CN108647161A - A kind of hardware observation circuit of record memory access address history - Google Patents
A kind of hardware observation circuit of record memory access address history Download PDFInfo
- Publication number
- CN108647161A CN108647161A CN201810340474.9A CN201810340474A CN108647161A CN 108647161 A CN108647161 A CN 108647161A CN 201810340474 A CN201810340474 A CN 201810340474A CN 108647161 A CN108647161 A CN 108647161A
- Authority
- CN
- China
- Prior art keywords
- address
- monitoring
- bit
- flag
- effective
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1605—Handling requests for interconnection or transfer for access to memory bus based on arbitration
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/1658—Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/18—Handling requests for interconnection or transfer for access to memory bus based on priority control
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Debugging And Monitoring (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
一种记录访存地址历史的硬件监测电路,包括监测区域配置单元、仲裁控制单元、标记存储器、标记位计数单元、优先级内容寻址存储体j、二级优先级内容寻址存储体、多路选择器。本发明与现有技术相比,通过采用两级优先级内容寻址存储体,可对多个非连续的地址空间进行监测,监测粒度可以进行配置选择,实现监测范围和监测粒度的折衷优化,为热备份计算机中单机失步后的数据信息同步过程提供了硬件支持。
A hardware monitoring circuit for recording memory access address history, including a monitoring area configuration unit, an arbitration control unit, a flag memory, a flag bit counting unit, a priority content addressable memory bank j, a secondary priority content addressable memory bank, and multiple path selector. Compared with the prior art, the present invention can monitor a plurality of non-continuous address spaces by adopting two-level priority content addressing storage bodies, and the monitoring granularity can be configured and selected to realize the compromise optimization between the monitoring range and the monitoring granularity. Provides hardware support for the synchronization process of data information after a single computer loses synchronization in a hot backup computer.
Description
技术领域technical field
本发明涉及一种记录访存地址历史的硬件监测电路。The invention relates to a hardware monitoring circuit for recording memory access address history.
背景技术Background technique
热备份计算机在连续稳定运行期间,会由于各种故障而造成某单机与其他单机失步,进而造成热备份计算机整机降级运行。为提高热备份计算机整机可靠性和寿命,将瞬时性故障造成的失步单机重新引入热备份计算机整机运行中意义很大。将其如何重新引入热备份计算机整机中重新工作问题即为并机问题。During the continuous and stable operation of the hot backup computer, due to various failures, one stand-alone computer will lose synchronization with other stand-alone computers, which will cause the whole hot backup computer to degrade. In order to improve the reliability and life of the hot backup computer, it is of great significance to reintroduce the out-of-synchronization stand-alone machine caused by the transient failure into the operation of the hot backup computer. The question of how to reintroduce it into the whole machine of the hot standby computer is the parallel machine problem.
空间站GNCC采用五机热备容错构型,其中四台单机具备处理能力,一台单机无处理能力。根据GNCC运行过程,失步单机与正常单机相比区别在于现场信息的不同。现场信息可分为两类:1)时间信息;2)数据信息。因此,失步单机必须完成时间同步和数据同步才能恢复和其他正常单机的同步,本发明主要通过访存地址分级监测方法为失步计算机的现场数据信息的同步提供硬件支持。The space station GNCC adopts a five-machine hot-standby fault-tolerant configuration, of which four stand-alone machines have processing capabilities, and one stand-alone machine has no processing capabilities. According to the GNCC operation process, the difference between the out-of-synchronization stand-alone and the normal stand-alone lies in the difference in field information. On-site information can be divided into two categories: 1) time information; 2) data information. Therefore, the out-of-synchronization stand-alone computer must complete time synchronization and data synchronization to restore synchronization with other normal stand-alone computers. The present invention mainly provides hardware support for the synchronization of on-site data information of the out-of-synchronization computer through a hierarchical monitoring method for memory access addresses.
发明内容Contents of the invention
本发明解决的技术问题是:克服现有技术的不足,提供了一种记录访存地址历史的硬件监测电路,克服传统的处理器访存地址历史无法记录的不足,通过采用两级优先级内容寻址存储体,可对多个非连续的地址空间进行监测,监测粒度可以进行配置选择,实现监测范围和监测粒度的折衷优化,为热备份计算机中单机失步后的数据信息同步过程提供了硬件支持。The technical problem solved by the present invention is to overcome the deficiencies of the prior art, provide a hardware monitoring circuit for recording the memory access address history, overcome the deficiency that the traditional processor access address history cannot be recorded, and adopt two levels of priority content Addressing memory banks can monitor multiple non-contiguous address spaces. The monitoring granularity can be configured and selected to achieve a compromise between monitoring range and monitoring granularity. It provides a data synchronization process for a single machine out of sync in a hot backup computer. Hardware support.
本发明的技术解决方案是:一种记录访存地址历史的硬件监测电路,包括监测区域配置单元、仲裁控制单元、标记存储器、标记位计数单元、优先级内容寻址存储体j、二级优先级内容寻址存储体、多路选择器;The technical solution of the present invention is: a hardware monitoring circuit for recording memory access address history, including a monitoring area configuration unit, an arbitration control unit, a flag memory, a flag bit counting unit, a priority content addressable memory bank j, a secondary priority level content addressable memory banks, multiplexers;
监测区域配置单元,根据外部任务需求对监测地址范围、监测地址粒度、监测使能进行配置,将配置后的监测地址范围、监测地址粒度、监测使能送至仲裁控制单元,并根据配置的监测地址范围配置监测起始地址寄存器、监测结束地址寄存器,且监测起始地址寄存器的配置值小于监测结束地址寄存器;所述的监测地址粒度为16n字节且不大于1024字节,其中,n为整数;The monitoring area configuration unit configures the monitoring address range, monitoring address granularity, and monitoring enablement according to external task requirements, and sends the configured monitoring address range, monitoring address granularity, and monitoring enablement to the arbitration control unit, and monitors according to the configuration. The address range configures the monitoring start address register and the monitoring end address register, and the configuration value of the monitoring start address register is less than the monitoring end address register; the monitoring address granularity is 16n bytes and not greater than 1024 bytes, where n is integer;
仲裁控制单元,监测待监测总线的写使能信号,当待监测地址处于监测地址范围且待监测总线的写使能信号有效时,判断监测地址粒度,若监测地址粒度为16n字节,生成标记存储器写信号有效且写地址为待监测地址的第n+8位到第n-1位,共10比特,将标记存储器写信号、写地址送至标记位计数单元、标记存储器;The arbitration control unit monitors the write enable signal of the bus to be monitored, and when the address to be monitored is in the monitored address range and the write enable signal of the bus to be monitored is valid, judges the granularity of the monitored address, and generates a flag if the granularity of the monitored address is 16n bytes The memory write signal is valid and the write address is the n+8th bit to the n-1th bit of the address to be monitored, a total of 10 bits, and the mark memory write signal and write address are sent to the mark bit counting unit and the mark memory;
标记存储器,接收标记存储器写信号、写地址进行标记存储,当标记存储器写信号有效时,将写地址对应的比特位置为1,其余比特位不变;将标记存储器中1024个比特位对应的数值送至标记位计数单元;将标记存储器中1024个比特位对应的数值连续32个进行打包,得到32个数值包并分别送至32优先级内容寻址存储体;Tag memory, receive the tag memory write signal and write address for tag storage, when the tag memory write signal is valid, set the bit position corresponding to the write address to 1, and the remaining bits remain unchanged; the value corresponding to the 1024 bits in the tag memory Send to the mark position counting unit; The numerical value corresponding to 1024 bits in the mark memory is packed 32 consecutively, obtain 32 numerical value packets and send to 32 priority content addressable storage bodies respectively;
标记位计数单元,接收标记存储器中1024个比特位对应的数值、标记存储器写信号、写地址,当写使能信号有效,且写地址对应的标记存储器中比特位为0时,标记位计数单元加1;The flag bit counting unit receives the value corresponding to 1024 bits in the flag memory, the flag memory write signal, and the write address. When the write enable signal is valid and the bit in the flag memory corresponding to the write address is 0, the flag bit counting unit plus 1;
优先级内容寻址存储体j,其中j为不大于32的正整数,接收32位的数值包,生成监测有效标志j、优先地址输出j;其中,如果优先级内容寻址存储体j接收的32位的数值包全为0,监测有效标志j为0,否则监测有效标志j为1;当32位的数值中第k位为1且第0位到k-1位为0时,优先地址输出j为k的5位二进制,其中,k=1,2,3,…,31,当32位的数值中第0位为1时,优先地址输出j为00000;将优先地址输出j送至多路选择器,将监测有效标志j送至二级优先级内容寻址存储体;Priority content addressable memory bank j, wherein j is a positive integer not greater than 32, receives a 32-bit value packet, generates monitoring valid flag j, and priority address output j; wherein, if the priority content addressable memory bank j receives The 32-bit value package is all 0, and the monitoring effective flag j is 0, otherwise the monitoring effective flag j is 1; when the kth bit in the 32-bit value is 1 and the 0th to k-1 bits are 0, the priority address Output j is the 5-bit binary of k, wherein, k=1, 2, 3, ..., 31, when the 0th bit in the 32-bit value is 1, the priority address output j is 00000; the priority address output j is sent to The way selector sends the monitoring effective flag j to the secondary priority content addressable storage body;
二级优先级内容寻址存储体,接收监测有效标志j,得到32位的监测有效标志数据包,生成监测有效标志、有效地址高位;其中,如果32位的监测有效标志数据包全为0,监测有效标志为0,否则监测有效标志为1;当32位的监测有效标志数据包第k位为1且第0位到k-1位为0时,有效地址高位为k的5位二进制,其中,k=1,2,3,…,31,当32位的监测有效标志数据包第0位为1时,有效地址高位为00000;将有效地址高位送至多路选择器;The second-level priority content addressing storage body receives the monitoring effective flag j, obtains the 32-bit monitoring effective flag data packet, and generates the monitoring effective flag and the high bit of the effective address; wherein, if the 32-bit monitoring effective flag data packet is all 0, The monitoring effective flag is 0, otherwise the monitoring effective flag is 1; when the kth bit of the 32-bit monitoring effective flag data packet is 1 and the 0th to k-1 bits are 0, the high bit of the effective address is the 5-bit binary of k, Wherein, k=1, 2, 3, ..., 31, when the 0th bit of the 32-bit monitoring effective flag data packet is 1, the high bit of the effective address is 00000; the high bit of the effective address is sent to the multiplexer;
多路选择器,接收有效地址高位、优先地址输出j,对有效地址高位对应的优先地址输出j作为有效地址低位。The multiplexer receives the high bit of the effective address and the output j of the priority address, and uses the output j of the priority address corresponding to the high bit of the effective address as the low bit of the effective address.
所述的监测粒度可进行配置,可对多个非连续的地址空间进行监测。The monitoring granularity can be configured, and multiple discontinuous address spaces can be monitored.
本发明与现有技术相比的优点在于:The advantage of the present invention compared with prior art is:
(1)本发明通过基于组合逻辑输出接口的C单元电路结构自动生成,解决了当前人工设计电路的时间开销过大的问题,具有可批量化实现组合逻辑单元的C单元加固的优点;(1) The present invention solves the problem of excessive time overhead of the current manual circuit design by automatically generating the C unit circuit structure based on the combinational logic output interface, and has the advantage of being able to realize the reinforcement of the C unit of the combinational logic unit in batches;
(2)本发明用两级优先级内容寻址存储体,与顺序查询模式相比,具有更高的查询效率,与直接并行查询的一级查询方法相比,具有更高的评估频率;(2) The present invention uses two-level priority content addressing storage body, compared with sequential query mode, has higher query efficiency, compared with the first-level query method of direct parallel query, has higher evaluation frequency;
(3)本发明结构可对多个非连续的地址空间进行监测,监测粒度可以进行配置选择,实现监测范围和监测粒度的折衷优化,具有很好的使用价值。(3) The structure of the present invention can monitor a plurality of discontinuous address spaces, and the monitoring granularity can be configured and selected to realize the compromise optimization between the monitoring range and the monitoring granularity, which has good use value.
附图说明Description of drawings
图1为本发明整体电路结构图;Fig. 1 is the overall circuit structure diagram of the present invention;
图2为内容寻址存储体结构示意图。FIG. 2 is a schematic diagram of the structure of a content addressable memory bank.
具体实施方式Detailed ways
一种记录访存地址历史的硬件监测电路,克服传统的处理器访存地址历史无法记录的上述不足,通过采用两级优先级内容寻址存储体,可对多个非连续的地址空间进行监测,监测粒度可以进行配置选择,实现监测范围和监测粒度的折衷优化,为热备份计算机中单机失步后的数据信息同步过程提供了硬件支持。A hardware monitoring circuit for recording the memory access address history, which overcomes the above-mentioned shortcomings that the traditional processor memory access address history cannot be recorded, and can monitor multiple discontinuous address spaces by using two-level priority content addressing memory banks , the monitoring granularity can be configured and selected to realize the compromise optimization between the monitoring range and the monitoring granularity, and provide hardware support for the data information synchronization process after the stand-alone out-of-sync in the hot backup computer.
监测区域配置单元,根据外部任务需求对监测地址范围、监测地址粒度、监测使能进行配置,将配置后的监测地址范围、监测地址粒度、监测使能送至仲裁控制单元,并根据配置的监测地址范围配置监测起始地址寄存器、监测结束地址寄存器,且监测起始地址寄存器的配置值小于监测结束地址寄存器;所述的监测地址粒度为16n字节且不大于1024字节,其中,n为整数;监测粒度可进行配置,可对多个非连续的地址空间进行监测。The monitoring area configuration unit configures the monitoring address range, monitoring address granularity, and monitoring enablement according to external task requirements, and sends the configured monitoring address range, monitoring address granularity, and monitoring enablement to the arbitration control unit, and monitors according to the configuration. The address range configures the monitoring start address register and the monitoring end address register, and the configuration value of the monitoring start address register is less than the monitoring end address register; the monitoring address granularity is 16n bytes and not greater than 1024 bytes, where n is Integer; monitoring granularity can be configured, and multiple non-contiguous address spaces can be monitored.
仲裁控制单元,监测待监测总线的写使能信号有效,当待监测地址处于监测地址范围且待监测总线的写使能信号有效时,判断监测地址粒度,若监测地址粒度为16n字节,生成标记存储器写信号有效且写地址为待监测地址的第n+8位到第n-1位,共10比特,将标记存储器写信号、写地址送至标记位计数单元、标记存储器。The arbitration control unit monitors that the write enable signal of the bus to be monitored is valid, and when the address to be monitored is in the monitored address range and the write enable signal of the bus to be monitored is valid, judges the granularity of the monitored address, and if the granularity of the monitored address is 16n bytes, generates The flag memory write signal is valid and the write address is the n+8th bit to the n-1th bit of the address to be monitored, a total of 10 bits, and the flag memory write signal and write address are sent to the flag bit counting unit and the flag memory.
标记存储器,接收标记存储器写信号、写地址进行标记存储,当标记存储器写信号有效时,将写地址对应的比特位置为1,其余比特位不变;将标记存储器中1024个比特位对应的数值送至标记位计数单元;将标记存储器中1024个比特位对应的数值连续32个进行打包,得到32个数值包并分别送至32优先级内容寻址存储体。Tag memory, receive the tag memory write signal and write address for tag storage, when the tag memory write signal is valid, set the bit position corresponding to the write address to 1, and the remaining bits remain unchanged; the value corresponding to the 1024 bits in the tag memory Sent to the tag bit counting unit; 32 consecutive values corresponding to 1024 bits in the tag memory are packed to obtain 32 value packets and sent to 32 priority content addressable memory banks respectively.
标记位计数单元,接收标记存储器中1024个比特位对应的数值、标记存储器写信号、写地址,当写使能信号有效,且写地址对应的标记存储器中比特位为0时,标记位计数单元加1。The flag bit counting unit receives the value corresponding to 1024 bits in the flag memory, the flag memory write signal, and the write address. When the write enable signal is valid and the bit in the flag memory corresponding to the write address is 0, the flag bit counting unit plus 1.
优先级内容寻址存储体j,其中j为不大于32的正整数,接收32位的数值包,生成监测有效标志j、优先地址输出j;其中,如果优先级内容寻址存储体j接收的32位的数值包全为0,监测有效标志j为0,否则监测有效标志j为1;当32位的数值中第k位为1且第0位到k-1位为0时,优先地址输出j为k的5位二进制,其中,k=1,2,3,…,31,当32位的数值中第0位为1时,优先地址输出j为00000;将优先地址输出j送至多路选择器,将监测有效标志j送至二级优先级内容寻址存储体。Priority content addressable memory bank j, wherein j is a positive integer not greater than 32, receives a 32-bit value packet, generates monitoring valid flag j, and priority address output j; wherein, if the priority content addressable memory bank j receives The 32-bit value package is all 0, and the monitoring effective flag j is 0, otherwise the monitoring effective flag j is 1; when the kth bit in the 32-bit value is 1 and the 0th to k-1 bits are 0, the priority address Output j is the 5-bit binary of k, wherein, k=1, 2, 3, ..., 31, when the 0th bit in the 32-bit value is 1, the priority address output j is 00000; the priority address output j is sent to The way selector sends the monitoring effective flag j to the secondary priority content addressable memory bank.
二级优先级内容寻址存储体,接收监测有效标志j,得到32位的监测有效标志数据包,生成监测有效标志、有效地址高位;其中,如果32位的监测有效标志数据包全为0,监测有效标志为0,否则监测有效标志为1;当32位的监测有效标志数据包第k位为1且第0位到k-1位为0时,有效地址高位为k的5位二进制,其中,k=1,2,3,…,31,当32位的监测有效标志数据包第0位为1时,有效地址高位为00000;将有效地址高位送至多路选择器。The second-level priority content addressing storage body receives the monitoring effective flag j, obtains the 32-bit monitoring effective flag data packet, and generates the monitoring effective flag and the high bit of the effective address; wherein, if the 32-bit monitoring effective flag data packet is all 0, The monitoring effective flag is 0, otherwise the monitoring effective flag is 1; when the kth bit of the 32-bit monitoring effective flag data packet is 1 and the 0th to k-1 bits are 0, the high bit of the effective address is the 5-bit binary of k, Among them, k=1, 2, 3, ..., 31, when the 0th bit of the 32-bit monitoring effective flag data packet is 1, the high bit of the effective address is 00000; the high bit of the effective address is sent to the multiplexer.
多路选择器,接收有效地址高位、优先地址输出j,对有效地址高位对应的优先地址输出j作为有效地址低位。例如有效地址高位为00000,则选择优先地址输出0作为有效地址低位,有效地址高位为00110,则选择优先地址输出6作为有效地址低位。下面结合附图对本发明进行详细的解释和说明。The multiplexer receives the high bit of the effective address and the output j of the priority address, and uses the output j of the priority address corresponding to the high bit of the effective address as the low bit of the effective address. For example, the high bit of the effective address is 00000, then select the priority address output 0 as the low bit of the effective address, and the high bit of the effective address is 00110, then select the priority address output 6 as the low bit of the effective address. The present invention will be explained and described in detail below in conjunction with the accompanying drawings.
本发明提出的访存地址分级监测方法的整体电路结构如图1所示,该电路结构包括监测区域配置单元、仲裁控制单元、标记位计数单元、标记存储器、两级优先级内容寻址存储体以及多路选择器构成。The overall circuit structure of the memory access address classification monitoring method proposed by the present invention is shown in Figure 1. The circuit structure includes a monitoring area configuration unit, an arbitration control unit, a flag bit counting unit, a flag memory, and a two-level priority content addressable memory bank. and a multiplexer.
监测过程如下:复位后,标记存储器的数值全为‘0’,当待监测写使能有效后,同时待监测写地址处于监测配置区域的有效区间内,则将待监测写地址对应的标记存储器映射位写为‘1’,标记位计数单元加1,待监测写地址与标记存储器映射位的关系可由监测粒度寄存器配置,当重复写同一个待监测地址时,标记位计数单元计数值不变,监测信息读出后,相应的RAM映射位写为‘0’。电路结构的每个单元详细设计如下:The monitoring process is as follows: After reset, the value of the tag memory is all '0', when the write enable to be monitored is enabled and the write address to be monitored is within the valid range of the monitoring configuration area, the tag memory corresponding to the write address to be monitored The mapping bit is written as '1', and the flag bit counting unit adds 1. The relationship between the write address to be monitored and the flag memory mapping bit can be configured by the monitoring granularity register. When the same address to be monitored is repeatedly written, the counting value of the flag bit counting unit remains unchanged , after the monitoring information is read out, the corresponding RAM mapping bit is written as '0'. The detailed design of each unit of the circuit structure is as follows:
(一)监测区域配置单元:(1) Monitoring area configuration unit:
该模块可对监测地址的范围、粒度、监测使能进行配置。每个监测区域配置单元包含一个监测起始地址寄存器和一个监测结束地址寄存器,监测起始地址寄存器的配置值小于监测结束地址寄存器时,配置有效,该结构便可监测配置地址范围内的地址写访问情况。如图1所示,有四个监测区域配置单元,每个监测区域配置单元有独立的使能配置位,则该结构可对四段非连续的地址空间进行监测。监测区域配置单元可配置监测粒度寄存器,该寄存器可在待监测范围和待监测粒度之间做合理折衷。待监测范围和待监测粒度以及标记存储器容量之间满足如下公式:This module can configure the range, granularity and monitoring enable of the monitoring address. Each monitoring area configuration unit includes a monitoring start address register and a monitoring end address register. When the configuration value of the monitoring start address register is less than the monitoring end address register, the configuration is valid, and the structure can monitor address writes within the configured address range. access situation. As shown in Figure 1, there are four monitoring area configuration units, and each monitoring area configuration unit has an independent enable configuration bit, so this structure can monitor four discontinuous address spaces. The monitoring area configuration unit can configure the monitoring granularity register, and the register can make a reasonable compromise between the scope to be monitored and the granularity to be monitored. The relationship between the range to be monitored, the granularity to be monitored, and the capacity of the tag memory satisfies the following formula:
监测范围/监测粒度=监测标记存储器容量Monitoring range/monitoring granularity = monitoring tag memory capacity
表1所示为监测粒度寄存器的配置值与监测粒度的关系:Table 1 shows the relationship between the configuration value of the monitoring granularity register and the monitoring granularity:
(二)仲裁控制单元:(2) Arbitration control unit:
该模块用于仲裁选通多个监测区域配置单元,通过监测总线写使能信号和写地址,生成标记存储器写信号和写地址,实现对标记存储器的读写控制。This module is used to arbitrate and select multiple monitoring area configuration units, and generate tag memory write signals and write addresses by monitoring the bus write enable signal and write address, so as to realize the read and write control of the tag memory.
(三)标记位计数单元:(3) Mark bit counting unit:
标记位是指待监测地址通过监测粒度寄存器而对应的标记存储器的特定比特位。标记位计数单元用于实现对待监测地址变化量的统计。当待监测写使能有效后,同时待监测写地址处于监测配置区域的有效区间内,同时该标记位尚未被写过,即对应的标记存储器数据为‘0’,则将标记位计数单元加1。The tag bit refers to a specific bit of the tag memory corresponding to the address to be monitored through the monitoring granularity register. The mark bit counting unit is used to realize the statistics of the change amount of the address to be monitored. When the write enable to be monitored is enabled, and the write address to be monitored is within the effective range of the monitoring configuration area, and the flag bit has not been written, that is, the corresponding flag memory data is '0', the flag bit counting unit is added to 1.
(四)标记存储器:(4) Tag memory:
该模块用于存储被更改过的待监测地址的信息,当待监测地址的内容被更改过,则标记存储器中对应的比特位为1,否则为0。This module is used to store the information of the address to be monitored that has been changed. When the content of the address to be monitored has been changed, the corresponding bit in the flag memory is 1, otherwise it is 0.
(五)优先级内容寻址存储体:(5) Priority content-addressable memory banks:
内容寻址存储体与传统的RAM存储结构有所不同,如图2所示。在本发明方案中,内容寻址存储体的匹配数据简单,只有一个比特,保证了优先级内容寻址存储体的结构不会过于复杂,同时,对于一个存储体中必然存在有多个地址与待匹配数据匹配的情况,其输出的有效地址为优先级最高的匹配地址,如果整个优先级内容寻址存储体中没有一个地址的数据可以匹配数据‘1’,则匹配标志输出‘0’,否则输出‘1’。The content-addressable memory bank is different from the traditional RAM memory structure, as shown in Figure 2. In the scheme of the present invention, the matching data of the content-addressable storage body is simple, only one bit, which ensures that the structure of the priority content-addressable storage body will not be too complicated, and at the same time, there must be multiple addresses and When the data to be matched is matched, the effective address output is the matching address with the highest priority. If there is no data at an address in the entire priority content addressable memory bank that can match the data '1', the match flag outputs '0', Otherwise output '1'.
本发明采用两级优先级内容寻址存储体,第一级的输入是标记存储器的数据,相当于对标记存储器数据进行分块统计,输出是低位有效地址和匹配标志;第二级优先级内容寻址存储体输入是多个第一级的优先级内容寻址存储体的匹配标志,输出是高位有效地址和整个系统的匹配标志。The present invention adopts two levels of priority content addressing storage body, the input of the first level is the data of the tag memory, which is equivalent to performing block statistics on the data of the tag memory, and the output is the low effective address and the matching flag; the content of the second level of priority The addressable memory bank input is the matching flag of multiple first-level priority content-addressable memory banks, and the output is the high-order effective address and the matching flag of the entire system.
与状态机控制下的顺序查询模式相比,本发明的两级优先级内容寻址存储体查询方法受益于查询的并行性,大大提高了查询的效率,而与直接并行查询的一级查询方法相比,本文方法合理的划分了查询流水级,提高了整个系统的工作频率。Compared with the sequential query mode under the control of the state machine, the two-level priority content addressing storage body query method of the present invention benefits from the parallelism of the query, greatly improving the efficiency of the query, and compared with the first-level query method of direct parallel query In contrast, the method in this paper divides the query pipeline reasonably and improves the working frequency of the whole system.
(六)多路选择器:(6) Multiplexer:
多路选择器模块用于将多个第一级的优先级内容寻址存储体的有效地址进行选通输出,其选通的依据来源于第二级优先级内容寻址存储体的高位地址输出。The multiplexer module is used to strobe and output the effective addresses of multiple first-level priority content-addressable memory banks, and the gating basis comes from the high-order address output of the second-level priority content-addressable memory banks .
本发明说明书中未作详细描述的内容属本领域技术人员的公知技术。The content that is not described in detail in the description of the present invention belongs to the well-known technology of those skilled in the art.
Claims (2)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810340474.9A CN108647161B (en) | 2018-04-17 | 2018-04-17 | Hardware monitoring circuit for recording access address history |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810340474.9A CN108647161B (en) | 2018-04-17 | 2018-04-17 | Hardware monitoring circuit for recording access address history |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108647161A true CN108647161A (en) | 2018-10-12 |
| CN108647161B CN108647161B (en) | 2020-07-14 |
Family
ID=63746629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810340474.9A Active CN108647161B (en) | 2018-04-17 | 2018-04-17 | Hardware monitoring circuit for recording access address history |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647161B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114297005A (en) * | 2021-12-29 | 2022-04-08 | 成都博尔微晶科技有限公司 | Small-area and repeated row redundancy replacement method for Norflash |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6539341B1 (en) * | 2000-11-06 | 2003-03-25 | 3Com Corporation | Method and apparatus for log information management and reporting |
| CN1794186A (en) * | 2004-12-22 | 2006-06-28 | 阿尔卡特公司 | Addressing error and address detection systems and methods |
| US20080240102A1 (en) * | 2007-03-27 | 2008-10-02 | Alcatel Lucent | Method and system for policy-based routing in a private network-to-network interface protocol based network |
| CN201570016U (en) * | 2009-12-25 | 2010-09-01 | 东南大学 | Dynamic command on-chip heterogenous memory resource distribution circuit based on virtual memory mechanism |
| CN102356607A (en) * | 2011-08-25 | 2012-02-15 | 华为技术有限公司 | Method and apparatus for learning media being accessed with control address |
| CN103414692A (en) * | 2013-07-15 | 2013-11-27 | 北京控制工程研究所 | High real-time spacecraft data transmission method based on priorities |
| US9460110B2 (en) * | 2011-09-21 | 2016-10-04 | Kevin Mark Klughart | File system extension system and method |
| CN106782633A (en) * | 2016-03-26 | 2017-05-31 | 深圳星忆存储科技有限公司 | The method of dynamic random access memory, storage data and reading and refreshing |
| US20170344746A1 (en) * | 2016-05-25 | 2017-11-30 | International Business Machines Corporation | Utilizing likely invariants for runtime protection of web services |
| US9922039B1 (en) * | 2016-03-31 | 2018-03-20 | EMC IP Holding Company LLC | Techniques for mitigating effects of small unaligned writes |
-
2018
- 2018-04-17 CN CN201810340474.9A patent/CN108647161B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6539341B1 (en) * | 2000-11-06 | 2003-03-25 | 3Com Corporation | Method and apparatus for log information management and reporting |
| CN1794186A (en) * | 2004-12-22 | 2006-06-28 | 阿尔卡特公司 | Addressing error and address detection systems and methods |
| US20080240102A1 (en) * | 2007-03-27 | 2008-10-02 | Alcatel Lucent | Method and system for policy-based routing in a private network-to-network interface protocol based network |
| CN201570016U (en) * | 2009-12-25 | 2010-09-01 | 东南大学 | Dynamic command on-chip heterogenous memory resource distribution circuit based on virtual memory mechanism |
| CN102356607A (en) * | 2011-08-25 | 2012-02-15 | 华为技术有限公司 | Method and apparatus for learning media being accessed with control address |
| US9460110B2 (en) * | 2011-09-21 | 2016-10-04 | Kevin Mark Klughart | File system extension system and method |
| CN103414692A (en) * | 2013-07-15 | 2013-11-27 | 北京控制工程研究所 | High real-time spacecraft data transmission method based on priorities |
| CN106782633A (en) * | 2016-03-26 | 2017-05-31 | 深圳星忆存储科技有限公司 | The method of dynamic random access memory, storage data and reading and refreshing |
| US9922039B1 (en) * | 2016-03-31 | 2018-03-20 | EMC IP Holding Company LLC | Techniques for mitigating effects of small unaligned writes |
| US20170344746A1 (en) * | 2016-05-25 | 2017-11-30 | International Business Machines Corporation | Utilizing likely invariants for runtime protection of web services |
Non-Patent Citations (1)
| Title |
|---|
| 肖瑞瑾: "《多核处理器层次化存储体系研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114297005A (en) * | 2021-12-29 | 2022-04-08 | 成都博尔微晶科技有限公司 | Small-area and repeated row redundancy replacement method for Norflash |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108647161B (en) | 2020-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7121875B1 (en) | Data integrity, such as persistent memory systems | |
| TWI475384B (en) | System and method for implementing a memory defect map | |
| CN101676886B (en) | Flash memory device and method for writing data thereto | |
| CN104809075B (en) | A kind of solid state record device and method for accessing real-time parallel processing | |
| US12153529B2 (en) | Memory system and computing system including the same | |
| CN102012791B (en) | Flash based PCIE (peripheral component interface express) board for data storage | |
| TW498343B (en) | Dynamic configuration of storage arrays | |
| TW201901690A (en) | Repair circuit for repair operation and memory device including repair circuit | |
| CN104331252A (en) | Isomeric NAND solid state disk structure and data reading management method of isomeric NAND solid state disk structure | |
| CN101710270A (en) | High-speed large-capacity memory based on flash memory and chip data management method | |
| US8392671B2 (en) | Memory controller, system, and method for accessing semiconductor memory | |
| US12124729B2 (en) | Controller to alter systems based on metrics and telemetry | |
| CN105373345B (en) | Memory device and module | |
| US20190108892A1 (en) | System and Method for Post-Package Repair Across DRAM Banks and Bank Groups | |
| US7809899B2 (en) | System for integrity protection for standard 2n-bit multiple sized memory devices | |
| US20250217053A1 (en) | Memory channel disablement | |
| CN103019624B (en) | Phase change memory device | |
| CN116701253A (en) | Storage controller for managing RAID information | |
| CN108647161B (en) | Hardware monitoring circuit for recording access address history | |
| CN116263645A (en) | Address Generation for Adaptive Dual-Device Data Correction Backup | |
| CN110968451B (en) | Memory access technology and computer system | |
| CN103870204A (en) | Data writing and reading method in cache as well as cache controller | |
| WO2024174121A1 (en) | Write data path structure and chip | |
| CN112466386B (en) | A fault classification-oriented memory testing system and method | |
| CN115543869A (en) | Multi-way set connection cache memory and access method thereof, and computer equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |