplifiers, read/latch stripes, subrow driver circuitry, subrow drivers, subrow decoders, mask registers, mode registers, and/or multiplexers, among other circuitry for subrow addressing, disclosed herein are expressly included within the scope of this disclosure.
Although specific embodiments have been illustrated and described herein, those of ordinary skill in the art will appreciate that an arrangement calculated to achieve the same results may be substituted for the specific embodiments shown. This disclosure is intended to cover adaptations or variations of one or more embodiments of the present disclosure. It is to be understood that the above description has been made in an illustrative fashion, and not a restrictive one. Combination of the above embodiments, and other embodiments not specifically described herein will be apparent to those of skill in the art upon reviewing the above description. The scope of the one or more embodiments of the present disclosure includes other applications in which the above structures and processes are used. Therefore, the scope of one or more embodiments of the present disclosure should be determined with reference to the appended claims, along with the full range of equivalents to which such claims are entitled.
In the foregoing Detailed Description, some features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the disclosed embodiments of the present disclosure have to use more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive subject matter lies in less than all features of a single disclosed embodiment. Thus, the following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment.
Claims
1. An apparatus, comprising:
- a plurality of subrows within a row of memory cells;
- a controller configured to selectably address and manage an activation state of each subrow of the plurality;
- subrow driver circuitry coupled to the controller, the subrow driver circuitry configured to maintain one or more subrows of the plurality in the activation state based at least in part on signaling from the controller; and
- a mask register configured to enable management of the activation state of the plurality of subrows;
- wherein the controller is further configured to manage the activation state of the plurality of subrows based at least in part on enablement by the mask register.
2. The apparatus of claim 1, wherein the activation state of a subrow comprises an active state or an inactive state.
3. (canceled)
4. The apparatus of claim 1, further comprising:
- a global row driver configured to maintain, in a first activation state, a first subrow of the plurality via a first subrow driver and maintain, in the first activation state, a second subrow of the plurality via a second subrow driver;
- a mask register configured to: selectably latch a number of selections from among the plurality of subrows to remain in a second activation state; and inhibit the first activation state via the first and second subrow drivers responsive to the latched number of selections.
5. The apparatus of claim 1, wherein the subrow driver circuitry comprises:
- a subrow driver coupled to the one or more subrows at a particular position, the subrow driver configured to control activation of the one or more subrows; and
- a subrow decoder coupled to the one or more subrows at the particular position, the subrow decoder configured to decode a signal from the controller based at least in part on the activation state of the one or more subrows.
6. The apparatus of claim 1, wherein the subrow driver circuitry comprises:
- a first subrow driver coupled to one of the subrows at a particular position within the row; and
- a second subrow driver coupled to another one of the subrows at a different position within the row;
- wherein the first and second subrow drivers are each configured to inhibit activation of the corresponding subrow responsive to signaling from the controller.
7. The apparatus of claim 6, wherein:
- the controller is further configured, in combination, to selectably address and direct that: the subrow at the particular position remains inactive responsive to signaling that inhibits activation of the first subrow driver; and the subrow at the different position remains inactive responsive to signaling that inhibits activation of the second subrow driver.
8. The apparatus of claim 6, wherein the first and second subrow drivers are each configured to be inactivated responsive to signaling from the controller.
9. The apparatus of claim 1, wherein the subrow driver circuitry comprises:
- a plurality of subrow drivers, and wherein: each subrow driver of the plurality is coupled to a respective one of the plurality of subrows; the plurality of subrow drivers are positioned between a number of the plurality of subrows; and the number of the plurality of subrows each comprises a predetermined number of memory cells.
10. An apparatus, comprising:
- a row of memory cells within an array, the row comprising a plurality of subrows;
- a plurality of subrow drivers corresponding to respective ones of the plurality of subrows;
- a controller configured to: provide a row activation signal to the row; and inhibit activation of a selected one of the plurality subrows responsive to the row activation signal via a signal received by the subrow driver corresponding to the selected one of the plurality of subrows; and
- a mask register configured to enable selection of a plurality of the subrows to remain inactive;
- wherein the controller is further configured to direct that a subrow selected via the mask register remains inactive.
11. The apparatus of claim 10, wherein:
- the controller addresses a subrow driver selectably coupled to a subrow at a particular position using a signal for selective activation of the subrow driver; and
- a subrow decoder is configured to decode the signal to select or deselect a gate for activation of the subrow driver.
12. (canceled)
13. The apparatus of claim 10, wherein:
- the row is one of a plurality of rows of memory cells within the array; and
- each of the plurality of rows comprises a plurality of subrows.
14. The apparatus of claim 13, wherein each subrow of the plurality includes a same quantity of memory cells.
15. The apparatus of claim 10, further comprising:
- a number of the plurality of subrow drivers positioned between subrows of each of a plurality of rows;
- each subrow driver coupled to one subrow of the plurality of subrows in each of the plurality of rows; and
- a predetermined number of memory cells that separate the number of the subrow drivers for each of the plurality of subrows.
16. A system, comprising:
- a subarray of memory cells in a memory array;
- a plurality of rows in the subarray, wherein each row comprises a plurality of memory cells corresponding to respective columns;
- a controller configured to selectably address a particular row and direct that a subrow at a particular position in the row remains inactive; and
- a mode register configured to enable selection between: all of the plurality of memory cells being activated in a row being addressed; and activation of selected subrows of the plurality of memory cells in the row being addressed.
17. The system of claim 16, further comprising:
- a plurality of sense amplifiers coupled to the respective plurality of columns;
- wherein the subrow at the particular position remaining inactive prevents access to a plurality of memory cells of the subrow by sense amplifiers corresponding to the plurality of memory cells.
18. The system of claim 17, wherein to prevent access includes to prevent storage of a number of data values by a plurality of sense amplifiers corresponding to the plurality of memory cells.
19. (canceled)
20. The system of claim 16, further comprising:
- a mask register configured to: enable inhibition of activation of selected subrows of memory cells in a row being addressed; and provide a row address that comprises an indication of which subrow driver for a particular subrow in the particular row is deselected for activation.
21. The system of claim 16, further comprising:
- a mask register configured to: enable inhibition of activation of a plurality of subrows in the particular row: and enable inhibition of activation of different subrows in each of the plurality of rows.
22. The system of claim 16, further comprising:
- a mask register configured to: inhibit performance of a refresh operation on the subrow at the particular position responsive to the row remaining inactive; and promote performance of a refresh operation on a subrow at a different position responsive to the row being activated.
23. A method for operating a memory device, comprising:
- using a mask register to enable management of an activation state of a plurality of subrows within a row of memory cells;
- using a controller to manage the activation state of the plurality of subrows based at least in part on enablement by the mask register;
- receiving a first signal by a first subrow driver coupled to a first subrow of the plurality of subrows; and
- deselecting activation of the first subrow driver to inhibit activation of the first subrow in response to receiving the first signal.
24. The method of claim 23, further comprising:
- receiving a second signal by a second subrow driver coupled to a second subrow of the plurality of subrows; and
- selecting activation of the second subrow driver to activate the second subrow in response to receiving the second signal.
25. The method of claim 23, further comprising:
- responsive to inhibiting activation of the first subrow: preventing access to a plurality of memory cells of the first subrow by sense amplifiers coupled to a plurality of columns corresponding to the plurality of memory cells; and preventing storage, by the sense amplifiers, of data values corresponding to the plurality of memory cells.
26. The method of claim 23, further comprising:
- responsive to inhibiting activation of a number of subrows: preventing access to the number of subrows; preventing storage of a number of data values corresponding to the number of subrows; and reducing an amount of energy used by the memory device, wherein the reduced amount of energy used corresponds to the number of subrows being inhibited from activation relative to energy used for activation of all of the plurality of subrows within the row in a different operating mode of the memory device.
27. An apparatus, comprising:
- a plurality of subrows within a row of memory cells;
- a controller configured to selectably address and manage an activation state of each subrow of the plurality;
- subrow driver circuitry coupled to the controller, the subrow driver circuitry configured to maintain one or more subrows of the plurality in the activation state based at least in part on signaling from the controller;
- a global row driver configured to maintain, in a first activation state, a first subrow of the plurality via a first subrow driver and maintain, in the first activation state, a second subrow of the plurality via a second subrow driver;
- a mask register configured to: selectably latch a number of selections from among the plurality of subrows to remain in a second activation state; and inhibit the first activation state via the first and second subrow drivers responsive to the latched number of selections.
28. A system, comprising:
- a subarray of memory cells in a memory array;
- a plurality of rows in the subarray, wherein each row comprises a plurality of memory cells corresponding to respective columns;
- a controller configured to selectably address a particular row and direct that a subrow at a particular position in the row remains inactive; and
- a mask register configured to: enable inhibition of activation of selected subrows of memory cells in a row being addressed; and provide a row address that comprises an indication of which subrow driver for a particular subrow in the particular row is deselected for activation.
29. A system, comprising:
- a subarray of memory cells in a memory array;
- a plurality of rows in the subarray, wherein each row comprises a plurality of memory cells corresponding to respective columns;
- a controller configured to selectably address a particular row and direct that a subrow at a particular position in the row remains inactive; and
- a mask register configured to: enable inhibition of activation of a plurality of subrows in the particular row: and enable inhibition of activation of different subrows in each of the plurality of rows.
30. A system, comprising:
- a subarray of memory cells in a memory array;
- a plurality of rows in the subarray, wherein each row comprises a plurality of memory cells corresponding to respective columns;
- a controller configured to selectably address a particular row and direct that a subrow at a particular position in the row remains inactive; and
- a mask register configured to: inhibit performance of a refresh operation on the subrow at the particular position responsive to the row remaining inactive; and promote performance of a refresh operation on a subrow at a different position responsive to the row being activated.
Type: Application
Filed: Dec 19, 2017
Publication Date: Jun 20, 2019
Inventors: Glen E. Hush (Boise, ID), Richard C. Murphy (Boise, ID)
Application Number: 15/846,410