TW556205B - Method for inspecting flash memory logic address - Google Patents
Method for inspecting flash memory logic address Download PDFInfo
- Publication number
- TW556205B TW556205B TW91106951A TW91106951A TW556205B TW 556205 B TW556205 B TW 556205B TW 91106951 A TW91106951 A TW 91106951A TW 91106951 A TW91106951 A TW 91106951A TW 556205 B TW556205 B TW 556205B
- Authority
- TW
- Taiwan
- Prior art keywords
- address
- area
- flash memory
- logical
- physical
- Prior art date
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Memory System (AREA)
Abstract
Description
556205556205
發明領域: 本發明係有關一種檢測快閃記憶體之邏輯位址的方 法,特別是關於一種將複數個區塊(bl〇ck)作為一區域 (zone ),以區域為單位製作成邏輯/實體位址對映表 (Address Mapping Table,AMT 或是L〇〇k — Up TaMe), 並藉此位址對映表來檢測邏輯位址。 發明背景: 、一按,在記憶體的運作原理中,當微處理器使用記憶 為資料儲存區時,計算的結果及資料都存放在這,如果程 j要加以取用時,也可以取用儲存區。而儲存資料和讀取 ,料%,微處理器便將會所需的資訊定義記憶體的位置, U處理器透過位置匯流排將位址送到記憶體,然後資料 會,,應的資料傳送到正確的位址上。而記憶體最重要的 便=4取的日守間,從彳政處理器發布指令到取得位址資料, 而記=體回應後到送資料給微處理器,至微處理器確實接 收到貝料為止,而這個程序所花的時間便是記憶體的存取 時間。 ,在快閃記憶體中係以若干byte組成之區塊(M〇ck )為單位進行資料儲存及讀取之使用,每一個可供資料存 取的區塊都具有一實體位址(Physical Address),、以代 f f决閃記憶體空間順序,如靜態隨機存取記憶體( )中實體記憶體的位置,同時每一區塊記錄磁碟機檔案作 業系統(File System)所標示的邏輯位址(L〇gical 麵 麵 第4頁 556205 五、發明說明(2)Field of the Invention: The present invention relates to a method for detecting a logical address of a flash memory, and more particularly to a method in which a plurality of blocks (bloc) are used as a zone, and the logic / entity is made by the unit of the zone. Address mapping table (Address Mapping Table, AMT or L00k — Up TaMe), and use this address mapping table to detect logical addresses. Background of the invention: One-click, in the working principle of the memory, when the microprocessor uses the memory as the data storage area, the calculation results and data are stored here. If the process j is to be accessed, it can also be accessed. Storage area. While storing data and reading data, the microprocessor will define the location of the memory by the required information. The U processor will send the address to the memory through the location bus, and then the data will be transmitted to the corresponding data. To the correct address. The most important part of the memory is the day of the fetch, from the instruction issued by the government processor to the address data, and the response of the body to the data sent to the microprocessor, until the microprocessor does receive the shell. The time taken by this process is the memory access time. In Flash memory, data is stored and read in units of blocks (Mock) composed of several bytes. Each block that can be accessed by data has a physical address. ), The flash memory space sequence is replaced by ff, such as the location of physical memory in static random access memory (), and each block records the logical bits marked by the file system of the drive Address (Page 4 of page 556205) 5. Description of the invention (2)
Address ),如第一(a)圖所示,且告 相對應的邏輯位址邏輯位址;然而,焉體位址各具有一 檔案作業系統存取資料所需的 在决閃機制中,由於 (LoglCal 實體位址推知邏輯位址之所在,知二係,無法直接由 到相對應之邏輯位址。 11通常有二種方式來得 第一種方式為在系統開機時先建立— 對映表,其係記錄邏輯位址和實體位 璉輯/實體位址 格,並設計有複數個SRAM來之間對映關係的表 一圖所示,在建立位址對映表時°,二二=址對映表,如第 為主,而將相對應的實體位置 ^铜1位址之排列順序 邏輯位址係呈不規則排列,因此必須% :二容中’由於原 才能夠建立一個如第一(b)圖所示之位m尾排序一次’ 對映表儲存在SRAM中,此種正向實體^映表,並將此 =:?建立一部份以達到節省別 立址對映表既 的位址對映表,進而快速得到相對搜哥在_中 址。此種利用仿科此主μ ; 對於5亥邏輯位址的實體位 址轉換時間短、_理速^ ^寸邏輯位址的方式雖然具有位 來記錄所二位之優點’但卻需要大量之咖 且因目前皆將l對關係’所佔有之空間面積相當大; 數目時整合在一單一系統晶片中, 、生忐蚀田…口疋者,無法依需求而隨時增加SRAM的數目, k成使用上的限制。 556205Address), as shown in the first (a) figure, and the corresponding logical address logical address; however, the carcass addresses each have a file operation system required to access data in the flash mechanism, because ( LoglCal's physical address knows where the logical address is, and the second system cannot be directly reached to the corresponding logical address. 11 There are usually two ways to get the first way is to build the system first when the system is turned on-the mapping table, which It records the logical address and the physical address series / physical address cell, and is designed as shown in the following table. There is a mapping relationship between multiple SRAMs. When the address mapping table is established °, 22 = address pair The mapping table is as the first one, and the corresponding physical position ^ The order of the logical address of the copper 1 address is irregularly arranged, so it must be%: Errongzhong, because the original one can build one as the first ( b) The map shown in the figure m is sorted once. The mapping table is stored in SRAM. This kind of forward entity ^ mapping table, and this = :? Create a part to save the existing mapping table. Address mapping table, and then quickly get the relative search brother in _ middle address. This kind of use imitates this master μ; For the physical address conversion time of the 5H logical address, the conversion time is short, and the method of logical address ^ ^ inch logical address has the advantages of recording the two bits, but it requires a lot of coffee and because of the current l The space occupied by the relationship is quite large; when the number is integrated into a single system chip, the field can be eroded, and the number of SRAMs cannot be increased at any time according to demand, which becomes a limitation on use. 556205
另一種由邏輯位址推知 計’亦無需建立任何位址對 母次要尋找實體位址所對應 頭開始搜尋,直至搜尋到所 係無S R A Μ之設計,所以不佔 址時都必須從頭開始搜尋, 資料之速度相當慢。 實體位址之方式係無SRAM的設 映表’僅需要利用軟體程式在 ^ 輯位址時,在記憶體中從 需之邏輯位址為止。此種方式 空間,但因每次要尋找邏輯位 搜尋時間慢,使得記憶體存取 …因;V本發明即在針對習知之缺失,在上述兩種實體 1址/邏輯位址轉換方式中’取得一有效的平衡點,使本 兼顧速度與空間上的問冑,以有效的提升資料存 發明目的與概述: 。本毛月之主要目的’係在提出一種檢測快閃記憶體邏 輯位址之方法,其係以區域(zone )為單位,先建立一等 比例縮減(Scaled Factor)之邏輯/實體位址對映表,並 透過該邏輯/實體位址對映表得到區域位址’進而搜尋到 相對於實體位址的邏輯位址,以便在有效的投資下,達到 速度的提升。 。本發明之另一目的,係在提出一種檢測快閃記憶體邏 輯位址之方法,其係利用可調整式的隨機存取記憶體,以 便在不增加太多空間的前提下,同時提升半導體磁碟裝置 的處理速度。 為達到上述之目的,本發明係先將快閃記憶體中之複數區Another type of logical address inference plan does not need to establish any address to search for the head corresponding to the physical address of the parent and secondary, until the design without the SRA M is found, so you must search from the beginning when not occupying the address. The speed of data is quite slow. The method of the physical address is the SRAM-free mapping table ', which only needs to use a software program to edit the address from the logical address required in the memory. This method has space, but because the search time for each logical bit is slow, the memory access is slowed down ... Because the present invention addresses the lack of knowledge in the above two physical 1-address / logical address conversion methods. To achieve an effective balance point, so that this book takes into account the speed and space problems, in order to effectively improve the purpose and summary of the invention of data storage:. The main purpose of this month is to propose a method for detecting the logical address of flash memory. It uses the zone as a unit and first establishes a scaled factor logical / physical address mapping. Table, and obtain the regional address through the logical / physical address mapping table, and then search for the logical address relative to the physical address, so as to achieve speed increase with effective investment. . Another object of the present invention is to provide a method for detecting a logical address of a flash memory, which uses an adjustable random access memory so as to simultaneously improve the semiconductor magnetism without adding too much space. The processing speed of the disc device. In order to achieve the above-mentioned object, the present invention first sets a plurality of regions in the flash memory.
第6頁 556205Page 6 556205
塊設定為 b& ss , 區域位址與其相對應邏輯位址 社糸、、充開機時先將實體 位址對映表,並將其儲存在一产關係’建立一等比例縮減 在存取資料時會先提供一待存取記憶體中;當系統 體,以根據該待查邏輯位址而 址至该隨機存取記憶 對之實體區域位址,進而搜尋今^位址對映表中得到一相 直至取得相對於該待查邏輯位‘ ::域内之每-區塊, 底下藉由具體實施例配合貝_位址為止。 容易瞭解本發明之目的、技^ ^的圖式詳加說明,當更 效。 T 谷、特點及其所達成之功 圖號說明: 10 待查邏輯位址 12 邏輯/實體位址對映表 14 實體區域 詳細說明: 機時=複數區塊作為一區域(zone) ’在系統開 Γ ίίϊ 對映表—PPlng Table 域實體位址對映表,並配合隨機存取 °中===!取之特性,搜尋在隨機存取記憶體 ^ r ^ 、义而知到實體區域,接續在該區域内尋找 而侍到相對於實體位址的邏輯位址。 個可i=ΐ憶體中進行資料儲存及讀取之使用時,每一 個可供資料存取的區塊都具有一實體位址,以代表在記憶The block is set to b & ss, the regional address corresponds to its corresponding logical address, and the physical address mapping table is stored at the time of power-on, and it is stored in the property relationship to establish a proportional reduction in access to data At first, a memory to be accessed will be provided; when the system body uses the logical address to be searched to obtain the physical area address of the random access memory pair, and then search the current address mapping table to get One phase is obtained until each logical block in the ':: domain' is obtained, and the bottom address is matched with a specific embodiment in the following. It is easy to understand the purpose of the present invention, and the drawings of the technique are explained in detail, which is more effective. T valley, characteristics and the figure of the work reached: 10 logical address to be checked 12 logical / physical address mapping table 14 detailed description of physical area: machine time = plural blocks as a zone 'in the system Open Γίίϊ Mapping Table—PPlng Table domain physical address mapping table, and with random access ° ===! To take the characteristics, search in the random access memory ^ r ^, meaning to know the physical area, It then looks for the logical address relative to the physical address in the area. When data can be stored and read in i = ΐ memory, each block that can be accessed by data has a physical address to represent the memory.
556205 五、發明說明(5) !::::實2!==:土發明係,個區域設定 的邏輯位址,如第_(af_u區域位置及其相對應 為非線性之對應匕(。)在圖二 根據第二(a)圖之關位本止糸之開機進仃初始化時’即會先 (Scaled ί 係表先複製建立-等比例縮減 第二⑻圖戶Γ示實體,址對映表,請同時參考 將其儲存在靜態隨機:取:上對映表之後’即 :;:rr 位址:==丄== 要佔用太多==縮;=7體位址對映表並不需 憶體之數目。 1所以可減少使用靜態隨機存取記 拖-,中丄上述由Ν個區域所組成之區域,該Ν值係為2η, 區域係由2"個區塊組成’較佳者,該η值係至 記,2 :開:ί ’若系統為一開放系統,如抽取式快閃 為不規則分佈’此時需要啟動2重組程 ^ Μ ^弟一圖所不,將邏輯位址相近似的磁區搬入同 磁ϋ ) ’例如:㉙區10〜19為同-區域z〇ne 〇, 磁〜4為區域Zone i,磁區3。〜39則同為區域Z繼 將斑Λ 近似的磁區搬入同一區域,則如圖所示 將磁㈣及13搬入區域2嶋〇,將磁區⑽“搬入區域 one 1,將磁區30搬入區域z〇ne k ;日後即使此抽取式快 第8頁 556205 五、發明說明(6) = ; = ;被移至其他系統存取,由於邏輯/實體位 == 規則,故可以保持相容性。*系統為封 閉糸、洗U夬閃圮憶體固定在系統内),一開始就 =方法,則可省略此重組程序。接著,系統會 t ;=f其相對應邏輯位址之關係’依系統SRAM的大 體位址ίΐί—如第二(b)圖所示之等比例縮減的邏輯/實 p機ϋ 並將此實體/邏輯位址對映表儲存在靜能 =查邏輯位址10,請參閱第四圖所示.二; A體位位址址的^份/訊U在靜態隨機存取記憶體内建之邏輯 Zone 〇 之實體區域二體Γ14之後,搜尋該ζ_ ° 位址之實體位址為I 直至取得相對於該待查邏輯 係將二快閃記憶體内寫入或更新資料時, 存空間’广,在同一個區域内’·若該區域内無足夠之儲 掣至另丄則必需將該區域内之原有資料連同新資料全部複 、至一區域内,以符合均勻讀寫之原則。 例d: it明之精神已說明完畢’以下特以-具體範 熟習此月上?快閃記憶體搜尋過程,並使 而據以實施。;σ >酌此範例之描述而獲得足夠的知識 此檢本發明檢測邏輯位址之流程圖,〜圖所示, 、、輯位址之方法係包括下列步驟:首先,如步驟 第9頁 556205 ·— — … 五 χχ»、發明說明(7) ' ' 11 — S 1 0所示,在系統開機時以箭 ^ ^ m ^ ^ , T 乂月11述之方式先將實體區域位址 與其相對應邏輯位址之關孫 實體位址對映表輯位址。建:-等比例縮減的邏;/ 記憶體中。 對映表,並將其儲存在一隨機存取 當系統如步驟S1 2所千德3 ^ ^ ^ ^ 斤不傳迗一待查邏輯位址至隨機存 作;二後,如二現機存'記憶體在接收到該待查邏輯位址 i 驟4所不,立刻根據該待查邏輯位址而 ==實/位址對映表中取得-相對於該邏輯位 S18所不比對判斷是否取的相 尤並二/二 終的實體位址,此時,罐f止之公位址’若獲付该取 束整個搜尋過·;;反:,P = : ’、如步娜所示結 驟S1 6,直至得到★亥實體巴找。号找到,即繼續進仃步 j °系貝體£域内相對之實體位U〇 由:本發明所快速建立的等比 位,二 係,所以者,故僅為部份之位址關 相對於邏輯位址,〜:己。體。且因係先找到 處筠1# ± 之貝體區域,在搜尋小範圍的择體區域, 處理速度較習知快。因A,本發 =的貝體= 使用少量的阡 糸利用區域之觀念配合 提下4記憶體,以便在不增加太多空間的前 度,故可d:導體磁碟ϊ置(快閃記憶體)的處理速 本發明同時兼:ϊ間之ί篁上取得一有效的平衡點,使 升糸統存取資料的性能。 叫點’以有效的如556205 V. Description of the Invention (5)! :::: Real 2! ==: Department of Soil and Inventory, the logical address set for each region, such as the position of the (__af_u) region and its corresponding non-linear correspondence. ) In Figure 2, according to the position of the second (a) figure, the system is initialized when it is initialized, that is, it will be scaled first. The scale is copied and established first. Mapping table, please also refer to storing it in the static random: take: after the mapping table 'that is:;: rr address: == 丄 == to occupy too much == contraction; = 7 body address mapping table and There is no need to recall the number of bodies. Therefore, the use of static random access can be reduced, and the above-mentioned area composed of N areas is 2η, and the area is composed of 2 " blocks. In the best case, the value of η is to be remembered, 2: Open: ‘If the system is an open system, such as the extraction flash is irregularly distributed’ At this time, it is necessary to start the 2 reorganization process. The magnetic regions with similar logical addresses are moved into the same magnetic region) 'For example: the regions 10 to 19 are the same-region zone 〇, the magnetic region 4 is the region Zone i, and the magnetic region 3. The region 39 is the same as the region Z. Move the magnetic field with the approximate Λ into the same area, then move the magnetic field and 13 into the area 2 嶋 〇, move the magnetic field ⑽ into the area one 1, and move the magnetic area 30 into the area zonek; This extraction is fast. Page 8 556205 V. Description of the invention (6) =; =; It has been moved to other systems for access. Because of the logic / physical bit == rule, compatibility can be maintained. * The system is closed. The U 夬 flash memory is fixed in the system). At the beginning, the method can be omitted. Then, the system will t; = f its corresponding logical address relationship according to the general address of the system SRAM. -Reduce the logical / real p machine proportionally as shown in the second (b) figure and store this physical / logical address mapping table at static energy = check logical address 10, please refer to the fourth figure Second; After the ^ copies of the A body address address / the U in the physical region of the logical zone 〇 built in the static random access memory, the second body Γ14 of the physical area, search for the physical address of the ζ_ ° address is I until the relative When the logic to be checked writes or updates data in the two flash memories, the storage space is wide, in the same area '· If there is not enough storage in the area to switch to another area, it is necessary to restore all the original data in the area together with the new data to an area to comply with the principle of uniform reading and writing. Example d: the spirit of it After the explanation is complete, the following special-specific examples are familiar with this month? The flash memory search process is implemented and implemented accordingly. Σ > According to the description of this example, sufficient knowledge is obtained. Check the present invention to detect the logical address The flow chart, as shown in the figure, includes the following steps: First, as in step 9 on page 556205 ····· 5χχ », invention description (7) '' 11 — S 1 0 It is shown that when the system is powered on, the physical address of the physical area and its corresponding logical address are mapped to the address list of the first place in the manner described by arrows ^ ^ m ^ ^, T 乂 11. Built:-Logical reduction logic; / in memory. Mapping table, and store it in a random access. When the system is as described in step S12, 2 ^ 3 ^ ^ ^ ^ Do not pass a logical address to be checked to the random storage; after the second, it is stored on the second machine. 'When the memory receives the logical address to be searched in step 4, it immediately obtains it from the real / address mapping table according to the logical address to be searched-to determine whether the logical bit S18 is not compared with the logical bit. Take the physical address of the second and second finals, at this time, the public address of the tank f ', if paid, the entire search has been done ;; Inverse :, P =:', as shown by Bruna Complete step S1 6 until you get ★ Hai Fangba. If the number is found, it will continue to advance. The relative physical position U in the shell body is in the same proportion. The second position is the same as the second position. Logical address, ~: self. body. In addition, because the corpus region of 1 # ± is found first, the processing speed is faster than that in the search of a small selection region. Because A, this hair = shell body = use a small number of ridges to use the concept of the area to remove 4 memory, so as not to add too much space before, so d: conductor disk set (flash memory The processing speed of the present invention also simultaneously: to achieve an effective balance point on the network, so as to improve the performance of the system to access data. Call the point ’to effectively
556205 五、發明說明(8) 以上所述之實施例僅係為說明本發明之技術思想及特 點,其目的在使熟習此項技藝之人士能夠瞭解本發明之内 容並據以實施,當不能以之限定本發明之專利範圍,即大 凡依本發明所揭示之精神所作之均等變化或修飾,仍應涵 蓋在本發明之專利範圍内。556205 V. Description of the invention (8) The above-mentioned embodiments are only for explaining the technical ideas and characteristics of the present invention. The purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. The limitation of the patent scope of the present invention, that is, all equal changes or modifications made according to the spirit disclosed by the present invention, should still be covered by the patent scope of the present invention.
556205 圖式簡單說明 圖式說明: 第一(a)圖及第一(b)圖分別為習知記憶體中實體位址與邏 輯位址之關係示意圖及利用該關係所建立之邏輯/實體位 址對映表。 第二(a)圖及第二(b)圖分別為記憶體中實體區域與邏輯位 址之關係示意圖以及利用本發明所建立之等比例縮減邏輯 /實體位址對映表。 第三圖為本發明於開放系統進行磁區重組程序之示意圖。 第四圖為本發明檢測邏輯位址之架構示意圖。 第五圖為本發明檢測邏輯位址之流程圖。556205 Schematic illustrations Schematic descriptions: The first (a) and the first (b) diagrams respectively show the relationship between the physical address and the logical address in the conventional memory and the logical / physical bit created by using the relationship Address mapping table. Figures 2 (a) and 2 (b) are the schematic diagrams of the relationship between the physical area and the logical address in the memory, and the proportional reduction logical / physical address mapping table established by the present invention. The third figure is a schematic diagram of the magnetic zone reorganization procedure in the open system of the present invention. The fourth figure is a schematic diagram of a structure for detecting a logical address according to the present invention. The fifth figure is a flowchart of detecting a logical address according to the present invention.
第12頁Page 12
Claims (1)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW91106951A TW556205B (en) | 2002-04-08 | 2002-04-08 | Method for inspecting flash memory logic address |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW91106951A TW556205B (en) | 2002-04-08 | 2002-04-08 | Method for inspecting flash memory logic address |
Publications (1)
Publication Number | Publication Date |
---|---|
TW556205B true TW556205B (en) | 2003-10-01 |
Family
ID=32228091
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW91106951A TW556205B (en) | 2002-04-08 | 2002-04-08 | Method for inspecting flash memory logic address |
Country Status (1)
Country | Link |
---|---|
TW (1) | TW556205B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI410795B (en) * | 2009-06-23 | 2013-10-01 | Phison Electronics Corp | Data writing method for flash memory and control circuit and storage system using the same |
-
2002
- 2002-04-08 TW TW91106951A patent/TW556205B/en not_active IP Right Cessation
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI410795B (en) * | 2009-06-23 | 2013-10-01 | Phison Electronics Corp | Data writing method for flash memory and control circuit and storage system using the same |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW564346B (en) | Memory device | |
CN103518193B (en) | Method and device for data deduplication | |
US20160253101A1 (en) | Memory Access and Detecting Memory Failures using Dynamically Replicated Memory | |
CN102841852B (en) | Wear leveling method, storing device and information system | |
US20170075718A1 (en) | Method and apparatus of accessing data of virtual machine | |
CN107784121B (en) | Lowercase optimization method of log file system based on nonvolatile memory | |
US9229714B2 (en) | Memory control apparatus, memory apparatus, information processing system, and processing method for use therewith | |
CN102239478B (en) | Data access processing method and device | |
CN106201331B (en) | Method and apparatus for writing data and storage medium | |
JP2008016003A (en) | Memory storage device and its method for reading and writing data | |
JP2002312130A5 (en) | ||
TW594740B (en) | Destructive-read random access memory system buffered with destructive-read memory cache | |
CN104077235A (en) | Method of partitioning physical block and memory system thereof | |
CN108604167A (en) | Technology for writing Z-operation | |
CN109117085A (en) | NAND flash memory device with NAND buffer | |
KR20180012565A (en) | Non-volatile memory system using volatile memory as cache | |
CN106844229B (en) | Organization method, system and device of solid state disk firmware mapping table | |
TWI569158B (en) | A database batch update method, data redo/undo log producing method and memory storage apparatus | |
CN102314321A (en) | Storage system and method and device for accessing data by using same | |
US11321002B2 (en) | Converting a virtual volume between volume types | |
JP3694501B2 (en) | Storage device | |
TW556205B (en) | Method for inspecting flash memory logic address | |
JP2003058417A (en) | Storage device | |
US11645202B2 (en) | Parsing method, parsing apparatus, electronic device, and computer storage medium | |
TW200931425A (en) | Method for managing flash memory blocks and controller using the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GD4A | Issue of patent certificate for granted invention patent | ||
MM4A | Annulment or lapse of patent due to non-payment of fees |