CN101740102B - Multichannel flash memory chip array structure and writing and reading method thereof - Google Patents
Multichannel flash memory chip array structure and writing and reading method thereof Download PDFInfo
- Publication number
- CN101740102B CN101740102B CN200810232221.6A CN200810232221A CN101740102B CN 101740102 B CN101740102 B CN 101740102B CN 200810232221 A CN200810232221 A CN 200810232221A CN 101740102 B CN101740102 B CN 101740102B
- Authority
- CN
- China
- Prior art keywords
- data
- channel
- flash memory
- writing
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000007547 defect Effects 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 19
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
Images
Landscapes
- Read Only Memory (AREA)
- Memory System (AREA)
Abstract
Description
技术领域 technical field
本发明涉及一种存储设备及其写入和读出方法,尤其涉及一种多通道闪存阵列结构及其写入和读出方法。The invention relates to a storage device and its writing and reading method, in particular to a multi-channel flash memory array structure and its writing and reading method.
背景技术 Background technique
闪存作为一种新的非易失性存储介质,以其存储密度大、携带方便、功耗低、掉电数据保持时间长及抗震性好等诸多优点,已经在消费类电子领域非常普及。在工业及军工领域,也越来越受到重视和欢迎。在一些大容量数据存储应用场合,往往会有多片闪存级联或者组成整列使用,以扩大存储空间和提高数据的吞吐量。但是,由于闪存在写入数据后需要进行较长时间的等待,以确保数据正确写入。典型的,一次写入需要等待200us,最大等待时间需要700us。如果按照正常的操作思路,向闪存中写入数据后就进行等待,数据的写入速度会很慢,无法满足实际的使用要求。As a new non-volatile storage medium, flash memory has become very popular in the field of consumer electronics due to its many advantages such as high storage density, easy portability, low power consumption, long data retention time after power failure, and good shock resistance. In the field of industry and military industry, it has also received more and more attention and popularity. In some large-capacity data storage applications, there are often multiple slices of flash memory cascaded or used as an entire column to expand storage space and improve data throughput. However, since the flash memory needs to wait for a long time after data is written, to ensure that the data is written correctly. Typically, a write needs to wait for 200us, and the maximum waiting time needs to be 700us. If you follow the normal operation idea and wait after writing data into the flash memory, the data writing speed will be very slow, which cannot meet the actual use requirements.
在闪存使用过程中,最核心的问题就是如何提高数据的读写速度。现有的解决方法中,有的也采用多个通道,但是公用一条数据总线,并不能实现真正意义上的并行处理;有的虽然使用多条总线,而且每条总线上也挂了多个芯片,但是每个通道上多个芯片的管理方法不能并行处理,并且为了取得高的写入速度牺牲每个芯片的写入等待时间,这样会使得数据的不可靠性降低。In the process of using flash memory, the core issue is how to improve the read and write speed of data. Among the existing solutions, some also use multiple channels, but share one data bus, which cannot realize parallel processing in the true sense; some use multiple buses, and each bus also has multiple chips , but the management method of multiple chips on each channel cannot be processed in parallel, and the write waiting time of each chip is sacrificed in order to achieve high write speed, which will reduce the unreliability of data.
发明内容 Contents of the invention
本发明的目的为解决现有技术中闪存数据读写速度慢的缺陷,提供了一种能够快速读写闪存的多通道闪存阵列结构及其写入和读出方法。The purpose of the present invention is to solve the defect of slow read and write speed of flash memory data in the prior art, and provide a multi-channel flash memory array structure capable of fast reading and writing of flash memory and its writing and reading methods.
本发明的技术解决方案为:Technical solution of the present invention is:
一种多通道闪存芯片阵列结构,包括由闪存芯片,以及与闪存芯片连接的数据总线、控制总线和片选信号线构成的多个通道,每个闪存芯片对应与一条片选信号线,每个芯片的片选信号线独立,其特殊之处是:A multi-channel flash memory chip array structure, including a plurality of channels composed of flash memory chips, a data bus connected to the flash memory chip, a control bus, and a chip selection signal line, each flash memory chip corresponds to a chip selection signal line, each The chip selection signal line of the chip is independent, and its special features are:
所述每个通道内的所有闪存芯片公用独立于其他通道的一条数据总线和一条控制总线。All the flash memory chips in each channel share a data bus and a control bus independent of other channels.
一种多通道闪存芯片阵列结构,包括由闪存芯片,以及与闪存芯片连接的数据总线、控制总线和片选信号线构成的多个通道,其特殊之处是:A multi-channel flash memory chip array structure, including a plurality of channels composed of flash memory chips, a data bus connected to the flash memory chip, a control bus, and a chip selection signal line, and its special features are:
所述每个通道内的闪存芯片构成闪存芯片组后进行级联,每个通道内的所有芯片组公用独立于其他通道的一条数据总线和一条控制总线,每个芯片组内的芯片共用控制总线和片选信号总线。The flash memory chips in each channel are cascaded after forming a flash memory chipset, and all chipsets in each channel share a data bus and a control bus independent of other channels, and the chips in each chipset share a control bus and chip select signal bus.
一种多通道闪存写入方法,其特殊之处是:A multi-channel flash memory writing method whose special features are:
包含以下步骤:Contains the following steps:
1]、命令和地址信息通过命令接口传递给命令解析单元;1], command and address information are passed to the command parsing unit through the command interface;
2]、命令解析单元收到命令后,启动数据接口管理单元和通道仲裁单元;2], after the order analysis unit receives the order, start the data interface management unit and the channel arbitration unit;
3]、通道仲裁单元根据地址信息进行通道的预分配,同时,数据接口管理单元从数据接口读取数据,并传递给通道仲裁单元和数据分配单元;3], the channel arbitration unit performs channel pre-allocation according to the address information, and at the same time, the data interface management unit reads data from the data interface and passes it to the channel arbitration unit and the data allocation unit;
4]、数据分配单元根据已经分配好的通道号,直接把数据写入对应通道的缓存中,然后由通道仲裁单元启动相应的闪存时序产生模块;4], the data allocation unit directly writes the data into the cache of the corresponding channel according to the allocated channel number, and then the channel arbitration unit starts the corresponding flash memory timing generation module;
5]、相应的闪存时序产生模块将该通道缓存中的数据通过与闪存时序产生模块相应的通道内的公用控制总线和数据总线写入闪存芯片中;5], the corresponding flash memory sequence generation module writes the data in the channel cache into the flash memory chip through the common control bus and data bus in the channel corresponding to the flash memory sequence generation module;
6]、当数据分配单元完成向某通道的数据写入后,仍有数据需要写入时,又可重新从数据接口读取数据,进行下一周期的数据写入工作。6], when the data distribution unit finishes writing data to a certain channel, when there is still data to be written, it can read data from the data interface again, and carry out the data writing work of the next cycle.
上述步骤5]中的数据写入闪存芯片时,包含以下步骤:When the data in the above step 5] is written into the flash memory chip, the following steps are included:
501]、占用相应通道内的第一个闪存芯片的实际写入时间,通过控制总线和数据总线向第一个闪存芯片写入数据;501], occupying the actual writing time of the first flash memory chip in the corresponding channel, and writing data to the first flash memory chip through the control bus and the data bus;
502]、利用当前闪存芯片的写入等待时间,通过控制总线和数据总线向下一个闪存芯片继续写入数据;502], using the write waiting time of the current flash memory chip, continue to write data to the next flash memory chip through the control bus and the data bus;
503]、依照步骤502]的方式,利用第一个闪存芯片的写入等待时间,依次向通道内剩余的闪存芯片写入数据,直至通道内的所有闪存芯片完成一次数据写入工作,而且当第一个闪存芯片的写入等待时间结束,相应通道内的写入操作一周完毕;503], according to the method of step 502], using the write waiting time of the first flash memory chip, write data to the remaining flash memory chips in the channel in turn, until all the flash memory chips in the channel complete a data writing work, and when The write waiting time of the first flash memory chip is over, and the write operation in the corresponding channel is completed for one week;
504]、返回步骤501],重新向第一个闪存芯片写入数据,开始新的写入操作,直至所有数据的写入工作完成,循环写入操作完毕。504], return to step 501], re-write data to the first flash memory chip, and start a new write operation until all data writing is completed, and the cycle writing operation is completed.
一种多通道闪存读出方法,其特殊之处是:A multi-channel flash memory readout method, its special features are:
包含以下步骤:Contains the following steps:
1]、命令和地址信息通过命令接口传递给命令解析单元;1], command and address information are passed to the command parsing unit through the command interface;
2]、命令解析单元分析数据和命令后,将转换后的命令和地址传递给通道仲裁和数据分配单元;2], after the command analysis unit analyzes the data and commands, the converted commands and addresses are passed to the channel arbitration and data allocation unit;
3]、通过通道仲裁单元单元启动相应通道的时序产生模块;3], start the timing generation module of the corresponding channel through the channel arbitration unit;
4]、通过与闪存时序产生模块相应的通道内的公用控制总线和数据总线从闪存芯片中读取数据,存放在相应的通道缓存中;4], read data from the flash memory chip through the common control bus and data bus in the channel corresponding to the flash memory timing generation module, and store it in the corresponding channel cache;
5]、通知通道仲裁和数据分配单元,从通道缓存中读取数据,通过数据接口管理单元将数据传给数据接口;5], notify the channel arbitration and data allocation unit, read data from the channel buffer, and pass the data to the data interface through the data interface management unit;
6]、当数据接口管理单元完成向某通道的数据读出后,仍有数据需要读出时,又可重新从闪存芯片读取数据,进行下一周期的数据读出工作。6] When the data interface management unit completes the data readout to a certain channel, when there is still data to be read out, it can read data from the flash memory chip again, and perform the data readout work of the next cycle.
本发明具有如下优点:The present invention has the following advantages:
1、闪存读写速度快。速度上可以突破闪存的理论读写速度,使得闪存的读写速度可以达到100MByte/s以上,以适应高速接口标准如PATA、SATA、1394和USB等。1. The read and write speed of flash memory is fast. In terms of speed, it can break through the theoretical reading and writing speed of flash memory, so that the reading and writing speed of flash memory can reach more than 100MByte/s, so as to adapt to high-speed interface standards such as PATA, SATA, 1394 and USB.
2、数据写入可靠性高。可靠性上严格按照闪存操作的时间特性进行等待,确保在各种条件下数据的可靠性。2. High reliability of data writing. In terms of reliability, wait strictly according to the time characteristics of flash memory operation to ensure the reliability of data under various conditions.
附图说明 Description of drawings
图1为本发明闪存芯片阵列结构示意图。FIG. 1 is a schematic diagram of the structure of a flash memory chip array according to the present invention.
图2为本发明每个通道内的闪存芯片读写流水线管理方法示意图。FIG. 2 is a schematic diagram of the method for managing the read-write pipeline of the flash memory chip in each channel of the present invention.
图3为本发明闪存芯片数据读写流程图。Fig. 3 is a flowchart of reading and writing data of a flash memory chip according to the present invention.
具体实施方式 Detailed ways
本发明的一种多通道闪存芯片阵列结构,包括由闪存芯片,以及与闪存芯片连接的数据总线、控制总线和片选信号线构成的多个通道,每个闪存芯片对应与一条片选信号线,每个芯片的片选信号线独立,每个通道内的所有闪存芯片公用独立于其他通道的一条数据总线和一条控制总线。当每个通道内的闪存芯片构成闪存芯片组级联时,芯片组由2个或者4个芯片拼合成16位或者32位宽的数据总线,构建芯片组的目的是为了增加位宽,提高单位时间内数据的写入速度,芯片组内的芯片共用控制总线和片选总线,数据总线不共用。A multi-channel flash memory chip array structure of the present invention includes a plurality of channels formed by flash memory chips, a data bus connected to the flash memory chip, a control bus, and a chip selection signal line, and each flash memory chip corresponds to a chip selection signal line , the chip select signal line of each chip is independent, and all flash memory chips in each channel share a data bus and a control bus independent of other channels. When the flash memory chips in each channel form a cascaded flash memory chipset, the chipset is composed of 2 or 4 chips into a 16-bit or 32-bit wide data bus. The purpose of building the chipset is to increase the bit width and increase the unit The writing speed of the data within the time, the chips in the chipset share the control bus and the chip select bus, and the data bus is not shared.
本发明采用多通道,即多维阵列的方法管理闪存芯片,闪存芯片阵列结构参见图1。由多片芯片组成行列结构。构成一个闪存阵列。例如,每行有16个闪存芯片,每两个闪存芯片为一对,构成16位的数据总线,每一对芯片共用控制总线和片选信号线。8对闪存芯片共享16位数据总线和公用的一条控制总线。每对芯片的片选信号线独立。这8对芯片所共享的一条16位数据总线和一条控制总线,以及每一对芯片对应的片选信号线共同构成一个独立的通道。The present invention adopts a multi-channel, that is, a multi-dimensional array method to manage flash memory chips, and the structure of the flash memory chip array is shown in FIG. 1 . A row-column structure is composed of multiple chips. Form a flash array. For example, there are 16 flash memory chips in each row, and every two flash memory chips form a pair to form a 16-bit data bus, and each pair of chips shares a control bus and a chip select signal line. Eight pairs of flash memory chips share a 16-bit data bus and a common control bus. The chip select signal lines of each pair of chips are independent. A 16-bit data bus and a control bus shared by the 8 pairs of chips, and chip select signal lines corresponding to each pair of chips together form an independent channel.
每个通道中的闪存芯片通过级联共享一条数据总线和控制总线,各芯片都有自己独立的片选信号。由于一条总线上有多个芯片,例如,可以采用如图1所示的多通道4对芯片级联,本发明的闪存芯片读写方法提高总线利用率,从而提高数据的写入速度。本发明中,每个闪存通道都有自己独立的通道缓存,独立的时序产生模块,有字节独立的数据和控制总线。这样多个通道可以并行工作,互不影响。并且,由硬件逻辑来实现数据到各个通道上的自动分配,大大的提高了数据的传输速度。整个闪存芯片阵列由多个通道构成,每增加一个通道,理论上数据的读写速度增加一倍,突破闪存的极限读写速度限制,在扩大存储容量的同时,也可以成倍的提高数据的吞吐率。阵列的通道个数可根据实际需要而定。理论上,如果控制闪存的MCU或者FPGA的资源够用,且前端没有速度瓶颈,则采用本发明所提供的方法,闪存的读写速度是没有上限的。The flash memory chips in each channel share a data bus and a control bus through cascading, and each chip has its own independent chip select signal. Since there are multiple chips on one bus, for example, multi-channel 4-pair chip cascading as shown in Figure 1 can be adopted, and the method for reading and writing flash memory chips of the present invention improves the bus utilization rate, thereby increasing the writing speed of data. In the present invention, each flash memory channel has its own independent channel cache, an independent timing generation module, and a byte-independent data and control bus. In this way, multiple channels can work in parallel without affecting each other. Moreover, the automatic allocation of data to each channel is realized by hardware logic, which greatly improves the data transmission speed. The entire flash memory chip array is composed of multiple channels. Theoretically, the reading and writing speed of data is doubled for each additional channel, which breaks through the limit of the ultimate reading and writing speed of flash memory. While expanding the storage capacity, it can also double the data throughput. Throughput rate. The number of channels of the array can be determined according to actual needs. Theoretically, if the resources of the MCU or FPGA controlling the flash memory are sufficient, and there is no speed bottleneck at the front end, then the read/write speed of the flash memory has no upper limit by adopting the method provided by the present invention.
本发明的闪存阵列管理方法基于FPGA来实现。图3中的虚线部分描述了多通道闪存阵列结构的管理单元,其中管理单元包含以下几个部分组成:命令解析单元、数据接口管理单元、通道仲裁和数据分配单元、各个通道缓存和每个通道的闪存时序产生单元。The flash memory array management method of the present invention is realized based on FPGA. The dotted line in Figure 3 describes the management unit of the multi-channel flash memory array structure, where the management unit consists of the following parts: command analysis unit, data interface management unit, channel arbitration and data allocation unit, each channel cache and each channel Flash timing generation unit.
每个通道都对应自己独立的通道缓存和闪存时序产生单元,每个闪存时序产生单元对应一条闪存的数据总线和数据总线,即控制着一行的闪存芯片。其中,Each channel corresponds to its own independent channel buffer and flash timing generation unit, and each flash timing generation unit corresponds to a flash memory data bus and data bus, that is, it controls a row of flash memory chips. in,
命令解析单元负责和外部微控制器进行通信,对外部传入的命令进行解析,并把这些命令传递到管理单元内部的模块,是整个管理单元的命令通道。The command parsing unit is responsible for communicating with the external microcontroller, parsing the incoming commands from the outside, and passing these commands to the internal modules of the management unit, which is the command channel of the entire management unit.
数据接口管理单元负责数据的传递,类似于一个DMA控制器的功能。DMA的中文名称叫做直接内存访问,是一种不经过CPU而直接与内存进行数据交换的计算机工作模式,它的好处在于降低了CPU的负担,且大大的提高了数据交互的速度The data interface management unit is responsible for data transfer, similar to the function of a DMA controller. The Chinese name of DMA is direct memory access. It is a computer working mode that directly exchanges data with memory without going through the CPU. Its advantage is that it reduces the burden on the CPU and greatly improves the speed of data interaction.
通道仲裁和数据分配单元负责数据到通道的分配。The channel arbitration and data allocation unit is responsible for the allocation of data to the channel.
数据接口负责从外部的数据存储器或者外部数据总线上读取数据或者向外部传输数据,是整个管理单元的数据通道。数据接口可以是外部存储器,也可以是微控制器的数据总线。The data interface is responsible for reading data from the external data memory or external data bus or transmitting data to the outside, and is the data channel of the entire management unit. The data interface can be an external memory or the data bus of the microcontroller.
命令接口可以是双口RAM存储器,也可以是微控制器的控制总线。The command interface can be a dual-port RAM memory, or the control bus of a microcontroller.
本发明的一种多通道闪存写入方法,包含有以下步骤:A kind of multi-channel flash memory writing method of the present invention comprises the following steps:
1]、命令和地址信息通过命令接口传递给命令解析单元;1], command and address information are passed to the command parsing unit through the command interface;
2]、命令解析单元收到命令后,启动数据接口管理单元和通道仲裁单元;2], after the order analysis unit receives the order, start the data interface management unit and the channel arbitration unit;
3]、通道仲裁单元根据地址信息进行通道的预分配,同时,数据接口管理单元从数据接口读取数据,并传递给通道仲裁单元和数据分配单元;3], the channel arbitration unit performs channel pre-allocation according to the address information, and at the same time, the data interface management unit reads data from the data interface and passes it to the channel arbitration unit and the data allocation unit;
4]、数据分配单元根据已经分配好的通道号,直接把数据写入对应通道的缓存中,然后由通道仲裁单元启动相应的闪存时序产生模块;4], the data allocation unit directly writes the data into the cache of the corresponding channel according to the allocated channel number, and then the channel arbitration unit starts the corresponding flash memory timing generation module;
5]、相应的闪存时序产生模块将该通道缓存中的数据通过与闪存时序产生模块相应的通道内的公用控制总线和数据总线写入闪存芯片中;5], the corresponding flash memory sequence generation module writes the data in the channel cache into the flash memory chip through the common control bus and data bus in the channel corresponding to the flash memory sequence generation module;
6]、当数据分配单元完成向某通道的数据写入后,仍有数据需要写入时,又可重新从数据接口读取数据,进行下一周期的数据写入工作。6], when the data distribution unit finishes writing data to a certain channel, when there is still data to be written, it can read data from the data interface again, and carry out the data writing work of the next cycle.
图3中通道缓存左侧的几个单元协同工作,只负责数据接口上数据向空闲通道缓存的搬移;通道缓存右侧的闪存时序产生模块负责数据向本通道内各个闪存芯片的写入。这两部分并行工作,互不影响,极大的提高了数据的吞吐率。In Figure 3, several units on the left side of the channel cache work together, and are only responsible for moving data on the data interface to the idle channel cache; the flash timing generation module on the right side of the channel cache is responsible for writing data to each flash memory chip in this channel. These two parts work in parallel without affecting each other, which greatly improves the data throughput.
每个通道的时序产生模块在管理该通道的芯片时,例如通道内有4个芯片,采用图2所示的每个通道内的闪存芯片读写流水线管理方法。当写完第一个芯片组后,让当前芯片组的芯片处于写等待状态,立即转入下一个芯片的操作,并不因为第一组芯片处于写等待状态而使得总线空闲。以此类推,当循环操作一周后回到第一个芯片组,刚好第一组芯片的写等待时间完成,又可进行下一轮的操作。这种流水线管理芯片组的方法保证每个通道的数据总线总是处于繁忙状态,从而提高了每个通道的数据传输速度。而且,由于每个芯片的写等待时间得到了保证,也就保证了数据写入的可靠性。When the timing generation module of each channel manages the chips of the channel, for example, there are 4 chips in the channel, the pipeline management method for reading and writing flash memory chips in each channel as shown in FIG. 2 is adopted. After writing the first chipset, let the chips of the current chipset be in the write waiting state, and immediately transfer to the operation of the next chip, and do not make the bus idle because the first group of chips is in the write waiting state. By analogy, when the loop operation returns to the first chip set after one week, the write waiting time of the first set of chips is completed, and the next round of operation can be performed. This method of pipeline management chipset ensures that the data bus of each channel is always in a busy state, thereby improving the data transmission speed of each channel. Moreover, since the write waiting time of each chip is guaranteed, the reliability of data writing is guaranteed.
因此,在上述的步骤5]中,每个单通道的数据写入还可以包含以下步骤:Therefore, in the above step 5], the data writing of each single channel may also include the following steps:
501]、占用相应通道内的第一个闪存芯片的实际写入时间,通过控制总线和数据总线向第一个闪存芯片写入数据;501], occupying the actual writing time of the first flash memory chip in the corresponding channel, and writing data to the first flash memory chip through the control bus and the data bus;
502]、利用第一个闪存芯片的写入等待时间,通过控制总线和数据总线向下一个闪存芯片继续写入数据;502], utilizing the write waiting time of the first flash memory chip, continue to write data to the next flash memory chip through the control bus and the data bus;
503]、依照步骤502]的方式,利用当前闪存芯片的写入等待时间,依次向通道内剩余的闪存芯片写入数据,直至通道内的所有闪存芯片完成一次数据写入工作,而且当第一个闪存芯片的写入等待时间结束,相应通道内的写入操作一周完毕;503], according to the method of step 502], using the write waiting time of the current flash memory chip, write data to the remaining flash memory chips in the channel in turn, until all the flash memory chips in the channel complete a data writing work, and when the first The write waiting time of a flash memory chip is over, and the write operation in the corresponding channel is completed for one week;
504]、返回步骤501],重新向第一个闪存芯片写入数据,开始新的写入操作,直至所有数据的写入工作完成,循环写入操作完毕。504], return to step 501], re-write data to the first flash memory chip, and start a new write operation until all data writing is completed, and the cycle writing operation is completed.
所述的步骤504]的含义为,当闪存芯片完成一个周期的写入后仍然有数据需要写入时,再依照步骤501]至503]循环进行数据写入,直至所有数据的写入工作完成。这样保证了每个通道内的数据总线总是处于繁忙状态,不会由于数据的写入等待时间而降低速度,从而提高每个通道的数据传输速度。The meaning of the step 504] is that when the flash memory chip still has data to be written after one cycle of writing is completed, the data is written in a cycle according to steps 501] to 503] until the writing of all data is completed . This ensures that the data bus in each channel is always in a busy state, and the speed will not be reduced due to the waiting time for data writing, thereby increasing the data transmission speed of each channel.
本发明的闪存数据读取方法同上述的闪存数据写入方法相反。The flash memory data reading method of the present invention is opposite to the above-mentioned flash memory data writing method.
本发明的一种多通道闪存读出方法,包含有以下步骤:A kind of multi-channel flash memory reading method of the present invention comprises the following steps:
1]、命令和地址信息通过命令接口传递给命令解析单元;1], command and address information are passed to the command parsing unit through the command interface;
2]、命令解析单元分析数据和命令后,将转换后的命令和地址传递给通道仲裁和数据分配单元;2], after the command analysis unit analyzes the data and commands, the converted commands and addresses are passed to the channel arbitration and data allocation unit;
3]、通过通道仲裁单元启动相应通道的时序产生模块;3], start the timing generation module of the corresponding channel through the channel arbitration unit;
4]、通过与闪存时序产生模块相应的通道内的公用控制总线和数据总线从闪存芯片中读取数据,存放在相应的通道缓存中;4], read data from the flash memory chip through the common control bus and data bus in the channel corresponding to the flash memory timing generation module, and store it in the corresponding channel cache;
5]、通知通道仲裁和数据分配单元,从通道缓存中读取数据,通过数据接口管理单元将数据传给数据接口;5], notify the channel arbitration and data allocation unit, read data from the channel buffer, and pass the data to the data interface through the data interface management unit;
6]、当数据接口管理单元完成向某通道的数据读出后,仍有数据需要读出时,又可重新从闪存芯片读取数据,进行下一周期的数据读出工作。6] When the data interface management unit completes the data readout to a certain channel, when there is still data to be read out, it can read data from the flash memory chip again, and perform the data readout work of the next cycle.
读取数据的时候,通道仲裁和数据分配单元有可以一次启动多个通道进行数据读操作,每个通道读取数据以后都存放在对应的通道缓存中,由通道仲裁和数据分配单元负责把数据读出向前传递。当通道缓存数据被读出以后,如果该通道还有等待读取的命令,又可进行下一次的读操作。所以,当读操作时,通道缓存右侧的单元只负责数据从闪存阵列读出,并写入到通道缓存中。通道缓存左侧的单元协同工作,负责数据从通道缓存中读出,传递到数据接口上。两部分仍然是并行工作,互不干扰。When reading data, the channel arbitration and data allocation unit can start multiple channels at one time for data read operation. After each channel reads data, it will be stored in the corresponding channel buffer. The channel arbitration and data allocation unit is responsible for the data. Read forward pass. After the channel cache data is read out, if the channel still has commands waiting to be read, the next read operation can be performed. Therefore, during a read operation, the unit on the right side of the channel cache is only responsible for reading data from the flash memory array and writing it into the channel cache. The units on the left side of the channel buffer work together to read data from the channel buffer and transfer it to the data interface. The two parts still work in parallel without interfering with each other.
本发明的闪存数据写入方法保证了每个芯片的操作时序正确,并且确保数据的写入闪存芯片的等待时间严格按照闪存芯片提供的参数最大值来操作,这样就保证了在恶劣的外部环境中,数据依然能够可靠的写入到闪存芯片中。具体的讲,就是在前一个芯片处于写等待的状态时,立刻转入到下一个所在位置闪存芯片进行写操作,依此类推,顺序写闪存芯片,当单通道内的芯片均完成了一次数据写操作后,单通道内的第一个芯片即第一次进行数据写入操作的闪存芯片正好处于等待完成状态,这样又可以进行下一次的数据写入。在保证了数据的写入速度的同时,有保证了数据的写入可靠性。而且,由于每个芯片的写等待时间得到了充足的保证,也就实现了数据写入的可靠性。而在有单通道构成的多通道闪存中,每增加一个本发明所述的单通道便会使得构成的多通道闪存的读写速度比原先的单通道读写速度增加一倍,从而大幅提高闪存的读写速度。The method for writing flash memory data of the present invention ensures that the operation timing of each chip is correct, and ensures that the waiting time for writing data into the flash memory chip operates strictly according to the maximum value of the parameters provided by the flash memory chip, thus ensuring , the data can still be reliably written into the flash memory chip. Specifically, when the previous chip is in the state of waiting for writing, it immediately transfers to the flash memory chip at the next location for writing operations, and so on, and writes the flash memory chips sequentially. When the chips in a single channel have completed a data After the write operation, the first chip in the single channel, that is, the flash memory chip that performs the data write operation for the first time, is just waiting for completion, so that the next data write can be performed. While ensuring the writing speed of data, it also ensures the writing reliability of data. Moreover, since the write waiting time of each chip is sufficiently guaranteed, the reliability of data writing is realized. And in the multi-channel flash memory that single channel is formed, every increase of a single channel described in the present invention will make the reading and writing speed of the formed multi-channel flash memory double than the original single-channel reading and writing speed, thereby greatly improving the flash memory. read and write speed.
Claims (3)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810232221.6A CN101740102B (en) | 2008-11-11 | 2008-11-11 | Multichannel flash memory chip array structure and writing and reading method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810232221.6A CN101740102B (en) | 2008-11-11 | 2008-11-11 | Multichannel flash memory chip array structure and writing and reading method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101740102A CN101740102A (en) | 2010-06-16 |
CN101740102B true CN101740102B (en) | 2014-03-26 |
Family
ID=42463439
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200810232221.6A Expired - Fee Related CN101740102B (en) | 2008-11-11 | 2008-11-11 | Multichannel flash memory chip array structure and writing and reading method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101740102B (en) |
Families Citing this family (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102023943B (en) * | 2010-11-25 | 2013-06-19 | 杭州晟元芯片技术有限公司 | Method for controlling as well as reading and writing NandFlash by four-path IO (input-output) interface |
TWI520152B (en) * | 2013-03-01 | 2016-02-01 | 慧榮科技股份有限公司 | Data storage device and flash memory control method |
IN2013MU02016A (en) * | 2013-06-13 | 2015-06-05 | Mediatek Inc | |
CN105892944B (en) * | 2016-03-30 | 2019-11-12 | 深圳忆联信息系统有限公司 | A kind of information processing method and electronic equipment |
CN105975416B (en) * | 2016-04-28 | 2018-11-20 | 西安电子科技大学 | Multichannel friction speed data Transmission system based on FPGA |
CN106502581B (en) * | 2016-09-30 | 2019-05-28 | 华为技术有限公司 | Flash controller, flash memory control method and solid state hard disk |
CN106528465B (en) * | 2016-11-10 | 2019-08-02 | 郑州云海信息技术有限公司 | A kind of Nand Flash controller and method |
CN107766270B (en) * | 2017-10-20 | 2020-05-26 | 深圳市风云实业有限公司 | Data reading management method and device for PCIe (peripheral component interface express) equipment |
CN108008919A (en) * | 2017-12-22 | 2018-05-08 | 中国电子科技集团公司第五十四研究所 | A kind of high-speed data handles SSD |
CN108563399B (en) * | 2018-03-13 | 2020-02-14 | 中山市江波龙电子有限公司 | Data reading method and device of storage device, terminal device and storage medium |
CN108932204A (en) * | 2018-06-13 | 2018-12-04 | 郑州云海信息技术有限公司 | A kind of multi-channel flash memory storage system |
US10777232B2 (en) * | 2019-02-04 | 2020-09-15 | Micron Technology, Inc. | High bandwidth memory having plural channels |
CN114328284B (en) * | 2021-12-30 | 2025-04-29 | 武汉喻芯半导体有限公司 | A storage master control architecture and flash memory particle control method |
CN114637711B (en) * | 2022-03-31 | 2024-06-18 | 深圳市洲明科技股份有限公司 | Chip control method, control data transmission method, device and computer equipment |
CN119201838B (en) * | 2024-11-26 | 2025-02-25 | 成都旋极历通信息技术有限公司 | A method for managing out-of-order reception of RapidIO Message transactions based on FPGA |
CN119473956B (en) * | 2025-01-15 | 2025-05-27 | 联芸科技(杭州)股份有限公司 | Controller of storage device and instruction scheduling method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7020737B2 (en) * | 2002-05-20 | 2006-03-28 | Micron Technology, Inc. | Pipelined burst memory access |
CN1790308A (en) * | 2005-12-27 | 2006-06-21 | 北京中星微电子有限公司 | Multi-channel flash memory transmission controller, chip and storage device |
CN101046725A (en) * | 2007-03-23 | 2007-10-03 | 忆正存储技术(深圳)有限公司 | Flash controller |
CN101082891A (en) * | 2007-05-10 | 2007-12-05 | 忆正存储技术(深圳)有限公司 | Paralleling flash memory controller |
-
2008
- 2008-11-11 CN CN200810232221.6A patent/CN101740102B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7020737B2 (en) * | 2002-05-20 | 2006-03-28 | Micron Technology, Inc. | Pipelined burst memory access |
CN1790308A (en) * | 2005-12-27 | 2006-06-21 | 北京中星微电子有限公司 | Multi-channel flash memory transmission controller, chip and storage device |
CN101046725A (en) * | 2007-03-23 | 2007-10-03 | 忆正存储技术(深圳)有限公司 | Flash controller |
CN101082891A (en) * | 2007-05-10 | 2007-12-05 | 忆正存储技术(深圳)有限公司 | Paralleling flash memory controller |
Also Published As
Publication number | Publication date |
---|---|
CN101740102A (en) | 2010-06-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101740102B (en) | Multichannel flash memory chip array structure and writing and reading method thereof | |
CN104407997B (en) | NAND Flash Single-Channel Synchronous Controller with Instruction Dynamic Scheduling | |
US8521945B2 (en) | Portable data storage using SLC and MLC flash memory | |
US11360711B2 (en) | Storage device temporarily suspending internal operation to provide short read response time for read request from host | |
CN101308697A (en) | Large-capacity FIFO burst buffer and data storage method based on SDRAM | |
US20230244615A1 (en) | Memory device, method for controlling memory device and memory system | |
CN108733580A (en) | Method for scheduling read commands | |
CN102968396B (en) | Dedicated data transmission module from Flash chip to sram chip | |
US20190354483A1 (en) | Controller and memory system including the same | |
US12008270B2 (en) | System, device, and method for memory interface including reconfigurable channel | |
US20250231890A1 (en) | Data burst queue management | |
CN112069095B (en) | DDR3 read-write transmission method and device | |
CN108139993B (en) | Memory device, memory controller, data cache device and computer system | |
CN116257191B (en) | Memory controller, component, electronic device and command scheduling method | |
KR20230148431A (en) | Semiconductor apparatus and semiconductor system having lun selection cycle, and an operating method of the semiconductor system | |
US11742004B2 (en) | Memory supporting multiple types of operations | |
US20070143534A1 (en) | Nonvolatile-memory-access control apparatus and nonvolatile-memory control system | |
CN104133798A (en) | Big data high-speed storage system and implementation method | |
WO2017156747A1 (en) | Memory access method and computer system | |
CN116301664B (en) | Memory controller, component, electronic device and command caching method | |
CN101751982B (en) | Connection method between flesh memory controller and flesh memory chip in flesh memory storing device | |
CN201374192Y (en) | Flash memory device | |
US20240419341A1 (en) | Storage device and method of operating the same | |
US20250224900A1 (en) | Storage device, storage controller, and operating method of storage controller | |
US20190096459A1 (en) | Memory devices for performing multiple write operations and operating methods thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
C56 | Change in the name or address of the patentee |
Owner name: XI'AN KEYWAY TECHNOLOGY CO., LTD. Free format text: FORMER NAME: XI'AN QIVI TEST + CONTROL TECHNOLOGY CO., LTD. |
|
CP03 | Change of name, title or address |
Address after: 710065 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Patentee after: XI'AN KEYWAY TECHNOLOGY Co.,Ltd. Address before: 710077 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Patentee before: Shenzhen Sangxia Solar Energy Co.,Ltd. |
|
CP03 | Change of name, title or address | ||
CP03 | Change of name, title or address |
Address after: 710077 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Patentee after: XI'AN KEYWAY TECHNOLOGY Co.,Ltd. Address before: 710065 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Patentee before: Xi'an Keyway Technology Co.,Ltd. |
|
TR01 | Transfer of patent right |
Effective date of registration: 20170920 Address after: 710077 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Co-patentee after: BIT RACO ELECTRONIC INFORMATION TECHNOLOGY CO.,LTD. Patentee after: XI'AN KEYWAY TECHNOLOGY Co.,Ltd. Address before: 710077 Xi'an high tech Zone, Jin Industrial Road, No., No. C Venture Park, No. 8, Patentee before: XI'AN KEYWAY TECHNOLOGY Co.,Ltd. |
|
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20211019 Address after: 710117 No. a1-134, building 4, phase II, information industry park, No. 526, Xitai Road, high tech Zone, Xi'an, Shaanxi Province Patentee after: Xi'an siduoruizhi Information Technology Co.,Ltd. Address before: 710077, No. 8, C zone, pioneering research and Development Park, 69 Kam Yip Road, hi tech Zone, Shaanxi, Xi'an Patentee before: XI'AN KEYWAY TECHNOLOGY Co.,Ltd. Patentee before: BIT RACO ELECTRONIC INFORMATION TECHNOLOGY Co.,Ltd. |
|
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20211130 Address after: 710117 a2-02, building 4, phase II, information industry park, No. 526, banxitai Road, Xinglong Street, high tech Zone, Xi'an, Shaanxi Province Patentee after: Leizhi digital system technology (Xi'an) Co.,Ltd. Address before: 710117 No. a1-134, building 4, phase II, information industry park, No. 526, Xitai Road, high tech Zone, Xi'an, Shaanxi Province Patentee before: Xi'an siduoruizhi Information Technology Co.,Ltd. |
|
TR01 | Transfer of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140326 |
|
CF01 | Termination of patent right due to non-payment of annual fee |