The exemplary and non-limiting embodiments of this invention relate generally to memory storage systems, methods, devices and computer programs and, more specifically, relate to mass memory devices, such as those containing non-volatile flash memory.
This section is intended to provide a background or context to the invention that is recited in the claims. The description herein may include concepts that could be pursued, but are not necessarily ones that have been previously conceived, implemented or described. Therefore, unless otherwise indicated herein, what is described in this section is not prior art to the description and claims in this application and is not admitted to be prior art by inclusion in this section.
The following abbreviations that may be found in the specification and/or the drawing figures are defined as follows:
Various types of flash-based mass storage memories currently exist. A basic premise of mass storage memory is to hide the flash technology complexity from the host system. A technology such as eMMC is one example. A managedNAND type of memory can be, for example, an eMMC, SSD, UFS or a microSD.
One non-limiting example of a flash memory controller construction is described in “A NAND Flash Memory Controller for SD/MMC Flash Memory Card”, Chuan-Sheng Lin and Lan-Rong Dung, IEEE Transactions of Magnetics, Vol. 43, No. 2, February 2007, pp. 933-935 (hereafter referred to as Lin et al.)
In a first aspect thereof the exemplary embodiments of this invention provide a method that comprises receiving a command originating from an initiator at a controller associated with a non-volatile mass memory coupled with a host device, the command being a command to write data that is currently resident in a memory of the host device to the non-volatile mass memory; moving the data that is currently resident in the memory of the host device from an original location to a portion of the memory allocated for use at least by the non-volatile mass memory; and acknowledging to the initiator that the command to write the data to the non-volatile mass memory has been executed.
In another aspect thereof the exemplary embodiments of this invention provide an apparatus that comprises a controller associated with a non-volatile memory module connected with a host device and an interface to the non-volatile mass memory module. The controller is configurable to process a command originating with an initiator, where the command is a command to write data that is currently resident in a memory of the host device to the non-volatile mass memory module. The controller is further configurable to move the data that is currently resident in the memory of the host device from an original location to a portion of the memory allocated for use at least by the non-volatile mass memory module; and to acknowledge to the initiator that the command to write the data to the non-volatile mass memory has been executed.
In another aspect thereof the exemplary embodiments of this invention provide a method that comprises providing a data processing system comprising at least one data processor and at least one memory comprised of dynamic random access memory, and autonomously performing an elementary memory operation in the memory.
In yet another aspect thereof the exemplary embodiments of this invention provide a memory module that comprises a controller; a non-volatile mass memory that is readable and writable by the controller; and an interface for connecting the memory module to a host device. The controller is configurable to receive a command originating from an initiator, where the command being a command to write data that is currently resident in a memory of the host device to the non-volatile mass memory module. The controller is further configurable to initiate the move of the data that is currently resident in the memory of the host device from an original location to a portion of the memory allocated for use at least by the non-volatile mass memory module; and to acknowledge to the initiator that the command to write the data to the non-volatile mass memory has been executed.
In the attached Drawing Figures:
Of interest to the ensuing description of the exemplary embodiments of this invention is commonly-assigned U.S. patent application Ser. No. 12/455,763, filed Jun. 4, 2009, “Apparatus and Method to Share Host System RAM with Mass Storage Memory RAM”, Olli Luukkainen, Kimmo Mylly and Jani Hyvonen (US 2010/0312947 A1), incorporated by reference herein. Before describing in detail the exemplary embodiments of this invention it will be useful to review at least a portion of the description of this commonly-assigned U.S. patent application Ser. No. 12/455,763.
As is stated, at present most mass storage memories provide LBA-based access, e.g., eMMC and different types of external memory cards such as SD. However, it may also be the case that the entire file system (FS) SW is embedded in the mass storage memory device.
When a mass storage memory is used in a high volume consumer device, such as a mobile wireless communication device, one important consideration is cost, and one factor affecting the cost is the amount of RAM in the mass storage memory device itself.
Another important consideration is performance. The overall performance depends on many factors. For example, for lengthy (time consuming) operations (in particular if the mass storage memory device contains an entire file system SW) there would be an advantage to include a substantial amount of RAM in the mass storage memory device. However, this can have a negative impact on the cost.
It may be the case that system context (metadata) would be stored in the flash memory of the mass storage memory device. However, this approach has several associated disadvantages. For example, repeatedly writing the system context (metadata) to the mass storage memory device raises wearing issues that can impact the usable life of the mass storage memory device. Also, writing data to the flash memory can be a relatively slow process.
Another important consideration is the power efficiency. To provide good power efficiency the mass storage memories are preferably shutdown (powered-off) when not needed (meaning also that the internal RAM of the device is preferably shutdown as well). However, and assuming that the RAM is volatile in nature, then whatever data is stored in the RAM is lost when the power is removed from the RAM. To then perform re-initialization after power-up all needed information (e.g., logical-to-physical mapping information and/or file system structures) need to be restored. A full re-initialization of a LBA mass storage memory may require a substantial (and user-noticeable) amount of time (e.g., up to one second with an SD card), and entire file system initialization (if the file system is resident in the mass storage memory) may take even longer. Therefore, it is desirable to retain internal device context over the power-off/power-on cycle.
The mass storage memory 20 includes a microcontroller or, more simply, a controller 22 that is connected via at least one internal bus 27 with a volatile RAM 24, a non-volatile mass memory 26 (e.g., a multi-gigabyte flash memory mass storage) and a MSMB interface (I/F) 28. The controller 22 operates in accordance with stored program instructions. The program instructions may be stored in the RAM 24 or in a ROM or in the mass memory 26. The mass storage memory 20 may be embodied as an MMC, eMMC or a SD device, as non-limiting examples, and may be external to (plugged into) the host device 10 or installed within the host device 10. Note that the mass memory 26 may, in some embodiments, store a file system (FS) 26A. In this case then the RAM 24 may store FS-related metadata 24A, such as one or more data structures comprised of bit maps, file allocation table data and/or other FS-associated information.
The embodiments of the invention described in commonly-assigned U.S. patent application Ser. No. 12/455,763 provide a technique to share the RAM 14 of the host device 10 with the mass storage memory device 20. It can be assumed that the host device 10 (e.g., a mobile computer, a cellular phone, a digital camera, a gaming device, a PDA, etc.) has the capability to allocate and de-allocate the RAM 14. The allocation of the RAM 14 may be performed dynamically or it may be performed statically. The allocation of a portion of the RAM may be performed in response to a request received at the host device 10, or at the initiative of the host device 10.
In the embodiments of the invention described in commonly-assigned U.S. patent application Ser. No. 12/455,763 the RAM 14 allocation is provided for the mass storage memory 20 (connected via the MSMB 18 to the host CPU 12), if the mass storage memory 20 has a need to extend its own RAM 24 space and/or if the mass storage memory 20 has a need for non-volatile RAM (the contents of which are not lost when the mass storage memory 20 is powered-off). The mass storage memory 20 can also read and/or write (R/W) allocated RAM 14 in the host device 10. The allocation/de-allocation and R/W access methods can be implemented by extensions to a command set used to communicate with the mass storage memory 20 via an applicable mass storage memory protocol.
In accordance with certain embodiments of the invention described in commonly-assigned U.S. patent application Ser. No. 12/455,763 the mass storage memory device 20 is provided with a mechanism to interrupt/send a message to host device 10 to initiate an allocation of space in the RAM 14. The interrupt/message is sent over the MSMB 18, and may be considered as an extension to current command sets. Referring to
The allocation of host RAM 14 may occur differently in some embodiments. For example, the host device 10 may allocate RAM 14 dynamically and pass a ‘pointer’ to the allocated RAM to the mass storage memory device 20. It is then up to the controller 22 of the mass storage memory device 20 how to utilize the allocated host RAM 14. Note that in this embodiment an explicit allocation request from the mass storage memory device 20 may not be sent to the host device 10. Instead, the host device 10 may on its own initiative allocate a portion of the RAM 14, such as when it first detects the presence of the mass memory storage device 20. Of course, subsequent signaling between the mass storage memory device 20 and the host device 10 may be used to change the size of the allocated RAM 14 if the initial allocation is not sufficient for the needs of the controller 22. As another example of RAM 14 allocation, a portion of the RAM 14 may be allocated by the host 10 in a static manner and the mass storage memory device 20 then simply uses the same portion of the RAM 14 each time it needs to extend the RAM 24. In this case the mass storage memory device 20 may already have knowledge of the location/size of the allocated RAM 14, and a pointer is not needed to be sent from the host device 10.
Note that while it may typically be the case that the mass storage memory device 20 will receive an allocation of host memory to store contents of the volatile RAM 24, in general the allocation may be for storing data for any read/write memory contained within the mass storage memory device 20.
Having thus provided an overview of various non-limiting and exemplary embodiments of the invention described in the commonly-assigned U.S. patent application Ser. No. 12/455,763, a description is now made of the exemplary embodiments of this invention.
In a managedNAND memory (e.g., eMMC, SSD, UFS, microSD) the memory controller (such as the controller 22 shown in
In a modern multi-tasking mobile device environment there may be many different types of accesses from different initiators that are queued to the mass storage at the same time. In the eMMC case the queued accesses would be served one-by-one since the eMMC device can only handle a single thread at a time. In the UFS case it is possible to send the accesses/commands to a UFS device queue. However, even though the commands can be queued this does not ensure that there necessarily are sufficient resources at the UFS device (especially in a cost optimized model) to immediately accommodate all of the data related to the queued accesses.
The above described commonly-assigned U.S. patent application Ser. No. 12/455,763 provides a model in which the mass storage memory 20 is provided with read/write access to the system DRAM 14. This concept is extended by the exemplary embodiments of this invention to enable the mass storage memory 20 to move data within the system DRAM, either logically (by the use of pointers) or physically. The actual move could occur within the DRAM 14 or the data could travel back and forth over the system DRAM bus 17 (e.g., between the system DRAM 14 and a Mass Memory Host Controller 13 DMA buffer). The Mass Memory Host Controller 13, shown in
The foregoing features can be realized through the use of several embodiments.
In a first embodiment a separate physical address space in the system DRAM 14 is reserved for the mass storage memory 20, or a logical space is reserved if the system DRAM 14 operates in a logical address space. The mass storage memory (such as the UFS mass storage memory 20) can utilize this address space freely, and is responsible for the management functions of this address space such as allocation/deallocation functions and other functions.
Reference can be made to
Assume that a source of data such as an application 14B, or a file system cache, or a file cache entity (as non-limiting examples) has data 14B′ to be stored into the mass memory module 20. The data 14B′ is moved to the transfer buffer 14D as the transfer data 14E by a file system/driver for subsequent delivery to the mass memory module 20. Optionally, as shown in
In accordance with an aspect of this invention the data 14E can be moved to the portion 14G allocated for the mass memory (UFS) to become the data 14E′ as described below.
An access may take place by the host device 10 as follows (it is assumed that the host device 10 has already correctly initiated the mass storage memory 20).
Note that the UFS memory module 20 need not process received access commands sequentially. For example, before processing the write command from the second initiator if another write command having a higher indicated priority arrives from a third initiator, where the write data has also been stored in the allocated portion 14G, the UFS memory module 20 could process the write command from the third initiator and then process the write command from the second initiator.
In another embodiment there need be no specific separate memory addresses reserved in the system DRAM 14 for the mass memory module 20. Instead the mass memory module 20 can have access to any (or almost any) location in the system DRAM 14. In this case instead of moving data physically in the system DRAM 14 the mass memory module 20 can control a list of memory pointers created by the host CPU 12. By modifying the lists of pointers (one list of pointers for host and another for the mass memory module 20) the mass memory module 20 can virtually “move” data from host CPU-controlled logical memory space to space controlled by the mass memory module 20. Note that in this case the transfer buffer 14D will/may still be present, however there is no need for the physical portion 14G allocated for the mass memory module 20.
Alternatively, the mass memory module 20 can use header information created by the host CPU 12. The header information can use separate bits for host valid/invalid, reserved and mass memory valid/invalid allocations and can be stored in the system
DRAM 14. In this case by modifying the header information (e.g., allocated/free) the mass memory module 20 could “move” data from host CPU 12 controlled addresses to addresses controlled by the mass memory module 20.
Either or both of these operations can be mediated by the host controller 13.
There can be a read command from an application to an address of which the data is still located in an area of the system DRAM 14 reserved for the non-volatile module 14G, i.e. the data stored earlier was not yet moved by the non-volatile memory module 20 to the non-volatile memory 26. In such a case the memory controller 22 would detect this read command and (instead of truly moving and storing the data into the non-volatile memory 26 before serving the read command) would copy (not move) the data from the memory area 14G to the location reserved for it in the file cache/file system cache 14C/transfer buffer 14D or application area 14B. The target address in the reserved location could be included in the access list 14F so that it is at least known by the host controller 13.
In the alternative embodiments related to modification of lists of pointers and headers correspondingly the above mentioned read case would be realized as copying of the data (related to the read request) to new target location and preparing the pointers/headers by the memory module/host controller so that the copied data will be immediately accessible and under control of the host system 10.
In all of these read-related embodiments it should be appreciated that the read data is copied and the data 14E′ still exists and will be written to the non-volatile memory 26 of the memory module 20 as soon as the memory controller 22 has the resources to perform the write operation.
It is also within the scope of the exemplary embodiments for the data to originate from a source other than the host CPU 12 (e.g., other than from the application processor). For example, the data could originate from an initiator that is embodied as a peripheral device 60 that is connected to the host device 10 either by a wired connection or a wireless connection. In this case instead of moving the data directly from the peripheral device 60 to the mass memory module 20 the mass memory module 20 causes the data to be stored in the system DRAM 14, such as in the allocated portion 14G. The data that originates at the peripheral device 60 is thus temporarily (and transparently to the peripheral device 60) cached in the system DRAM 14.
Further in accordance with the exemplary embodiments of this invention the operation of the DRAM 14 is enhanced by providing additional functionality to locally perform operations in the DRAM 14.
By way of introduction, one of the processing capacity bottlenecks of a wireless terminal (e.g., a smartphone) is the execution memory, especially the memory bus. While the execution memory, also known as work memory, is a separate component (e.g., external DRAM), it is connected to a host ASIC by a bus that has a limited transfer capacity (bandwidth).
The means of improving the memory bandwidth are limited. In most cases the cost of several memory buses and memory devices respectively is too high. Also the bus itself takes a major portion of the pin/pad budget of the host ASIC, meaning that at a given pad pitch, the package containing the host ASIC silicon, would became unnecessarily large with multiple buses, creating problems for miniaturization efforts.
A common problem is that the memory bandwidth, or transaction capacity, is not sufficient for many desired use cases. The host ASIC can contain several memory masters (e.g., processors, accelerators, and DMA engines) and one DRAM controller 11 that provides the external memory bus. At the other end of this bus is the memory device (DRAM component). Memory masters issue read and write requests to the memory. The DRAM is built of several banks (4 to 8 typically) that could operate in parallel. Thus, while parallelism exists at both ends of the memory bus, the memory bus itself is a single-threaded arrangement.
The exemplary aspects of this invention include providing a means to fully exploit the memory device by enabling internal operations to be performed locally. This improves the overall memory subsystem performance, enhances CPU capacity by off-loading some operations from the CPU, and decreases power consumption. The exemplary aspects of this invention provide for some elementary operations to be carried out by the memory device through the use of an enhanced command set.
It should be noted that while these aspects of the exemplary embodiments of this invention are described in the context of enhancing the operation of the system DRAM 14 these aspects of the invention could be applied as well to enhancing the operation of the mass storage memory device 20. For example, the resident memory controller 22 of the memory module 20 could be modified so as to provide local autonomous operations within the mass memory 26.
Further in accordance with the exemplary embodiments of this invention some basic, elemental functions such as memory fill and memory move are handled internally to the system DRAM 14. These operations can be handled, for example, by system DRAM memory controller and/or by a control module or function 15 embedded in the system DRAM 14 (shown in
In this embodiment the system DRAM 14, with or without assistance of the system DRAM memory controller 11, can perform, for example, memory fill and move function(s) independently and autonomously without active intervention/control by the CPU 12. This reduces the CPU 12 and system DRAM bus 17 load due at least to the mass memory module/mass memory host controller 13 controlling the physical data move from the buffer 14B/14D to the allocated portion 14G.
It can be shown that many memory transactions are simple fills of memory address space with data (e.g., zeroes) or moving data chunks from one memory location into another memory location, where the locations can be inside the same physical memory device or inside two different memory devices. This conventionally requires activity at the CPU level, which means that system processing efficiency and power consumption are less than optimal.
In accordance with these further aspects of the invention at least some elementary in-memory operations are removed from the responsibility of the CPU 12 and transferred to the system DRAM 14, such as to the controller 15 that can be embedded with the system DRAM 14.
In this exemplary embodiment the controller 15 functionality includes “smart memory” features. Alternatively these smart memory features could be included in the host controller 13 shown in
In these embodiments system DRAM controller 11 or the system DRAM module (control 15) may analyse commands issued for it and initiate a “smart memory” operation such as a memory fill or a memory move operation.
The additional functionality added to, for example, the controller 15 may include at least a data fill feature and a data move feature. The additional functionality optionally includes security functions for the system DRAM 14, pack/unpack functions between the system DRAM 14 and the mass memory 26, data transaction functions between the DRAM 14 and the mass memory 26 and, for example, mass memory related error detection and correction and wear levelling functions.
As an example of the use of the smart memory features, a memory block reset is typically performed by writing zeroes into the memory. In accordance with this invention a ‘data fill’ function supports at least a Fill operation using any specified bit pattern (e.g., Fill: Data), where Data could be any specified bit pattern such as ‘0000’, ‘1111’, ‘0101’, etc.
A Data Move operation can be a read/write (copy) sequence followed by a Fill operation to the region of system DRAM 14 from which the data was moved (read). The Fill operation could set the memory locations to some predetermined value (e.g., zero), or to a value specified by the Data Move operation command in a manner analogous to the Fill operation discussed above. The Fill operation can be optional, so that the Data Move operation functions as a Data Copy operation (i.e., the data is replicated to another location while the original data is unchanged).
A non-limiting example of the implementation of the smart memory functionality at the command level is as follows.
Low Power Double Data Rate (LPDDR1 (LP1)) as specified by JEDEC (e.g., see JEDEC Standard, Low Power Double Data Rate (LPDDR) SDRAM Standard, JESD209B, February 2010) does not provide a straightforward method of adding these types of smart features. An exemplary implementation may be to use a sequence of multiple conventional Mode Register Set (MRS) commands (MRS with bank address BA0=BA1=‘1’ is available) using address pins A0-Ax to deliver the following information to the memory device:
The number of needed MRS commands in the sequence is dependent on addressing, data granularity and operation size, which are implementation dependant factors. In LPDDR2 and LPDDR3 Mode Register Write commands could be used in a similar fashion. It should be understood that conventional LPDDRx solutions allow MRS and MRW operation only from DRAM idle state, but other solutions are technically possible. One such implementation may be to require that the data which is the source or the target of the move or fill operation is located in open DRAM pages.
It should also be understood that with future DRAM interfaces new dedicated commands are possible to be used for these operations—and these operations could be made much more efficient. Fundamentally from a system viewpoint some or all of the following operations may be delivered to the memory:
Full command and address delivery may occur in a single or a multiple command sequence—a certain command code may, for example, indicate multicycle delivery via cmd/address/data bus as suits the implementation. The particular implementation dictates the requirements for the data and address granularity, requirements for the memory state prior to and after the smart memory feature operations, whether the memory internally uses some intermediate buffers to move or fill the data or whether open rows (DRAM sense amps) are reserved during the entire operation, any timing requirements, etc.
The command could be issued by the, for example, the CPU 12 (e.g., by an OS or file system utility) or by the mass memory host controller 13 and then autonomously executed by the controller 15 associated with the system DRAM 14 or by the DRAM memory controller 11.
There are a number of advantages and technical effects that can be realized by the use of the exemplary embodiments of this invention as described above. For example, a faster response to the initiator (Host CPU 12, a process, an initiator, an application) can be provided that an access has been served (actually taken into the ManagedNAND side for further processing). This enhances the user experience. Further, no changes need be made to the upper layer software (e.g., file system and/or applications) as moving data occurs at a low layer of hardware/software. Further, the allocated portion 14G of the system DRAM 14 for the ManagedNAND functionality need not be excessively large. For example, for many uses of interest some hundreds of kilobytes could be sufficient as in many cases it would be most advantageous to temporarily cache just small random accesses, and to store directly to the non-volatile mass memory 26 larger sequential accesses. In addition, the use of the further aspects of this invention enable certain elementary memory operations (e.g., fill, move) to be accomplished locally (e.g., within the system DRAM module 14) without active involvement of the CPU 12.
As an example, image data captured by the camera 128 can be stored in the mass storage memory device 20 under control of a camera application and can thus benefit from the use of the embodiments of this invention. As another example, audio data captured by the microphone 124 can be stored in the mass storage memory device 20 under control of an audio application and can thus also benefit from the use of the embodiments of this invention.
Within the sectional view of
Signals going to and from the camera 128 may pass through an image/video processor 44 that encodes and decodes the various image frames. A separate audio processor 46 may also be present controlling signals to and from the speakers 34 and the microphone 124. The graphical display interface 120 is refreshed from a frame memory 48 as controlled by a user interface chip 50 which may process signals to and from the display interface 20 and/or additionally process user inputs from the keypad 22 and elsewhere.
Certain embodiments of the UE 10 may also include one or more secondary radios such as a wireless local area network radio WLAN 37 and a Bluetooth® radio 39, which may incorporate an antenna on the chip or be coupled to an antenna off the chip. Throughout the apparatus are various memories such as random access memory RAM, which can include the system DRAM 14, read only memory ROM 45, and in some embodiments removable memory such as the illustrated memory card 20 on which various programs and data may be stored. All of these components within the UE 10 are normally powered by a portable power supply such as a battery 49.
The processors 38, 40, 42, 44, 46, 50, if embodied as separate entities in a UE 10, may operate in a slave relationship to the main processor (CPU) 12, which may then be in a master relationship to them. Certain embodiments may be disposed across various chips and memories as shown, or disposed within another processor that combines some of the functions described above for
In this exemplary embodiment the CPU 12 of the UE 10 (the host device) operates with the memory card 20 (the mass storage memory device) as described above with respect to
An aspect of the exemplary embodiments of this invention is a method, shown in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
In the method described in
The various blocks shown in
In general, the various exemplary embodiments may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. For example, some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device, although the invention is not limited thereto. While various aspects of the exemplary embodiments of this invention may be illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.
It should thus be appreciated that at least some aspects of the exemplary embodiments of the inventions may be practiced in various components such as integrated circuit chips and modules, and that the exemplary embodiments of this invention may be realized in an apparatus that is embodied as an integrated circuit. The integrated circuit, or circuits, may comprise circuitry (as well as possibly firmware) for embodying at least one or more of a data processor or data processors, a digital signal processor or processors, baseband circuitry and radio frequency circuitry that are configurable so as to operate in accordance with the exemplary embodiments of this invention.
An apparatus in accordance with the exemplary embodiments includes means for sending a command from a host device to a controller associated with a non-volatile mass memory coupled with the host device, the command being a command to write data that is currently resident in a memory of the host device to the non-volatile mass memory; means for moving the data that is currently resident in the memory of the host device from an original location to a portion of the memory allocated for use by the non-volatile mass memory; and means for acknowledging to the host device that the command to write the data to the non-volatile mass memory has been executed.
Various modifications and adaptations to the foregoing exemplary embodiments of this invention may become apparent to those skilled in the relevant arts in view of the foregoing description, when read in conjunction with the accompanying drawings. However, any and all modifications will still fall within the scope of the non-limiting and exemplary embodiments of this invention.
It should be noted that the terms “connected,” “coupled,” or any variant thereof, mean any connection or coupling, either direct or indirect, between two or more elements, and may encompass the presence of one or more intermediate elements between two elements that are “connected” or “coupled” together. The coupling or connection between the elements can be physical, logical, or a combination thereof. As employed herein two elements may be considered to be “connected” or “coupled” together by the use of one or more wires, cables and/or printed electrical connections, as well as by the use of electromagnetic energy, such as electromagnetic energy having wavelengths in the radio frequency region, the microwave region and the optical (both visible and invisible) region, as several non-limiting and non-exhaustive examples.
Furthermore, some of the features of the various non-limiting and exemplary embodiments of this invention may be used to advantage without the corresponding use of other features. As such, the foregoing description should be considered as merely illustrative of the principles, teachings and exemplary embodiments of this invention, and not in limitation thereof.
Number | Name | Date | Kind |
---|---|---|---|
5586291 | Lasker et al. | Dec 1996 | A |
5701516 | Cheng et al. | Dec 1997 | A |
5802069 | Coulson | Sep 1998 | A |
6067300 | Baumert et al. | May 2000 | A |
6513094 | Magro | Jan 2003 | B1 |
6665747 | Nazari | Dec 2003 | B1 |
6842829 | Nichols et al. | Jan 2005 | B1 |
7136963 | Ogawa et al. | Nov 2006 | B2 |
7181574 | Lele | Feb 2007 | B1 |
7321958 | Hofstee et al. | Jan 2008 | B2 |
7480749 | Danilak | Jan 2009 | B1 |
7571295 | Sakarda et al. | Aug 2009 | B2 |
8190803 | Hobson et al. | May 2012 | B2 |
20020108014 | Lasser | Aug 2002 | A1 |
20030028737 | Kaiya et al. | Feb 2003 | A1 |
20030137860 | Khatri et al. | Jul 2003 | A1 |
20040010671 | Sampsa et al. | Jan 2004 | A1 |
20040203670 | King et al. | Oct 2004 | A1 |
20040221124 | Beckert et al. | Nov 2004 | A1 |
20050010738 | Stockdale et al. | Jan 2005 | A1 |
20050071570 | Takasugl et al. | Mar 2005 | A1 |
20050097280 | Hofstee et al. | May 2005 | A1 |
20060041888 | Radulescu et al. | Feb 2006 | A1 |
20060069899 | Schoinas et al. | Mar 2006 | A1 |
20060075147 | Schoinas et al. | Apr 2006 | A1 |
20060075395 | Lee et al. | Apr 2006 | A1 |
20070088867 | Cho et al. | Apr 2007 | A1 |
20070234006 | Radulescu et al. | Oct 2007 | A1 |
20070283078 | Li et al. | Dec 2007 | A1 |
20080127131 | Gao et al. | May 2008 | A1 |
20080162792 | Wu et al. | Jul 2008 | A1 |
20080228984 | Yu et al. | Sep 2008 | A1 |
20080281944 | Vorne et al. | Nov 2008 | A1 |
20090106503 | Lee et al. | Apr 2009 | A1 |
20090157950 | Selinger | Jun 2009 | A1 |
20090182940 | Matsuda et al. | Jul 2009 | A1 |
20090182962 | Khmelnitsky et al. | Jul 2009 | A1 |
20090198871 | Tzeng | Aug 2009 | A1 |
20090198872 | Tzeng | Aug 2009 | A1 |
20090216937 | Yasufuku | Aug 2009 | A1 |
20090222629 | Yano et al. | Sep 2009 | A1 |
20090307377 | Anderson et al. | Dec 2009 | A1 |
20090327584 | Tetrick et al. | Dec 2009 | A1 |
20100005281 | Buchmann et al. | Jan 2010 | A1 |
20100037012 | Yano et al. | Feb 2010 | A1 |
20100100648 | Madukkarumukumana et al. | Apr 2010 | A1 |
20100106886 | Marcu et al. | Apr 2010 | A1 |
20100115193 | Manus et al. | May 2010 | A1 |
20100161882 | Stern et al. | Jun 2010 | A1 |
20100169558 | Honda et al. | Jul 2010 | A1 |
20100172180 | Paley et al. | Jul 2010 | A1 |
20100250836 | Sokolov et al. | Sep 2010 | A1 |
20100293420 | Kapil et al. | Nov 2010 | A1 |
20100312947 | Luukkainen et al. | Dec 2010 | A1 |
20110082967 | Deshkar et al. | Apr 2011 | A1 |
20110087804 | Okaue et al. | Apr 2011 | A1 |
20110208914 | Winokur | Aug 2011 | A1 |
20110264860 | Hooker et al. | Oct 2011 | A1 |
20110296088 | Duzly et al. | Dec 2011 | A1 |
20120102268 | Smith et al. | Apr 2012 | A1 |
20120131263 | Yeh | May 2012 | A1 |
20120131269 | Fisher et al. | May 2012 | A1 |
20120151118 | Flynn et al. | Jun 2012 | A1 |
20120210326 | Torr et al. | Aug 2012 | A1 |
20130138840 | Kegel et al. | May 2013 | A1 |
20130145055 | Kegel et al. | Jun 2013 | A1 |
20140068140 | Mylly | Mar 2014 | A1 |
20150039819 | Luukkainen et al. | Feb 2015 | A1 |
20160077963 | Luukkainen et al. | Mar 2016 | A1 |
Number | Date | Country |
---|---|---|
101887350 | Nov 2010 | CN |
1091283 | Apr 2001 | EP |
1094392 | Apr 2001 | EP |
WO9965193 | Dec 1999 | WO |
Entry |
---|
Tanenbaum, A. “Structured Computer Organization”, 1984, Prentice-Hall, Inc., p. 11. |
Lin, et al., “A NAND Flash Memory Controller for SD/MMC Flash Memory Card”, IEEE Transactions of Magnetics, vol. 43, No. 2, (Feb. 2, 2007), (pp. 933-935). |
Apostolakis, et al., “Software-Based Self Testing of Symmetric Shared-Memory Multiprocessors”, IEEE Transactions on Computers, vol. 58, No. 12, Dec. 2009, 13 pages. |
“Embedded MultiMediaCard (eMMC) Mechanical Standard”, JESD84-C43, JEDEC Standard, JEDEC Solid State Technology Association, Jun. 2007, 13 pages. |
The PCT Search Report and Written Opinion mailed Sep. 5, 2013 for PCT application No. PCT/US13/37298, 9 pages. |
JEDEC Standard, “Embedded MultiMediaCard (eMMC) Product Standard, High Capacity,” JESD84-A42, Jun. 2007, 29 pages. |
JEDEC Standard, “Embedded ZmultiMediaCard(eMMC) eMMC/Card Product Standard, high Capacity, Including Reliable Write, Boot, and Sleep Modes,” (MMCA, 4.3), JSEDD84-A43, Nov. 2007, 166 pages. |
JEDEC Standard, “Embedded MultiMediaCard (eMMC) Mechanical Standard,” JESD84-C43, Jun. 2007, 13 pages. |
Li, et al., “A Method for Improving Concurrent Write Performance by Dynamic Mapping Virtual Storage System Combined with Cache Management”, 2011 IEEE 7th International Conference of Parallel Distributed System, Dec. 7-8, 2011, 10 pages. |
Numonyz, “How to boot an embedded system from an eMMCTM equipped with a Microsoft FAT file system.” Application note AN2539, Nov. 2008, pp. 1-25. |
Office Action for U.S. Appl. No. 14/520,030, mailed on Dec. 4, 2014, Olli Luukkainen, “Apparatus and Method to Share Host System RAM with Mass Storage Memory RAM”, 6 pages. |
Office Action for U.S. Appl. No. 13/596,480, mailed on Mar. 13, 2014, Kimmo J. Mylly, “Dynamic Central Cache Memory”, 15 pages. |
Office action for U.S. Appl. No. 12/455,763, mailed on Mar. 14, 2014, Luukkainen et al., “Apparatus and method to share host system ram with mass storage memory ram”, 6 pages. |
Office action for U.S. Appl. No. 12/455,763, mailed on Aug. 1, 2013, Luukkainen et al., “Apparatus and method to share host system ram with mass storage memory ram”, 28 pages. |
The PCT Search Report and Written Opinion mailed Mar. 6, 2014 for PCT application No. PCT/US13/56980, 11 pages. |
Final Office Action for U.S. Appl. No. 14/520,030, mailed on May 20, 2015, Olli Luukkaninen, “Apparatus and Method to Share Host System RAM with Mass Storage Memory RAM”, 6 pages. |
The Extended European Search Report mailed Aug. 31, 2015 for European patent application No. 13741595.6, a counterpart foreign application of U.S. Appl. No. 13/358,806, 8 pages. |
Office action for U.S. Appl. No. 14/945,757, mailed on Jan. 21, 2016, Luukkainen et al., “Apparatus and Method to Share Host System RAM with Mass Storage Memory RAM”, 16 pages. |
Chinese Office Action mailed Mar. 28, 2016 for Chinese Patent Application No. 201380006769.8, a counterpart foreign application of U.S. Appl. No. 13/358,806, 36 pages. |
Number | Date | Country | |
---|---|---|---|
20130198434 A1 | Aug 2013 | US |