CN118277306B - Data eye training method and device for physical layer of memory - Google Patents
Data eye training method and device for physical layer of memory Download PDFInfo
- Publication number
- CN118277306B CN118277306B CN202410687052.4A CN202410687052A CN118277306B CN 118277306 B CN118277306 B CN 118277306B CN 202410687052 A CN202410687052 A CN 202410687052A CN 118277306 B CN118277306 B CN 118277306B
- Authority
- CN
- China
- Prior art keywords
- sample data
- data signals
- eye
- delay
- sampling clock
- 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.)
- Active
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
-
- 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
- G06F13/38—Information transfer, e.g. on bus
- G06F13/42—Bus transfer protocol, e.g. handshake; Synchronisation
- G06F13/4204—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
- G06F13/4234—Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being a memory bus
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
Abstract
The invention provides a data eye diagram training method and a device for a physical layer of a memory, which are characterized in that the initial delay corresponding to each of a plurality of sample data signals is determined based on a first eye diagram corresponding to each of the plurality of sample data signals, a delay chain corresponding to each of the plurality of sample data signals is pushed based on the initial delay corresponding to each of the plurality of sample data signals, then right alignment is carried out on a second eye diagram corresponding to each of the plurality of sample data signals, when a pair Ji Yan corresponding to each of the plurality of sample data signals is obtained, and meanwhile, the delay of a data sampling clock signal relative to each of the sample data signals is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagram corresponding to each of the plurality of sample data signals, so that the eye diagram alignment eye diagram of each of the sample data signals can be realized only by training for two to three times, the data training efficiency is effectively improved, and meanwhile, the eye diagram width and eye height of each sample data signal are also improved.
Description
Technical Field
The present invention relates to the field of memory technologies, and in particular, to a method and apparatus for training a data eye for a physical layer of a memory.
Background
The DRAM (Dynamic Random-Access Memory) or FLASH (FLASH) granules need to be trained before performing read and write operations to calibrate the timing difference (skew) between DQS (data strobe) and DQ (data line) due to line delay on PCB (Printed Circuit Board ) and other factors, so as to ensure the stability of PHY (PHYSICAL LAYER ) when data is read and written at high speed. The conventional training method is to push the delay chain of each DQ to the maximum value and then move DQs to train the left and right eyes to obtain the DQ signal, but since the DQ delay is not calculated by the algorithm, multiple training is required, the order of magnitude is about tens of times (depending on the size of the delay chain of DQ), so that it takes a long time and the training efficiency is low. In addition, this approach, after training, when the timing difference between the data lines is large, results in overlapping of the data at the time of sampling, thereby reducing the width and height of the eye pattern of each DQ, and failing to obtain the eye pattern optimal sampling point.
Disclosure of Invention
The invention provides a data eye diagram training method and device for a physical layer of a memory, which are used for solving the defects that the training efficiency is low and the width and the height of an eye diagram of each DQ are small when the time sequence difference between data lines is large in the prior art.
The invention provides a data eye training method for a physical layer of a memory, which comprises the following steps:
Acquiring a data sampling clock signal and a plurality of sample data signals;
Scanning the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
determining initial delays corresponding to the plurality of sample data signals based on the first eye diagrams corresponding to the plurality of sample data signals, and pushing delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
scanning the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
And right-aligning the second eye patterns corresponding to the sample data signals to obtain a pair Ji Yanshi corresponding to the sample data signals, and determining the delay of the data sampling clock signal relative to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the sample data signals.
According to the method for training a data eye diagram of a physical layer of a memory, the right alignment is performed on the second eye diagrams corresponding to the plurality of sample data signals to obtain alignment delay of the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagrams corresponding to the plurality of sample data signals, which further includes:
Pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye patterns corresponding to the plurality of sample data signals respectively;
Based on the respective third eye patterns of the plurality of sample data signals, a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined.
According to the method for training a data eye diagram of a physical layer of a memory, the determining a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to each sample data signal based on a third eye diagram corresponding to each of the plurality of sample data signals specifically includes:
And right-aligning the third eye patterns corresponding to the sample data signals to obtain fine alignment delay of the sample data signals, and determining the fine delay of the data sampling clock signal relative to the sample data signals based on the middle position of the third eye pattern with the narrowest eye width in the third eye patterns corresponding to the sample data signals.
According to the method for training a data eye diagram of a physical layer of a memory, the right alignment is performed on the second eye diagrams corresponding to the plurality of sample data signals to obtain the pair Ji Yanshi corresponding to the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagrams corresponding to the plurality of sample data signals, which specifically includes:
Determining a second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals as a reference eye pattern;
determining a distance between a right eye of a second eye diagram corresponding to each of the plurality of sample data signals and a right eye of the reference eye diagram as a pair Ji Yanshi corresponding to each of the plurality of sample data signals;
A delay of the data sampling clock signal relative to each sample data signal is determined based on a left eye position of a second eye pattern corresponding to each of the plurality of sample data signals and a distance between a right eye of the second eye pattern corresponding to each of the plurality of sample data signals and a right eye of the reference eye pattern, and an eye width of the reference eye pattern.
According to the method for training a data eye diagram of a physical layer of a memory, the determining initial delays corresponding to the plurality of sample data signals respectively specifically includes:
And determining the maximum delay of each delay chain of the plurality of sample data signals as the initial delay corresponding to the corresponding sample data signal.
The invention also provides a data eye training device for a physical layer of a memory, which comprises:
A signal acquisition unit for acquiring a data sampling clock signal and a plurality of sample data signals;
the first scanning unit is used for scanning the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
An initial delay unit, configured to determine initial delays corresponding to the plurality of sample data signals based on first eye patterns corresponding to the plurality of sample data signals, and push delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
The second scanning unit is used for scanning the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
And the eye pattern alignment unit is used for right-aligning the second eye patterns corresponding to the sample data signals to obtain pairs Ji Yanshi corresponding to the sample data signals, and determining the delay of the data sampling clock signal relative to the sample data signals based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the sample data signals.
According to the data eye diagram training device for a physical layer of a memory, the right alignment is performed on the second eye diagrams corresponding to the plurality of sample data signals to obtain the alignment delay of the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagrams corresponding to the plurality of sample data signals, which further includes:
Pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye patterns corresponding to the plurality of sample data signals respectively;
Based on the respective third eye patterns of the plurality of sample data signals, a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined.
According to the data eye training device for a physical layer of a memory, the determining the fine alignment delay of the plurality of sample data signals and the fine delay of the data sampling clock signal relative to each sample data signal based on the third eye corresponding to each of the plurality of sample data signals specifically includes:
And right-aligning the third eye patterns corresponding to the sample data signals to obtain fine alignment delay of the sample data signals, and determining the fine delay of the data sampling clock signal relative to the sample data signals based on the middle position of the third eye pattern with the narrowest eye width in the third eye patterns corresponding to the sample data signals.
According to the data eye diagram training device for a physical layer of a memory, the right alignment is performed on the second eye diagrams corresponding to the plurality of sample data signals to obtain the pair Ji Yanshi corresponding to the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagrams corresponding to the plurality of sample data signals, which specifically includes:
Determining a second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals as a reference eye pattern;
determining a distance between a right eye of a second eye diagram corresponding to each of the plurality of sample data signals and a right eye of the reference eye diagram as a pair Ji Yanshi corresponding to each of the plurality of sample data signals;
A delay of the data sampling clock signal relative to each sample data signal is determined based on a left eye position of a second eye pattern corresponding to each of the plurality of sample data signals and a distance between a right eye of the second eye pattern corresponding to each of the plurality of sample data signals and a right eye of the reference eye pattern, and an eye width of the reference eye pattern.
According to the data eye training device for a physical layer of a memory, the determining initial delays corresponding to the plurality of sample data signals specifically includes:
And determining the maximum delay of each delay chain of the plurality of sample data signals as the initial delay corresponding to the corresponding sample data signal.
According to the data eye diagram training method and device for the physical layer of the memory, the first eye diagrams corresponding to the sample data signals are obtained by scanning the sample data signals, initial delay corresponding to the sample data signals is determined based on the first eye diagrams corresponding to the sample data signals, delay chains corresponding to the sample data signals are pushed based on the initial delay corresponding to the sample data signals, then the sample data signals are scanned to obtain the second eye diagrams corresponding to the sample data signals, right alignment is carried out on the second eye diagrams corresponding to the sample data signals to obtain the pair Ji Yan corresponding to the sample data signals, and meanwhile, delay of a data sampling clock signal relative to the sample data signals is determined based on the middle position of the second eye diagram with the narrowest eye width in the second eye diagram corresponding to the sample data signals.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for training a data eye for a physical layer of a memory according to the present invention;
FIG. 2 is a schematic diagram of a data signal provided by the present invention;
FIG. 3 is a schematic flow chart of the right alignment method provided by the invention;
fig. 4 is a schematic diagram of a data eye training device for a physical layer of a memory according to the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Fig. 1 is a flow chart of a data eye training method for a physical layer of a memory, as shown in fig. 1, the method includes:
step 110, acquiring a data sampling clock signal and a plurality of sample data signals;
step 120, scanning the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
Step 130, determining initial delays corresponding to the plurality of sample data signals based on the first eye patterns corresponding to the plurality of sample data signals, and pushing delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
step 140, scanning the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
And step 150, right-aligning the second eye patterns corresponding to the plurality of sample data signals to obtain the pairs Ji Yanshi corresponding to the plurality of sample data signals, and determining the delay of the data sampling clock signal relative to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals.
Specifically, a plurality of sample data signals DQ0, DQ1, DQn on the data sampling clock signal DQS and DQ lines are acquired, and each sample data signal is scanned to obtain a first eye diagram corresponding to each sample data signal. However, as shown in fig. 2 (for example, DQ0, DQ1, DQ2, and DQ 3), the initial phase of each sample data signal has a case where valid data of a part of the sample data signal appears on the left side of the DQs scan along the start point due to the line delay of the PCB, skew between DQs and DQ, and difference between DQs, and the part of the sample data signal is found to have no left eye after the eye pattern of each sample data signal is obtained by pushing the delay of DQs. Accordingly, an initial delay corresponding to each of the plurality of sample data signals may be determined based on the respective first eye diagrams of each of the plurality of sample data signals, and the respective delay chains of the plurality of sample data signals may be driven based on the respective initial delays of the plurality of sample data signals. Wherein the initial delay of any sample data signal is used to obtain a complete eye diagram of the sample data signal based on the data sampling clock signal. That is, by pushing the RX delay of each sample data signal DQi (i=0, 1,..n) to push DQi to the right of the DQS sampling edge, the initial delay of each DQi can be calculated, and then the delay value of each DQi is backfilled based on the initial delay, so that scanning again can result in a complete eye diagram of each sample data signal DQi. In some embodiments, the maximum delay of each delay chain of the plurality of sample data signals may be determined as the corresponding initial delay of the respective sample data signal, i.e. the delay chain of each sample data signal is pushed directly to a maximum value.
In order to ensure that the eye width and eye height of the eye pattern of the effective data of all sample data signals are maximum, scanning the plurality of sample data signals again to obtain second eye patterns corresponding to the plurality of sample data signals respectively, then right-aligning the second eye patterns corresponding to the plurality of sample data signals respectively to obtain pairs Ji Yanshi corresponding to the plurality of sample data signals respectively, and determining the delay of the data sampling clock signal relative to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals respectively, so that the aligned delay corresponding to the plurality of sample data signals respectively and the delay backfill of the data sampling clock signal relative to each sample data signal are realized. The eye patterns of the respective sample data signals may be aligned based on the respective alignment delays of the respective sample data signals, while also improving the eye width and eye height of the eye patterns of the respective sample data signals. Therefore, the whole scheme can realize the eye alignment of each sample data signal only by performing training twice (corresponding to the initial delay in step 130 and the right alignment of the second eye pattern of each sample data signal in step 150 respectively), thereby effectively improving the training efficiency of the data eye pattern.
In some embodiments, as shown in fig. 3, the second eye patterns corresponding to the plurality of sample data signals are right aligned to obtain aligned delays of the plurality of sample data signals, and the delays of the data sampling clock signals relative to the respective sample data signals are determined based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals, which further includes:
Step 310, pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on the alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye diagrams corresponding to the plurality of sample data signals respectively;
Step 320, determining fine alignment delays of the plurality of sample data signals and fine delays of the data sampling clock signal relative to the respective sample data signals based on the respective third eye diagrams of the plurality of sample data signals.
Specifically, right alignment is performed on the second eye patterns corresponding to the plurality of sample data signals, so as to obtain pairs Ji Yanshi of the plurality of sample data signals, and delay time of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals. At this time, the obtained alignment delay of the plurality of sample data signals and the delay of the data sampling clock signal relative to each sample data signal are sufficient to realize the eye alignment of each sample data signal, thereby effectively improving the efficiency of the data eye training. However, considering that the alignment delay of each sample data signal is a theoretical calculation value, a delay chain of each sample data signal may introduce a small amount of error due to temperature or voltage when actually performing delay, so that an eye pattern of each sample data signal may not be precisely aligned after the alignment delay based on each sample data signal and the delay of the corresponding sample data signal with respect to each sample data signal based on the data sampling clock signal.
Therefore, in order to improve the accuracy of the data eye training so that the eye patterns of the respective sample data signals are precisely aligned, the alignment delay of the respective sample data signals and the corresponding sample data signals based on the delay of the data sampling clock signals relative to the respective sample data signals may be backfilled onto the respective delay chains, thereby pushing the delay chains of the plurality of sample data signals and the data sampling clock signals, respectively, based on the alignment delays of the pairs Ji Yan of the plurality of sample data signals and the data sampling clock signals, and acquiring the respective third eye patterns of the plurality of sample data signals. Then, a fine pair Ji Yan of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined based on the respective third eye diagrams of the plurality of sample data signals.
In some embodiments, the third eye patterns corresponding to the plurality of sample data signals may be right aligned to obtain a fine pair Ji Yanshi of the plurality of sample data signals, and the fine delay of the data sampling clock signal with respect to each sample data signal may be determined based on the intermediate position of the third eye pattern with the narrowest eye width among the third eye patterns corresponding to the plurality of sample data signals. It can be seen that the alignment of the third eye pattern corresponding to each of the plurality of sample data signals is similar to the alignment of the second eye pattern corresponding to each of the plurality of sample data signals.
In some embodiments, to achieve alignment of the second eye patterns corresponding to the respective plurality of sample data signals, a second eye pattern having a narrowest eye width among the second eye patterns corresponding to the respective plurality of sample data signals may be determined as a reference eye pattern, and a distance (i.e., dqn_right_eye) between a right eye (dqn_right_eye) of the second eye pattern corresponding to the respective plurality of sample data signals and a right eye (dqn_right_min) of the reference eye pattern may be determined as a pair Ji Yanshi corresponding to the respective plurality of sample data signals. Then, a delay of the data sampling clock signal with respect to each of the sample data signals is determined based on the left eye position of the second eye corresponding to each of the plurality of sample data signals and a distance between the right eye of the second eye corresponding to each of the plurality of sample data signals and the right eye of the reference eye, and an eye width of the reference eye. Wherein the delay of the data sampling clock signal with respect to any of the sample data signals DQi can be determined by:
DQSC_delay=DQi_left_eye－DQi_shift+1/2eye_margin
Wherein DQSC _delay is the delay of the data sampling clock signal relative to any sample data signal DQi, dqi_left_eye is the left eye position of the second eye diagram corresponding to the sample data signal DQi, dqi_shift is the distance between the right eye of the second eye diagram corresponding to the sample data signal DQi and the right eye of the reference eye diagram, and eye_margin is the eye width of the reference eye diagram.
In other embodiments, in order to achieve alignment of the third eye patterns corresponding to the plurality of sample data signals, the third eye pattern having the narrowest eye width among the third eye patterns corresponding to the plurality of sample data signals may be determined as the reference eye pattern, and the distance between the right eye of the third eye pattern corresponding to the plurality of sample data signals and the right eye of the reference eye pattern may be determined as the fine pair Ji Yanshi corresponding to the plurality of sample data signals. Then, a fine delay of the data sampling clock signal with respect to each of the sample data signals is determined based on the left eye position of the third eye corresponding to each of the plurality of sample data signals and a distance between the right eye of the third eye corresponding to each of the plurality of sample data signals and the right eye of the reference eye, and an eye width of the reference eye. The calculation manner of the fine alignment delay of the sample data signal/the fine delay of the data sampling clock signal relative to the sample data signal is similar to the calculation manner of the delay of Ji Yan time/the data sampling clock signal relative to the sample data signal of the sample data signal provided in the above embodiment, and is not repeated herein.
In summary, the method provided by the embodiment of the invention scans the plurality of sample data signals to obtain the first eye patterns corresponding to the plurality of sample data signals, determines the initial delay corresponding to the plurality of sample data signals based on the first eye patterns corresponding to the plurality of sample data signals, pushes the delay chain corresponding to the plurality of sample data signals based on the initial delay corresponding to the plurality of sample data signals, scans the plurality of sample data signals to obtain the second eye patterns corresponding to the plurality of sample data signals, right aligns the second eye patterns corresponding to the plurality of sample data signals to obtain the pair Ji Yan corresponding to the plurality of sample data signals, and determines the delay of the data sampling clock signal relative to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye pattern corresponding to the plurality of sample data signals.
The data eye training device for the physical layer of the memory provided by the invention is described below, and the data eye training device for the physical layer of the memory described below and the data eye training method for the physical layer of the memory described above can be referred to correspondingly.
Based on any of the above embodiments, fig. 4 is a schematic structural diagram of a data eye training device for a physical layer of a memory according to the present invention, as shown in fig. 4, the device includes:
A signal acquisition unit 410 for acquiring a data sampling clock signal and a plurality of sample data signals;
The first scanning unit 420 is configured to scan the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
An initial delay unit 430, configured to determine initial delays corresponding to the plurality of sample data signals based on the first eye diagrams corresponding to the plurality of sample data signals, and push delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
a second scanning unit 440, configured to scan the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
An eye pattern alignment unit 450, configured to right align the second eye patterns corresponding to the plurality of sample data signals, obtain the pair Ji Yanshi corresponding to each of the plurality of sample data signals, and determine the delay of the data sampling clock signal with respect to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to each of the plurality of sample data signals.
According to the device provided by the embodiment of the invention, the first eye patterns corresponding to the sample data signals are obtained by scanning the sample data signals, so that the initial delay corresponding to the sample data signals is determined based on the first eye patterns corresponding to the sample data signals, the delay chains corresponding to the sample data signals are pushed based on the initial delay corresponding to the sample data signals, then the sample data signals are scanned to obtain the second eye patterns corresponding to the sample data signals, the second eye patterns corresponding to the sample data signals are aligned to the right, and when the pair Ji Yan corresponding to the sample data signals is obtained, the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the sample data signals is determined, the delay of the data sampling clock signal relative to the sample data signals is determined.
Based on any of the foregoing embodiments, the right-aligning the second eye patterns corresponding to the plurality of sample data signals to obtain aligned delays of the plurality of sample data signals, and determining delays of the data sampling clock signal with respect to each sample data signal based on an intermediate position of a second eye pattern with a narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals, where the steps further include:
Pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye patterns corresponding to the plurality of sample data signals respectively;
Based on the respective third eye patterns of the plurality of sample data signals, a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined.
Based on any of the foregoing embodiments, the determining, based on the third eye diagrams corresponding to the plurality of sample data signals, the fine alignment delay of the plurality of sample data signals and the fine delay of the data sampling clock signal with respect to each sample data signal specifically includes:
And right-aligning the third eye patterns corresponding to the sample data signals to obtain fine alignment delay of the sample data signals, and determining the fine delay of the data sampling clock signal relative to the sample data signals based on the middle position of the third eye pattern with the narrowest eye width in the third eye patterns corresponding to the sample data signals.
Based on any of the foregoing embodiments, the right-aligning the second eye patterns corresponding to the plurality of sample data signals to obtain the pair Ji Yanshi corresponding to the plurality of sample data signals, and determining the delay of the data sampling clock signal with respect to each sample data signal based on the middle position of the second eye pattern with the narrowest eye width in the second eye pattern corresponding to the plurality of sample data signals specifically includes:
Determining a second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals as a reference eye pattern;
determining a distance between a right eye of a second eye diagram corresponding to each of the plurality of sample data signals and a right eye of the reference eye diagram as a pair Ji Yanshi corresponding to each of the plurality of sample data signals;
A delay of the data sampling clock signal relative to each sample data signal is determined based on a left eye position of a second eye pattern corresponding to each of the plurality of sample data signals and a distance between a right eye of the second eye pattern corresponding to each of the plurality of sample data signals and a right eye of the reference eye pattern, and an eye width of the reference eye pattern.
Based on any one of the foregoing embodiments, the determining the initial delays corresponding to the plurality of sample data signals specifically includes:
And determining the maximum delay of each delay chain of the plurality of sample data signals as the initial delay corresponding to the corresponding sample data signal.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. A method for training a data eye for a physical layer of a memory, comprising:
Acquiring a data sampling clock signal and a plurality of sample data signals;
Scanning the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
determining initial delays corresponding to the plurality of sample data signals based on the first eye diagrams corresponding to the plurality of sample data signals, and pushing delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
scanning the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
Right alignment is carried out on the second eye patterns corresponding to the sample data signals, so that pairs Ji Yanshi corresponding to the sample data signals are obtained, and the delay of the data sampling clock signal relative to the sample data signals is determined based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to the sample data signals;
The right alignment of the second eye patterns corresponding to the plurality of sample data signals is performed to obtain a pair Ji Yanshi corresponding to each of the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to each of the plurality of sample data signals, which specifically includes:
Determining a second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals as a reference eye pattern;
determining a distance between a right eye of a second eye diagram corresponding to each of the plurality of sample data signals and a right eye of the reference eye diagram as a pair Ji Yanshi corresponding to each of the plurality of sample data signals;
Determining a delay of the data sampling clock signal relative to each sample data signal based on a left eye position of a second eye pattern corresponding to each of the plurality of sample data signals and a distance between a right eye of the second eye pattern corresponding to each of the plurality of sample data signals and a right eye of the reference eye pattern, and an eye width of the reference eye pattern;
The determining the initial delay corresponding to each of the plurality of sample data signals specifically includes:
And determining the maximum delay of each delay chain of the plurality of sample data signals as the initial delay corresponding to the corresponding sample data signal.
2. The method for training a data eye diagram of a physical layer of a memory according to claim 1, wherein right-aligning the second eye diagrams corresponding to the plurality of sample data signals to obtain aligned delays of the plurality of sample data signals, and determining delays of the data sampling clock signals relative to the respective sample data signals based on intermediate positions of the second eye diagrams with the narrowest eye width among the second eye diagrams corresponding to the plurality of sample data signals, further comprises:
Pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye patterns corresponding to the plurality of sample data signals respectively;
Based on the respective third eye patterns of the plurality of sample data signals, a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined.
3. The method according to claim 2, wherein determining the fine alignment delay of the plurality of sample data signals and the fine delay of the data sampling clock signal with respect to each sample data signal based on the respective third eye pattern of the plurality of sample data signals comprises:
And right-aligning the third eye patterns corresponding to the sample data signals to obtain fine alignment delay of the sample data signals, and determining the fine delay of the data sampling clock signal relative to the sample data signals based on the middle position of the third eye pattern with the narrowest eye width in the third eye patterns corresponding to the sample data signals.
4. A data eye training apparatus for a physical layer of a memory, comprising:
A signal acquisition unit for acquiring a data sampling clock signal and a plurality of sample data signals;
the first scanning unit is used for scanning the plurality of sample data signals to obtain a first eye pattern corresponding to each of the plurality of sample data signals;
An initial delay unit, configured to determine initial delays corresponding to the plurality of sample data signals based on first eye patterns corresponding to the plurality of sample data signals, and push delay chains corresponding to the plurality of sample data signals based on the initial delays corresponding to the plurality of sample data signals; the initial delay of any sample data signal is used for acquiring a complete eye diagram of the any sample data signal based on the data sampling clock signal;
The second scanning unit is used for scanning the plurality of sample data signals to obtain second eye patterns corresponding to the plurality of sample data signals;
An eye pattern alignment unit, configured to right align the second eye patterns corresponding to the plurality of sample data signals, obtain a pair Ji Yanshi corresponding to each of the plurality of sample data signals, and determine a delay of the data sampling clock signal with respect to each sample data signal based on a middle position of a second eye pattern with a narrowest eye width in the second eye patterns corresponding to each of the plurality of sample data signals;
The right alignment of the second eye patterns corresponding to the plurality of sample data signals is performed to obtain a pair Ji Yanshi corresponding to each of the plurality of sample data signals, and the delay of the data sampling clock signal relative to each sample data signal is determined based on the middle position of the second eye pattern with the narrowest eye width in the second eye patterns corresponding to each of the plurality of sample data signals, which specifically includes:
Determining a second eye pattern with the narrowest eye width in the second eye patterns corresponding to the plurality of sample data signals as a reference eye pattern;
determining a distance between a right eye of a second eye diagram corresponding to each of the plurality of sample data signals and a right eye of the reference eye diagram as a pair Ji Yanshi corresponding to each of the plurality of sample data signals;
Determining a delay of the data sampling clock signal relative to each sample data signal based on a left eye position of a second eye pattern corresponding to each of the plurality of sample data signals and a distance between a right eye of the second eye pattern corresponding to each of the plurality of sample data signals and a right eye of the reference eye pattern, and an eye width of the reference eye pattern;
The determining the initial delay corresponding to each of the plurality of sample data signals specifically includes:
And determining the maximum delay of each delay chain of the plurality of sample data signals as the initial delay corresponding to the corresponding sample data signal.
5. The apparatus for training a data eye for a physical layer of a memory of claim 4, wherein right-aligning the second eye patterns corresponding to the plurality of sample data signals to obtain aligned delays of the plurality of sample data signals, and determining delays of the data sampling clock signals with respect to the respective sample data signals based on intermediate positions of the second eye patterns with the narrowest eye widths among the second eye patterns corresponding to the plurality of sample data signals, further comprises:
Pushing delay chains of the plurality of sample data signals and the data sampling clock signal respectively based on alignment delay of the plurality of sample data signals and the pair Ji Yanshi of the data sampling clock signal, and obtaining third eye patterns corresponding to the plurality of sample data signals respectively;
Based on the respective third eye patterns of the plurality of sample data signals, a fine alignment delay of the plurality of sample data signals and a fine delay of the data sampling clock signal relative to the respective sample data signals are determined.
6. The apparatus of claim 5, wherein the determining the fine alignment delay of the plurality of sample data signals and the fine delay of the data sampling clock signal with respect to each sample data signal based on the respective third eye pattern of the plurality of sample data signals comprises:
And right-aligning the third eye patterns corresponding to the sample data signals to obtain fine alignment delay of the sample data signals, and determining the fine delay of the data sampling clock signal relative to the sample data signals based on the middle position of the third eye pattern with the narrowest eye width in the third eye patterns corresponding to the sample data signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410687052.4A CN118277306B (en) | 2024-05-30 | 2024-05-30 | Data eye training method and device for physical layer of memory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410687052.4A CN118277306B (en) | 2024-05-30 | 2024-05-30 | Data eye training method and device for physical layer of memory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118277306A CN118277306A (en) | 2024-07-02 |
| CN118277306B true CN118277306B (en) | 2024-08-16 |
Family
ID=91633813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410687052.4A Active CN118277306B (en) | 2024-05-30 | 2024-05-30 | Data eye training method and device for physical layer of memory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118277306B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI887090B (en) * | 2024-08-21 | 2025-06-11 | 慧榮科技股份有限公司 | Method and computer program product and apparatus for calibrating signals with flash interface |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108646984A (en) * | 2018-05-16 | 2018-10-12 | 华为技术有限公司 | A DQS position adjustment method and device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9639495B2 (en) * | 2014-06-27 | 2017-05-02 | Advanced Micro Devices, Inc. | Integrated controller for training memory physical layer interface |
| CN108922570B (en) * | 2018-07-13 | 2020-11-13 | 豪威科技(上海)有限公司 | Phase offset detection method, training method, circuit and system for reading DQS signal |
| CN114356206B (en) * | 2020-09-30 | 2024-11-26 | 华为技术有限公司 | Memory training method, memory controller, processor and electronic device |
| CN113178223A (en) * | 2021-04-27 | 2021-07-27 | 珠海全志科技股份有限公司 | Data training method of memory, computer device and computer readable storage medium |
| CN115762590A (en) * | 2022-11-21 | 2023-03-07 | 上海奎芯集成电路设计有限公司 | Eye pattern judgment circuit applied to memory interface and eye pattern judgment method thereof |
-
2024
- 2024-05-30 CN CN202410687052.4A patent/CN118277306B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108646984A (en) * | 2018-05-16 | 2018-10-12 | 华为技术有限公司 | A DQS position adjustment method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118277306A (en) | 2024-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118277306B (en) | Data eye training method and device for physical layer of memory | |
| US6434081B1 (en) | Calibration technique for memory devices | |
| CN1838310A (en) | Memory subsystem and method for generating latch clock | |
| KR100930401B1 (en) | Semiconductor memory device | |
| CN113178223A (en) | Data training method of memory, computer device and computer readable storage medium | |
| US9355696B1 (en) | Calibration in a control device receiving from a source synchronous interface | |
| US6889357B1 (en) | Timing calibration pattern for SLDRAM | |
| US20030191995A1 (en) | System for communicating with synchronous device | |
| US12027197B2 (en) | Signal skew in source-synchronous system | |
| CN118331900A (en) | Method and device for searching left and right boundaries based on CS bus training | |
| US20160293237A1 (en) | Information processing apparatus, information processing system and phase adjusting method | |
| US7889824B2 (en) | System and method for alignment of clock to data | |
| KR20180050816A (en) | Semiconductor device and system including the same | |
| CN114255819A (en) | Method for evaluating performance of interface circuit and related equipment | |
| CN114627918B (en) | Write leveling of memory devices using write DLL circuitry | |
| US20070245096A1 (en) | Finding a data pattern in a memory | |
| US6658604B1 (en) | Method for testing and guaranteeing that skew between two signals meets predetermined criteria | |
| US8370568B2 (en) | Memory interface and adaptive data access method | |
| CN111710354B (en) | CMD (cross-talk) delay compensation method, device, equipment and medium for DDR3 | |
| US7454559B2 (en) | Filtering bit position in a memory | |
| CN118866042B (en) | Method and device for determining data strobe point | |
| CN111916122B (en) | Memory system and method of operating the same | |
| CN118568024B (en) | An ONFI PHY training method suitable for multiple types of delay line structures | |
| CN118038929A (en) | Method and device for training DQS (DQS) gating signal based on preamble training mode | |
| US20250285699A1 (en) | Test device and test method for measuring timing specifications of memory device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |