Upload cuda_matmul_trace_vllm.json with huggingface_hub
Browse files- cuda_matmul_trace_vllm.json +629 -0
cuda_matmul_trace_vllm.json
ADDED
|
@@ -0,0 +1,629 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
{
|
| 3 |
+
"schemaVersion": 1,
|
| 4 |
+
"deviceProperties": [
|
| 5 |
+
{
|
| 6 |
+
"id": 0, "name": "AMD Instinct MI300X VF", "totalGlobalMem": 205571227648,
|
| 7 |
+
"computeMajor": 9, "computeMinor": 4,
|
| 8 |
+
"maxThreadsPerBlock": 1024, "maxThreadsPerMultiprocessor": 2048,
|
| 9 |
+
"regsPerBlock": 65536, "warpSize": 64,
|
| 10 |
+
"sharedMemPerBlock": 65536, "numSms": 304
|
| 11 |
+
, "maxSharedMemoryPerMultiProcessor": 65536
|
| 12 |
+
}
|
| 13 |
+
],
|
| 14 |
+
"roctracer_version": 4.1,
|
| 15 |
+
"hip_runtime_version": 60342131,
|
| 16 |
+
"hip_driver_version": 60342131,
|
| 17 |
+
"trace_id": "3B55A95D004243E49AFCA3BA14757EE7",
|
| 18 |
+
"traceEvents": [
|
| 19 |
+
{
|
| 20 |
+
"ph": "X", "cat": "cpu_op", "name": "aten::matmul", "pid": 455, "tid": 455,
|
| 21 |
+
"ts": 7429064525509.327, "dur": 297396.921,
|
| 22 |
+
"args": {
|
| 23 |
+
"External id": 1,"Record function id": 0, "Fwd thread id": 0, "Sequence number": 0, "finished": true, "Ev Idx": 0
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"ph": "X", "cat": "cpu_op", "name": "aten::mm", "pid": 455, "tid": 455,
|
| 28 |
+
"ts": 7429064525539.719, "dur": 297357.448,
|
| 29 |
+
"args": {
|
| 30 |
+
"External id": 2,"Record function id": 0, "Fwd thread id": 0, "Sequence number": 0, "finished": true, "Ev Idx": 1
|
| 31 |
+
}
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"ph": "X", "cat": "cpu_op", "name": "aten::empty", "pid": 455, "tid": 455,
|
| 35 |
+
"ts": 7429064525887.174, "dur": 39.320,
|
| 36 |
+
"args": {
|
| 37 |
+
"External id": 3,"Record function id": 0, "finished": true, "Ev Idx": 2
|
| 38 |
+
}
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipStreamIsCapturing", "pid": 455, "tid": 455,
|
| 42 |
+
"ts": 7429064525586.815, "dur": 2.314,
|
| 43 |
+
"args": {
|
| 44 |
+
"External id": 2, "cid": 290, "correlation": 1
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"ph": "f", "id": 1, "pid": 455, "tid": 455, "ts": 7429064525586.815,
|
| 49 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipMalloc", "pid": 455, "tid": 455,
|
| 53 |
+
"ts": 7429064525591.437, "dur": 75.397,
|
| 54 |
+
"args": {
|
| 55 |
+
"External id": 2, "bytes": 65011712, "cid": 108, "correlation": 2, "ptr": "0x55fa15629e50"
|
| 56 |
+
}
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"ph": "s", "id": 2, "pid": 455, "tid": 455, "ts": 7429064525591.437,
|
| 60 |
+
"cat": "ac2g", "name": "ac2g"
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipMalloc", "pid": 455, "tid": 455,
|
| 64 |
+
"ts": 7429064525740.086, "dur": 35.336,
|
| 65 |
+
"args": {
|
| 66 |
+
"External id": 2, "bytes": 2621440, "cid": 108, "correlation": 3, "ptr": "0x7ffe7048af50"
|
| 67 |
+
}
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"ph": "s", "id": 3, "pid": 455, "tid": 455, "ts": 7429064525740.086,
|
| 71 |
+
"cat": "ac2g", "name": "ac2g"
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipMemset", "pid": 455, "tid": 455,
|
| 75 |
+
"ts": 7429064525782.260, "dur": 14.254,
|
| 76 |
+
"args": {
|
| 77 |
+
"External id": 2, "cid": 154, "correlation": 4
|
| 78 |
+
}
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"ph": "f", "id": 4, "pid": 455, "tid": 455, "ts": 7429064525782.260,
|
| 82 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 86 |
+
"ts": 7429064525799.123, "dur": 0.763,
|
| 87 |
+
"args": {
|
| 88 |
+
"External id": 2, "cid": 370, "correlation": 5
|
| 89 |
+
}
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"ph": "f", "id": 5, "pid": 455, "tid": 455, "ts": 7429064525799.123,
|
| 93 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipStreamIsCapturing", "pid": 455, "tid": 455,
|
| 97 |
+
"ts": 7429064525892.788, "dur": 0.470,
|
| 98 |
+
"args": {
|
| 99 |
+
"External id": 3, "cid": 290, "correlation": 6
|
| 100 |
+
}
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"ph": "f", "id": 6, "pid": 455, "tid": 455, "ts": 7429064525892.788,
|
| 104 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipMalloc", "pid": 455, "tid": 455,
|
| 108 |
+
"ts": 7429064525893.749, "dur": 29.083,
|
| 109 |
+
"args": {
|
| 110 |
+
"External id": 3, "bytes": 79691776, "cid": 108, "correlation": 7, "ptr": "0x7ffe7048a140"
|
| 111 |
+
}
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"ph": "s", "id": 7, "pid": 455, "tid": 455, "ts": 7429064525893.749,
|
| 115 |
+
"cat": "ac2g", "name": "ac2g"
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 119 |
+
"ts": 7429064525932.854, "dur": 0.304,
|
| 120 |
+
"args": {
|
| 121 |
+
"External id": 2, "cid": 370, "correlation": 8
|
| 122 |
+
}
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"ph": "f", "id": 8, "pid": 455, "tid": 455, "ts": 7429064525932.854,
|
| 126 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 130 |
+
"ts": 7429064525994.881, "dur": 0.238,
|
| 131 |
+
"args": {
|
| 132 |
+
"External id": 2, "cid": 370, "correlation": 9
|
| 133 |
+
}
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"ph": "f", "id": 9, "pid": 455, "tid": 455, "ts": 7429064525994.881,
|
| 137 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 141 |
+
"ts": 7429064525996.286, "dur": 0.169,
|
| 142 |
+
"args": {
|
| 143 |
+
"External id": 2, "cid": 370, "correlation": 10
|
| 144 |
+
}
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"ph": "f", "id": 10, "pid": 455, "tid": 455, "ts": 7429064525996.286,
|
| 148 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipModuleLoad", "pid": 455, "tid": 455,
|
| 152 |
+
"ts": 7429064528594.963, "dur": 103603.691,
|
| 153 |
+
"args": {
|
| 154 |
+
"External id": 2, "cid": 170, "correlation": 11
|
| 155 |
+
}
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"ph": "f", "id": 11, "pid": 455, "tid": 455, "ts": 7429064528594.963,
|
| 159 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 163 |
+
"ts": 7429064632214.111, "dur": 1.114,
|
| 164 |
+
"args": {
|
| 165 |
+
"External id": 2, "cid": 370, "correlation": 12
|
| 166 |
+
}
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"ph": "f", "id": 12, "pid": 455, "tid": 455, "ts": 7429064632214.111,
|
| 170 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 174 |
+
"ts": 7429064745562.959, "dur": 2.839,
|
| 175 |
+
"args": {
|
| 176 |
+
"External id": 2, "cid": 370, "correlation": 13
|
| 177 |
+
}
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"ph": "f", "id": 13, "pid": 455, "tid": 455, "ts": 7429064745562.959,
|
| 181 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipGetDevicePropertiesR0600", "pid": 455, "tid": 455,
|
| 185 |
+
"ts": 7429064745586.023, "dur": 0.696,
|
| 186 |
+
"args": {
|
| 187 |
+
"External id": 2, "cid": 370, "correlation": 14
|
| 188 |
+
}
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"ph": "f", "id": 14, "pid": 455, "tid": 455, "ts": 7429064745586.023,
|
| 192 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipModuleLoad", "pid": 455, "tid": 455,
|
| 196 |
+
"ts": 7429064745602.158, "dur": 77217.980,
|
| 197 |
+
"args": {
|
| 198 |
+
"External id": 2, "cid": 170, "correlation": 15
|
| 199 |
+
}
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"ph": "f", "id": 15, "pid": 455, "tid": 455, "ts": 7429064745602.158,
|
| 203 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipExtModuleLaunchKernel", "pid": 455, "tid": 455,
|
| 207 |
+
"ts": 7429064822844.054, "dur": 20.100,
|
| 208 |
+
"args": {
|
| 209 |
+
"External id": 2, "kernel": "Cijk_Ailk_Bljk_SB_Bias_AS_SAV_UserArgs_MT128x64x32_MI32x32x1_SN_LDSB0_AFC1_AFEM1_AFEM1_ASEM1_CLR1_CADS0_EPS0_GRVWA4_GRVWB4_GSUAMB_ISA942_IU1_K1_LBSPPA0_LBSPPB128_LBSPPM0_LPA0_LPB4_LPM0_LRVW4_LWPMn1_MIAV0_MIWT2_1_MO1_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB1_WS64_WG64_4_1_WGMXCC1_WGMXCCG0", "cid": 65, "correlation": 16, "grid": [516096, 1, 1], "block": [256, 1, 1], "shared memory": 0
|
| 210 |
+
}
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"ph": "s", "id": 16, "pid": 455, "tid": 455, "ts": 7429064822844.054,
|
| 214 |
+
"cat": "ac2g", "name": "ac2g"
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipDeviceSynchronize", "pid": 455, "tid": 455,
|
| 218 |
+
"ts": 7429064823009.505, "dur": 1348.012,
|
| 219 |
+
"args": {
|
| 220 |
+
"cid": 48, "correlation": 17
|
| 221 |
+
}
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"ph": "f", "id": 17, "pid": 455, "tid": 455, "ts": 7429064823009.505,
|
| 225 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"ph": "X", "cat": "cuda_runtime", "name": "hipDeviceSynchronize", "pid": 455, "tid": 455,
|
| 229 |
+
"ts": 7429064824390.626, "dur": 0.870,
|
| 230 |
+
"args": {
|
| 231 |
+
"cid": 48, "correlation": 18
|
| 232 |
+
}
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"ph": "f", "id": 18, "pid": 455, "tid": 455, "ts": 7429064824390.626,
|
| 236 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"ph": "X", "cat": "gpu_memset", "name": "Memset (Device)", "pid": 2, "tid": 0,
|
| 240 |
+
"ts": 7429064525818.773, "dur": 5.652,
|
| 241 |
+
"args": {
|
| 242 |
+
"External id": 2, "device": 2, "stream": 0, "correlation": 4, "kind": "Device"
|
| 243 |
+
}
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"ph": "f", "id": 4, "pid": 2, "tid": 0, "ts": 7429064525818.773,
|
| 247 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"ph": "X", "cat": "kernel", "name": "Cijk_Ailk_Bljk_SB_Bias_AS_SAV_UserArgs_MT128x64x32_MI32x32x1_SN_LDSB0_AFC1_AFEM1_AFEM1_ASEM1_CLR1_CADS0_EPS0_GRVWA4_GRVWB4_GSUAMB_ISA942_IU1_K1_LBSPPA0_LBSPPB128_LBSPPM0_LPA0_LPB4_LPM0_LRVW4_LWPMn1_MIAV0_MIWT2_1_MO1_NTn1_NTA0_NTB0_NTC0_NTD0_NTM0_NEPBS0_NLCA1_NLCB1_ONLL1_PGR2_PLR1_PKA1_SIA3_SS1_SPO0_SRVW0_SSO0_SVW1_TLDS1_ULSGRO0_USL1_UIOFGRO0_USFGROn1_VSn1_VWA1_VWB1_WSGRA0_WSGRB1_WS64_WG64_4_1_WGMXCC1_WGMXCCG0", "pid": 2, "tid": 0,
|
| 251 |
+
"ts": 7429064822886.347, "dur": 1475.199,
|
| 252 |
+
"args": {
|
| 253 |
+
"External id": 2, "device": 2, "stream": 0, "correlation": 16, "kind": "Dispatch Kernel", "grid": [516096, 1, 1], "block": [256, 1, 1]
|
| 254 |
+
}
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"ph": "f", "id": 16, "pid": 2, "tid": 0, "ts": 7429064822886.347,
|
| 258 |
+
"cat": "ac2g", "name": "ac2g", "bp": "e"
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 0,
|
| 262 |
+
"args": {
|
| 263 |
+
"name": "python"
|
| 264 |
+
}
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 0,
|
| 268 |
+
"args": {
|
| 269 |
+
"labels": "CPU"
|
| 270 |
+
}
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 0,
|
| 274 |
+
"args": {
|
| 275 |
+
"sort_index": 455
|
| 276 |
+
}
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 0, "tid": 0,
|
| 280 |
+
"args": {
|
| 281 |
+
"name": "python"
|
| 282 |
+
}
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 0, "tid": 0,
|
| 286 |
+
"args": {
|
| 287 |
+
"labels": "GPU 0"
|
| 288 |
+
}
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 0, "tid": 0,
|
| 292 |
+
"args": {
|
| 293 |
+
"sort_index": 5000000
|
| 294 |
+
}
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 1, "tid": 0,
|
| 298 |
+
"args": {
|
| 299 |
+
"name": "python"
|
| 300 |
+
}
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 1, "tid": 0,
|
| 304 |
+
"args": {
|
| 305 |
+
"labels": "GPU 1"
|
| 306 |
+
}
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 1, "tid": 0,
|
| 310 |
+
"args": {
|
| 311 |
+
"sort_index": 5000001
|
| 312 |
+
}
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 2, "tid": 0,
|
| 316 |
+
"args": {
|
| 317 |
+
"name": "python"
|
| 318 |
+
}
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 2, "tid": 0,
|
| 322 |
+
"args": {
|
| 323 |
+
"labels": "GPU 2"
|
| 324 |
+
}
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 2, "tid": 0,
|
| 328 |
+
"args": {
|
| 329 |
+
"sort_index": 5000002
|
| 330 |
+
}
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 3, "tid": 0,
|
| 334 |
+
"args": {
|
| 335 |
+
"name": "python"
|
| 336 |
+
}
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 3, "tid": 0,
|
| 340 |
+
"args": {
|
| 341 |
+
"labels": "GPU 3"
|
| 342 |
+
}
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 3, "tid": 0,
|
| 346 |
+
"args": {
|
| 347 |
+
"sort_index": 5000003
|
| 348 |
+
}
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 4, "tid": 0,
|
| 352 |
+
"args": {
|
| 353 |
+
"name": "python"
|
| 354 |
+
}
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 4, "tid": 0,
|
| 358 |
+
"args": {
|
| 359 |
+
"labels": "GPU 4"
|
| 360 |
+
}
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 4, "tid": 0,
|
| 364 |
+
"args": {
|
| 365 |
+
"sort_index": 5000004
|
| 366 |
+
}
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 5, "tid": 0,
|
| 370 |
+
"args": {
|
| 371 |
+
"name": "python"
|
| 372 |
+
}
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 5, "tid": 0,
|
| 376 |
+
"args": {
|
| 377 |
+
"labels": "GPU 5"
|
| 378 |
+
}
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 5, "tid": 0,
|
| 382 |
+
"args": {
|
| 383 |
+
"sort_index": 5000005
|
| 384 |
+
}
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 6, "tid": 0,
|
| 388 |
+
"args": {
|
| 389 |
+
"name": "python"
|
| 390 |
+
}
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 6, "tid": 0,
|
| 394 |
+
"args": {
|
| 395 |
+
"labels": "GPU 6"
|
| 396 |
+
}
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 6, "tid": 0,
|
| 400 |
+
"args": {
|
| 401 |
+
"sort_index": 5000006
|
| 402 |
+
}
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 7, "tid": 0,
|
| 406 |
+
"args": {
|
| 407 |
+
"name": "python"
|
| 408 |
+
}
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 7, "tid": 0,
|
| 412 |
+
"args": {
|
| 413 |
+
"labels": "GPU 7"
|
| 414 |
+
}
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 7, "tid": 0,
|
| 418 |
+
"args": {
|
| 419 |
+
"sort_index": 5000007
|
| 420 |
+
}
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 8, "tid": 0,
|
| 424 |
+
"args": {
|
| 425 |
+
"name": "python"
|
| 426 |
+
}
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 8, "tid": 0,
|
| 430 |
+
"args": {
|
| 431 |
+
"labels": "GPU 8"
|
| 432 |
+
}
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 8, "tid": 0,
|
| 436 |
+
"args": {
|
| 437 |
+
"sort_index": 5000008
|
| 438 |
+
}
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 9, "tid": 0,
|
| 442 |
+
"args": {
|
| 443 |
+
"name": "python"
|
| 444 |
+
}
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 9, "tid": 0,
|
| 448 |
+
"args": {
|
| 449 |
+
"labels": "GPU 9"
|
| 450 |
+
}
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 9, "tid": 0,
|
| 454 |
+
"args": {
|
| 455 |
+
"sort_index": 5000009
|
| 456 |
+
}
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 10, "tid": 0,
|
| 460 |
+
"args": {
|
| 461 |
+
"name": "python"
|
| 462 |
+
}
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 10, "tid": 0,
|
| 466 |
+
"args": {
|
| 467 |
+
"labels": "GPU 10"
|
| 468 |
+
}
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 10, "tid": 0,
|
| 472 |
+
"args": {
|
| 473 |
+
"sort_index": 5000010
|
| 474 |
+
}
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 11, "tid": 0,
|
| 478 |
+
"args": {
|
| 479 |
+
"name": "python"
|
| 480 |
+
}
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 11, "tid": 0,
|
| 484 |
+
"args": {
|
| 485 |
+
"labels": "GPU 11"
|
| 486 |
+
}
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 11, "tid": 0,
|
| 490 |
+
"args": {
|
| 491 |
+
"sort_index": 5000011
|
| 492 |
+
}
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 12, "tid": 0,
|
| 496 |
+
"args": {
|
| 497 |
+
"name": "python"
|
| 498 |
+
}
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 12, "tid": 0,
|
| 502 |
+
"args": {
|
| 503 |
+
"labels": "GPU 12"
|
| 504 |
+
}
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 12, "tid": 0,
|
| 508 |
+
"args": {
|
| 509 |
+
"sort_index": 5000012
|
| 510 |
+
}
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 13, "tid": 0,
|
| 514 |
+
"args": {
|
| 515 |
+
"name": "python"
|
| 516 |
+
}
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 13, "tid": 0,
|
| 520 |
+
"args": {
|
| 521 |
+
"labels": "GPU 13"
|
| 522 |
+
}
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 13, "tid": 0,
|
| 526 |
+
"args": {
|
| 527 |
+
"sort_index": 5000013
|
| 528 |
+
}
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 14, "tid": 0,
|
| 532 |
+
"args": {
|
| 533 |
+
"name": "python"
|
| 534 |
+
}
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 14, "tid": 0,
|
| 538 |
+
"args": {
|
| 539 |
+
"labels": "GPU 14"
|
| 540 |
+
}
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 14, "tid": 0,
|
| 544 |
+
"args": {
|
| 545 |
+
"sort_index": 5000014
|
| 546 |
+
}
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"name": "process_name", "ph": "M", "ts": 7429064519991.639, "pid": 15, "tid": 0,
|
| 550 |
+
"args": {
|
| 551 |
+
"name": "python"
|
| 552 |
+
}
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"name": "process_labels", "ph": "M", "ts": 7429064519991.639, "pid": 15, "tid": 0,
|
| 556 |
+
"args": {
|
| 557 |
+
"labels": "GPU 15"
|
| 558 |
+
}
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 15, "tid": 0,
|
| 562 |
+
"args": {
|
| 563 |
+
"sort_index": 5000015
|
| 564 |
+
}
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"name": "thread_name", "ph": "M", "ts": 7429064519991.639, "pid": 2, "tid": 0,
|
| 568 |
+
"args": {
|
| 569 |
+
"name": "stream 0 "
|
| 570 |
+
}
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"name": "thread_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 2, "tid": 0,
|
| 574 |
+
"args": {
|
| 575 |
+
"sort_index": 0
|
| 576 |
+
}
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"name": "thread_name", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 455,
|
| 580 |
+
"args": {
|
| 581 |
+
"name": "thread 455 (python)"
|
| 582 |
+
}
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"name": "thread_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 455,
|
| 586 |
+
"args": {
|
| 587 |
+
"sort_index": 455
|
| 588 |
+
}
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"name": "thread_name", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 455,
|
| 592 |
+
"args": {
|
| 593 |
+
"name": "thread 455 (python)"
|
| 594 |
+
}
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"name": "thread_sort_index", "ph": "M", "ts": 7429064519991.639, "pid": 455, "tid": 455,
|
| 598 |
+
"args": {
|
| 599 |
+
"sort_index": 455
|
| 600 |
+
}
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"ph": "X", "cat": "Trace", "ts": 7429064519947.431, "dur": 304458.658,
|
| 604 |
+
"pid": "Spans", "tid": "PyTorch Profiler",
|
| 605 |
+
"name": "PyTorch Profiler (0)",
|
| 606 |
+
"args": {
|
| 607 |
+
"Op count": 0
|
| 608 |
+
}
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"name": "process_sort_index", "ph": "M", "ts": 7429064519947.431,
|
| 612 |
+
"pid": "Spans", "tid": 0,
|
| 613 |
+
"args": {
|
| 614 |
+
"sort_index": 536870912
|
| 615 |
+
}
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"name": "Iteration Start: PyTorch Profiler", "ph": "i", "s": "g",
|
| 619 |
+
"pid": "Traces", "tid": "Trace PyTorch Profiler", "ts": 7429064519947.431
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"name": "Record Window End", "ph": "i", "s": "g",
|
| 623 |
+
"pid": "", "tid": "", "ts": 7429064824552.924
|
| 624 |
+
}
|
| 625 |
+
],
|
| 626 |
+
"traceName": "cuda_matmul_trace.json",
|
| 627 |
+
"displayTimeUnit": "ms",
|
| 628 |
+
"baseTimeNanoseconds": 1743521598000000000
|
| 629 |
+
}
|