ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 14:59:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

implement device batched gemm b scale for wmma (#2825)

Browse Source 
* rebased on top of develop

* fixed missing shuffeling and wrong indexing

* added tests for batched_b_scale

* added missing files

* fixed wrong stride computation and removed k batching (for now) due to precision issues

* reinstated k-batching with PRNG constrained to -1..1

* added specialization of GeneratorTensor_3 for int4 and fixed internal overflow

* added k-batching to reference and increased tolerances for test

* changed gemm_b_scale and gemm_universal tests to use correct parameters

* adressed review commentsd

* ported fixes back to non-batched version of b_scale

* adressed review comments

* run clang-format on older commits

* add type-conversion to AccDataType and then to CDataType to exactly mimic GPU's behavior

* added newline at end of file

* reflected changes from muitl-abd branch in batched b_scale

* fixed gfx11 issue

* changed range for pki4 to -1...1 (-0.5...0.5 never really made sense for i4 anyway and always should have caused compiler errors, but since there was no int4 specialization of GeneratorTensor3 until now, this passed

* run clang format

* set range of i4 generation to 0...1 for upstream tests to pass. This replicated previous behavior, which however means that it is NOT properly tested.

* reduced range for pk_i4 even further to 0..0

* removed failing xld instances. Failure now uncovered now that tests were fixed

* removed generation of int4 values entierly

* divide B buffer by BPackedSize

---------

Co-authored-by: Kevin Abraham <kevin.abraham@streamhpc.com>
This commit is contained in:
kabrahamAMD
2025-10-16 20:00:42 +02:00
committed by GitHub
parent d7278cc664
commit c4b2da9cbd
22 changed files with 1352 additions and 97 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
include/ck/library/utility/host_tensor_generator.hpp
View File

@@ -264,7 +264,7 @@ struct GeneratorTensor_2<ck::pk_i4_t>
{
int hi = std::rand() % (max_value - min_value) + min_value + 8;
int lo = std::rand() % (max_value - min_value) + min_value + 8;
ck::pk_i4_t r = ((hi << 4) + lo) & 0xff;
ck::pk_i4_t r = (((hi & 0xf) << 4) + (lo & 0xf));
return r;
}
};
@@ -436,6 +436,22 @@ struct GeneratorTensor_3<ck::f4x2_pk_t>
}
};
template <>
struct GeneratorTensor_3<ck::pk_i4_t>
{
int min_value = 0;
int max_value = 1;
template <typename... Is>
ck::pk_i4_t operator()(Is...)
{
int hi = std::rand() % (max_value - min_value) + min_value + 8;
int lo = std::rand() % (max_value - min_value) + min_value + 8;
ck::pk_i4_t r = (((hi & 0xf) << 4) + (lo & 0xf));
return r;
}
};
template <>
struct GeneratorTensor_3<ck::f6x32_pk_t>
{