"git@developer.sourcefind.cn:orangecat/ollama.git" did not exist on "6f26e9322fd4639b4e414f8890b0213783e74d7c"
  • Jianfeng Yan's avatar
    Conv3d new (#94) · 6dfb92bb
    Jianfeng Yan authored
    
    
    * conv3d compiles but has memory error
    
    * conv3d works
    
    * fix performance issue by using __builtin_amdgc_readfirstlane
    
    * change MakeBlock2CTileMap to MakeDefaultBlock2CTileMap; change c_blockid_to* to cblockid_to*
    
    * clang-format
    
    * remove CK_EXPERIMENTAL_PASS_TENSOR_DECRIPTOR_BY_*; moved wrapper into DeviceConv3d
    
    * format
    
    * remove useless marc
    
    * add comment
    Co-authored-by: default avatarChao Liu <chao.liu2@amd.com>
    6dfb92bb
integral_constant.hpp 1.54 KB