Multi-group CachePool with coherence-free write-through private L1 cache - #24
Open
HHT1228 wants to merge 39 commits into
Open
Multi-group CachePool with coherence-free write-through private L1 cache#24HHT1228 wants to merge 39 commits into
HHT1228 wants to merge 39 commits into
Conversation
DiyouS
requested changes
Jul 22, 2026
DiyouS
left a comment
DiyouS
left a comment
Collaborator
There was a problem hiding this comment.
See comments t each file.
General Issues:
- Bender.lock seems completely broken now. Need fix
- Unused/comment codes needs to be removed
- Why HPD cache is configured through peripherals? It eats a huge area and at super bad performance! They should be configured directly from the core, using CSR instead!
| import reqrsp_pkg::*; | ||
| #( | ||
| // Core-side topology | ||
| parameter int unsigned NrTCDMPortsPerCore = 5, // Spatz lanes + 1 Snitch |
Collaborator
There was a problem hiding this comment.
Just a coding style note: We usually do not accept end-of-line comments. It is cleaner to put comments in the line above
| input l2_cache_rsp_t l2_rsp_i, | ||
| output logic l2_rsp_ready_o, // response ready -> interco core_rsp_ready_i | ||
|
|
||
| // Performance events |
Collaborator
There was a problem hiding this comment.
Maybe better to pack all performance related signals into a structure
| } | ||
| // if(core_id != 0) { | ||
| // while(1){} | ||
| // } |
| CXX ?= /usr/pack/gcc-11.2.0-af/linux-x64/bin/g++ | ||
| CC ?= /usr/pack/gcc-11.2.0-af/linux-x64/bin/gcc | ||
| CXX := /usr/pack/gcc-11.2.0-af/linux-x64/bin/g++ | ||
| CC := /usr/pack/gcc-11.2.0-af/linux-x64/bin/gcc |
Collaborator
There was a problem hiding this comment.
Keep it ?=. The written path only works in ETH environment
DiyouS
reviewed
Jul 24, 2026
DiyouS
requested changes
Jul 24, 2026
| common_cells: | ||
| revision: 9ca8a7655f741e7dd5736669a20a301325194c28 | ||
| version: 1.39.0 | ||
| revision: 1281545696eb3fcba50ec5b4275993476a3c710e |
Collaborator
There was a problem hiding this comment.
any reason of switching to a new version?
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 a coherence-free per-core private L1 cache (LP1) with write-through mode. Includes end-to-end AMO support and a per-core CMO runtime function (
lp1_inval/lp1_wt_flush) for explicit software data protection. Data protection applied and tested on selected kernels (e.g., fdotp, fmatmul, fft, spin/mcs-lock, gemv).See WTL1_note.md for details.
Note: two Bender.yml deps (Insitu-Cache, hpdcache) still point to my personal forks.