osaca 0.3.1.dev1

OSACA

Open Source Architecture Code Analyzer

This tool allows automatic instruction fetching of assembly code, auto-generating of testcases for assembly instructions creating latency and throughput benchmarks on a specific instruction form and throughput analysis and throughput prediction for a innermost loop kernel.

Getting started

pip install --user osaca

python ./setup.py install

Design

A schematic design of OSACA’s workflow is shown below:

Usage

The usage of OSACA can be listed as:

osaca [-h] [-V] [--arch ARCH] [--export-graph GRAPHNAME] FILEPATH

-h, --help

prints out the help message.

-V, --version

shows the program’s version number.

--arch ARCH

needs to be replaced with the wished architecture abbreviation. This flag is necessary for the throughput analysis (default function) and the inclusion of an ibench output (-i). Possible options are SNB, IVB, HSW, BDW, SKX and CSX for the latest Intel micro architectures starting from Intel Sandy Bridge and ZEN1 for AMD Zen (17h family) architecture. Furthermore, VULCAN for Marvell`s ARM-based ThunderX2 architecture is available.

--insert-marker

OSACA calls the Kerncraft module for the interactively insertion of IACA marker in suggested assembly blocks.

--db-check

Run a sanity check on the by “–arch” specified database. The output depends on the verbosity level. Keep in mind you have to provide a (dummy) filename in anyway.

--export-graph EXPORT_PATH

Output path for .dot file export. If “.” is given, the file will be stored as “./osaca_dg.dot”. After the file was created, you can convert it to a PDF file using dot: dot -Tpdf osaca_dg.dot -o osaca_dependency_graph.pdf

The FILEPATH describes the filepath to the file to work with and is always necessary

Hereinafter OSACA’s scope of function will be described.

movl    $111,%ebx       #IACA/OSACA START MARKER
.byte   100,103,144     #IACA/OSACA START MARKER
Loop:
  # ...
movl    $222,%ebx       #IACA/OSACA END MARKER
.byte   100,103,144     #IACA/OSACA END MARKER

mov x1, #111            // OSACA START
.byte 213,3,32,31       // OSACA START
  \\ ...
mov x1, #222            // OSACA END
.byte 213,3,32,31       // OSACA END

Example

For clarifying the functionality of OSACA a sample kernel is analyzed for an Intel CSX core hereafter:

double a[N], double b[N];
double s;

// loop
for(int i = 0; i < N; ++i)
    a[i] = s * b[i];

The code shows a simple scalar multiplication of a vector b and a floating-point number s. The result is written in vector a. After including the OSACA byte marker into the assembly, one can start the analysis typing

osaca --arch CSX PATH/TO/FILE

in the command line.

The output is:

Open Source Architecture Code Analyzer (OSACA) - v0.3
Analyzed file:      scale.s.csx.O3.s
Architecture:       csx
Timestamp:          2019-10-03 23:36:21

 P - Throughput of LOAD operation can be hidden behind a past or future STORE instruction
 * - Instruction micro-ops not bound to a port
 X - No throughput/latency information for this instruction in data file


Throughput Analysis Report
--------------------------
                              Port pressure in cycles
     |  0   - 0DV  |  1   |  2   -  2D  |  3   -  3D  |  4   |  5   |  6   |  7   |
-----------------------------------------------------------------------------------
 170 |             |      |             |             |      |      |      |      |   .L22:
 171 | 0.50        | 0.50 | 0.50   0.50 | 0.50   0.50 |      |      |      |      |   vmulpd        (%r12,%rax), %ymm1, %ymm0
 172 |             |      | 0.50        | 0.50        | 1.00 |      |      |      |   vmovapd       %ymm0, 0(%r13,%rax)
 173 | 0.25        | 0.25 |             |             |      | 0.25 | 0.25 |      |   addq  $32, %rax
 174 | 0.25        | 0.25 |             |             |      | 0.25 | 0.25 |      |   cmpq  %rax, %r14
 175 |             |      |             |             |      |      |      |      | * jne   .L22

       1.00          1.00   1.00   0.50   1.00   0.50   1.00   0.50   0.50


Latency Analysis Report
-----------------------
 171 |  8.0 | | vmulpd      (%r12,%rax), %ymm1, %ymm0
 172 |  5.0 | | vmovapd     %ymm0, 0(%r13,%rax)

       13.0


Loop-Carried Dependencies Analysis Report
-----------------------------------------
173 |  1.0 | addq   $32, %rax                      | [173]

It shows the whole kernel together with the average port pressure of each instruction form and the overall port binding. Furthermore, the critical path of the loop kernel and all loop-carried dependencies, each with a list of line numbers being part of this dependency chain on the right.

Credits

Implementation: Jan Laukemann

License

AGPL-3.0