ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 06:49:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_BUILDER] Convolution forward transfer concepts. (#3535)

Browse Source 
* Rename member variable to better reflect its actuall meaning.

* Add transfer checks for conv fwd xdl.

* Validate tensor layouts & vector size conv fwd v3.

* Add combined transfer concepts.

* Add transfer concepts for conv fwd factories.

* Fix clang format

* Add helper instruction to get max mem vector instruction width.

* Apply review comments.

* Rename thread cluster access(->arrange) order concept

* FIx merge artifacts.

* Add generic access order limits into block transfer concept.
This commit is contained in:
Adam Osewski
2026-01-19 10:54:10 +01:00
committed by GitHub
parent fe40a5d139
commit 1a6d1b59ef
13 changed files with 570 additions and 157 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
include/ck_tile/core/arch/arch.hpp
View File

@@ -1173,4 +1173,11 @@ enum LLVMSchedGroupMask : int32_t
DS_WRITE = 1 << 9,
ALL = (DS_WRITE << 1) - 1,
};
CK_TILE_HOST_DEVICE static constexpr auto get_max_mem_vec_inst_width()
{
// Currently on all arch max memory vector instruction width is 16 bytes.
return 16;
}
} // namespace ck_tile