PYNQ: Python productivity on ZYNQ

PYNQ: Python productivity on ZYNQ

---

기본적인 설명

The PYNQ-Z1 board is designed to be used with PYNQ, a new open-source framework that enables embedded programmers to exploit the capabilities of Xilinx Zynq All Programmable SoCs (APSoCs) without having to design programmable logic circuits.

• YNQ XC7Z020-1CLG400C

  • 650MHz dual-core Cortex-A9 processor
  • DDR3 memory controller with 8 DMA channels and 4 High Performance AXI3 Slave ports
  • High-bandwidth peripheral controllers: 1G Ethernet, USB 2.0, SDIO
  • Low-bandwidth peripheral controller: SPI, UART, CAN, I2C
  • Programmable from JTAG, Quad-SPI flash, and microSD card
  • Programmable logic equivalent to Artix-7 FPGA
    • 13,300 logic slices, each with four 6-input LUTs and 8 flip-flops
    • 630 KB of fast block RAM
    • 4 clock management tiles, each with a phase-locked loop (PLL) and mixed-mode clock manager (MMCM)
    • 220 DSP slices
    • On-chip analog-to-digital converter (XADC)

• 512 MB DDR3

• Wide range of USB, Ethernet, Video and Audio connectivity
• Arduino shield and Pmod connectors for adding-on hardware devices

• Programmable from JTAG, Quad-SPI flash, and microSD card

• Key FPGA Specifications
  

다른 비싼것 FPGA 성능
ZYNQ-ZCU104 (약 100만원)

• Logic slices 504,000
• memory 38MB
• DSP slices 1,728

ZYNQ-ZCU104 (약 1,000만원)

• Logic slices 930,000
• memory 60.5B
• DSP slices 4,272

보드 이미지 다운로드
http://www.pynq.io/board

보드 비교

dd command를 이용한 Disk image fussing on micro sd card

공식문서
https://pynq.readthedocs.io/en/v1.3/17_appendix.html

확인 및 unmount
diskutil list
`diskutil unmountDisk /dev/disk``

플래시 방법
sudo dd bs=1m if=pynq_z1_v2.4.img of=/dev/rdisk4

중간 중간 progress확인은 ctrl+t (SIGINFO signal)을 이용해서 확인 한다.

load: 7.21  cmd: dd 6329 uninterruptible 0.00u 0.00s
30+0 records in
29+0 records out
30408704 bytes transferred in 3.649711 secs (8331812 bytes/sec)
load: 7.21  cmd: dd 6329 uninterruptible 0.00u 0.01s
36+0 records in
35+0 records out
36700160 bytes transferred in 4.396652 secs (8347297 bytes/sec)
load: 5.83  cmd: dd 6329 uninterruptible 0.00u 0.29s
1033+0 records in
1032+0 records out
1082130432 bytes transferred in 128.452542 secs (8424360 bytes/sec)
load: 6.31  cmd: dd 6329 uninterruptible 0.00u 0.53s
1885+0 records in
1884+0 records out
1975517184 bytes transferred in 235.902680 secs (8374289 bytes/sec)
5401+1 records in
5401+1 records out
5664169984 bytes transferred in 679.525499 secs (8335478 bytes/sec)

정상적으로 했다면 아래와 같이 변경된다.
diskutil list

/dev/disk4 (internal, physical):
  #:                       TYPE NAME                    SIZE       IDENTIFIER
  0:     FDisk_partition_scheme                        *7.8 GB     disk4
  1:                 DOS_FAT_32                         7.8 GB     disk4s1

/dev/disk4 (internal, physical):
   #:                       TYPE NAME                    SIZE       IDENTIFIER
   0:     FDisk_partition_scheme                        *7.8 GB     disk4
   1:             Windows_FAT_32 NO NAME                 104.9 MB   disk4s1
   2:                      Linux                         5.6 GB     disk4s2

Getting Started with PYNQ support

• 동영상 가이드 존재
  https://pynq.readthedocs.io/en/v2.0/getting_started.html

설치후 OS 버전 정보

xilinx@pynq:~$ lsb_release -a
No LSB modules are available.
Distributor ID:    pynqlinux
Description:    PYNQ Linux, based on Ubuntu 18.04
Release:    v2.4
Codename:    provo

접속 기본 정보

공식 사이트: https://pynq.readthedocs.io/en/latest/getting_started.html

공유기에 따라서 IP Address는 다름

초기 계정 비번 xilinx / xilinx

jupyter notebook 가능

• http://192.168.0.3:9090/tree
• http://192.168.2.99

SAMBA 공유 파일 시스템 접속 방법 (OSX)

고유 연결 방법 Finder에서 이동->연결->URL 입력

• mac/linux: smb://pynq/xilinx

참고 사이트

• pynq.io
• pynq.readthedocs.org
• github.com/Xilinx/pynq
• digilentinc.com/pynq
• pynq.io/support