xwb_fofb_shaper_filt.vhd: fix wrong internal sizes #36
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
ericonr
reviewed
May 17, 2024
Although the ABI supports up to 10 biquads, internal signals were sized to hold g_NUM_BIQUADS coefficients only. This can lead to wrong register accesses depending on the dimensions. So, proper size these signals to avoid this.
guilhermerc
force-pushed
the
fix-shaper-filt-regs
branch
from
1a8d930
to
4ee6a32
Compare
|
👍 |
ericonr
added a commit
to lnls-dig/uhal
that referenced
this pull request
May 17, 2024
The static assertion for the amount of coefficients was written to make sure the sizeof expressions were correct, but it also works as a check for if the ABI ever changes. The discrepancy in the amount of coefficients for Core and Controller is intentional: for Core, we have only the actually relevant coefficients; for Controller, we support the highest amount, so library clients don't need to implement logic based on num_biquads. However, we can loop only through the valid coefficients in both cases, for a slight optimization. When actually writing to hardware, only the valid coefficients can be written: both the 3 unused coefficients and the unused biquads must be skipped. This was a bug noticed during development of the library, and might be fixed in the future [1]. [1] lnls-dig/fofb-ctrl-gw#36
ericonr
added a commit
to lnls-dig/uhal
that referenced
this pull request
May 17, 2024
The static assertion for the amount of coefficients was written to make sure the sizeof expressions were correct, but it also works as a check for if the ABI ever changes. The discrepancy in the amount of coefficients for Core and Controller is intentional: for Core, we have only the actually relevant coefficients; for Controller, we support the highest amount, so library clients don't need to implement logic based on num_biquads. However, we can loop only through the valid coefficients in both cases, for a slight optimization. When actually writing to hardware, only the valid coefficients can be written: both the 3 unused coefficients and the unused biquads must be skipped. This was a bug noticed during development of the library, and might be fixed in the future [1]. [1] lnls-dig/fofb-ctrl-gw#36
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.
Although the ABI supports up to 10 biquads, internal signals were sized to hold g_NUM_BIQUADS coefficients only. This can lead to wrong register accesses depending on the dimensions. So, proper size these signals to avoid this.