CN112737713B - Automatic adjusting method and device for PCIe link equalization coefficient - Google Patents
Automatic adjusting method and device for PCIe link equalization coefficient Download PDFInfo
- Publication number
- CN112737713B CN112737713B CN202011643886.3A CN202011643886A CN112737713B CN 112737713 B CN112737713 B CN 112737713B CN 202011643886 A CN202011643886 A CN 202011643886A CN 112737713 B CN112737713 B CN 112737713B
- Authority
- CN
- China
- Prior art keywords
- pcie link
- equalization
- parameters
- optimal
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000010586 diagram Methods 0.000 claims description 29
- 230000003247 decreasing effect Effects 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 3
- 239000002243 precursor Substances 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03878—Line equalisers; line build-out devices
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention provides a PCIe link equalization coefficient automatic adjustment method, which comprises the following steps: performing quick coarse adjustment on the PCIe link to obtain an optimal preset value preset of the PCIe link; determining a micro-adjustment threshold range according to the optimal preset value preset; training the PCIe link by adopting parameters in the fine adjustment threshold range to obtain an equalization result corresponding to each parameter in the fine adjustment threshold range; and selecting a fine adjustment parameter corresponding to the optimal equalization result as the equalization parameter of the PCIe link according to the equalization result. The technical scheme provided by the invention can quickly find out the optimal PCIe link equalization coefficient value, reduce the time spent by complete exhaustion of the coefficients, and avoid the uncertainty caused by random searching, thereby improving the link stability.
Description
Technical Field
The invention relates to the field of CPU processors, in particular to an automatic adjusting method and device for PCIe link equalization coefficients.
Background
The invention relates to the field of CPU processors, in particular to an automatic adjusting method for a PCIe link equalization coefficient. With the rise of PCIe bandwidth and frequency, after the signal passes through a long transmission line and a connector, inter-code crosstalk is more likely to occur, and signal loss is very serious. When PCIe link signals are poor, unexpected phenomena may result that devices cannot be identified, errors occur, speed drops, etc. under the system. Signal equalization adjustment is therefore particularly important. In the existing PCIe equalization method, the optimal equalization parameters are generally found out through full traversal of the equalization parameters, and this process needs to take more time.
Disclosure of Invention
The automatic adjusting method and the device for the PCIe link equalization coefficient can quickly find out the optimal PCIe link equalization coefficient value, reduce the time spent by completely exhausting the coefficients, avoid uncertainty caused by random searching, and improve the link stability.
In a first aspect, the present invention provides a method for automatically adjusting a PCIe link equalization coefficient, including:
performing quick coarse adjustment on the PCIe link to obtain an optimal preset value preset of the PCIe link;
determining a micro-adjustment threshold range according to the optimal preset value preset;
training the PCIe link by adopting parameters in the fine adjustment threshold range to obtain an equalization result corresponding to each parameter in the fine adjustment threshold range;
and selecting a fine adjustment parameter corresponding to the optimal equalization result as the equalization parameter of the PCIe link according to the equalization result.
Optionally, determining the fine adjustment threshold range according to the optimal preset value preset includes:
acquiring three-dimensional parameters corresponding to the optimal preset value preset, wherein the three-dimensional parameters comprise a pre-cursor, a post-cursor and a main cursor;
and taking the three-dimensional parameter corresponding to the optimal preset value preset as a center, and determining a fine adjustment threshold range containing a plurality of three-dimensional parameters.
Optionally, training the PCIe link using parameters within the fine-tuning threshold range includes:
training the PCIe link with parameters within the fine-tuning threshold range;
after training is completed, an eye diagram tool is used to obtain an eye diagram corresponding to each parameter.
Optionally, selecting, according to the equalization result, a fine adjustment parameter corresponding to the optimal equalization result as an equalization parameter of the PCIe link includes:
comparing the eye patterns corresponding to each parameter to obtain an eye pattern with optimal data;
and taking the parameters corresponding to the eye pattern with optimal data as equalization parameters of the PCIe link.
Optionally, training the PCIe link with parameters within the fine-tuning threshold range to obtain an eye diagram corresponding to each parameter includes:
sequentially arranging the parameters, and recording the number of parameters to be trained;
training each parameter in turn to obtain an eye diagram corresponding to each parameter and storing the eye diagram;
the number of parameters to be trained is decreased along with the training of the parameters;
and when the number of parameters to be trained is reduced to zero, ending the training of the parameters.
Optionally, selecting, according to the equalization result, a fine adjustment parameter corresponding to the optimal equalization result as an equalization parameter of the PCIe link includes: and selecting parameters corresponding to the optimal equalization result as equalization parameters of the equipment transmitting end.
Optionally, before the quickly coarsening the PCIe link, further includes:
performing signal quality inspection on the PCIe link;
when the signal quality of the PCIe link meets the requirements, maintaining the current equalization parameters of the PCIe link;
and when the signal quality of the PCIe link is not satisfactory, performing quick coarse adjustment on the PCIe link.
Optionally, performing fast coarse adjustment on the PCIe link includes:
determining a plurality of preset values according to PCIe specifications;
performing balanced training on the PCIe link by adopting a plurality of preset values to obtain training results of the PCIe link by the plurality of preset values;
and obtaining an optimal preset value preset according to the training result.
In a second aspect, the present invention provides an automatic PCIe link equalization coefficient adjustment device, including:
the fast adjusting module is used for fast and coarsely adjusting the PCIe link to obtain an optimal preset value preset of the PCIe link;
the threshold range determining module is used for determining a micro-adjustment threshold range according to the optimal preset value preset;
the equalization training module is used for training the PCIe link by adopting parameters in the fine adjustment threshold range so as to obtain equalization results corresponding to the parameters in the fine adjustment threshold range;
and the parameter determining module is used for selecting a fine adjustment parameter corresponding to the optimal equalization result as the equalization parameter of the PCIe link according to the equalization result.
Optionally, the threshold range determining module is configured to: acquiring three-dimensional parameters corresponding to the optimal preset value preset, wherein the three-dimensional parameters comprise a pre-cursor, a post-cursor and a main cursor; and determining a fine adjustment threshold range containing a plurality of three-dimensional parameters by taking the three-dimensional parameter corresponding to the optimal preset value preset as a center.
In the technical scheme provided by the invention, the quick coarse adjustment is quick, but the adjustment precision is low, so that the selection of the optimal preset value preset is firstly carried out by adopting the quick coarse adjustment, the fine adjustment threshold range of the fine adjustment is determined according to the preset value, the precision of the fine adjustment parameter determined by adopting the optimal preset value preset is high, the range is small, the optimal PCIe link equalization coefficient can be quickly found out, the time spent by the complete exhaustion of the coefficient is reduced, the uncertainty caused by random search is avoided, and the link stability is improved. In the technical scheme provided by the invention, the fast coarse adjustment has a faster speed, the optimal preset value can be found out quickly, and the fine adjustment is the equalization parameter adjustment with higher precision further carried out on the basis of the fast coarse adjustment, so that the invention can give consideration to the adjustment speed and the adjustment precision, and can quickly obtain the equalization coefficient which enables the PCIe link to have high stability.
Drawings
FIG. 1 is a flowchart of a PCIe link equalization coefficient automatic adjustment method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a fine-tuning threshold range determination for an automatic adjustment method for PCIe link equalization coefficients according to another embodiment of the present invention;
FIG. 3 is a flowchart of determining optimal equalization parameters for an automatic adjustment method for equalization coefficients of a PCIe link according to another embodiment of the present invention;
FIG. 4 is a flowchart of parameter training for an automatic PCIe link equalization coefficient adjustment method according to another embodiment of the present invention;
FIG. 5 is a signal quality check flowchart of an automatic PCIe link equalization coefficient adjustment method according to another embodiment of the present invention;
FIG. 6 is a flowchart of a coarse-tuning throttle for PCIe link equalization coefficient automatic adjustment in accordance with another embodiment of the present invention;
fig. 7 is a schematic diagram of an apparatus for automatically adjusting PCIe link equalization coefficients according to another embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only 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.
The embodiment of the invention provides a PCIe link equalization coefficient automatic adjustment method, as shown in fig. 1, comprising the following steps:
step 100, performing fast coarse adjustment on the PCIe link to obtain an optimal Preset value Preset of the PCIe link;
in some embodiments, the fast rough adjustment of the PCIe link refers to fast rough adjustment of the PCIe link according to a preset value given in the PCIe specification, and the preset value with the best balancing result is selected as the rough adjustment optimal preset value preset through the preset value given in the traversing specification.
Step 200, determining a micro-adjustment threshold range according to the optimal preset value preset;
in some embodiments, the fine tuning threshold range refers to a plurality of sets of parameters selected centering around the optimal preset value preset, and in some preferred embodiments, for example, a plurality of sets of parameters between two sets of parameters corresponding to two preset values adjacent to the optimal preset value preset may be used as the fine tuning threshold range.
Step 300, training the PCIe link by adopting parameters within the micro-adjustment threshold range to obtain an equalization result corresponding to each parameter within the micro-adjustment threshold range;
in some embodiments, multiple groups of parameters within the fine adjustment threshold range are trained, and in the training process, the equalization result is saved, so that the subsequent judgment of the quality of the equalization result of each parameter is facilitated.
Step 400, selecting the fine adjustment parameters corresponding to the optimal equalization result as the equalization parameters of the PCIe link according to the equalization result.
In some embodiments, the equalization results of the parameters in the micro-adjustment threshold range are compared, and the parameter with the optimal equalization result is used as the equalization parameter of the PCIe link, so that the current equalization result is the optimal equalization result of the parameters in the micro-adjustment threshold range, and meanwhile, because the optimal preset value preset corresponding to the micro-adjustment threshold range is the best equalization result of all preset values, the current equalization result of the PCIe link can be determined to be the optimal equalization result of all the parameters by performing micro-adjustment around the optimal preset value preset.
In the technical scheme provided by the embodiment, the fast coarse adjustment is fast, but the adjustment precision is low, so that the selection of the optimal preset value preset is firstly performed by adopting the fast coarse adjustment, the threshold range of the fine adjustment is determined according to the preset value, the parameter precision of the fine adjustment parameter determined by adopting the optimal preset value preset is high, the range is small, and therefore, even if each parameter in the fine adjustment threshold range is traversed later, less time is required.
Based on the embodiment shown in fig. 1, as shown in fig. 2, step 200 includes:
step 210, obtaining three-dimensional parameters with pre-cursor, post-cursor and main cursor parameters corresponding to the optimal preset value preset;
in some embodiments, because the parameters actually required in the equalization process are three parameters of the pre-cursor, the post-cursor and the main cursor, after the optimal preset value preset is selected, the optimal preset value preset is mapped into three-dimensional parameters including the pre-cursor, the post-cursor and the main cursor, so as to facilitate the selection of a subsequent fine adjustment threshold range.
Step 220, centering on the three-dimensional parameter corresponding to the optimal preset value preset, determining a fine adjustment threshold range including a plurality of three-dimensional parameters.
In some embodiments, the process of determining the fine adjustment threshold range is to select a plurality of parameters distributed around the three-dimensional parameter corresponding to the optimal preset value preset with the three-dimensional parameter corresponding to the optimal preset value preset as a center, and the selected set of parameters is the fine adjustment threshold range.
In this embodiment, the optimal preset value preset is mapped into the three-dimensional parameter, and then the fine adjustment threshold range is determined, so that the method is more visual and simpler.
As an alternative embodiment, step 300 includes: and training the PCIe link by using the parameters in the fine adjustment threshold range, and obtaining an eye diagram corresponding to each parameter by using an eye diagram tool after training is completed. Since the eye diagram is a test manner capable of directly representing the equalization result, in this embodiment, the equalization result of each parameter can be determined by performing link training on the parameter in the fine adjustment threshold range and forming an eye diagram using an eye diagram tool.
Based on the embodiment shown in fig. 1, as shown in fig. 3, step 400 includes:
step 410, comparing the eye patterns corresponding to each parameter to obtain the eye pattern with optimal data;
in some embodiments, eye height and eye width data should be considered in the comparison of eye patterns, e.g., training results for each parameter are determined based on eye height and eye width in the specification.
And step 420, taking the parameters corresponding to the eye pattern with optimal data as equalization parameters of the PCIe link.
In some embodiments, since the eye diagram characterizes the equalization result, the data-optimized eye diagram, i.e., the parameter corresponding to the best equalization result, can be selected by the eye diagram.
In this embodiment, the parameters in the fine adjustment threshold range are selected through the eye diagram, and the optimal equalization parameters are intuitively and conveniently selected.
Based on the embodiment shown in fig. 1, as shown in fig. 4, step 300 includes:
step 310, arranging the parameters in sequence, and recording the number of parameters to be trained;
in some embodiments, the parameters are sequentially arranged, i.e., the parameters are given a sequence, for example, the parameters may be ordered according to a sequence stored by default in the process of searching the parameters, or the three-dimensional parameters may be sequentially ordered according to the size of the primary label cursor, or the pre-or post-label cursor may be ordered according to the size of the pre-or post-label cursor, or other manners. Ordering the parameters can ensure that training is performed in an inherent order in the subsequent parameter training process.
Step 320, training each parameter in turn to obtain an eye diagram corresponding to each parameter and storing the eye diagram;
in some embodiments, since the eye diagram is an intuitive representation of the equalization result, the eye diagram is generated during training of each parameter to facilitate subsequent evaluation of the equalization result.
Step 330, the number of parameters to be trained is decreased with the training of the parameters;
in some embodiments, the number of parameters to be trained is reduced by one for each complete training of a parameter in order to provide an end condition for the parameter training process, avoiding repeated training.
Step 340, ending the training of the parameters when the number of parameters to be trained is decremented to zero.
In some embodiments, when the number of parameters to be trained is decremented to zero, which indicates that training of all parameters is currently completed, the training of the parameters may be ended.
In this embodiment, by training the training parameters sequentially, repeated training of the same parameter is avoided, and the conditions for stopping training can be accurately obtained by decreasing the number of parameters to be trained.
As a preferred embodiment, step 400 includes selecting, as the equalization parameter of the device transmitting end, a parameter corresponding to the optimal equalization result. The signal transmitting end is a source of signal transmission, and is beneficial to improving the quality of the transmitted signal.
On the basis of the embodiment shown in fig. 1, as shown in fig. 5, step 100 further includes:
step 010, performing signal quality inspection on the PCIe link;
in some embodiments, the PCIe link needs to be checked for signal quality before being adjusted, and when the signal quality is good, the PCIe link may not be adjusted, so that computing resources and time can be saved, and when the signal quality is bad, the PCIe link does not meet the user's requirements, the PCIe link needs to be adjusted, so as to improve the signal quality. For example, whether the signal quality of the current PCIe link meets the requirement is determined according to the signal quality requirement in the specification, or the signal quality check may be performed with a higher requirement, for example, with a standard higher than 10% of the signal quality requirement in the specification.
Step 011, when the signal quality of the PCIe link meets the requirement, maintaining the current equalization parameters of the PCIe link;
in some embodiments, the quality of the signal is satisfactory, indicating that the quality of the signal is good, and that it can be used without adjustment. The PCIe link is not regulated, so that not only can the calculation resources required during regulation be saved, but also the time required for regulation can be saved.
And step 012, executing quick coarse adjustment to the PCIe link when the signal quality of the PCIe link is not satisfactory.
When the signal quality is not satisfactory, the PCIe link is adjusted to improve the signal quality in order to ensure the signal quality in the subsequent use process.
In the embodiment, the purpose of considering both the calculation efficiency and the signal quality is achieved by increasing the process of signal quality detection, when the signal quality meets the requirements, the calculation resources and the calculation time are saved as much as possible, and when the signal quality does not meet the requirements, the signal quality is improved as much as possible.
Based on the embodiment shown in fig. 1, as shown in fig. 6, step 100 includes:
step 110, determining a plurality of preset values according to PCIe specifications;
in some embodiments, 11 preset values are specified in the PCIe specification for the user to adjust the PCIe link. In this embodiment, the plurality of preset values means that 11 preset values in the PCIe specification can be directly selected.
Step 120, performing equalization training on the PCIe link by using a plurality of preset values preset to obtain training results of the PCIe link by using the plurality of preset values;
in some embodiments, the training result is obtained by performing the equalization training on the plurality of preset values, for example, since the eye diagram may represent the result of the equalization training, the result of the eye diagram may be used as the result of the equalization training.
Step 130, obtaining an optimal preset value preset according to the training result.
In some embodiments, since the eye diagram represents the equalization result, the preset value corresponding to the optimal eye diagram may be used as the optimal preset value preset in the eye diagram obtained by performing equalization training on a plurality of preset values.
In this embodiment, the optimal preset value is obtained by traversing a plurality of preset values and comparing eye patterns, so that the process is intuitive and simple, and the operation is facilitated.
The embodiment of the invention also provides a PCIe link equalization device, as shown in fig. 7, including:
a fast adjustment module 1010, configured to perform fast coarse adjustment on the PCIe link to obtain an optimal preset value preset of the PCIe link;
in some embodiments, the fast rough adjustment of the PCIe link refers to fast rough adjustment of the PCIe link according to a preset value given in the PCIe specification, and the preset value with the best equalization result is selected as the optimal preset value preset after traversing the preset value given in the PCIe specification.
A threshold range determining module 1020, configured to determine a fine adjustment threshold range according to the optimal preset value preset;
in some embodiments, the fine tuning threshold range refers to a plurality of sets of parameters selected centering around the optimal preset value preset, and in some preferred embodiments, for example, a plurality of sets of parameters between two sets of parameters corresponding to two preset values adjacent to the optimal preset value preset may be used as the fine tuning threshold range.
The equalization training module 1030 is configured to train the PCIe link using parameters within the fine adjustment threshold range, so as to obtain an equalization result corresponding to each parameter within the fine adjustment threshold range;
in some embodiments, multiple groups of parameters within the fine adjustment threshold range are trained, and in the training process, the equalization result is saved, so that the subsequent judgment of the quality of the equalization result of each parameter is facilitated.
And the parameter determining module 1040 is configured to select, according to the equalization result, a fine adjustment parameter corresponding to the optimal equalization result as an equalization parameter of the PCIe link.
In some embodiments, the equalization results of the parameters in the micro-adjustment threshold range are compared, and the parameter with the optimal equalization result is used as the equalization parameter of the PCIe link, so that the current equalization result is the optimal equalization result of the parameters in the micro-adjustment threshold range, and meanwhile, because the optimal preset value preset corresponding to the micro-adjustment threshold range is the best equalization result of all preset values, the current equalization result of the PCIe link can be determined to be the optimal equalization result of all the parameters by performing micro-adjustment around the optimal preset value preset.
In the technical scheme provided by the embodiment, the fast coarse adjustment is fast, but the adjustment precision is low, so that the selection of the optimal preset value preset is firstly performed by adopting the fast coarse adjustment, the fine adjustment threshold range of the fine adjustment is determined according to the preset value, and the parameter precision of the fine adjustment parameter determined by adopting the optimal preset value preset is high, the range is small, so that less time is required to be spent even if each parameter in the fine adjustment threshold range is traversed in the follow-up.
On the basis of the embodiment shown in fig. 7, the threshold range determining module 1020 is configured to: acquiring three-dimensional parameters corresponding to the optimal preset value preset, wherein the three-dimensional parameters comprise a pre-cursor, a post-cursor and a main cursor; and determining a fine adjustment threshold range containing a plurality of three-dimensional parameters by taking the three-dimensional parameter corresponding to the optimal preset value preset as a center. In some embodiments, because the parameters actually required in the equalization process are three parameters of the pre-cursor, the post-cursor and the main cursor, after the optimal preset value preset is selected, the optimal preset value preset is mapped into three-dimensional parameters including the pre-cursor, the post-cursor and the main cursor, so that the subsequent selection of the fine adjustment threshold range is facilitated. In some embodiments, the process of determining the fine adjustment threshold range is to select a plurality of parameters distributed around the three-dimensional parameter corresponding to the optimal preset value preset with the three-dimensional parameter corresponding to the optimal preset value preset as a center, and the selected set of parameters is the fine adjustment threshold range.
In this embodiment, the optimal preset value preset is mapped into the three-dimensional parameter, and then the fine adjustment threshold range is determined, so that the method is more visual and simpler.
Those skilled in the art will appreciate that implementing all or part of the above-described method embodiments may be accomplished by way of computer programs, which may be stored on a computer readable storage medium, which when executed may comprise the steps of the method embodiments described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like.
The foregoing is merely illustrative of the present invention, and the present invention is not limited to the above embodiments, and any changes or substitutions easily contemplated by those skilled in the art within the technical scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (7)
1. The automatic PCIe link equalization coefficient adjusting method is characterized by comprising the following steps of:
performing quick coarse adjustment on the PCIe link to obtain an optimal preset value preset of the PCIe link;
taking the parameters between two groups of parameters corresponding to two adjacent preset values of the optimal preset value preset as a micro-adjustment threshold range;
training the PCIe link in a way of traversing parameters in the fine adjustment threshold range to obtain an equalization result corresponding to each parameter in the fine adjustment threshold range;
selecting parameters corresponding to the optimal equalization result as equalization parameters of the PCIe link according to the equalization result;
the fast coarsening of the PCIe link includes: performing balanced training on a PCIe link by traversing a plurality of preset values given in PCIe specifications to obtain training results of the PCIe link by the plurality of preset values; and obtaining an optimal preset value preset according to the training result.
2. The PCIe link equalization coefficient automatic adjustment method according to claim 1, wherein training the PCIe link using parameters within the fine adjustment threshold range comprises:
training the PCIe link with parameters within the fine-tuning threshold range;
after training is completed, an eye diagram tool is used to obtain an eye diagram corresponding to each parameter.
3. The PCIe link equalization coefficient automatic adjustment method according to claim 2, wherein selecting, as the equalization parameter of the PCIe link, a parameter corresponding to an optimal equalization result according to the equalization result comprises:
comparing the eye patterns corresponding to each parameter to obtain an eye pattern with optimal data;
and taking the parameters corresponding to the eye pattern with optimal data as equalization parameters of the PCIe link.
4. The PCIe link equalization coefficient automatic adjustment method according to claim 1, wherein training the PCIe link with parameters within the fine adjustment threshold range to obtain an eye diagram corresponding to each parameter comprises:
sequentially arranging the parameters, and recording the number of parameters to be trained;
training each parameter in turn to obtain an eye diagram corresponding to each parameter and storing the eye diagram;
the number of parameters to be trained is decreased along with the training of the parameters;
and when the number of parameters to be trained is reduced to zero, ending the training of the parameters.
5. The PCIe link equalization coefficient automatic adjustment method according to claim 1, wherein selecting, as the equalization parameter of the PCIe link, a parameter corresponding to an optimal equalization result according to the equalization result comprises: and selecting parameters corresponding to the optimal equalization result as equalization parameters of the equipment transmitting end.
6. The PCIe link equalization coefficient automatic adjustment method according to claim 1, further comprising, before performing the fast coarse adjustment on the PCIe link:
performing signal quality inspection on the PCIe link;
when the signal quality of the PCIe link meets the requirements, maintaining the current equalization parameters of the PCIe link;
and when the signal quality of the PCIe link is not satisfactory, performing quick coarse adjustment on the PCIe link.
7. Automatic adjusting device of balanced coefficient of PCIe link, characterized by including:
the fast adjusting module is used for fast and coarsely adjusting the PCIe link to obtain an optimal preset value preset of the PCIe link;
the threshold range determining module takes parameters between two groups of parameters corresponding to two preset values adjacent to the optimal preset value preset as micro-adjustment threshold ranges;
the equalization training module is used for training the PCIe link in a way of traversing parameters in the fine adjustment threshold range so as to obtain equalization results corresponding to the parameters in the fine adjustment threshold range;
the parameter determining module is used for selecting the parameter corresponding to the optimal equalization result as the equalization parameter of the PCIe link according to the equalization result;
the fast adjustment module is further configured to perform balanced training on a PCIe link by using a manner of traversing a plurality of preset values given in a PCIe specification, so as to obtain a training result of the PCIe link by the plurality of preset values; and obtaining an optimal preset value preset according to the training result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011643886.3A CN112737713B (en) | 2020-12-30 | 2020-12-30 | Automatic adjusting method and device for PCIe link equalization coefficient |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011643886.3A CN112737713B (en) | 2020-12-30 | 2020-12-30 | Automatic adjusting method and device for PCIe link equalization coefficient |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112737713A CN112737713A (en) | 2021-04-30 |
| CN112737713B true CN112737713B (en) | 2023-12-19 |
Family
ID=75609236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011643886.3A Active CN112737713B (en) | 2020-12-30 | 2020-12-30 | Automatic adjusting method and device for PCIe link equalization coefficient |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112737713B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113722265A (en) * | 2021-08-19 | 2021-11-30 | 飞腾信息技术有限公司 | Debugging optimization method and device for interconnection channel in multi-CPU system |
| CN114416636A (en) * | 2021-12-17 | 2022-04-29 | 飞腾信息技术有限公司 | PCIE equipment link rate matching method, system on chip and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885907A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Equalization Coefficient Search Algorithm |
| CN103885911A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Multipass approach for performing channel equalization training |
| CN106817258A (en) * | 2017-01-13 | 2017-06-09 | 郑州云海信息技术有限公司 | A kind of method and device chosen and verify PCIE link equalization parameters |
| CN108418582A (en) * | 2018-02-11 | 2018-08-17 | 华为技术有限公司 | Transmit method, driver and the system of signal |
| CN110069370A (en) * | 2019-04-11 | 2019-07-30 | 苏州浪潮智能科技有限公司 | A kind of PCH Uplink parameter optimization method and system |
-
2020
- 2020-12-30 CN CN202011643886.3A patent/CN112737713B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885907A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Equalization Coefficient Search Algorithm |
| CN103885911A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Multipass approach for performing channel equalization training |
| CN106817258A (en) * | 2017-01-13 | 2017-06-09 | 郑州云海信息技术有限公司 | A kind of method and device chosen and verify PCIE link equalization parameters |
| CN108418582A (en) * | 2018-02-11 | 2018-08-17 | 华为技术有限公司 | Transmit method, driver and the system of signal |
| CN110069370A (en) * | 2019-04-11 | 2019-07-30 | 苏州浪潮智能科技有限公司 | A kind of PCH Uplink parameter optimization method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112737713A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112737713B (en) | Automatic adjusting method and device for PCIe link equalization coefficient | |
| CN110417463B (en) | Consistency test method, consistency test device and storage medium | |
| CN103744974B (en) | Method and device for selecting local interest points | |
| CN106888060B (en) | Anti-interference test method and system for wireless communication module | |
| CN106658701A (en) | Positioning method and apparatus | |
| US20230034994A1 (en) | Channel Identification Method and Apparatus, Transmission Method, Transmission Device, Base Station, and Medium | |
| CN110702986B (en) | Method and system for dynamically generating self-adaptive signal search threshold in real time | |
| US8331428B2 (en) | Signal transmission apparatus and signal transmission control method | |
| CN112566170B (en) | Network quality evaluation method, device, server and storage medium | |
| CN107799831B (en) | Method and device for determining cell datum line | |
| CN109560978B (en) | Network flow detection method, device and system and computer readable storage medium | |
| CN109308360A (en) | Channel parameter finding method, device, storage medium and computer equipment | |
| CN113204642A (en) | Text clustering method and device, storage medium and electronic equipment | |
| US20220272711A1 (en) | Method and apparatus for carrier aggregation optimization | |
| CN116403631A (en) | Memory parameter calibration method and memory parameter calibration device | |
| CN107038330B (en) | Data missing compensation method and device | |
| CN114389751B (en) | Automatic equalization method and system for error code test equipment | |
| CN112860495A (en) | Debugging method of I2C slave device, I2C master device and storage medium | |
| CN115407915A (en) | Eye pattern tool implementation method, system, electronic equipment and storage medium | |
| CN112069767B (en) | Transmission line wiring length estimation method, device, equipment and medium | |
| CN113095373B (en) | Ship detection method and system based on self-adaptive position prediction and capable of detecting any rotation angle | |
| CN115311526A (en) | Defect sample generation method and system based on improved Cycle GAN network | |
| CN110990561B (en) | Automatic classification realization method for defect text of power equipment | |
| CN105099963B (en) | A kind of device and method of offset estimation | |
| CN107659466B (en) | System and method for quantifying and calculating throughput capacity service level of cloud system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |