System and method for implementing read-after-write commands in memory interface

Abstract

The invention relates to a system and method for implementing read-after-write commands in a memory interface. An interface of a memory subsystem may receive a write command addressed to a first address and a read command addressed to a second address, and may receive data corresponding to the write command. The interface may determine whether the first address matches the second address, in response to determining that the first address matches the second address, may drop the read command and the second address, and may provide data to the host.

Description

[0001] Information about divisional applications

[0002] This case is a divisional application. The parent case of the divisional application is an invention patent application with application date of December 29, 2020, application number 202011589045.9, and invention name "System and method for implementing write-after-read commands in a memory interface".

[0003] Priority information

[0004] This application claims priority to U.S. Provisional Application Serial No. 62 / 954,869, filed on December 30, 2019, the contents of which are incorporated herein by reference. Technical Field

[0005] Embodiments of the present disclosure relate generally to memory subsystems, and more particularly, to implementing read-after-write commands in an interface. Background Art

[0006] The memory subsystem may include one or more memory devices that store data. The memory devices may be, for example, non-volatile memory devices and volatile memory components. Typically, the host system may utilize the memory subsystem to store data at the memory devices and retrieve data from the memory devices. Summary of the invention

[0007] According to one embodiment, a system for implementing a read-after-write command in a memory interface is provided. The system includes: a memory device; a processing device, the processing device is coupled to the memory device and configured to control a cache memory for the memory device; and an interface, the interface is coupled to the processing device and configured to: receive a write command addressed to a first address and a read command addressed to a second address; receive data corresponding to the write command; determine whether the first address matches the second address; and in response to determining that the first address matches the second address: discard the read command and the second address; and provide the data to a host.

[0008] According to another embodiment, a method for implementing a read-after-write command in a memory interface is provided. The method includes: receiving a write command addressed to a first address and a read command addressed to a second address at an interface of a memory subsystem; determining whether the first address matches the second address; determining whether data corresponding to the write command is damaged; in response to determining that the first address matches the second address and the data is not damaged: discarding the read command and the second address; and providing the data to a host; and in response to determining that the first address does not match the second address or the data is damaged, providing the read command and the second address to a processing device.

[0009] According to a further embodiment, a system for implementing a read-after-write command in a memory interface is provided. The system includes: a memory device; a processing device, the processing device coupled to the memory device and configured to control a cache memory for the memory device; an interface, the interface coupled to the processing device and configured to: receive a write command addressed to a first address and a first read command and a second read command addressed to a second address; receive data corresponding to the write command; in response to determining that the first address matches the second address and the data is not corrupted: discard the first read command, the second read command and the second address; and provide the data to a host; and in response to determining that the first address does not match the second address or the data is corrupted, provide the first read command, the second read command and the second address to the processing device for processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The present disclosure can be more fully understood through the detailed description given below and through the accompanying drawings of various embodiments of the present disclosure.

[0011] Figure 1 An exemplary computing system is shown that includes a memory subsystem including an interface and a cache controller according to some embodiments of the present disclosure.

[0012] Figure 2 An example of an interface of a computing system according to some embodiments of the present disclosure is shown.

[0013] Figure 3 is a flow chart of an exemplary method for interfacing according to some embodiments of the present disclosure.

[0014] Figure 4 is a block diagram of an exemplary computer system in which embodiments of the present disclosure may operate. DETAILED DESCRIPTION

[0015] Various aspects of the present disclosure relate to a memory subsystem that implements a read-after-write command in a memory interface. The memory subsystem may be a storage device, a memory module, or a combination of a storage device and a memory module. Figure 1 Examples of storage devices and memory modules are described. In general, a host system may utilize a memory subsystem that includes one or more devices, such as memory devices that store data. The host system may provide data to be stored at the memory subsystem and may request retrieval of data from the memory subsystem.

[0016] The host system may send access requests (e.g., write commands, read commands) to the memory subsystem, such as to store data on a memory device at the memory subsystem and read data from a memory device on the memory subsystem. The data to be read or written specified by the host request is referred to as "host data" hereinafter. The host request may include logical address information (e.g., logical block address (LBA), namespace) for the host data, which is a location that the host system associates with the host data. The logical address information may be part of the metadata of the host data.

[0017] The memory subsystem may receive a read command after receiving a write command. The read command may be received before executing a write command and after receiving a write command. The read command and the write command may address the same address and / or the same logical block of a memory device of the memory subsystem. In some cases, it may be beneficial to process the read command without retrieving data from the memory device. For example, a conventional cache controller of the memory device may be able to retrieve data from the cache memory to service the read command. However, retrieving data from the cache memory may utilize a processing device of the cache controller.

[0018] Aspects of the present disclosure address the above and other deficiencies. An interface of a memory subsystem may receive a read command after receiving a write command to the same address. As used herein, the same address may reference the same logical address, the same physical address, and / or the same logical block. The interface may provide data corresponding to the write command to process the read command. The data may be provided to the host in response to receiving the read command. Processing the write command in the interface of the memory subsystem may transfer work from the cache controller to the interface, which may free up resources of the cache controller to process other commands. By retrieving data from the cache memory and / or the memory device of the memory subsystem, the read command manipulated (e.g., processed) in the interface may be manipulated with less delay than the read command processed in the cache controller.

[0019] The figures herein follow a numbering convention in which the first digit or digits correspond to the figure number and the remaining digits identify the element or component in the figure. Similar elements or components between different figures may be identified by using similar numerals. For example, in Figure 1 109 can be used to index the component "09". Figure 2 Similar elements in the drawings may be referenced as 209. It should be understood that the elements shown in the various embodiments herein may be added, exchanged and / or eliminated to provide multiple additional embodiments of the present disclosure. In addition, it should be understood that the proportions and relative scales of the elements provided in the drawings are intended to illustrate certain embodiments of the present invention and should not be considered limiting.

[0020] Figure 1 An exemplary computing system 100 is shown that includes a memory subsystem 103, in accordance with some embodiments of the present disclosure. The memory subsystem 103 may include media such as one or more volatile memory devices, one or more non-volatile memory devices, or a combination thereof.

[0021] The memory subsystem 103 may be a storage device, a memory module, or a mixture of a storage device and a memory module. Examples of storage devices include solid-state drives (SSDs), flash drives, universal serial bus (USB) flash drives, embedded multimedia controllers (eMMC) drives, universal flash storage (UFS) drives, secure digital (SD) cards, and hard disk drives (HDDs). Examples of memory modules include dual in-line memory modules (DIMMs), small outline DIMMs (SO-DIMMs), and various types of non-volatile dual in-line memory modules (NVDIMMs).

[0022] The computing system 100 may be a computing device such as a desktop computer, a laptop computer, a network server, a mobile device, a vehicle (e.g., an airplane, drone, train, automobile, or other transportation), an Internet of Things (IoT) enabled device, an embedded computer (e.g., a computer included in a vehicle, an industrial device, or a networked commercial device), or such a computing device including a memory and a processing device.

[0023] Computing system 100 may include a host system 102 coupled to one or more memory subsystems 103. In some embodiments, host system 102 is coupled to memory subsystems 103 of different types. Figure 1 An example of a host system 102 coupled to one memory subsystem 103 is shown. As used herein, "coupled to" or "coupled with" generally refers to a connection between components, which can be an indirect communication connection or a direct communication connection (e.g., without intervening components), whether wired or wireless, including connections such as electrical connections, optical connections, magnetic connections, etc.

[0024] The host system 102 may include a processor chipset and a software stack executed by the processor chipset. The processor chipset may include one or more cores, one or more caches, a memory controller (e.g., an NVDIMM controller), and a storage protocol controller (e.g., a PCIe controller, a SATA controller). The host system 102 uses the memory subsystem 103, for example, to write data to the memory subsystem 103 and read data from the memory subsystem 103.

[0025] The host system 102 may be coupled to the memory subsystem 103 via a physical host interface 109. The interface 109 may be internal to the memory subsystem 103. Examples of the physical host interface include, but are not limited to, a Serial Advanced Technology Attachment (SATA) interface, a Peripheral Component Interconnect Express (PCIe) interface, a Universal Serial Bus (USB) interface, a Fibre Channel, a Serial Attached SCSI (SAS), a Small Computer System Interface (SCSI), a Double Data Rate (DDR) memory bus, a Dual In-line Memory Module (DIMM) interface (e.g., a DIMM slot interface supporting Double Data Rate (DDR)), an Open NAND Flash Interface (ONFI), Double Data Rate (DDR), Low Power Double Data Rate (LPDDR), or any other interface. The interface 109 may be used to transmit commands, addresses, and / or data between the host system 102 and the memory subsystem 103. When the memory subsystem 103 is coupled to the host system 102 via a PCIe interface, the host system 102 may further utilize an NVM Express (NVMe) interface to access components (e.g., the memory device 105). The physical host interface may provide an interface for passing control, address, data, and other signals between the memory subsystem 103 and the host system 102 . Figure 1 Memory subsystem 103 is shown as an example. In general, host system 102 can access multiple memory subsystems via the same communication connection, multiple separate communication connections, and / or a combination of communication connections.

[0026] The memory devices 105, 112 may include any combination of different types of non-volatile memory devices 105 and / or volatile memory devices 112. Volatile memory devices (e.g., memory device 112) may be, but are not limited to, random access memory (RAM) such as dynamic random access memory (DRAM) and synchronous dynamic random access memory (SDRAM).

[0027] Some examples of non-volatile memory devices (e.g., memory device 105) include NAND-type flash memory and write-in-place memory, such as a three-dimensional cross-point ("3D cross-point") memory device, which is a cross-point array of non-volatile memory cells. The cross-point array of non-volatile memory can be combined with a stackable cross-grid data access array to perform bit storage based on changes in body resistance. In addition, compared to many flash-based memories, cross-point non-volatile memory can perform write-in-place operations, in which non-volatile memory cells can be programmed without having to erase the non-volatile memory cells in advance. NAND-type flash memory includes, for example, two-dimensional NAND (2D NAND) and three-dimensional NAND (3D NAND).

[0028] Each of the memory devices 105 may include one or more memory cell arrays. One type of memory cell, such as a single-level cell (SLC), may store one bit per cell. Other types of memory cells, such as multi-level cells (MLC), three-level cells (TLC), four-level cells (QLC), and five-level cells (PLC), may store multiple bits per cell. In some embodiments, each of the memory devices 105 and 112 may include one or more memory cell arrays, such as SLC, MLC, TLC, QLC, or any combination thereof. In some embodiments, a particular memory device may include an SLC portion, an MLC portion, a TLC portion, a QLC portion, or a PLC portion of a memory cell. The memory cells of the memory devices 105 and 112 may be grouped into pages, which may refer to logical units of a memory device for storing data. With certain types of memory (e.g., NAND), pages may be grouped to form blocks.

[0029] Although a non-volatile memory device 105 such as a 3D cross-point array of non-volatile memory cells and a NAND-type memory (e.g., 2D NAND, 3D NAND) and a 3D cross-point array of non-volatile memory cells is described, the memory device 105 can be based on any other type of non-volatile memory, such as read-only memory (ROM), phase change memory (PlocCM), self-selected memory, other chalcogenide-based memory, ferroelectric transistor random access memory (FeTRAM), ferroelectric random access memory (FeRAM), magnetic random access memory (MRAM), spin transfer torque (STT)-MRAM, conductive bridging RAM (CBRAM), resistive random access memory (RRAM), oxide-based RRAM (OxRAM), NOR (NOR) flash memory, and electrically erasable programmable read-only memory (EEPROM).

[0030] The memory subsystem controller 104 (or, for simplicity, the controller 104) can communicate with the memory device 105 to perform operations such as reading data, writing data, or erasing data at the memory device 105, as well as other such operations. The memory subsystem controller 104 can include hardware such as one or more integrated circuits and / or discrete components, buffer memory, or a combination thereof. The memory subsystem controller 104 can be a microcontroller, a dedicated logic circuit system (e.g., a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc.), or other suitable processor.

[0031] The memory subsystem controller 104 may include a processor 107 (e.g., a processing device) configured to execute instructions stored in a local memory 108 to perform the operations described herein. In the example shown, the local memory 108 of the memory subsystem controller 104 includes an embedded memory configured to store instructions for performing various processes, operations, logic flows, and routines that control the operation of the memory subsystem 103, including handling communications between the memory subsystem 103 and the host system 102.

[0032] In some embodiments, local memory 108 may include memory registers that store memory pointers, retrieved data, etc. For example, local memory 108 may also include read-only memory (ROM) for storing microcode. Figure 1 The exemplary memory subsystem 103 in FIG. 1 has been shown as including a memory subsystem controller 104, but in another embodiment of the present disclosure, the memory subsystem 103 does not include a memory subsystem controller 104 and may instead rely on external control (e.g., control provided by an external host or by a processor or controller separate from the memory subsystem).

[0033] In general, the memory subsystem controller 104 may receive commands or operations from the host system 102, and may convert the commands or operations into instructions or appropriate commands to achieve the desired access to the memory device 105 and / or the memory device 112. The memory subsystem controller 104 may be responsible for other operations such as wear leveling operations, garbage collection operations, error detection and error correction code (ECC) operations, encryption operations, cache operations, and address conversion between logical addresses (e.g., logical block addresses (LBA), namespaces) and physical addresses (e.g., physical block addresses) associated with the memory device 105. The memory subsystem controller 104 may further include a host interface circuit system to communicate with the host system 102 via a physical host interface. The host interface circuit system may convert commands received from the host system into command instructions to access the memory device 105 and / or the memory device 112, and convert responses associated with the memory device 105 and / or the memory device 112 into information for the host system 102.

[0034] The memory subsystem 103 may also include additional circuitry or components not shown. In some embodiments, the memory subsystem 103 may include a cache or buffer (e.g., DRAM) and address circuitry (e.g., row decoders and column decoders) that can receive addresses from the memory subsystem controller 104 and decode the addresses to access the memory device 105.

[0035] In some embodiments, the memory device 105 includes a local media controller 110 that operates in conjunction with the memory subsystem controller 104 to perform operations on one or more memory cells of the memory device 105. An external controller (e.g., the memory subsystem controller 104) can manage the memory device 105 externally (e.g., perform media management operations on the memory device 105). In some embodiments, the memory device 105 is a managed memory device, which is a raw memory device combined with a local controller (e.g., the local controller 110) for media management within the same memory device package. An example of a managed memory device is a managed NAND (MNAND) device.

[0036] In at least one embodiment, the interface 109 is compatible with the DDR-4 NVDIMM-P protocol and has a pin compatible with the DDR-4 NVDIMM-P physical slot. For example, the interface 109 may include an NVDIMM-P CA physical layer (PHY), an NVDIMM-P event PHY, and an NVDIMM-P DQ PHY. The NVDIMM-P command / address (CA) PHY may be coupled to a command decoder for processing host commands. The interface 109 may include a memory (e.g., a content addressable memory) for storing commands and addresses associated therewith. The interface 109 may compare commands received from the host system 102 to determine whether there are two or more write commands to the same address of the memory device 105 in other possible memory devices.

[0037] NVDIMM-P CA PHY can be configured to share a single alarm signal for parity errors and uncorrectable ECC (UECC) errors. However, the waiting time and pulse width of parity errors and UECC errors on a single alarm signal are different. In some embodiments, the NVDIMM-P CAPHY can be coupled to a parity register and an error register. The parity register can be configured to track the waiting time and pulse width of a single alarm signal for a parity error. The error register can be configured to track the waiting time and pulse width of a UECC error. A single alarm generation logic can use the parity register and the error register based on the type of error (parity or UECC) to generate an alarm signal with an appropriate assertion time (e.g., starting from the waiting time) and pulse width. The register can be programmed with the corresponding waiting time and pulse width.

[0038] The NVDIMM-P DQ PHY may be coupled to the error circuitry and, via the error circuitry, to the write buffer for writing data to the memory subsystem 103. The error circuitry may be included in the interface 109. Figure 2 The error circuitry is described in more detail. The NVDIMM-P DQ PHY may also be coupled to a read buffer for receiving data from the memory subsystem 103.

[0039] The NVDIMM-P DQ PHY may be coupled to a command decoder to provide an indication to the host that the memory subsystem 103 is ready to receive commands from it. In some embodiments, the command decoder may provide such an indication based on the amount of write credits available. The interface 109 may have a memory dedicated to storing write credits from the cache controller 111. In at least one embodiment, each write command is equal to one write credit. For each write command provided by the interface 109 to the cache controller 111, one write credit is consumed. Once the cache controller 111 has processed the write command, it may release the write credit by signaling the interface 109 to indicate that the write is complete. As a non-limiting example, the interface 109 may store 10 write credits, which means that the interface 109 may have up to 10 outstanding write commands for the cache controller, after which the interface 109 will indicate to the host 102 that it cannot receive additional write commands from it.

[0040] Relative to Figure 2 Some additional details of interface 109 are shown and described. Ultimately, interface 109 is coupled to memory subsystem controller 104 (and thereby to memory device 105) and to cache controller 111 (and thereby to cache memory 112, since cache controller 111 is coupled to cache memory 112). Cache controller 111 may also be coupled to memory subsystem controller 104.

[0041] In order to provide large capacity and persistent storage, the back end of the memory subsystem 103 includes a 3D cross point array of non-volatile memory cells as a memory device 105. However, the latency of the 3D cross point memory is much higher than that of a typical DRAM. Therefore, in order to narrow the latency gap between a typical DRAM and a 3D cross point memory, the memory subsystem 103 includes a higher speed cache memory 112 between the back end storage and the interface 109 provided by the memory device 105. The cache memory 112 may include DRAM, SRAM, reduced latency DRAM third generation (RLDRAM3), or a combination thereof. RLDRAM3 is a DRAM with an SRAM-like interface. RLDRAM3 generally has lower latency and larger capacity than SRAM. RLDRAM3 generally has better performance than DRAM, especially for back-to-back read and write access or fully random access.

[0042] The NVDIMM-P event PHY can be coupled to an event manager for receiving reports of events in the memory subsystem 103. For example, the event manager can receive a ready signal from the memory subsystem 103 (or a transfer manager associated therewith). The ready signal can indicate any response or event from any source (e.g., an emergency event, an interrupt, a read ready, etc.). For different sources, such responses or events can occur simultaneously. Responses or events can be serialized by the event manager. The event manager can use a single response counter to track responses and events. Whenever any response is ready or an event occurs, the single response counter increments by one (even if they occur simultaneously). A single response counter can set a flag for a pending emergency event. In response to setting the flag, once the currently occurring event or response is completed, the earliest marked emergency event can be processed. Emergency events can be given priority through the operation of a single response counter.

[0043] The cache controller 111 may manage commands, addresses, and / or data (for the memory device 105) received from the host 102 and store them in the cache memory 112 before executing the commands and / or storing the data in the memory device 105. The associated commands and / or addresses may be stored in a memory internal to the interface 109 and manipulated by a command builder and / or command manager (not specifically shown) before being transmitted to the cache controller 111 as described herein. In some embodiments, the cache memory 112 may store data retrieved from the memory device 105 before sending it to the host 102. The cache memory 112 may be read and written faster than the memory device 105. Thus, reading and writing to the cache memory 112 rather than the memory device 105 may be beneficial in providing low latency when executing write commands and / or read commands.

[0044] The interface 109 may mark a write command addressed to the same address as a read command. The interface may mark a command by adding or modifying metadata associated with the command. For example, the interface may add a bit to the metadata associated with the command indicating that the command has been marked. The marked command may be used to determine whether to forward data corresponding to the marked command to a read buffer of the interface 109. The read buffer may be used to provide data to the host system 102. In response to providing the data stored in the read buffer to the host system 102, the interface 109 may discard (e.g., delete) the read command. The interface 109 may discard the read command without providing the read command to the cache controller 111 and / or processing the read command using the cache controller 111. The interface 109 may, in response to receiving a read command addressed to the same address as a write command (for which the accompanying data for the write command is still in the interface (e.g., in the write buffer)), cause the accompanying data to be stored in the read buffer, cause the accompanying data to be transmitted from the read buffer to the host to satisfy the read command, and delete the read command (without actually executing the read command relative to the cache memory 112 or the memory device 105).

[0045] Marking write commands in interface 109 may reduce the workload of cache controller 111. In some embodiments, write commands may be marked from outside of interface 109 and cache controller 111 and discarded by interface 109.

[0046] In various cases, cache controller 111 may move data from memory device 105 to cache memory 112 without using memory subsystem controller 104. Cache controller 111 may also move data from memory device 105 to cache memory 112 with the use of memory subsystem controller 104.

[0047] Figure 2 FIG. 2 shows an example of an interface 209 of a computing system according to some embodiments of the present disclosure. The interface 209 is similar to Figure 1 Interface 209 may include a comparator 221, a command buffer 222, a write buffer 224, and a read buffer 230, generally referred to as "buffers 222, 224, and 230". The interface may also include error correction circuitry, which is labeled as "ECC circuitry 223" for convenience. ECC circuitry 223 may be error detection and / or correction circuitry, such as correction code (ECC) circuitry, cyclic redundancy check (CRC) circuitry, other error detection and / or correction circuitry, or a combination thereof.

[0048] As used herein, buffers 222, 224, and 230 and / or error memory 229 may include data structures stored in a memory such as an SRAM or a different type of memory. For example, buffers 222, 224, and 230 and error memory 229 may be stored in registers. Buffers 222, 224, and 230 may be first-in-first-out (FIFO) data structures that maintain an order associated with the receipt of commands (e.g., read commands and / or write commands), addresses, and / or data.

[0049] Interface 209 may receive commands, addresses, and / or data. For example, commands received by interface 209 may be decoded by a command decoder. The received commands may be used to send data to a memory subsystem (e.g., Figure 1 For example, a command may be used to write data received by interface 209 to an address also received by interface 209. The address may be the address of a memory cell of a memory device of the memory subsystem hosting the interface.

[0050] Data received by the interface 209 may be provided to the cache controller 211 for temporary storage in the cache memory 212 before being provided to the memory subsystem for storage in the memory device of the memory subsystem. In this way, a write command to an address of a memory device may be implemented by storing the data corresponding to the write command in the cache memory 212. When the memory subsystem is ready, the data stored in the cache memory 212 may be provided to the memory device according to the address associated with the write command. In at least one embodiment, the host sending the write command and the associated address may not be aware that the data is temporarily stored in the cache memory 212 before the data is stored in the addressed location.

[0051] In various examples, an identification (ID) 225 may be assigned to a write command and / or a read command before the interface 209 receives the write command and / or the read command. For example, ID 225 may be assigned to a write command and / or a read command by a host. A read ID may be assigned to a read command, and a write ID may be assigned to a write command. Command buffer 222 may store ID 225 and address 226 corresponding to a write command and / or a read command. For example, each entry in command buffer 222 may contain a write ID or a read ID and an address, wherein the write ID and the address correspond to a write command, and the read ID and the address correspond to a read command. Address 226 corresponding to a read command may identify a memory cell to be read. Address 226 corresponding to a write command may identify a memory cell to be written. Address 226 may be a physical address and / or a logical address, as well as other types of memory addresses that may be used. In at least one embodiment, command buffer 222 may store ID 225 and address 226 until the memory subsystem associated with the relevant address has executed the corresponding write command. For example, the command buffer 222 may store a specific write ID 225 and address 226 before the write command and accompanying data have been provided to the cache controller 211 and before the write command and accompanying data have been provided to the memory subsystem. Such an embodiment may be beneficial in allowing the interface 209 to provide write data from the interface 109 to the host and discard read commands from the command pipeline before the read commands are provided to or executed by the memory subsystem.

[0052] Comparator 221 may compare addresses 226 to determine whether any two or more of addresses 226 are the same address or overlapping addresses. For example, comparator 221 may compare a first write ID having a first address with a second read ID having a second address. If the first address and the second address match, comparator 221 may determine that commands having the first write ID and the second read ID may be implemented to write to and read from the same address of a memory device, respectively. The determination that multiple commands are written to and read from the same address may be used to identify write commands and read commands that write to and read from the memory device of the memory subsystem, respectively, to the same address.

[0053] The command IDs may be sequentially assigned based on the order in which the commands are received from the host. In various cases, the comparator 221 may also determine whether a read command is received after a write command is received. For example, the comparator 221 may compare the IDs of the write command and the read command to determine whether the number of the write ID is less than the read ID. In some embodiments, the comparator 221 may determine whether a write command is inserted into the command buffer 222 before inserting the read command into the command buffer 222. The determination that a write command is addressed to the same address as a read command and that a write command is received before a read command is received may be used to identify a write command and / or a read command.

[0054] In response to determining that the identified command writes to the same address, the comparator 221 can mark the command and / or the corresponding ID 225. The comparator 221 can cause the ECC circuit system 223 to mark the identified write command and / or read command by sending a signal indicating the above situation to the ECC circuit system 223. In such an example, the ECC circuit system 223 can mark the read command received later as an error, regardless of whether the read command (or accompanying data) is error-checked and / or regardless of whether the read command (or accompanying data) passes the error check when executed. Marking the read command can include setting a bit in the metadata of the read command. The interface 209, the cache controller 211, and / or the memory subsystem controller can be configured to read the bit from the metadata of the read command, and in response to the bit being set, the read command can be deleted.

[0055] ECC circuit system 223 may include error memory 229 and circuit system 228 configured to operate according to error correction code ("ECC 228"). Error memory 229 may store ID 225, ECC result 227 of ECC 228, and a determination as to whether ID 225 corresponds to a write command addressed to the same address as a read command. Conventional ECC circuit system 223 does not store a determination as to whether ID 225 corresponds to a write command addressed to the same address as a read command. ECC circuit system 223 may operate according to ECC 228 to detect and / or correct corruption of data corresponding to a command received by interface 209. ECC circuit system 223 may identify uncorrupted data and / or data containing corruption according to ECC 228. ECC circuit system 223 may utilize error memory 229 to identify corrupted data. Error memory 229 may store entries including a write ID (e.g., ID 225) and ECC result 227. Associating the write ID 225 with the ECC result 227 of the data received by the interface 209 can allow the data to be associated with the write command. For example, each entry in the error memory 229 can contain the write ID corresponding to the write command and the result of the ECC 228. The ECC 228 can determine whether the data is corrupted. For example, the ECC 228 can determine whether the data corresponding to the write ID is corrupted. The determination can be stored as the ECC result 227 in the error memory 229.

[0056] The error memory 229 may also store a determination of whether a write command with the write ID 225 has the same address as a corresponding read command with the read ID. The error memory 229 may store the determination in a field representing a read-after-write result 235. For example, the comparator 221 may store the read-after-write result 235 in the error memory 229 in response to identifying a read-after-write dependency. As used herein, a read-after-write dependency describes receiving a read command after receiving a write command, where the read command and the write command address the same address. For example, a bit with a particular value may indicate that there is no read-after-write dependency, while a bit with a different value may indicate that there is a read-after-write dependency.

[0057] After the ECC 228 has determined that the write data is not corrupted, the write data received by the interface 209 may be stored in the write buffer 222. For example, data may be provided to the write buffer 222 from the ECC circuit system 223. Each entry of the write buffer 222 may contain a write ID and data corresponding to a write command. If the error memory 229 indicates that there is no read-after-write dependency, the result of the ECC 228, the data 231, the write ID 225, the read ID 225, and / or the address 226 may be provided to the cache controller 211. If there is a read-after-write dependency, the ECC 228, the data 231, the write ID 225, and the corresponding address 226 may be provided to the cache controller 211. In response to the read-after-write result 235 indicating that the read command and the write command address the same memory address, the read command and any associated address may not be provided to the cache controller 211 from the command buffer 222.

[0058] In various cases, the read-after-write result 235 and the ECC result 227 can be used to determine whether to provide a write command and / or a read command to the cache controller 211. For example, if the ECC result 227 indicates that the data corresponding to the write command is corrupted, both the write command and the read command addressed to the same address and / or different addresses can be provided to the cache controller 211. The indication that the write data is corrupted as indicated by the ECC result 227 makes the write data inaccessible and thus avoids being used to implement the corresponding read command. If the ECC result 227 indicates that the data corresponding to the write command is not corrupted, and the read-after-write result 235 indicates that the write command is addressed to the same address addressed by the read command, the interface can provide a write command to the cache controller 211 instead of a read command. Assuming that the write data will be stored to the same address that will be read to process the read command, the write data can be used to implement the read command. If the ECC result 227 indicates that the data corresponding to the write command is not corrupted, and the read-after-write result 235 indicates that the write command addresses a different address than the read command, the interface 209 may provide the write command and the read command to the cache controller 211. However, assuming that an address different from the address where the write data will be stored will be read to implement the read command, the interface 209 may not implement the read command using the write data. The cache controller 211 may execute the read and write commands it receives.

[0059] The write data 231 stored in the write buffer 224 may be provided to the read buffer 230 in response to the identification of the read-after-write dependency and the lack of indication that the write data 231 is corrupted. The write data 231 may be provided to the read buffer 230 via one or more lines 234. The lines 234 may couple the write buffer 224 to the read buffer 230 so that data may be provided from the write buffer 224 to the read buffer 230. As such, the read-after-write result 235 and the ECC result 227 may be used to determine whether to forward data from the write buffer 224 to the read buffer 230.

[0060] If data is provided from write buffer 224 to read buffer 230, the write command and / or associated write ID may be discarded from command buffer 222, assuming that the data may be provided to the host system from read buffer 230. Thus, in this case, processing the read command includes determining whether there is a read-after-write dependency, storing data corresponding to the write command from write buffer 224 in read buffer 230, and providing the data from read buffer 230 to the host. The read command may be processed without providing the read command to cache controller 211, without accessing cache memory 212, without providing the read command to a memory subsystem controller, and / or without providing access to a memory device of the memory subsystem.

[0061] In some examples, comparator 221 may include circuitry configured to determine whether there is a read-after-write dependency (e.g., read-after-write result 235) as identified by error memory 229. If a read-after-write dependency is identified, comparator 221 may identify a write ID 225 associated with the read-after-write dependency from error memory 229. In response to determining that the write command and the read command address the same address, the comparator may cross-reference the write ID 225 to the read ID 225 stored by comparator 221. Comparator 221 may then access write data 231 from write buffer 224 using the write ID, and store the write data as read data in read buffer 230 using the read ID. For example, comparator 221 may cause a new entry to be created in read buffer 230, the new entry including the read ID and the write data as read data. The write data may be stored in read buffer 230 without transferring the data from cache memory 212 to read buffer 230, or without transferring the data from a different memory device of a memory subsystem to the read buffer.

[0062] The comparator 221 can utilize the ECC circuit system 223 to mark the command read from the same address as the pending write command. For example, the comparator 221 can mark the command read from the same address as the pending write command in the error memory 229. For example, the comparator 221 can identify the write ID and read ID corresponding to the write command and the read command addressed to the same address. The identified write ID can be used to modify the entry of the error memory 229. The write ID can be used to update the post-write read result 235. The comparator 221 can store the write ID and read ID internally or in a memory external to the comparator but internal to the interface 209. The stored command, write ID and / or read ID can be used to retrieve the write data 231 under the write ID from the write buffer 224 and store the write data 231 to the read buffer 230 under the write ID.

[0063] Error memory 229 may be used to provide both information about the results of ECC 228 and information about which commands write and read data from the same address. Information may be provided to cache controller 211 and circuitry configured to store write data as read data in read buffer 230.

[0064] The error memory 229 may also be used for the purpose of determining whether to release a write ID. For example, if a read-after-write dependency is identified and marked using the read-after-write result 235, the write ID associated with the read-after-write dependency may be used to prevent the write ID from being released. Conventionally, the write ID may be released by providing a write command (e.g., a write ID) and corresponding write data from the write buffer 224 and / or the command buffer 222 to the cache controller 211. After the write ID, the associated address 226, and / or the write data 231 are provided to the cache controller 211, the write ID may then be discarded (e.g., deleted) from the write buffer 224, the command buffer 222, and / or the error memory 229. Discarding the write ID from the write buffer 224, the command buffer 222, and / or the error memory 229 may make the write ID available for use with a different incoming write command.

[0065] Although not shown, comparator 221 may be implemented external to interface 209 and cache controller 211. External comparator 221 may mark commands in error memory 229. Marking commands externally from interface 209 and / or cache controller 211 may cause read commands to be processed using the write data of the marked command without providing the read command to cache controller 211.

[0066] The cache controller 211 may implement untagged commands. For example, the cache controller 211 may write data to the cache memory and / or may provide the data to a memory subsystem controller (e.g., Figure 1 The memory subsystem controller 104 in the memory subsystem is used to write to the memory device of the memory subsystem.

[0067] Commands, addresses, and / or data may be received via different streams. ECC circuitry 223 may be configured to utilize error memory 229 to describe the attributes of both the address of the command and the data corresponding to the command. The attributes may include whether the read command and the write command store data to the same address or read data from the same address. The attributes may also include whether the data corresponding to the write command is damaged. Thus, the mark write command may include attributes describing the command stream received using command buffer 222, the attributes of the data stream stored in write buffer 224, and / or the attributes of the data stream stored in read buffer 230, assuming that the data is stored from write buffer 224 to read buffer 230.

[0068] In various examples, comparator 221 may identify two or more write commands addressed to the same address as a read command. Comparator 221 may store a write ID and a read ID corresponding to the write command and the read command, respectively. Each of the write commands may be marked in error memory 229 with the write ID corresponding to the write command. Error memory 229 may be used to determine whether to store data corresponding to the write command in read buffer 230.

[0069] In some examples, comparator 221 can identify the last (e.g., most recent) write command received from the marked write commands. Comparator 221 can transmit write data 231 corresponding to the received and identified most recent write command. Write data 231 can be stored from write buffer 224 to read buffer 230 using a read ID corresponding to the read command. In this way, an entry of read buffer 230 can include read ID 225 and read data 233. In response to processing the read command, the ID of the identified write command can be released.

[0070] In some examples, a write command and a plurality of read commands may be addressed to the same address. Comparator 221 may identify a write command and a read command that are being addressed to the same address. Write ID and read ID may be stored by comparator 221. Comparator 221 may mark a write command using error memory 229 and a write-after-read result 235 of the comparison performed by comparator 221. Interface 209 may utilize lines to store write data to read buffer 230. For example, interface 209 may generate a first entry into read buffer 230, the first entry including a first read ID and write data. In an example where two read commands and a write command are addressed to the same address, interface 209 may then generate a second entry into read buffer 230, the second entry including a second read ID and write data. In various examples, providing data from read buffer 230 to a host may include providing data from read buffer 230 to an input / output line to provide data to the host.

[0071] Figure 3 is a flow chart of an exemplary method 380 for interfacing according to some embodiments of the present disclosure. The method 380 may be performed by processing logic, which may include hardware (e.g., a processing device, a circuit system, dedicated logic, programmable logic, microcode, hardware of a device, an integrated circuit, etc.), software (e.g., instructions running or executed on a processing device), or a combination thereof. In some embodiments, the method 380 is performed by Figure 1The processes are performed by the interface 109 of the embodiment. Although shown in a particular order or sequence, unless otherwise specified, the order of the processes can be modified. Therefore, the embodiments shown should be understood as examples only, and the processes shown can be performed in different orders, and some processes can be performed in parallel. In addition, one or more processes can be omitted in various embodiments. Therefore, not all processes are required in every embodiment. Other processing flows are also possible.

[0072] At operation 381, a write command addressed to a first address and a read command addressed to a second address may be received at an interface of a memory subsystem. At operation 382, ​​a determination may be made as to whether the first address matches the second address. At operation 383, a determination may be made as to whether data corresponding to the write command is corrupted. At operation 384, in response to determining that the first address matches the second address and the data is not corrupted, the read command and the second address may be discarded, and the data may be provided to a host. At operation 385, in response to determining that the first address does not match the second address or the data is corrupted, the read command and the second address may be provided to a cache controller.

[0073] In various examples, method 380 may further include, in response to determining that the first address matches the second address, identifying a write ID of a write command addressed to the first address and a read ID of a read command addressed to the second address. The data may be stored in a read buffer. The data may also be associated with the read ID in the read buffer.

[0074] In response to determining that the first address does not match the second address and the data is not corrupted, the write command and the first address may be provided to the cache controller. In response to determining that the data is corrupted, the write command and the first address may be provided to the cache controller to be discarded by the cache controller. For example, the cache controller may discard the command and / or the corresponding data in response to determining that the data is corrupted. The data may be identified as corrupted using the error correction circuitry of the interface.

[0075] In various cases, a different write command may be received having a write ID different from the write ID corresponding to the write command. A determination may be made that the write command was received before the different write command was received. The determination may be stored in the error correction circuitry using the write ID.

[0076] A determination may also be made as to whether a write command was received before a different write command. In response to determining that a different write command was received after the write command, different data may be provided to the host using the read ID. Providing data to the host may also include providing data to the host in response to determining that a different write command was received before the write command.

[0077] Figure 4An exemplary machine of a computer system 490 is shown, in which a set of instructions for causing the machine to perform one or more of the methods discussed herein may be executed. In some embodiments, the computer system 490 may correspond to a host system (e.g., Figure 1 102) that includes, is coupled to, or utilizes a memory subsystem (e.g., Figure 1 103), or can be used to execute the operation of the controller (for example, execute the operating system to execute the corresponding Figure 1 In some embodiments, the machine may be connected (e.g., using a network) to other machines in a LAN, an intranet, an extranet, and / or the Internet. The machine may operate in the capacity of a server or a client user machine in server-client user network environment, as a peer machine in a peer-to-peer (or distributed) network environment, or as a server or a client user machine in a cloud computing infrastructure or environment.

[0078] The machine may be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a network appliance, a server, a network router, a switch or a bridge, or any machine capable of executing a set of instructions (sequential instructions or otherwise) that specify actions to be taken by the machine. Further, while a single machine is illustrated, the term "machine" should also be construed to include a collection of machines that individually or collectively execute a set (or multiple sets) of instructions to perform one or more of the methodologies discussed herein.

[0079] The exemplary computer system 490 includes a processing device 492, a main memory 494 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM)), etc.), a static memory 498 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage system 499, which communicate with each other via a bus 497.

[0080] Processing device 492 represents one or more general processing devices, such as a microprocessor, a central processing unit, etc. More specifically, the processing device can be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor that implements other instruction sets or a processor that implements a combination of instruction sets. Processing device 492 can also be one or more special processing devices, such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a network processor, etc. Processing device 492 is configured to execute instructions 493 for performing the operations and steps discussed herein. Computer system 490 may further include a network interface device 495 to communicate through a network 496.

[0081] The data storage system 499 may include a machine-readable storage medium 491 (also referred to as a computer-readable medium) on which is stored one or more sets of instructions 493 or software embodying any one or more of the methods or functions described herein. During the execution of the instructions 493 by the computer system 490, the instructions may also reside in whole or at least in part in the main memory 494 and / or in the processing device 492, which also constitute machine-readable storage media. The machine-readable storage medium 491, the data storage system 499, and / or the main memory 494 may correspond to Figure 1 Memory subsystem 103.

[0082] In one embodiment, instructions 493 include instructions for implementing a method corresponding to a computing component (e.g., Figure 1 The instructions may include instructions for interfaces with the memory subsystem (e.g., Figure 1 The interface 109 in the embodiment of the present invention is used to merge the command instruction 489 associated with the write command. Although the machine-readable storage medium 491 is shown as a single medium in the exemplary embodiment, the term "machine-readable storage medium" should be considered to include a single medium or multiple media storing one or more sets of instructions. The term "machine-readable storage medium" should also be considered to include a medium capable of storing or encoding a set of instructions for machine execution and causing the machine to perform one or more of the methods of the present disclosure. Therefore, the term "machine-readable storage medium" should be considered to include, but not limited to, solid-state memory, optical media, and magnetic media.

[0083] Some portions of the foregoing detailed description have been presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means by which those skilled in the art of data processing most effectively communicate the substance of their work to others skilled in the art. Here, an algorithm is generally considered to be a self-consistent sequence of operations that leads to a desired result. An operation is an operation that requires physical manipulation of physical quantities. Typically, although not necessarily, these quantities take the form of electrical or magnetic signals that can be stored, combined, compared, and otherwise manipulated. Mainly for general reasons, it has proven convenient to sometimes refer to these signals as bits, values, elements, symbols, characters, terms, numbers, etc.

[0084] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. The present disclosure may involve the actions and processes of computer systems or similar electronic computing devices that manipulate and transform data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system's memories or registers or other such information storage systems.

[0085] The present disclosure also relates to an apparatus for performing the operations herein. The apparatus may be specially constructed for the intended purpose, or it may comprise a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but not limited to, various types of disks (including floppy disks, optical disks, CD-ROMs, and magneto-optical disks), read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of medium suitable for storing electronic instructions, each coupled to a computer system bus.

[0086] The algorithms and displays presented herein are not inherently related to a particular computer or other device. Various general purpose systems may be used together with the programs taught herein, or it may prove convenient to construct more specialized equipment to perform the method. As described above, structures for various of these systems will occur. In addition, the present disclosure is not described with reference to a particular programming language. It should be understood that various programming languages ​​may be used to implement the teachings of the present disclosure as described herein.

[0087] The present disclosure may be provided as a computer program product or software, which may include a machine-readable medium having instructions stored thereon, and the instructions may be used to program a computer system (or other electronic device) to perform a process according to the present disclosure. The machine-readable medium includes a mechanism for storing information in a machine (e.g., computer) readable form. In some embodiments, the machine-readable (e.g., computer-readable) medium includes a machine (e.g., computer) readable storage medium, such as a read-only memory ("ROM"), a random access memory ("RAM"), a disk storage medium, an optical storage medium, a flash memory component, etc.

[0088] In the foregoing description, embodiments of the present disclosure have been described with reference to specific exemplary embodiments of the present disclosure. It will be apparent that various modifications may be made thereto without departing from the broader spirit and scope of the embodiments of the present disclosure as set forth in the following claims. Therefore, the description and drawings should be regarded as illustrative rather than restrictive.

Claims

1. A memory subsystem for implementing a read-after-write command in a memory interface, comprising: Memory device; a processing device coupled to the memory device and configured to control a cache memory for the memory device to store the data prior to storing the data on the memory device; and an interface of the memory subsystem, the interface comprising error correction circuitry and a comparator, wherein the interface is coupled to the processing device and configured to: receiving a write command addressed to a first address, a read command addressed to a second address from a host, and receiving data corresponding to the write command from the host; In response to determining that the first address matches the second address and the write command is inserted into the command buffer of the interface before the read command is inserted into the command buffer: discarding the read command and the second address from the command buffer by: sending a signal from the comparator to the error correction circuitry to cause the error correction circuitry to mark the read command as erroneous regardless of whether error checking is performed on the read command and if performed regardless of whether the read command passes the error checking, wherein the memory subsystem controller is configured to delete the read command in response to the read command being marked by the error correction circuitry; as well as The data is provided to the host from a read buffer of the interface. 2 . The memory subsystem of claim 1 , wherein the interface is further configured to provide the read command and the second address to the processing device in response to determining that the first address does not match the second address.

3. The memory subsystem of any one of claims 1 to 2, wherein the interface is further configured to: in response to determining that the first address matches the second address: providing the data to the host without providing the read command and the second address to the processing device; and The read command and the second address are discarded from the command buffer before providing the command in the command buffer to the processing device.

4. The memory subsystem of claim 3, wherein the interface is further configured to: providing the data from the write buffer to the read buffer to provide the data to the host; and The data from the read buffer is provided to an input / output line to provide the data to the host.

5. The memory subsystem of claim 3, wherein the interface is further configured to provide the data from the write buffer to an input / output line to provide the data to the host.

6. A method for implementing a read-after-write command in a memory interface, comprising: receiving, at an interface of the memory subsystem and from a host, a write command addressed to a first address and a read command addressed to a second address; storing the write command and the read command in a command buffer of the interface; determining, using a comparator of the interface, whether the first address matches the second address; determining whether data stored in a write buffer of the interface and corresponding to the write command is corrupted; In response to determining that the first address matches the second address and the data is not corrupted: discarding the read command and the second address from the command buffer at the interface by: sending a signal from the comparator to error correction circuitry of the interface to cause the error correction circuitry to mark the read command as erroneous without regard to whether error checking is performed on the read command and if performed without regard to whether the read command passes error checking, wherein a memory subsystem controller is configured to delete the read command in response to the read command being marked by the error correction circuitry; as well as providing the data from a read buffer of the interface to the host; as well as In response to determining that the first address does not match the second address or the data is corrupted, providing the read command and the second address to a processing device coupled to a memory device of the memory subsystem and configured to control a cache memory for the memory device to store the data before storing the data on the memory device. 7 . The method of claim 6 , further comprising identifying the write ID of the write command addressed to the first address and the read ID of the read command addressed to the second address in response to determining that the first address matches the second address.

8. The method of claim 7, further comprising storing the data in the read buffer and associating the data in the read buffer with the read ID.

9. The method of claim 6, further comprising providing the write command and the first address to the processing device in response to determining that the first address does not match the second address and the data is not corrupted.

10. The method according to any one of claims 6 to 9, further comprising: in response to determining that the first address matches the second address and the data is not corrupted, providing the read command and the read address to the processing device; In response to determining that the data is corrupted, providing the write command and the first address to the processing device to be discarded by the processing device; as well as Wherein determining whether the data corresponding to the write command is corrupted is performed using the error correction circuit system.

11. The method according to claim 10, further comprising: receiving a different write command having a write ID different from the write ID corresponding to the write command; determining that the write command is received before the different write command; as well as The determination of whether the first address matches the second address is stored in the error correction circuitry using the different write ID.

12. The method according to claim 11, further comprising: determining whether the write command was received before the different write command; In response to determining that the different write command is received after the write command, providing different data to the host using the read ID; and Wherein providing the data to the host further comprises providing the data to the host in response to determining that the different write command is received before the write command.

13. A memory subsystem for implementing a read-after-write command in a memory interface, comprising: Memory device; a processing device coupled to the memory device and configured to control a cache memory for the memory device to store the data prior to storing the data on the memory device; and an interface of the memory subsystem, the interface comprising error correction circuitry and a comparator, wherein the interface is coupled to the processing device and configured to: receiving from a host a write command addressed to a first address, a first read command and a second read command addressed to a second address, and receiving from the host data corresponding to the write command; In response to determining that the first address matches the second address and the data is not corrupted: discarding the first read command, the second read command, and the second address from the command buffer by: sending a signal from the comparator to the error correction circuitry to cause the error correction circuitry to mark the first read command and the second read command as erroneous without regard to error checking being performed on the read commands and without regard to whether the read commands pass the error checking if performed, wherein the memory subsystem controller is configured to delete the first read command and the second read command in response to the first read command and the second read command being marked by the error correction circuitry; as well as providing the data from a read buffer of the interface to the host; as well as In response to determining that the first address does not match the second address or the data is corrupted, the first read command, the second read command, and the second address are provided to the processing device for processing.

14. The memory subsystem of claim 13, wherein the interface is further configured to: storing a first determination that the data is not corrupted in the error correction circuitry using the write ID of the write command; storing in the error correction circuitry a second determination that the first address matches the second address using the write ID; and Responsive to the second determination, regardless of whether the write command has been provided to the processing device, avoiding releasing the write ID until the data has been stored to the read buffer using the first read ID and the second read ID corresponding to the first read command and the second read command, respectively.

15. The memory subsystem of any one of claims 13 to 14, wherein the interface is further configured to: storing in the error correction circuitry a determination that the data is corrupted utilizing the write ID of the write command; and In response to providing the write command to the processing device regardless of whether the first address matches the second address, the write ID is released.

Images