- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I use Cyclone V development kit with no Ethernet connected to the FPGA. One Ethernet connected to HPS.
I tested CPU load and data rate with simple program:for(i1 = 0; i1 < i2; i1++)
{
sent_size += sendto(sock, buf_frame, packet_size, 0,
(struct sockaddr *) &echoclient,
sizeof(echoclient));
//usleep(sleep_us); /// commented to test max rate, uncommented to reduce CPU load
}
I've get 86.4 MB/s with packets 8192 B and 100 % CPU load. But I need CPU for others tasks. So, I've uncomment usleep(sleep_us) and got with sleep_us = 450, packet_size=8192 only 13.56 MB/s with 17 % CPU. 17 % CPU load is too much, 14 MB/s is too few. Is it possible to send more data for less CPU time on HPS of Cyclone V SoC ?
Link Copied
6 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Did you verify you're transmitting Jumbo frames?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The LAN where device is connected has MTU = 1500
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Fix that and re-test.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
It is desirable to avoid the restrictions that MTU is differ from commonly acceptable.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
--- Quote Start --- It is desirable to avoid the restrictions that MTU is differ from commonly acceptable. --- Quote End --- Your original question in your original post was: --- Quote Start --- Is it possible to send more data for less CPU time on HPS of Cyclone V SoC ? --- Quote End --- The answer to that question is: yes, by sending fewer but larger packets. Your current test is sending (6) 1500b packets instead of (1) 8192b packet, so your processor overhead is 6x what it needs to be.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
--- Quote Start --- Your original question in your original post was: The answer to that question is: yes, by sending fewer but larger packets. Your current test is sending (6) 1500b packets instead of (1) 8192b packet, so your processor overhead is 6x what it needs to be. --- Quote End --- I'm not a Linux guru or anything, but I thought the stmmac driver included DMA support, so that a non-blocking socket to ethernet should be able to reduce overhead. I haven't looked in to it in any depth, but it would make sense. To reduce the CPU load on such a tight loop, though, you'd need some sort of select event that triggers when the driver is finished sending, with the thread sleeping in between (pthread)
![](/skins/images/872293744008A34B36F8ABF94A46CC66/responsive_peak/images/icon_anonymous_message.png)
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page