Use HMB for FPGA Timestamps #5499
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Current timestamp read code uses FPGA register reads. Through testing, this read was slower then clock_gettime by about 4-5x. However, another method of reading the FPGA time is available, using HMB. HMB is memory mapped IO from RAM to the FPGA. So to code side, reading the value is just a memory barrier and a memory read. There is some latency on the write side, so a very small artifical delay (5us) is added to avoid register reads such as interrupts being ahead of current timestamps, which could cause issues. Below is read times for 1000 calls to clock_gettime, register reads and hmb reads. ``` Clock: Rise 1.72939400 s Fall 1.72990700 s Delta 0.00051300 s FPGA : Rise 1.72999000 s Fall 1.73429300 s Delta 0.00430300 s HMB : Rise 1.73466800 s Fall 1.73481900 s Delta 0.00015100 s ```
calcmogul
reviewed
Aug 3, 2023
calcmogul
reviewed
Aug 3, 2023
PeterJohnson
reviewed
Aug 3, 2023
|
|
|
/format |
PeterJohnson
approved these changes
Aug 4, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Current timestamp read code uses FPGA register reads. Through testing, this read was slower then clock_gettime by about 4-5x. However, another method of reading the FPGA time is available, using HMB. HMB is memory mapped IO from RAM to the FPGA. So to code side, reading the value is just a memory barrier and a memory read.
There is some latency on the write side, so a very small artifical delay (5us) is added to avoid register reads such as interrupts being ahead of current timestamps, which could cause issues.
Below is read times for 1000 calls to clock_gettime, register reads and hmb reads.