Add more GPU architectures support (#112)

* Add more GPU architectures support * Update layout.py * Optimize performance, Add SM90 support, Add 1D2D SM100 support * Add fmtlib submodule at commit 553ec11 --------- Co-authored-by: fzyzcjy <5236035+fzyzcjy@users.noreply.github.com>
2025-07-18 11:32:22 +08:00
parent 03d0be3d2d
commit 9da4a23561
67 changed files with 5586 additions and 2965 deletions
--- a/csrc/utils/format.hpp
+++ b/csrc/utils/format.hpp
@@ -0,0 +1,6 @@
+#pragma once
+
+// Just a wrapper for the `fmt` headers
+#define FMT_HEADER_ONLY
+#include <fmt/base.h>
+#include <fmt/format.h>