issues
search
NVIDIA
/
Fuser
A Fusion Code Generator for NVIDIA GPUs (commonly known as "nvFuser")
Other
225
stars
42
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Benchmark failure in lit-gpt falcon-7b model
#2264
xwang233
opened
11 hours ago
0
bank conflicts due to L1 cache
#2263
liqiangxl
opened
13 hours ago
0
Ensure that non-matmul schedulers reject matmul fusions
#2262
jacobhinkle
opened
17 hours ago
2
Invalid loop promotion due to propagation of incomplete promotion
#2261
naoyam
opened
19 hours ago
0
Multi-dimensional DeviceMesh
#2260
cowanmeg
opened
1 day ago
2
Implement User Scheduling in Python Frontend
#2259
rdspring1
opened
1 day ago
0
fp8 recipe support with full reduction
#2258
jjsjann123
opened
1 day ago
0
Five distributed tests are failing on one GPU.
#2257
wujingyue
opened
1 day ago
1
Remove CommParams and make the fields there data attributes.
#2256
wujingyue
opened
1 day ago
2
Make the naming style consistent.
#2255
wujingyue
opened
1 day ago
1
Remove createParams* functions.
#2254
wujingyue
opened
1 day ago
1
IdModel buildAllGraphs() fails in presegment pass
#2253
jjsjann123
opened
1 day ago
3
reduce IdModel graph build in AllocationOrderInference
#2252
jjsjann123
closed
1 day ago
3
Generate a DOT representation of fusion transform.
#2251
naoyam
closed
9 hours ago
2
Remove `is_root_in_mesh` and `team` from CommParams.
#2250
wujingyue
opened
1 day ago
7
Revert "Allocation order refactor (#2168)"
#2249
jjsjann123
closed
1 day ago
2
removing obsolete error message
#2248
jjsjann123
closed
1 day ago
2
Allow FusionExecutorCache to take preallocated outputs
#2247
samnordmann
opened
1 day ago
2
Fix MatmulOp IterDomain mapping
#2246
jacobhinkle
closed
21 hours ago
2
Fix bug in scatter
#2245
samnordmann
closed
2 days ago
8
Minor fixes and cleanups.
#2244
wujingyue
closed
1 day ago
2
Host IRs: add Communications
#2243
samnordmann
closed
20 hours ago
2
NoOp scheduler claims zero-dimensional non-trivial fusions
#2241
jacobhinkle
opened
2 days ago
0
`LinearOp` IR node
#2240
Priya2698
opened
3 days ago
2
Map sibling domains in the LOOP graph
#2239
naoyam
opened
3 days ago
1
[WIP] IdModel-based indexing
#2238
naoyam
opened
3 days ago
0
Print runtime information for debugging.
#2237
wujingyue
closed
2 days ago
2
[WIP] Translate matmul op
#2236
jacobhinkle
opened
3 days ago
1
allocation domain propagation to support reshape operations
#2235
jjsjann123
opened
3 days ago
0
Improve perf of `isResharding(Expr*)`
#2234
samnordmann
closed
23 hours ago
3
Clone of #2185 before rebase for reference.
#2233
wujingyue
opened
6 days ago
0
IdModel: loop promotion cleanup
#2232
naoyam
opened
6 days ago
3
Remove unnecessary `default:`.
#2231
wujingyue
closed
6 days ago
1
shmoo test of iter grouped grid reduction
#2230
liqiangxl
closed
6 days ago
1
Add test that external matmul plugin is loaded properly
#2229
jacobhinkle
closed
6 days ago
3
DeviceMesh helper functions
#2228
cowanmeg
closed
3 days ago
3
Kill `MmaOp::layout`
#2227
zasdfgbnm
opened
1 week ago
2
Support allocation domain in the matmul scheduler
#2226
protonu
opened
1 week ago
3
[WIP] Check that matmul epilogue contains only pointwise ops
#2225
jacobhinkle
opened
1 week ago
0
Initial top-down presegmentation phase
#2224
jacobhinkle
opened
1 week ago
2
Simplify Communication::toString.
#2223
wujingyue
closed
1 week ago
1
Add `TensorView::flatten` as a convenient method for merging multiple dims
#2222
zasdfgbnm
closed
1 week ago
2
Remove outdated checks from the matmul scheduler.
#2221
wujingyue
closed
1 week ago
3
IdModel: Step 5 of the loop promotion analysis
#2220
naoyam
closed
6 days ago
3
Skip tests using FusionProfiler when compute-sanitizer is used
#2219
naoyam
closed
1 week ago
2
[WIP] IdModel-based indexing
#2218
naoyam
closed
3 days ago
0
add a inner_outer_reduction scheduler
#2217
liqiangxl
opened
1 week ago
0
SS
#2216
zasdfgbnm
closed
1 week ago
0
Correct matmul benchmark name
#2215
Priya2698
closed
1 week ago
0
Temporary hopper case for getMmaOp
#2214
cowanmeg
closed
1 week ago
2
Next