Adding correct vlmul shifts for all vpu modes#215
Merged
xalbertoisorna merged 2 commits intoxmos:developfrom Apr 16, 2026
Merged
Adding correct vlmul shifts for all vpu modes#215xalbertoisorna merged 2 commits intoxmos:developfrom
xalbertoisorna merged 2 commits intoxmos:developfrom
Conversation
Contributor
Author
|
andrewstanfordjason
approved these changes
Apr 15, 2026
| if (NN_ARCH == TARGET_ARCH_VX4A){ | ||
| res = res >> 1; | ||
| } | ||
| int32_t res = ((int32_t)vpu->vR.s8[i] * (int32_t)val + (1L<<(shift - 1))) >> shift; |
Contributor
There was a problem hiding this comment.
as this pattern (X + (1<<Y))>>Y comes up a lot for the rounding it might be a good idea to make it a macro if it's not already
Contributor
Author
|
@uvvpavel fyi, will be integrated in next release |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds the correct right shift for vlmul in vpu_sim.c in different vpu modes for vx4a other than s16 (s8, s32).
It could be implemented directly with macros, but since it will end up as an if statement at the architecture level anyway, it feels more natural to use a switch and compute each case explicitly.
I’m happy to replace the function with macros or an alternative approach if preferred; my main goal here is to ensure correct VPU rounding and saturation for all modes so I have a vpu sim to test against.