[refactor] Separate SNode read/write kernels into a dedicated class #2205
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Outdated suggestions cannot be applied.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.
Related issue = #2196
Purpose
This is one of the steps to decouple
SNode
fromtaichi/program
. Because many files insidetaichi/ir
depend on SNode, they transitively depend ontaichi/program
as well.taichi/program
is very high level, and almost depends on everything in the codebase. So without breaking the tie, it is very hard for us to separate out IR.Note that this PR alone is not enough (actually, far from enough) for modularizing the IR. I'm thinking of moving out
SNode::expr
later (we may have to have a similar mapping fromSNode
toexpr
). The reason is similar:expr
is anExpr
, which depends on the expression and the frontend IRtaichi/program
.As a short-term goal, I think it will be good enough if we can make
ir.h/cpp
,statements.h/cpp
,snode.h/cpp
self contained.[Click here for the format server]