Teledyne SP Devices ADQ7 Скачать руководство пользователя страница 11

Classification

Revision

Public

PA1

Document ID

Print date

19-2241

2019-05-02

Figure 5: Screen capture from the CUDA example.

5.2.1

Running the Example

1.

Make sure cuda toolkit including examples are installed

2.

Make sure your user is in the ‘adq’ group:

groups

3.

Make sure the ADQ7 device driver is loaded:

ls /dev/adq*

4.

Go to ADQ7_GPUDirect_example/source/gdrdrv

5.

Build kernel module for GPUDirect:

make

6.

Load kernel module:

sudo ./insmod.sh

7.

Go to directory ADQ7_GPUDirect_example/source

8.

Build example:

make

9.

Run example:

./cuda_example

Once the example is running press

h

for information about mouse and keyboard controls and

i

for

information.

5.2.2

Adjusting Example Settings

The file

gpu_streaming_defines.h

contains macros for settings and debug printouts. When two-channel

mode is active (

STREAM_CHANNELS = 2

) the channel used for generating graphics can be chosen by re-

ADQ7 GPU Peer-To-Peer – User Guide

www.teledyne-spdevices.com

Page 10 of 11

Содержание ADQ7

Страница 1: ...ADQ7 GPU Peer To Peer User Guide Author s Teledyne SP Devices Document ID 19 2241 Classification Public Revision PA1 Print date 2019 05 02...

Страница 2: ...a Valid 7 4 3 Set up P2P GPU with SetupDMAP2p2D 7 4 3 1 Nvidia 7 4 3 2 AMD 8 4 4 Wait for a Completed Buffer 8 4 4 1 Nvidia 8 4 4 2 AMD 8 4 5 Detect and Handle Overflows 8 4 6 Process Received Data an...

Страница 3: ...sing unit Graphics card OCT Optical coherence tomography P2P Peer to peer UL2 User logic 2 open FPGA area in the ADQ7 firmware 2 Prerequisites Hardware ADQ7 digitizer Peer to peer capable GPU Windows...

Страница 4: ...he right hand side of the data buffer the data valid buffer is shown containing the number one for each valid line A A A A A A A A B B A A A A A A A A B B A A A A A A A A B B 1 1 1 1 1 1 Figure 1 Reco...

Страница 5: ...e attached example code 3 3 Backplane Peer To Peer Transfer Data is written from the digitizer directly to the GPU without going through the host CPU or host memory This reduces requirement on the hos...

Страница 6: ...05 02 a switch it is not a mandatory function of the PCIe standard Therefore make sure that there is a PCIe switch between the two endpoints or that the root complex supports peer to peer transfer AD...

Страница 7: ...be used Initialize GPU driver Allocate and pin buffers in GPU Initialize ADQ Set up triggers Set up P2P GPU with SetupDMAP2p2D Start streaming Wait for a completed buffer Detect and handle overflows P...

Страница 8: ...with data processing active 32 128 512 2048 8192 32768 131072 524288 2097152 8388608 5000 5200 5400 5600 5800 6000 6200 6400 6600 6800 7000 7200 1ch 2ch Record length samples MiB s Figure 4 Maximum t...

Страница 9: ...AMD Completed buffers are detected by GPU The function clEnqueueWaitSignalAMD is used to make GPU wait until the marker associated with the specified buffer is equal to or greater than the specified v...

Страница 10: ...d on the ADQ7 backplate Table 2 Signal connections for ADQ7 device Signal ADQ7 input Analog channel A A Analog channel B B optional A trigger Trig B trigger Sync optional 5 2 Nvidia Example The exampl...

Страница 11: ...el module for GPUDirect make 6 Load kernel module sudo insmod sh 7 Go to directory ADQ7_GPUDirect_example source 8 Build example make 9 Run example cuda_example Once the example is running press h for...

Страница 12: ...OpenCL queue which blocks the queue until a marker write is detected At that time a kernel enqueued after the marker wait can run This sequence is repeated in an alternating pattern for two buffers T...

Страница 13: ...ngen 6 SE 583 30 Link ping Sweden Phone 46 0 13 645 0600 Fax 46 0 13 991 3044 Email spd_info teledyne com Copyright 2019 Teledyne Signal Processing Devices Sweden AB All rights reserved including thos...

Отзывы: