fix: avoid vector loads during restore#12
Open
good-circle wants to merge 1 commit into
Open
Conversation
There was a problem hiding this comment.
Pull request overview
This PR updates the RVV restore path to avoid vector memory loads from the flash/MMIO-backed checkpoint region (which XiangShan can’t handle), by switching to scalar loads plus vector register construction using vector ALU instructions.
Changes:
- Add an initial
vsetvlito establish a temporarySEW=64, LMUL=1, VL=2configuration for restore-time vector ops. - Replace
vl1re64.vwhole-register loads withld+vmv.v.x/vslide1down.vxto populate vector registers without vector memory accesses. - Refactor the vector restore sequence using a new
RESTORE_VECTOR_REGmacro.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
XiangShan does not support vector loads from the flash/MMIO region used by GCPT restore. Restoring vector registers with vl1re64.v from that region can fail during checkpoint restore. Load each saved vector register with scalar loads first, then move the data into the vector register with vector move/slide instructions. This keeps the restore path compatible with both flash and DRAM backed checkpoints.
good-circle
force-pushed
the
20260628-cpt
branch
from
a96a2ba to
a482770
Compare
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
XiangShan does not support vector loads from the flash/MMIO region used by GCPT restore. Restoring vector registers with vl1re64.v from that region can fail during checkpoint restore.
Load each saved vector register with scalar loads first, then move the data into the vector register with vector move/slide instructions. This keeps the restore path compatible with both flash and DRAM backed checkpoints.