[GitHub] [incubator-tvm] FrozenGene commented on a change in pull request #6042: [TIR] Add an option to limit the maximum extent of explicitly unrolled loops

[GitBox <gi...@apache.org>]

FrozenGene commented on a change in pull request #6042:
URL: https://github.com/apache/incubator-tvm/pull/6042#discussion_r453482210



##########
File path: src/tir/transforms/unroll_loop.cc
##########
@@ -57,6 +58,9 @@ struct UnrollLoopConfigNode : public tvm::AttrsNode<UnrollLoopConfigNode> {
     TVM_ATTR_FIELD(explicit_unroll)
         .describe("Whether to explicitly unroll the loop instead of setting a pragma")
         .set_default(true);
+    TVM_ATTR_FIELD(explicit_unroll_max_extent)
+        .describe("The maximum extent of a loop that can be unrolled explicitly (-1 for infinite)")
+        .set_default(32);

Review comment:
       How about change the default value to `16`? As like ARM CPU (especially arm 32 bits), 32 is still a little bigger for them. `16` should cover common hardwares.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org