issues
search
ROCm
/
hipBLASLt
hipBLASLt is a library that provides general matrix-matrix operations with a flexible API and extends functionalities beyond a traditional BLAS library
https://rocm.docs.amd.com/projects/hipBLASLt/en/latest/index.html
MIT License
63
stars
88
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add support for fallback from compute type f16 to f32
#1263
cliffxzx
opened
4 weeks ago
2
Revert "Revised output logic YAML to use efficiency instead of gflops."
#1262
jichangjichang
closed
4 weeks ago
0
Fix: setting incorrect winner index make 3_LibraryLogic break
#1261
jichangjichang
closed
4 weeks ago
0
Bump rocm-docs-core from 1.8.2 to 1.8.3 in /docs/sphinx
#1260
dependabot[bot]
closed
3 weeks ago
0
Add packaging in requirements.txt
#1259
KKyang
closed
4 weeks ago
0
dot2 fp16 mac kernel for gfx942
#1258
boringmorning
opened
4 weeks ago
0
Fix workspace arg of hipblaslt-bench
#1257
AndySu12
closed
3 weeks ago
1
Fix temp register alignment in stream-k alpha check
#1256
AlexBrownAMD
closed
3 weeks ago
0
Clarify release and version number for 6.3.0
#1255
amd-jnovotny
closed
4 weeks ago
0
Optimized on ds read scheduling if numItersPLR == 0
#1254
Serge45
closed
1 week ago
2
Update gfx942 Equality 1004_hhs_bmm sizes
#1253
AndySu12
closed
1 month ago
1
make tensile detect local device correctly
#1252
fsx950223
closed
3 weeks ago
1
BBS NT Gridbase
#1251
Jinp800125
closed
1 month ago
1
Code refactoring
#1250
KKyang
closed
4 weeks ago
2
Enable test option for stream-k full tile or remainder tile
#1249
AlexBrownAMD
closed
3 weeks ago
1
update centos package install approach
#1248
fsx950223
closed
2 weeks ago
0
Remove function emitLdChangeReference
#1247
KKyang
closed
1 month ago
2
feature: DTV with Swizzling (tensorA)
#1246
solaslin
opened
1 month ago
0
Refine local read scheduling logic.
#1245
hcman2
closed
4 weeks ago
3
Cherry-pick to release-staging/rocm-rel-6.3: Convert changelog to new format (#1240)
#1244
amd-jnovotny
closed
1 month ago
0
[Issue]: "Attempting to use hipBLASLt on a unsupported architecture!"
#1243
nktice
closed
3 weeks ago
6
Fix clang compilation error
#1242
KKyang
closed
1 month ago
0
gridbased tuning BBS TN gfx942
#1241
aazz44ss
closed
1 month ago
2
Convert changelog to new format
#1240
amd-jnovotny
closed
1 month ago
0
Use the same addrVgpr for all vectors in global store
#1239
KKyang
closed
1 month ago
2
Add a check for unsupported transpose and datatype
#1238
cliffxzx
closed
3 weeks ago
2
Add Navi32 TN_BBS/NN_BBS Kernels
#1237
wenchuanchen
closed
3 weeks ago
3
remove previous tf32 20 cu logic files
#1236
m-kim
closed
1 month ago
0
fix: hipblaslt-bench output "non-supported type" when bias_type is no…
#1235
jichangjichang
closed
1 month ago
0
Conditionally populate _format9 array in scoped function
#1234
ellosel
opened
1 month ago
0
Aquavanjaram942X BBS TN GEMM sizes tuned
#1233
aferoz21
closed
1 month ago
2
Update gfx942 20cu tf32 NN
#1232
KKyang
closed
1 month ago
1
Fix incorrect global read of i8 and f8 on gfx1201
#1231
cliffxzx
closed
4 weeks ago
2
enable DirectToVgpr + input type conversion
#1230
nakajee
closed
1 month ago
2
fortran used by bench and test clients (#1160)
#1229
TorreZuk
closed
1 month ago
1
re-enable DirectToLds
#1228
nakajee
closed
1 month ago
2
use of min(int, size_t) causes ambiguity with clang
#1227
yxsamliu
closed
3 weeks ago
1
Update doc and lib version
#1226
jichangjichang
closed
1 month ago
0
Revert "Add a check for unsupported transpose and datatype"
#1225
jichangjichang
closed
1 month ago
0
update changelog for 6.3
#1224
jichangjichang
closed
1 month ago
0
[Issue]: Where is the `Module.addInst` method defined?
#1223
103yiran
closed
3 weeks ago
2
Enable variant builds via device ID and cu count
#1222
bstefanuk
opened
1 month ago
2
hip_fp8.h doesn't exist before hip 6.2
#1221
m-kim
opened
1 month ago
2
Update gfx942 BBS NT/NN/TN GridBased yaml for 20240911_BBS_non_llm_ra…
#1220
Jinp800125
closed
1 month ago
1
updating BBS_TN library for aquavanjaram942_20cu
#1219
babakpst
closed
1 month ago
1
Fix mismatch issue with GSU>255
#1218
nakajee
closed
1 month ago
2
Update gfx942 BBS NN/NT/TN Equality yamls for amending 0911 sizes
#1217
AndySu12
closed
1 month ago
0
findBLIS should check local build/dep
#1216
m-kim
closed
1 month ago
0
added kernels for BBS/HHS TN/NT/NN to equality for aquavanjaram942 (#…
#1215
smalekta
closed
1 month ago
0
modify WGMXCC to correct value
#1214
aazz44ss
closed
1 month ago
1
Previous
Next