[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r435383834 ## File path: src/target/llvm/codegen_llvm.cc ## @@ -736,7 +736,40 @@ llvm::Function* CodeGenLLVM::GetIntrinsicDecl(llvm::Intrinsic::ID id,

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r434689864 ## File path: src/target/llvm/codegen_llvm.cc ## @@ -736,7 +736,40 @@ llvm::Function* CodeGenLLVM::GetIntrinsicDecl(llvm::Intrinsic::ID id,

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r429998819 ## File path: topi/python/topi/cuda/softmax.py ## @@ -53,13 +54,62 @@ def schedule_softmax(outs): raise ValueError('Tag is expected to be

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r426735612 ## File path: src/tir/transforms/lower_warp_memory.cc ## @@ -213,9 +213,13 @@ class WarpAccessRewriter : protected StmtExprMutator {

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r426370564 ## File path: src/tir/transforms/lower_warp_memory.cc ## @@ -213,9 +213,13 @@ class WarpAccessRewriter : protected StmtExprMutator {

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r426350654 ## File path: src/tir/transforms/lower_warp_memory.cc ## @@ -213,9 +213,13 @@ class WarpAccessRewriter : protected StmtExprMutator {

[GitHub] [incubator-tvm] wpan11nv commented on a change in pull request #5600: [TOPI] Improve CUDA softmax scheduling

wpan11nv commented on a change in pull request #5600: URL: https://github.com/apache/incubator-tvm/pull/5600#discussion_r426314749 ## File path: src/tir/transforms/lower_warp_memory.cc ## @@ -213,9 +213,13 @@ class WarpAccessRewriter : protected StmtExprMutator {