TWI342566B - Motherboard fault-finding testing method - Google Patents

Motherboard fault-finding testing method Download PDF

Info

Publication number
TWI342566B
TWI342566B TW96133544A TW96133544A TWI342566B TW I342566 B TWI342566 B TW I342566B TW 96133544 A TW96133544 A TW 96133544A TW 96133544 A TW96133544 A TW 96133544A TW I342566 B TWI342566 B TW I342566B
Authority
TW
Taiwan
Prior art keywords
memory
motherboard
value
error
error correction
Prior art date
Application number
TW96133544A
Other languages
Chinese (zh)
Other versions
TW200912937A (en
Inventor
Eric Li
Fred Cf Chen
Michael-Yh Chen
Yan-Min Wang
Original Assignee
Inventec Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Inventec Corp filed Critical Inventec Corp
Priority to TW96133544A priority Critical patent/TWI342566B/en
Publication of TW200912937A publication Critical patent/TW200912937A/en
Application granted granted Critical
Publication of TWI342566B publication Critical patent/TWI342566B/en

Links

Landscapes

  • Techniques For Improving Reliability Of Storages (AREA)
  • For Increasing The Reliability Of Semiconductor Memories (AREA)

Description

1342566 九、發明說明: ^【發明所屬之技術領域】 本發明係有關於一種主機板測試技術,更詳而言之, 係關於一種應用於具有至少一記憶體之電腦設備中之主 機板糾錯測試方法。 【先前技術】 電腦設備中配置之記憶體係用以配合中央處理單元 (Central process〇r Unit ; cpu)高速儲存正在執行之 鲁程式及資料,所以作為電腦系統中重要之元件,其可靠性 和容錯能力一直係業界研究之課題。 t 目則’業界已將錯誤檢查與糾錯技術(Err〇r1342566 IX. Description of the invention: ^ [Technical field to which the invention pertains] The present invention relates to a motherboard testing technique, and more particularly to a motherboard error correction applied to a computer device having at least one memory testing method. [Prior Art] The memory system configured in the computer equipment is used to cope with the central processing unit (Cpu) to store the programs and data being executed at a high speed. Therefore, it is an important component of the computer system, and its reliability and fault tolerance. Competence has always been the subject of industry research. t目目' Industry has error detection and error correction technology (Err〇r

Correcting Code ; ECC)應用於例如伺服器、工作站等電 .腦設備中,以透過該電腦設備之主機板測試配置於該電腦 口又備中之記憶體性能。而且,應用上述錯誤檢查與糾錯技 術之同時需配合使用一記憶體位元錯誤產生裝置,以透過 鲁垓6己憶體位元錯誤產生裝置令一記憶體產生記憶體位元 錯誤,俾供後續透過該主機板並應用該錯誤檢查與糾錯技 術執行記憶體性能測試作業。 请參閱第1圖,係顯示上述記憶體位元錯誤產生骏置 1應用於一電腦設備之記憶體2中之配置示意圖,其中, 该兄憶體2具有複數晶片單元,具體而,該記憶體即為 雙面針腳定義記憶體模組(Dual In_line Mem〇ryThe Correction Code (ECC) is applied to, for example, a server, a workstation, etc., in a brain device, to test the performance of the memory configured in the computer port through the motherboard of the computer device. Moreover, the application of the above error checking and error correction technology requires the use of a memory bit error generating device to cause a memory bit error to be generated by a memory device through a reckless 6 memory bit error generating device. The motherboard performs the memory performance test operation using the error checking and error correction techniques. Please refer to FIG. 1 , which is a schematic diagram showing the configuration of the memory bit error generating device 1 applied to the memory 2 of a computer device, wherein the brother cell 2 has a plurality of chip units, and specifically, the memory is Define a memory module for double-sided pins (Dual In_line Mem〇ry

Modules ; DIMM),而該晶片單元即為動態隨機存取記憶 體(Dynamic Random Access Memory ; DRAM),且各今曰 110327 5 1342566 :片單元具有一輸出引腳以及對應該輸出引腳之資料引 '腳,如圖所示,該記憶體位元錯誤產生裝置1係應用於該 記憶體2之一晶片單元21a、21b之輸出引腳211a、211b 與資料引腳2i3a、213b之間,以產生雙位元記憶體位元 錯誤,但不以此為限,亦可依該記憶體位元錯誤產生裝置 1應用於該記憶體2中之晶片單元之數量予以確定欲產生 之έ己憶體錯误位數。相應地,目前之主機板糾錯測試方法 係首先將該記憶體插入至該電腦設備主機板中之一設定 鲁δ己憶體插槽後’啟動遺電腦設備;接著,於該電腦設備啟 動後,透過配置於該電腦設備主機板中之記憶體控制器 (Memory Controller Hub ; MCH) 31 (如第 1 圖所示)向 該記憶體之一位址段連續寫入數值丨;隨後,透過上述記 憶體位元錯誤產生裝置斷開該記憶體之晶片單元之輸出 引腳與資料引腳之間之傳輸路徑以產生記憶體位元錯 块,然後,透過該記憶體控制器讀取自該位址段輸出之數 鲁值,並於自該位址段讀取之數值由寫入時之數值丨變為〇 時,透過該主機板中之記憶體控制器並應用該錯誤檢查與 糾錯技術記錄該記憶體出錯之相關訊息。 此外,如第1圖所示,由於配置於該電腦設備中之纪 憶體2之各該晶片單元之資料引腳透過一資料匯流排 (Data Bus ) 30並行連接至記憶體控制器3丨,以由該圮 憶體控制器31對該記憶體2之晶片單元進行資料讀寫作 業。然,當透過上述記憶體位元錯誤產生裝置〗斷開該記 憶體2之晶片單元21a、21b之輸出引腳211a、211b與資 110327 6 1342566 :料引腳213a、213b之間之傳輸路徑時,該記憶體2之晶 片單兀21 a、21 b之資料引腳21 3a、213b處於懸空狀態, •即此時’該資料引腳213a及213b之數值取決於藉由該記 憶體位7C錯誤產生裝置i斷開該輸出引腳2丨1 a、2 n b與 各δ亥資料引腳213a、213b之間之傳輸路徑時,該輸出引 腳211&、2111)輸出至該資料引腳2133、2131)之數值,由 於斷開時,該輸出引腳211 a、211 b輸出之數值係為亂數, 造成戎資料引腳213a、213b之數值亦為亂數,可能為數 #值0 ’亦可能為數值1。 如此’則透過該主機板糾錯測試方法預先經由該記憶 體控制器31向該記憶體2之一位址段連續寫入數值J 後,再由該記憶體控制器31自該位址段中讀取該記憶體 ‘ 2之各晶片單元之資料引腳輸出之數值,此時,該位址段 中對應該記憶體2中未配置該記憶體位元錯誤產生裝置j 之晶片單元之資料引腳之輸出數值仍舊為數值丨,而該位 #址段中對應該記憶體2中配置該記憶體位元錯誤產生裝 置1之晶片單元21a、21b之資料引腳213a、213b輸出之 數值則可能為數值1亦或數值〇,由於該資料引腳21 % 及213b之數值係為亂數,當透過該記憶體位元錯誤產生 裝置1斷開各該輸出引腳211a' 211b與各該資料引腳 213a、213b之間之傳輸路徑時,該資料弓丨腳213a、21北 之數值恰為數值〇時,則自該位址段讀取到之對應該資料 引腳ma、213b之輸出數值係為〇,此時,該主機板糾 、〗式方法可正確记錄泫圮憶體出錯相關訊息;而若斷開 110327 7 1342566 :時’該資料引腳213a、213b之數值恰為數值!時,則自 該位址段讀取到之對應該資料引腳213a、213b之輸出數 •值ϋΪ 1,,時’由於透過該主機板糾錯測試方法自該位 ^ /、項取之數值均為1 ,則無法偵測該記憶體2是 否出錯進而得到錯誤之測試結果。事實上,按照第1 圖之配置方式,5己憶體2上係產生雙位元記憶體錯誤,記 憶體位元錯誤測試結果應記錄該電腦設備之該記憶體2 位置產生有雙位元記憶體錯誤等相關訊息,而透過上述主 機板糾錯測試方法得到之記憶體出錯訊息係為一不確定 因素,極大影響該測試結果。 .^ 、’不上所述’如何提出一種可生成正確記憶體錯誤測試 -結果=主機板糾錯測試方法,以解決習知技術之缺失,實 為目前亟欲解決之技術問題。 、 【發明内容】 址鑒於上述習知技術之缺點,本發明之主要目的在於提 鲁仪種主機板糾錯測試方法,以獲取正確記憶體錯誤測試 結果’俾提升測試準確率。 為達上述目的及其他目的,本發明提供一種應用於具 有至少一記憶體之電腦設備中之主機板糾錯測試方法。^ $明之主機板糾錯測試方法係包括:將該記憶體與該電腦 设備電性連接,並啟動該電腦設備,透過該電腦設備之主 機板向4 s己憶體之一第一位址段連續寫入第一數值,同時 向該記憶體之-第二位址段連續寫入第二數值,該第_數 值與該第二數值為不同數值;於該記憶體上產生記憶體位 110327 8 1342566 體控制器分別同時向該記憶體之第一及第二位址段分別 連續寫入第一及第二數值,接著,於該記憶體上產生記憶 體位元錯誤,隨後,透過該主機板中之記憶體控制器讀取 自該第一及第二位址段輸出之數值’並分別比對自該第一 位址段以及該第二位址段寫入與讀取之數值,且於該第一 及該第二其中之一位址段寫入與讀取數值不一致時,記錄 6己體出錯相關訊息,藉此,無論透過一記憶體位元錯誤 產生裝置斷開该§己憶體之晶片單元之輸出引腳與資料引' 鲁腳之間之傳輸路徑的時,該記憶體之晶片單元之資料引腳 輸出之數值係為數值1亦或數值0,均可由其中一位址段 取得5己憶體出錯相關訊息,以得到正確測試結果。 【實施方式】 , 以下係藉由特定的具體實例說明本發明之實施方 式,熟悉此技藝之人士可由本說明書所揭示之内容輕易地 瞭解本發明之其他優點與功效。本發明亦可藉由其他不同 φ的具體實例加以施行或制,本說明書中的各項細節亦可 基於不同觀點與應用,在不恃離本發明之精神下進行各種 修飾與變更。 。請參閱第2圖,係顯示本發明之主機板糾錯測試方法 之操作流程示意圖。請同時參閱第1圖,本發明之主機板 糾錯測試方法係應用於具有至少一記憶體之電腦設備 中’於本實施例中’該電腦設備中係為配置有記憶體卜 2 ’而-記憶體位元錯誤產生裝置i係應用於該記憶體2 如第1圖所不),但不以此為限,其中,該記憶體2 110327 10 1342566 、具有複數晶片單元21a ( 21b等),且各該晶片單元2U (21b)具有一輸出引腳211a (211b)以及對應該輸出引 腳211a ( 211b)之資料引腳213a ( 213b);此外,該電 腦設備係為搭載有記憶體錯誤檢查與糾錯(Modules; DIMMs, and the wafer unit is a Dynamic Random Access Memory (DRAM), and each of the current 110327 5 1342566: chip units have an output pin and data corresponding to the output pin 'Foot, as shown, the memory bit error generating device 1 is applied between the output pins 211a, 211b of one of the chip units 21a, 21b of the memory 2 and the data pins 2i3a, 213b to generate a double The bit memory bit is wrong, but not limited thereto, and the number of bit errors to be generated may be determined according to the number of the chip cells used by the memory bit error generating device 1 in the memory 2. . Correspondingly, the current motherboard error correction test method first inserts the memory into one of the computer device motherboards to set the Lu δ 忆 体 体 slot, and then starts the legacy computer device; then, after the computer device is started, And continuously writing a value to one of the address segments of the memory through a memory controller (MCH) 31 (shown in FIG. 1) disposed in the motherboard of the computer device; The memory bit error generating device disconnects a transmission path between the output pin of the memory unit of the memory and the data pin to generate a memory bit error block, and then reads the address bit from the memory controller through the memory controller The number of outputs is a Lu value, and when the value read from the address segment is changed from the value at the time of writing to 〇, the memory controller in the motherboard is used and the error checking and error correction technique is applied to record the value. A message about a memory error. In addition, as shown in FIG. 1, the data pins of the respective chip units of the memory device 2 disposed in the computer device are connected in parallel to the memory controller 3 through a data bus (Data Bus) 30, The data read and write operation is performed on the wafer unit of the memory 2 by the memory controller 31. However, when the transmission path between the output pins 211a and 211b of the wafer units 21a and 21b of the memory 2 and the resources 110327 6 1342566: the material pins 213a and 213b are disconnected by the memory bit error generating means. The data pins 21 3a, 213b of the wafer unit 21 a, 21 b of the memory 2 are in a floating state, that is, the value of the data pins 213a and 213b at this time depends on the error generating device by the memory bit 7C. i disconnects the output path between the output pins 2丨1 a, 2 nb and each of the δ 资料 data pins 213a, 213b, and the output pins 211 & 2111) are output to the data pins 2133, 2131) The value of the output pin 211 a, 211 b is a random number when it is disconnected, so that the value of the data pin 213a, 213b is also a random number, which may be a number # value 0 ' may also be a numerical value 1. Thus, the value J is continuously written to an address segment of the memory 2 via the memory controller 31 through the error correction test method of the motherboard, and then the memory controller 31 is from the address segment. Reading the data pin output value of each chip unit of the memory '2, at this time, the data pin of the chip unit corresponding to the memory bit error generating device j in the memory 2 corresponding to the address field The output value is still a value 丨, and the value of the data pin 213a, 213b of the chip unit 21a, 21b of the memory unit 2 in the memory 2 corresponding to the memory bit error generating device 1 may be a value. 1 or the value 〇, since the values of the data pins 21% and 213b are random numbers, when the memory bit error generating device 1 disconnects each of the output pins 211a' 211b and each of the data pins 213a, In the transmission path between 213b, when the value of the north of the data pin 213a, 21 is exactly the value 〇, the output value corresponding to the data pin ma, 213b read from the address segment is 〇, At this time, the motherboard correcting and arranging method Hyun properly destroyed, memory and recording the error-related messages; 11032771342566 and disconnected if: when 'the data pins 213a, 213b of the numerical value is exactly! When the number of outputs corresponding to the data pins 213a, 213b is ϋΪ 1, from the address segment, the value is taken from the bit ^ /, by the error correction test method of the motherboard. If both are 1, it is impossible to detect whether the memory 2 is in error or not and obtain an incorrect test result. In fact, according to the configuration of Figure 1, the double-bit memory error is generated on the 5 memory, and the memory bit error test result should record the memory device 2 position of the computer device to generate double-bit memory. Errors and other related information, and the memory error message obtained through the above-mentioned motherboard error correction test method is an uncertain factor, which greatly affects the test result. .^, 'Not on the above' How to propose a fault test that can generate correct memory - Result = Motherboard error correction test method to solve the lack of the prior art, which is the technical problem that is currently being solved. SUMMARY OF THE INVENTION In view of the above-mentioned shortcomings of the prior art, the main purpose of the present invention is to improve the error detection test method of the motherboard of the instrument type to obtain the correct memory error test result. To achieve the above and other objects, the present invention provides a motherboard error correction test method for use in a computer device having at least one memory. ^ The clear motherboard test method includes: electrically connecting the memory to the computer device, and starting the computer device, through the motherboard of the computer device to the first address of the 4 s memory The segment continuously writes the first value, and simultaneously writes a second value to the second address segment of the memory, the _value is different from the second value; and the memory bit 110327 is generated on the memory. 1342566 The body controller simultaneously writes the first and second values to the first and second address segments of the memory, respectively, and then generates a memory bit error on the memory, and then passes through the motherboard. The memory controller reads the value outputted from the first and second address segments and compares the values written and read from the first address segment and the second address segment, respectively, and When the first and the second address segment writes are inconsistent with the read value, the 6-bit error related information is recorded, thereby disconnecting the DRAM from the memory through the memory bit error generating device. Unit output pin and data lead 'Lu foot When the transmission path is between, the value of the data pin output of the chip unit of the memory is a value of 1 or a value of 0, and one of the address segments can obtain 5 error-related information for correct test. result. The embodiments of the present invention are described below by way of specific examples, and those skilled in the art can readily appreciate the other advantages and effects of the present invention from the disclosure herein. The present invention may be embodied or carried out by other specific examples of the φ, and various modifications and changes may be made without departing from the spirit and scope of the invention. . Referring to Fig. 2, there is shown a flow chart showing the operation of the error correction test method for the motherboard of the present invention. Please also refer to FIG. 1 , the method for testing the error correction of the motherboard of the present invention is applied to a computer device having at least one memory. In the present embodiment, the computer device is configured with a memory device 2 ' The memory bit error generating device i is applied to the memory 2 as shown in FIG. 1 , but not limited thereto, wherein the memory 2 110327 10 1342566 has a plurality of chip units 21a ( 21b, etc.), and Each of the chip units 2U (21b) has an output pin 211a (211b) and a data pin 213a (213b) corresponding to the output pin 211a (211b); and the computer device is equipped with a memory error check and Error correction

Correcting Code ; ECC)測試軟體之作業系統;再者,該 電腦設備之主機板復配置有記憶體控制器(Mem〇ryCorrecting Code; ECC) test software operating system; in addition, the motherboard of the computer device is configured with a memory controller (Mem〇ry

Controller Hub ; MCH )、複數資料匿流排⑼ataBus)、 以及複數供插接該記憶體之記憶體插槽(未圖示),由於 #本發明之主機板糾錯測試方法均可應用於習知電腦設備 之所有記憶體與記憶體控制器之連接方式中,即該電腦設 備之各該記憶體係透過各該資料匯流排並行連接至該記 憶體控制器,以由該記憶體控制器對各該記憶體2、2,之 •晶片單元進行資料讀寫作業,故為簡化說明及圖式,並使 本發明之特徵及結構更為清晰易懂,乃於圖式中僅顯示出 與本發明直接關聯之結構,其餘部份則省略。以下將一併 配合第1圖詳細說明本發明之主機板糾錯測試方法之具 體操作步驟。 如第2圖所示,首先進行步驟Si〇,將該記憶體插入 至該電腦設備之主機板之一設定記憶體插槽後,啟動該電 月a 6又備。其中,該記憶體之記憶體插槽之設定規則係為當 該電腦設備《各該記憶體插#中僅插接單一記憶體(僅為 一配置有該記憶體位元錯誤產生裝置丨之記憶體2存在) 時:係將離該記憶體控制器最遠之記憶體插槽中作為該設 定汜憶體插槽,以插接該記憶體,俾供產生合理之記憶體 π ]10327 1342566 配置由於此記憶體配置方式係為所屬技術領域熟知技 術,在此不再為文贅述。此外,於本實施例中,啟動該電 月句°又備係指啓動該電腦設備之主機板並進入該電腦設備 業系統後,再運行該記憶體錯誤檢查與糾錯測試軟體 之操作步驟。接著進行步驟S20。Controller Hub; MCH), complex data stream (9) ataBus), and a plurality of memory slots (not shown) for plugging the memory, since the motherboard error correction test method of the present invention can be applied to conventional In the connection mode of all the memory devices of the computer device and the memory controller, the memory system of the computer device is connected to the memory controller through the data busbars in parallel, so that the memory controller The memory unit 2, 2, and the chip unit perform data reading and writing operations, so that the description and drawings are simplified, and the features and structures of the present invention are more clearly understood, and only the direct description of the present invention is shown in the drawings. The structure of the association is omitted. The specific operation steps of the error correction test method for the motherboard of the present invention will be described in detail below with reference to Fig. 1. As shown in Fig. 2, first, step Si is performed, and the memory is inserted into one of the motherboards of the computer device to set the memory slot, and then the battery is started. Wherein, the setting rule of the memory slot of the memory is that when the computer device "only inserts a single memory into the memory plug # (only a memory configured with the memory bit error generating device) When the body 2 exists): the memory slot farthest from the memory controller is used as the setting memory slot to insert the memory for generating a reasonable memory π ] 10327 1342566 configuration Since this memory configuration is well known in the art, it will not be described herein. In addition, in the embodiment, the operation of the memory error check and the error correction test software is performed after the motherboard of the computer device is started and the computer equipment system is started. Next, step S20 is performed.

於步驟S20中,於該電腦設備啟動後,透過該電腦設 備之主機板向該記憶體之一第一位址段連續寫入第一數 值同時向該記憶體之一第二位址段連續寫入第二數值。 更洋而s之,係透過配置於該主機板中之記憶體控制器向 5亥圮憶體之第一及第二位址段分別寫入第一及第二數 ,亦即,該第一數值及該第二數值係為相異之二進位數 f '數值0或1),於一實施例中’該第一數值係為數值 八/第一數值係為數值1,以32位元的記憶體為例,即 :二別同時向該第一位址段寫入0x00000000(32位元資 =二為〇),向該第二位址段寫入〇xffffffff(32位元資 /,全為1);於另一實施例中,亦可係該第一數值為數值 同^亥第二*數值為數值〇,以32位記憶體為例,即為分別 υΒ、向该第一位址段寫入位元資料全為 向5亥第二位址段寫入0x00000000(32位元資料全為 1此外,該第一位址段與該第二位址段之間距不小於4 疋組。接著進行步驟S30。 於^驟S3G中,於該記憶體上產生記憶體位元錯誤。 過一而言’如第1圖所示,該記憶體位元錯誤之產生係透 〇隐體位元錯誤產生裝置1應用於該記憶體2之晶片 110327 12 丄:> 舛/:)〇〇 甲—2 2m、2ib之輸出引腳2ua、2ub與資料引腳ma、 斷開奸Φ予以實現者’以由該記憶體位元錯誤產生裝置1 的傳輸二弓:腳211a、211b與資料引腳213a、213b之間 g . ,向今、琢記憶體2產生記憶體位元錯誤,亦 輪心丨㈣對應騎Μ腳之資料引腳之間呈縣 =時’而該產生記憶體位元錯誤之記憶體係J電: :肴中配置該憶體位元錯誤產生裝置!之記憶體 進行步驟S40。 一於步驟S40中,透過該電腦設備之主機板讀取自該第 ^及^二位址段輸出之數值。更詳而言之,係透過配置於 。玄電知设備之該主機板中之記憶體控制器利用各該資料 匯流排分別訪問該記憶體2之第一及第二位址段,以自該 第及第一位址段讀取該記憶體2之各晶片單元之資料 引腳輸出之數值。接著進行步驟S5〇。 於步驟S 5 0中,分別比對自該第一位址段以及該第二 籲位址段寫入與讀取之數值,並於該第一及第二位址段兑中 之一者寫入與讀取數值不一致時,記錄記憶體出錯之相關 訊息。為更明確瞭解應用本發明之主機板糾錯測試方法如 何實現準確記錄記㈣出錯之相關訊息,纟此將以向該記 憶體寫入之第-數值為數值〇、第二數值為數值】(但不 以此為限)為範例進-步說明。如第i圖所示,#透過該 έ己憶體位兀錯誤產生裝置1斷開該記憶體2之晶片單元 21a、21b之各該輸出引腳2lla、2iib與各該資料引腳 213a、213b之間之傳輸路徑時’該資料引腳213a、以扑 110327 13 1342566 >之數值恰均為數值〇時,則自該第一及第二位址段讀取到 ·之對應該資料引腳213a、213b之輸出數值分別為數值〇、 -〇,此時,即使比對該第一位址段中寫入與讀取數值均為 〇 ;未能從中偵測到該記憶體2出錯,亦可由比對該第二 位址段中寫入與讀取數值不一致,偵測到該記憶體\出 錯;倘若透過該記憶體位元錯誤產生裝置丨斷開各該輸出 引腳211&、2111)與各該資料引腳213&、21礼之間之傳輸 路徑時,該資料引腳213a、213b之數值恰均為數值丨時^ •則自該第一及第二位址段讀取到之對應該資料引腳 213a、213b之輸出數值則分別為數值}、i,此時,即使 比對該第二位址段中寫入與讀取數值均為丨,未能從中偵 測到該記憶體2出錯,亦可由比對該第一位址段中寫入與 •讀取數值不一致,偵測到該記憶體2出錯。亦即,無論透 過泫§己憶體位元錯誤產生裝置丨斷開各該輸出引腳 211a、211b與各該資料引腳213a、21扑之間之傳輸路徑 •時,该記憶體2之晶片單元21a、21b之資料引腳2仏、 213b輸出之數值係為數值丨亦或數值〇,均可透過本發明 =機板糾錯測試方法由其中一位址段獲取正確之測試 綜上所述,本發明之主機板糾錯測試方法係先行將記 電腦設備主機板之一設定記憶體插槽後’啟動 違電月㈣備’之後,再透過配置該主機板中之記 器^別同時向該記憶體之第一及第二位址段分料續工寫制 入弟-及第二數值’接著,於該記憶體上產生記憶體位元 110327 14 1342566 錯誤;隨後,復透過該記憶體控制器讀取自該第一及第二 位址段輸出之數值,並分別比對自該第—位址段以及該第 •二位址段寫入與讀取之數值,且於該第—及該第二其中之 位址段寫入與讀取數值不一致時,記錄記憶體出錯相關 訊息,藉此,無論透過一記憶體錯誤位元產生裝置斷開該 記憶體之晶片單兀之輸出引腳與資料引 徑時,該記憶體之晶片單元之資料弓丨_^= 值卜亦或數值0,均可由其中一位址段取得記憶體出錯相 鲁關訊息,以得到正確測試結果,極大地提高了測試正確率。 上述實施例僅例示性說明本發明之原理及其功效,而 非用於限制本發明。任何熟習此項技藝之人士均可在不^ '背本發明之精神及範訂,對上述實施例進行修飾斑改逆 '變。因此,本發明之權利保護範圍,ys & # > ^ 範圍所列。 應如後逑之申請專利 【圖式簡單說明】In step S20, after the computer device is started, the first value of one of the first address segments of the memory is continuously written by the motherboard of the computer device, and the second address segment of the memory is continuously written. Enter the second value. Further, the first and second numbers are respectively written to the first and second address segments of the 5H memory through the memory controller disposed in the motherboard, that is, the first The value and the second value are the distinct binary digits f 'value 0 or 1). In an embodiment, the first numerical value is a value of eight / the first numerical value is a value of 1, and is 32 bits. The memory is taken as an example, that is, two: at the same time, write 0x00000000 (32-bit element = two for 〇) to the first address segment, and write 〇xffffffff to the second address segment (32-bit yuan/, all In another embodiment, the first value may be a numerical value and the second value of the second value is a numerical value, and the 32-bit memory is taken as an example, that is, respectively, to the first address. The segment write bit data is all written to 0x00000000 to the second address segment of 5H (32-bit data is all 1 in addition, the distance between the first address segment and the second address segment is not less than 4 疋 group. Then, step S30 is performed. In step S3G, a memory bit error is generated on the memory. In other words, as shown in FIG. 1, the memory bit error is generated through the hidden body bit error. The generating device 1 is applied to the wafer 110327 of the memory 2 12 丄: > 舛 / :) armor - 2 2m, 2ib output pin 2ua, 2ub and data pin ma, disconnected Φ to achieve ' The memory bit error is generated by the transmission two bows of the memory bit error generating device 1 between the legs 211a and 211b and the data pins 213a and 213b, and the memory bit error is generated in the memory 2, and the wheel center (4) corresponds. The data between the pins of the riding foot is county=time' and the memory system that generates the memory bit error J:: The memory bit error generating device is arranged in the dish! The memory proceeds to step S40. In step S40, the motherboard outputted from the first and second address segments is read by the motherboard of the computer device. In more detail, it is configured by . The memory controller in the motherboard of the Xundian device accesses the first and second address segments of the memory 2 by using the data busbars respectively, to read the first and second address segments from the first address segment The value of the data pin output of each chip unit of the memory 2. Then proceed to step S5. In step S50, the values written and read from the first address segment and the second address segment are respectively compared, and one of the first and second address segments is written. When the input and read values are inconsistent, the memory related error message is recorded. In order to more clearly understand how the error correction test method of the motherboard to which the present invention is applied is to accurately record the information related to the error (4), the first value written to the memory is the value 〇 and the second value is the value] But not limited to this) for the example step-by-step instructions. As shown in the figure i, the error generating means 1 disconnects the output pins 2lla, 2iib of the chip units 21a, 21b of the memory 2 and the respective data pins 213a, 213b. In the case of the transmission path between the data pin 213a and the value of the 110327 13 1342566 > are all numerical values, the corresponding data pin 213a is read from the first and second address segments. The output values of 213b are respectively 〇, -〇. At this time, even if the value written and read in the first address segment is 〇; if the memory 2 is not detected from the error, it may be Comparing the written and read values in the second address segment, the memory\error is detected; if the memory bit error generating device is used, the output pins 211&, 2111) and each are disconnected. When the data pin 213&, 21 is in the transmission path, the data pins 213a, 213b are all numerical values ^ ^ then read from the first and second address segments corresponding to The output values of the data pins 213a, 213b are respectively the values}, i, at this time, even if compared to the second The value written and read in the address segment is 丨, the memory 2 error is not detected from it, and the memory is detected by inconsistent with the written and read values in the first address segment. Body 2 error. That is, the memory unit of the memory 2 is disconnected when the transmission path between the output pins 211a and 211b and each of the data pins 213a and 21 is disconnected by the device. 21a, 21b data pin 2仏, 213b output value is a numerical value or a value 〇, can be obtained through the invention = machine board error correction test method from one of the address segments to obtain the correct test summary, The method for correcting the error of the motherboard of the present invention is to first record the memory slot of one of the computer device motherboards and then start the power-off month (four) preparation, and then configure the recorder in the motherboard to simultaneously The first and second address segments of the memory are successively written into the younger-and second value. Then, the memory bit 110327 14 1342566 is generated on the memory; then, the memory controller is repeatedly transmitted through the memory controller. Reading the values output from the first and second address segments, and respectively comparing the values written and read from the first address segment and the second address segment, and in the first and the When the second address segment is written and inconsistent with the read value, Recording a memory error related message, whereby the data of the chip unit of the memory is broken when the output pin and the data path of the memory chip of the memory are disconnected by a memory error bit generating device. ^= Value or value 0, can get the memory error phase information from one of the address segments to get the correct test results, greatly improving the test accuracy. The above-described embodiments are merely illustrative of the principles of the invention and its effects, and are not intended to limit the invention. Anyone skilled in the art can modify the above-described embodiments without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is listed in the range of ys &#> ^. Should apply for a patent as the latter [Simplified illustration]

第1圖係顯示記憶體位元錯誤產生裝置 體中之配置示意圖;以及 °憶 第2圖係顯示本發明之主機板糾錯測 流程示意圖。 ίτ之鉍作 【主要元件符號說明】 1 圮憶體位元錯誤產生裝置 2' 2, 記憶體 21a、21b 晶片單元 211a' 211b輸出引腳 110327 15 1342566 v 213a、213b 數據引腳 • 30 資料匯流排 '31 記憶體控制器Fig. 1 is a view showing a configuration of a memory bit error generating device; and Fig. 2 is a view showing a flow chart of error correction of the motherboard of the present invention. ί 铋 【 主要 主要 主要 主要 主要 主要 主要 主要 主要 主要 主要 主要 主要 主要 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 体 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 110 '31 memory controller

SiO〜S50 步驟SiO~S50 step

16 11032716 110327

Claims (1)

1342566 、申請專利範面: 種主機板糾錯測試方法,係應用於且有至I、— 體之電腦設備中,該主機板糾錯測試;:法係^記憶 將該記憶體與該電腦設備電性連接’並 h 腦設備,透過該電腦設備之主機板向該記憶體之 一位址段連續寫人第一數值,同時向該記憶體卜ί -位址段連績寫人第二數值,該第—數值與 = 值為不同數值; —數 於该記憶體上產生記憶體位元錯誤; 透過該電腦設備之主機板讀取自該第__及第二 位址段輸出之數值;以及 分別比對自該第一位址段以及該第二位址 入與讀取之數值,並於該第—及”二位址段其中 4致時’記錄該記憶體產生該 位元錯誤的相關訊息。 Λ 2. 如申請專利範圍第i項之主機板糾錯測試方法,其 中’該電腦設備係搭載有記憶體錯誤檢查斑叫錯、 (ΕΓΓ〇Γ C〇rreCting Code1CC)測試軟體之;業系 統。 3· 如中請專利範Μ 2項之主機板糾錯測試方法,其 中,啟動該電腦設備之步驟係'為啓動該電腦設狀主 機板並進人該電腦設備之作㈣統後再運行該記憶 體錯誤檢查與糾錯測試軟體之操作步驟。 如申請專利第1項之主機板糾錯測試方法,其 110327 17 4.1342566, patent application paradigm: a kind of motherboard error correction test method, which is applied to the computer equipment of the I, the body, the error correction test of the motherboard;: the legal system ^ memory the memory and the computer equipment Electrically connecting the 'h brain device, through the motherboard of the computer device, continuously writes the first value to one of the address segments of the memory, and writes the second value to the memory address of the memory block. The first value and the = value are different values; - the number of memory bit errors is generated on the memory; the value output from the first __ and the second address segment is read by the motherboard of the computer device; The value of the input and the read from the first address segment and the second address is recorded, and the memory generates a related message of the bit error in the first and second address segments. Λ 2. For the error correction test method of the motherboard in the scope of patent application i, in which 'the computer equipment is equipped with a memory error check spot error, (ΕΓΓ〇Γ C〇rreCting Code1CC) test software; industry system. 3· If you please Fan Wei 2 item motherboard error correction test method, wherein the step of starting the computer device is to start the computer to set up the motherboard and enter the computer device (4) and then run the memory error check and error correction The test procedure of the test software. For example, the error correction test method of the motherboard of claim 1 is 110327 17 4. 二位址段之間距不小於4 位 中,該第一位址段與該第 元組。 6. ^申請專利範圍第丨項之主機板糾錯測試方法,其 ^該記憶體具有複數晶片單元,且各該晶片單^且 有一輸出引腳以及對應該輸出引腳之資料引腳。 如申請專利範圍第5項之主機板糾錯測試方法,其 »亥。己憶體位7〇錯誤之產生係使該輸出引腳與對應 h輪出引腳之資料引腳之間呈懸浮狀態時,而^ : 憶體產生記憶體位元錯誤。 X〇 如申請專利範圍第5項之主機板糾錯測試方法,其 中,忒έ己憶體之各該晶片單元之資料引腳透過一資料 匯流排(Data Bus)分別連接至該主機板上的一記憶 體控制器(Memory Controller Hub ; MCH)中,以^ 泫记憶體控制器對該記憶體之各該晶片 引腳進行》料存取。 8.如申請專利範圍第7項之主機板糾錯測試方法,其 中,該電腦設備之主機板設置有一用以插接該記憶體 之設定記憶體插槽,當該電腦設備僅插接單—記憶體 時,係將離該記憶體控制器最遠之記憶體插槽作^該 設定記憶體插槽。 ~ 9.如申請專利範圍第1項之主機板糾錯測試方法,其 中,該第一數值與該第二數值係為數值〇或數值】。 110327 18The distance between the two address segments is not less than 4 bits, and the first address segment is the same as the first group. 6. The method of claim 4, wherein the memory has a plurality of chip units, and each of the chips has an output pin and a data pin corresponding to the output pin. For example, the error correction test method for the motherboard of the fifth application patent scope is: Hai. The memory of the 7-bit error is caused by the floating state between the output pin and the data pin of the corresponding h-wheeling pin, and ^: the memory element generates a memory bit error. X. For example, the motherboard error correction test method of claim 5, wherein the data pins of each of the wafer units are connected to the motherboard through a data bus (Data Bus). In a memory controller (MCH), the memory controller of the memory is accessed by the memory controller. 8. The error correction test method for a motherboard according to claim 7, wherein the motherboard of the computer device is provided with a setting memory slot for inserting the memory, and when the computer device is only plugged in - In the case of the memory, the memory slot farthest from the memory controller is used as the setting memory slot. ~ 9. The main board error correction test method of claim 1, wherein the first value and the second value are numerical values or numerical values. 110327 18
TW96133544A 2007-09-07 2007-09-07 Motherboard fault-finding testing method TWI342566B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW96133544A TWI342566B (en) 2007-09-07 2007-09-07 Motherboard fault-finding testing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW96133544A TWI342566B (en) 2007-09-07 2007-09-07 Motherboard fault-finding testing method

Publications (2)

Publication Number Publication Date
TW200912937A TW200912937A (en) 2009-03-16
TWI342566B true TWI342566B (en) 2011-05-21

Family

ID=44725061

Family Applications (1)

Application Number Title Priority Date Filing Date
TW96133544A TWI342566B (en) 2007-09-07 2007-09-07 Motherboard fault-finding testing method

Country Status (1)

Country Link
TW (1) TWI342566B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI689812B (en) * 2018-11-30 2020-04-01 英業達股份有限公司 Method of locating the location of an error of a memory device

Also Published As

Publication number Publication date
TW200912937A (en) 2009-03-16

Similar Documents

Publication Publication Date Title
CN107430538B (en) Dynamic application of ECC based on error type
US10204698B2 (en) Method to dynamically inject errors in a repairable memory on silicon and a method to validate built-in-self-repair logic
US10169126B2 (en) Memory module, memory controller and systems responsive to memory chip read fail information and related methods of operation
US11200961B1 (en) Apparatus, system and method to log memory commands and associated addresses of a memory array
US20160117219A1 (en) Device, system and method to restrict access to data error information
US8862953B2 (en) Memory testing with selective use of an error correction code decoder
WO2016160274A1 (en) Extracting selective information from on-die dram ecc
US20200243159A1 (en) Stacked memory apparatus using error correction code and repairing method thereof
TWI359424B (en) Memory, repair system and method for testing the s
US10606692B2 (en) Error correction potency improvement via added burst beats in a dram access cycle
CN102339647A (en) Detection method and apparatus for error checking and correcting (ECC) check module
US9437327B2 (en) Combined rank and linear address incrementing utility for computer memory test operations
US10725672B2 (en) Memory module, memory controller and systems responsive to memory chip read fail information and related methods of operation
US9009548B2 (en) Memory testing of three dimensional (3D) stacked memory
US7293221B1 (en) Methods and systems for detecting memory address transfer errors in an address bus
US20220139485A1 (en) Memory device capable of outputting fail data in parallel bit test and memory system including the memory device
CN111078462B (en) Data verification method and circuit
CN102339648A (en) Error-detection/error-correction verification module detection method and device
TWI342566B (en) Motherboard fault-finding testing method
US9110796B2 (en) Apparatus and circuitry for memory-based collection and verification of data integrity information
US20230289270A1 (en) Host controlled electronic device testing
EP4180960A1 (en) Error correction circuit, memory system, and error correction method
TWI353608B (en) Bit-error generating device
US10229057B2 (en) Method and apparatus for avoiding bus contention after initialization failure
CN101387978A (en) Main board error correction testing method

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees