ECE 408 sem project
This commit is contained in:
Aryan Singh committed
1 parent
1d79ccd2e3
commit
fa3efcaf07
139 files changed
+11634
-1
No files matched your search
@@ -0,0 +1,106 @@
|
||||
test-gpt2-kernels - TF32 Mode is turned OFF.
|
||||
test-gpt2 - [State]
|
||||
Batch Size: 4
|
||||
Sequence Length: 64
|
||||
GPT2 - Allocated 221 MiB for activations
|
||||
test-gpt2 - Printing 10 Expected vs Actual Logits:
|
||||
-43.431618, -43.431660
|
||||
-39.836346, -39.836384
|
||||
-43.065910, -43.065952
|
||||
-42.828045, -42.828083
|
||||
-43.529541, -43.529587
|
||||
-44.318398, -44.318439
|
||||
-41.227425, -41.227470
|
||||
-41.270760, -41.270802
|
||||
-42.541393, -42.541447
|
||||
-42.394997, -42.395035
|
||||
---------------- Logits OK! (max err 0.000088, RMSE 0.000002) ----------------
|
||||
_________________________ All tests passed! _________________________
|
||||
Collecting data...
|
||||
Generating '/tmp/nsys-report-b36a.qdstrm'
|
||||
␍[1/8] [0% ] report5.nsys-rep␍[1/8] [0% ] report5.nsys-rep␍[1/8] [====25% ] report5.nsys-rep␍[1/8] [=========46% ] report5.nsys-rep␍[1/8] [==========48% ] report5.nsys-rep␍[1/8] [===========50% ] report5.nsys-rep␍[1/8] [===========51% ] report5.nsys-rep␍[1/8] [======34% ] report5.nsys-rep␍[1/8] [===========50% ] report5.nsys-rep␍[1/8] [===========51% ] report5.nsys-rep␍[1/8] [===========52% ] report5.nsys-rep␍[1/8] [===========53% ] report5.nsys-rep␍[1/8] [============54% ] report5.nsys-rep␍[1/8] [============55% ] report5.nsys-rep␍[1/8] [============56% ] report5.nsys-rep␍[1/8] [============57% ] report5.nsys-rep␍[1/8] [=============58% ] report5.nsys-rep␍[1/8] [=============59% ] report5.nsys-rep␍[1/8] [==================77% ] report5.nsys-rep␍[1/8] [======================91% ] report5.nsys-rep␍[1/8] [========================100%] report5.nsys-rep␍[1/8] [========================100%] report5.nsys-rep
|
||||
[2/8] [0% ] report5.sqlite
[2/8] [1% ] report5.sqlite
[2/8] [2% ] report5.sqlite
[2/8] [3% ] report5.sqlite
[2/8] [4% ] report5.sqlite
[2/8] [5% ] report5.sqlite
[2/8] [6% ] report5.sqlite
[2/8] [7% ] report5.sqlite
[2/8] [8% ] report5.sqlite
[2/8] [9% ] report5.sqlite
[2/8] [10% ] report5.sqlite
[2/8] [11% ] report5.sqlite
[2/8] [12% ] report5.sqlite
[2/8] [13% ] report5.sqlite
[2/8] [14% ] report5.sqlite
[2/8] [=15% ] report5.sqlite
[2/8] [=16% ] report5.sqlite
[2/8] [=17% ] report5.sqlite
[2/8] [==18% ] report5.sqlite
[2/8] [==19% ] report5.sqlite
[2/8] [==20% ] report5.sqlite
[2/8] [==21% ] report5.sqlite
[2/8] [===22% ] report5.sqlite
[2/8] [===23% ] report5.sqlite
[2/8] [===24% ] report5.sqlite
[2/8] [====25% ] report5.sqlite
[2/8] [====26% ] report5.sqlite
[2/8] [====27% ] report5.sqlite
[2/8] [====28% ] report5.sqlite
[2/8] [=====29% ] report5.sqlite
[2/8] [=====30% ] report5.sqlite
[2/8] [=====31% ] report5.sqlite
[2/8] [=====32% ] report5.sqlite
[2/8] [======33% ] report5.sqlite
[2/8] [======34% ] report5.sqlite
[2/8] [======35% ] report5.sqlite
[2/8] [=======36% ] report5.sqlite
[2/8] [=======37% ] report5.sqlite
[2/8] [=======38% ] report5.sqlite
[2/8] [=======39% ] report5.sqlite
[2/8] [========40% ] report5.sqlite
[2/8] [========41% ] report5.sqlite
[2/8] [========42% ] report5.sqlite
[2/8] [=========43% ] report5.sqlite
[2/8] [=========44% ] report5.sqlite
[2/8] [=========45% ] report5.sqlite
[2/8] [=========46% ] report5.sqlite
[2/8] [==========47% ] report5.sqlite
[2/8] [==========48% ] report5.sqlite
[2/8] [==========49% ] report5.sqlite
[2/8] [===========50% ] report5.sqlite
[2/8] [===========51% ] report5.sqlite
[2/8] [===========52% ] report5.sqlite
[2/8] [===========53% ] report5.sqlite
[2/8] [============54% ] report5.sqlite
[2/8] [============55% ] report5.sqlite
[2/8] [============56% ] report5.sqlite
[2/8] [============57% ] report5.sqlite
[2/8] [=============58% ] report5.sqlite
[2/8] [=============59% ] report5.sqlite
[2/8] [=============60% ] report5.sqlite
[2/8] [==============61% ] report5.sqlite
[2/8] [==============62% ] report5.sqlite
[2/8] [==============63% ] report5.sqlite
[2/8] [==============64% ] report5.sqlite
[2/8] [===============65% ] report5.sqlite
[2/8] [===============66% ] report5.sqlite
[2/8] [===============67% ] report5.sqlite
[2/8] [================68% ] report5.sqlite
[2/8] [================69% ] report5.sqlite
[2/8] [================70% ] report5.sqlite
[2/8] [================71% ] report5.sqlite
[2/8] [=================72% ] report5.sqlite
[2/8] [=================73% ] report5.sqlite
[2/8] [=================74% ] report5.sqlite
[2/8] [==================75% ] report5.sqlite
[2/8] [==================76% ] report5.sqlite
[2/8] [==================77% ] report5.sqlite
[2/8] [==================78% ] report5.sqlite
[2/8] [===================79% ] report5.sqlite
[2/8] [===================80% ] report5.sqlite
[2/8] [===================81% ] report5.sqlite
[2/8] [===================82% ] report5.sqlite
[2/8] [====================83% ] report5.sqlite
[2/8] [====================84% ] report5.sqlite
[2/8] [====================85% ] report5.sqlite
[2/8] [=====================86% ] report5.sqlite
[2/8] [=====================87% ] report5.sqlite
[2/8] [=====================88% ] report5.sqlite
[2/8] [=====================89% ] report5.sqlite
[2/8] [======================90% ] report5.sqlite
[2/8] [======================91% ] report5.sqlite
[2/8] [======================92% ] report5.sqlite
[2/8] [=======================93% ] report5.sqlite
[2/8] [=======================94% ] report5.sqlite
[2/8] [=======================95% ] report5.sqlite
[2/8] Line truncated
|
||||
[3/8] Executing 'nvtx_sum' stats report
|
||||
[4/8] Executing 'osrt_sum' stats report
|
||||
|
||||
Time (%) Total Time (ns) Num Calls Avg (ns) Med (ns) Min (ns) Max (ns) StdDev (ns) Name
|
||||
-------- --------------- --------- ---------- --------- -------- --------- ----------- ----------------------
|
||||
72.8 691449484 17 40673499.1 9775260.0 1583 100131438 45632127.3 poll
|
||||
12.5 118880952 591 201152.2 23013.0 1032 24864504 1071706.0 ioctl
|
||||
10.3 97973444 6 16328907.3 63448.5 8215 89350283 35931749.3 fread
|
||||
1.5 14522344 38 382166.9 6252.0 3516 8002510 1594225.9 fopen
|
||||
0.9 8531381 1 8531381.0 8531381.0 8531381 8531381 0.0 pthread_cond_wait
|
||||
0.7 7052981 1 7052981.0 7052981.0 7052981 7052981 0.0 pthread_cond_broadcast
|
||||
0.5 4782327 32 149447.7 2219.0 1342 4255792 753563.6 fclose
|
||||
0.2 2370702 25 94828.1 8777.0 5971 1658446 328407.0 mmap64
|
||||
0.2 2034120 11 184920.0 25638.0 1142 842979 290924.4 write
|
||||
0.1 633095 17 37240.9 4338.0 1904 388908 93374.4 mmap
|
||||
0.0 365446 2 182723.0 182723.0 177223 188223 7778.2 pthread_create
|
||||
0.0 330916 55 6016.7 5962.0 2335 10970 2241.2 open64
|
||||
0.0 86322 1 86322.0 86322.0 86322 86322 0.0 fgets
|
||||
0.0 58760 1 58760.0 58760.0 58760 58760 0.0 fflush
|
||||
0.0 41479 8 5184.9 4508.5 2685 11542 2877.0 munmap
|
||||
0.0 40978 6 6829.7 6848.0 3146 9859 2589.1 open
|
||||
0.0 23634 3 7878.0 9628.0 4248 9758 3144.3 pipe2
|
||||
0.0 18855 2 9427.5 9427.5 7614 11241 2564.7 socket
|
||||
0.0 18135 9 2015.0 1904.0 1012 3897 1071.0 read
|
||||
0.0 9648 1 9648.0 9648.0 9648 9648 0.0 connect
|
||||
0.0 6322 1 6322.0 6322.0 6322 6322 0.0 fopen64
|
||||
0.0 3285 2 1642.5 1642.5 1372 1913 382.5 fwrite
|
||||
0.0 2184 1 2184.0 2184.0 2184 2184 0.0 bind
|
||||
0.0 1042 1 1042.0 1042.0 1042 1042 0.0 listen
|
||||
|
||||
[5/8] Executing 'cuda_api_sum' stats report
|
||||
|
||||
Time (%) Total Time (ns) Num Calls Avg (ns) Med (ns) Min (ns) Max (ns) StdDev (ns) Name
|
||||
-------- --------------- --------- ---------- ---------- -------- --------- ----------- ---------------------------------
|
||||
95.3 290852270 3 96950756.7 37077310.0 19757 253755203 137054884.4 cudaMemcpy
|
||||
2.0 6229529 12 519127.4 5315.0 191 2251928 798535.2 cudaFree
|
||||
0.8 2467101 1 2467101.0 2467101.0 2467101 2467101 0.0 cudaFreeHost
|
||||
0.6 1704733 1 1704733.0 1704733.0 1704733 1704733 0.0 cudaGetDeviceProperties_v2_v12000
|
||||
0.5 1553469 1 1553469.0 1553469.0 1553469 1553469 0.0 cudaMallocHost
|
||||
0.3 1039319 171 6077.9 2826.0 2705 389279 29892.9 cudaLaunchKernel
|
||||
0.3 938599 6 156433.2 140363.0 3266 449833 164344.2 cudaMalloc
|
||||
0.1 173923 838 207.5 181.0 90 4588 181.4 cuGetProcAddress_v2
|
||||
0.0 42730 1 42730.0 42730.0 42730 42730 0.0 cudaMemset
|
||||
0.0 13035 4 3258.8 3111.0 1072 5741 1916.5 cudaDeviceSynchronize
|
||||
0.0 12044 18 669.1 291.0 260 2676 789.3 cudaEventCreateWithFlags
|
||||
0.0 6834 18 379.7 270.5 210 1814 374.7 cudaEventDestroy
|
||||
0.0 5199 4 1299.8 1317.5 1202 1362 69.6 cuInit
|
||||
0.0 1433 3 477.7 120.0 110 1203 628.2 cuModuleGetLoadingMode
|
||||
0.0 1233 2 616.5 616.5 481 752 191.6 cudaGetDriverEntryPoint_v11030
|
||||
|
||||
[6/8] Executing 'cuda_gpu_kern_sum' stats report
|
||||
|
||||
Time (%) Total Time (ns) Instances Avg (ns) Med (ns) Min (ns) Max (ns) StdDev (ns) Name
|
||||
-------- --------------- --------- --------- --------- -------- -------- ----------- ----------------------------------------------------------------------------------------------------
|
||||
99.0 243425601 49 4967869.4 4539787.0 1227363 73783097 10137706.5 matmul_forward_kernel(float *, const float *, const float *, const float *, int, int, int, int)
|
||||
0.5 1135204 12 94600.3 94560.5 94273 95360 268.7 preattention_scores_kernel(float *, float *, int, int, int, int, float *)
|
||||
0.1 345442 12 28786.8 28752.5 28224 29696 465.2 attention_v_kernel(float *, float *, float *, int, int, int, int)
|
||||
0.1 226976 12 18914.7 18912.0 18688 19072 88.9 gelu_forward_kernel(float *, const float *, int)
|
||||
0.1 146081 24 6086.7 6112.0 5664 6625 369.0 residual_forward_kernel(float *, float *, float *, int)
|
||||
0.1 141856 12 11821.3 11808.0 11712 11968 88.9 softmax_forward_kernel(float *, float, const float *, int, int)
|
||||
0.1 131456 25 5258.2 5120.0 4704 6016 438.5 layernorm_forward_kernel(float *, float *, float *, const float *, const float *, const float *, in…
|
||||
0.0 117888 12 9824.0 9824.0 9664 10144 133.7 permute_kernel(float *, float *, float *, const float *, int, int, int, int)
|
||||
0.0 90304 1 90304.0 90304.0 90304 90304 0.0 encoder_forward_kernel(float *, const int *, const float *, const float *, int, int, int)
|
||||
0.0 70368 12 5864.0 5856.0 5792 5920 33.8 unpermute_kernel(float *, float *, int, int, int, int)
|
||||
|
||||
[7/8] Executing 'cuda_gpu_mem_time_sum' stats report
|
||||
|
||||
Time (%) Total Time (ns) Count Avg (ns) Med (ns) Min (ns) Max (ns) StdDev (ns) Operation
|
||||
-------- --------------- ----- ---------- ---------- -------- -------- ----------- ----------------------------
|
||||
80.8 36866268 2 18433134.0 18433134.0 896 36865372 26067121.0 [CUDA memcpy Host-to-Device]
|
||||
18.3 8354069 1 8354069.0 8354069.0 8354069 8354069 0.0 [CUDA memcpy Device-to-Host]
|
||||
0.9 406945 1 406945.0 406945.0 406945 406945 0.0 [CUDA memset]
|
||||
|
||||
[8/8] Executing 'cuda_gpu_mem_size_sum' stats report
|
||||
|
||||
Total (MB) Count Avg (MB) Med (MB) Min (MB) Max (MB) StdDev (MB) Operation
|
||||
---------- ----- -------- -------- -------- -------- ----------- ----------------------------
|
||||
497.905 2 248.952 248.952 0.001 497.904 352.070 [CUDA memcpy Host-to-Device]
|
||||
232.392 1 232.392 232.392 232.392 232.392 0.000 [CUDA memset]
|
||||
51.511 1 51.511 51.511 51.511 51.511 0.000 [CUDA memcpy Device-to-Host]
|
||||
|
||||
Generated:
|
||||
/u/asingh41/sp26_ece408_Temperature1.5/report5.nsys-rep
|
||||
/u/asingh41/sp26_ece408_Temperature1.5/report5.sqlite
|
||||
Reference in new issue
Block a user