CN-122019335-A - RISC-V basic software ecological maturity scoring method and device

CN122019335ACN 122019335 ACN122019335 ACN 122019335ACN-122019335-A

Abstract

The embodiment of the invention relates to the technical field of basic software ecological maturity measurement, and provides an ecological maturity scoring method and device of RISC-V basic software, wherein the method comprises the steps of respectively installing and operating RISC-V basic software to be tested on a first target hardware platform and a second target hardware platform; and calculating the maturity score of the RISC-V basic software to be tested based on the maturity score of each dimension, wherein the maturity score of the RISC-V basic software to be tested is calculated based on the running data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform. Therefore, the method realizes the quantitative measurement of the ecological development level of RISC-V basic software, and has comprehensiveness, expandability and operability.

Inventors

JI CHENCHEN
YU JIAGENG
WU YANJUN
MA XIAOHAN

Assignees

中国科学院软件研究所

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (10)

1. A method for scoring the ecological maturity of RISC-V base software, comprising: respectively installing and running RISC-V basic software to be tested on a first target hardware platform and a second target hardware platform; Calculating maturity scores of multiple dimensions based on operation data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform; and calculating the ecological maturity score of the RISC-V basic software to be tested based on the maturity score of each dimension.
2. The method of claim 1, wherein the plurality of dimensions includes performance advance, compilation quality, and memory advance; the calculating the maturity score of multiple dimensions based on the running data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform comprises the following steps: acquiring the speed ratio of time of a plurality of test cases on task execution walls of the first target hardware platform and the second target hardware platform, and calculating to obtain a first geometric average value; performing weighted calculation based on the first geometric mean value and a first standardized weight coefficient to obtain a performance advanced degree score; acquiring the acceleration ratio of the CPU instruction numbers of the plurality of test cases on the first target hardware platform and the second target hardware platform, and calculating to obtain a second geometric average value; Weighting calculation is carried out on the basis of the second geometric mean value and a second standardized weight coefficient, so that a compiling quality score is obtained; The acceleration ratio of the cache read missing quantity of the plurality of test cases on the first target hardware platform and the second target hardware platform is obtained, and a third geometric average value is obtained through calculation; and carrying out weighted calculation based on the third geometric mean value and a third standardized weight coefficient to obtain the memory advanced score.
3. The method of claim 1, wherein the plurality of dimensions further comprises functional integrity, open source development liveness, and heterogeneous computing support; The calculating the maturity score of multiple dimensions based on the running data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform further comprises: Acquiring the execution passing rate of the same test case on the first target hardware platform and the second target hardware platform respectively; Calculating the ratio of the execution passing rate, and carrying out weighted calculation by combining a fourth standardized weight coefficient to obtain a functional integrity score; Obtaining the patch quantity of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform respectively; calculating the patch quantity ratio, and carrying out weighted calculation by combining a fifth standardized weight coefficient to obtain an open source development liveness score; Acquiring heterogeneous computing support quantity of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform respectively; and calculating the heterogeneous calculation support quantity ratio, and carrying out weighted calculation by combining with a sixth standardized weight coefficient to obtain a heterogeneous calculation support score.
4. The method according to claim 2, wherein the method further comprises: Calculating a performance first progress score by a first formula: ; Wherein, the For the first normalized weight coefficient, Is a first geometric mean; Calculating a compilation quality score by a second formula, the second formula being: ; Wherein, the For the second normalized weight coefficient, Is a second geometric mean; Calculating a memory advance score by a third formula, the third formula being: ; Wherein, the For the third normalized weight coefficient, Is a third geometric mean.
5. A method according to claim 3, characterized in that the method further comprises: Calculating a functional integrity score by a fourth formula: ; Wherein, the For the fourth normalized weight coefficient, Representing the execution pass rate of the same test case on the second target hardware platform, Representing the execution passing rate of the same test case on the first target hardware platform; calculating an open source development liveness score through a fifth formula, wherein the fifth formula is as follows: ; Wherein, the For the fifth normalized weight coefficient, Representing the number of patches of the RISC-V foundation software to be tested on the second target hardware platform, Representing the patch number of the RISC-V basic software to be tested on the first target hardware platform; Calculating a heterogeneous computation support score by a sixth formula: ; Wherein, the For the sixth normalized weight coefficient, Representing the heterogeneous computing support quantity of the RISC-V foundation software to be tested on the second target hardware platform, Representing the heterogeneous computing support quantity of the RISC-V base software to be tested on the first target hardware platform.
6. The method of any one of claims 1-5, wherein calculating an ecological maturity score for the RISC-V base software to be tested based on the maturity score for each dimension comprises: Normalizing the maturity score of each dimension and drawing a spider web map; and determining the total ecological maturity score of the RISC-V basic software to be tested based on the spider web graph, wherein the total ecological maturity score is used for transversely comparing multiple basic software or longitudinally tracking the ecological progress of the same software in different version periods.
7. An ecological maturity scoring device for RISC-V base software, comprising: The installation module is used for installing and running RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform respectively; The computing module is used for computing maturity scores of multiple dimensions based on the running data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform; And the scoring module is used for calculating and obtaining the ecological maturity score of the RISC-V basic software to be tested based on the maturity score of each dimension.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the ecological maturity scoring method of the RISC-V base software of any one of claims 1 to 6 when the computer program is executed.
9. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the ecological maturity scoring method of RISC-V base software of any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements a method of ecological maturity scoring of RISC-V base software as claimed in any one of claims 1 to 6.

Description

RISC-V basic software ecological maturity scoring method and device Technical Field The invention relates to the technical field of basic software ecological maturity measurement, in particular to a method and a device for scoring the ecological maturity of RISC-V basic software. Background RISC-V is an open Instruction Set Architecture (ISA) that is being gradually applied to various fields of high-performance computing, data centers, embedded systems, etc., due to its openness, scalability, and low cost characteristics. However, the software ecology of RISC-V is still in a development stage, and compared with mature architectures such as ARM, obvious gaps exist in the aspects of performance optimization, compiling adaptation, functional support, open source liveness, heterogeneous computing support and the like. Most of the existing RISC-V software ecological assessment methods are concentrated on a single performance index, lack a multi-dimensional, quantifiable and horizontally-longitudinally-comparable assessment system, and are difficult to comprehensively reflect the maturity of an ecological system. In addition, the lack of systematic comparison with the mainstream maturation architecture (e.g., ARM) also limits the objectivity and operability of the evaluation results. Therefore, there is an urgent need for a method for evaluating the ecological maturity of RISC-V based software based on multidimensional index, quantifiable calculation, and comparable to a reference architecture, to comprehensively, objectively and extendably scale the ecological development level of RISC-V software. Disclosure of Invention The invention provides an ecological maturity scoring method and device for RISC-V base software, which are used for solving the defect that in the prior art, RISC-V software is only evaluated on one side qualitatively, and a quantized, comparable and multidimensional maturity scale is lacked, realizing the quantized measurement of the ecological development level of RISC-V base software, and having comprehensiveness, expandability and operability. The invention provides an ecological maturity scoring method of RISC-V basic software, which comprises the following steps: respectively installing and running RISC-V basic software to be tested on a first target hardware platform and a second target hardware platform; Calculating maturity scores of multiple dimensions based on operation data of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform; and calculating the ecological maturity score of the RISC-V basic software to be tested based on the maturity score of each dimension. In one possible embodiment, the method further comprises: The plurality of dimensions comprise performance first progress, compiling quality and memory advance; acquiring the speed ratio of time of a plurality of test cases on task execution walls of the first target hardware platform and the second target hardware platform, and calculating to obtain a first geometric average value; performing weighted calculation based on the first geometric mean value and a first standardized weight coefficient to obtain a performance advanced degree score; acquiring the acceleration ratio of the CPU instruction numbers of the plurality of test cases on the first target hardware platform and the second target hardware platform, and calculating to obtain a second geometric average value; Weighting calculation is carried out on the basis of the second geometric mean value and a second standardized weight coefficient, so that a compiling quality score is obtained; The acceleration ratio of the cache read missing quantity of the plurality of test cases on the first target hardware platform and the second target hardware platform is obtained, and a third geometric average value is obtained through calculation; and carrying out weighted calculation based on the third geometric mean value and a third standardized weight coefficient to obtain the memory advanced score. In one possible embodiment, the method further comprises: The multiple dimensions further comprise functional integrity, open source development liveness and heterogeneous computing support; Acquiring the execution passing rate of the same test case on the first target hardware platform and the second target hardware platform respectively; Calculating the ratio of the execution passing rate, and carrying out weighted calculation by combining a fourth standardized weight coefficient to obtain a functional integrity score; Obtaining the patch quantity of the RISC-V basic software to be tested on the first target hardware platform and the second target hardware platform respectively; calculating the patch quantity ratio, and carrying out weighted calculation by combining a fifth standardized weight coefficient to obtain an open source development liveness score; Acquiring heterogeneous computing support quantity of the RIS