new compute kernel design, less mem read delay

This commit is contained in:
Nitload
2026-04-07 01:04:05 +08:00
parent d702b3224c
commit 12a216d404
47 changed files with 225 additions and 145 deletions
@@ -16,10 +16,6 @@
<Filter Include="Document">
<UniqueIdentifier>{fbd74b53-5314-4d9d-a788-07bd1b413f74}</UniqueIdentifier>
</Filter>
<Filter Include="模板文件">
<UniqueIdentifier>{C8E0D8C1-42A1-4F7E-9C35-7E6D32D49A7B}</UniqueIdentifier>
<Extensions>h;hh;hpp;hxx;h++;hm;inl;inc;ipp;xsd</Extensions>
</Filter>
</ItemGroup>
<ItemGroup>
<ClCompile Include="sn-src\dllmain.c">
@@ -43,7 +39,7 @@
<ClCompile Include="sn-src\load_glfw.c">
<Filter>源文件</Filter>
</ClCompile>
<ClCompile Include="sn-src\m.transpose.x86.c">
<ClCompile Include="sn-src\m.basic.x86.c">
<Filter>源文件</Filter>
</ClCompile>
<ClCompile Include="sn-src\sk.fma.x86.c">
@@ -64,15 +60,12 @@
<ClCompile Include="sn-src\mk.fma.x86.c">
<Filter>源文件</Filter>
</ClCompile>
<ClCompile Include="sn-src\thread.windows.c">
<ClCompile Include="sn-src\t.affinity.c">
<Filter>源文件</Filter>
</ClCompile>
<ClCompile Include="sn-src\simdid_compare.x86.c">
<Filter>源文件</Filter>
</ClCompile>
<ClCompile Include="sn-src\thread.linux.c">
<Filter>源文件</Filter>
</ClCompile>
</ItemGroup>
<ItemGroup>
<ClInclude Include="sn-head\gl.h">