ECE 408 sem project

This commit is contained in:
Aryan Singh committed 2026-06-07 13:43:20 -05:00
1 parent 1d79ccd2e3
commit fa3efcaf07
139 files changed
+11634 -1

No files matched your search

+107
View File
@@ -0,0 +1,107 @@
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, -43.431740
-39.836346, -39.836452
-43.065910, -43.066021
-42.828045, -42.828144
-43.529541, -43.529644
-44.318398, -44.318497
-41.227425, -41.227528
-41.270760, -41.270863
-42.541393, -42.541523
-42.394997, -42.395119
---------------- Logits OK! (max err 0.000043, RMSE 0.000002) ----------------
_________________________ All tests passed! _________________________
Collecting data...
Generating '/tmp/nsys-report-e0c8.qdstrm'
␍[1/8] [0% ] report4.nsys-rep␍[1/8] [0% ] report4.nsys-rep␍[1/8] [====25% ] report4.nsys-rep␍[1/8] [==========49% ] report4.nsys-rep␍[1/8] [===========50% ] report4.nsys-rep␍[1/8] [=======36% ] report4.nsys-rep␍[1/8] [===========50% ] report4.nsys-rep␍[1/8] [===========51% ] report4.nsys-rep␍[1/8] [===========52% ] report4.nsys-rep␍[1/8] [===========53% ] report4.nsys-rep␍[1/8] [============54% ] report4.nsys-rep␍[1/8] [============55% ] report4.nsys-rep␍[1/8] [============56% ] report4.nsys-rep␍[1/8] [============57% ] report4.nsys-rep␍[1/8] [=============58% ] report4.nsys-rep␍[1/8] [=============59% ] report4.nsys-rep␍[1/8] [=============60% ] report4.nsys-rep␍[1/8] [===================79% ] report4.nsys-rep␍[1/8] [======================90% ] report4.nsys-rep␍[1/8] [========================100%] report4.nsys-rep␍[1/8] [========================100%] report4.nsys-rep
[2/8] [0% ] report4.sqlite [2/8] [1% ] report4.sqlite [2/8] [2% ] report4.sqlite [2/8] [3% ] report4.sqlite [2/8] [4% ] report4.sqlite [2/8] [5% ] report4.sqlite [2/8] [6% ] report4.sqlite [2/8] [7% ] report4.sqlite [2/8] [8% ] report4.sqlite [2/8] [9% ] report4.sqlite [2/8] [10% ] report4.sqlite [2/8] [11% ] report4.sqlite [2/8] [12% ] report4.sqlite [2/8] [13% ] report4.sqlite [2/8] [14% ] report4.sqlite [2/8] [=15% ] report4.sqlite [2/8] [=16% ] report4.sqlite [2/8] [=17% ] report4.sqlite [2/8] [==18% ] report4.sqlite [2/8] [==19% ] report4.sqlite [2/8] [==20% ] report4.sqlite [2/8] [==21% ] report4.sqlite [2/8] [===22% ] report4.sqlite [2/8] [===23% ] report4.sqlite [2/8] [===24% ] report4.sqlite [2/8] [====25% ] report4.sqlite [2/8] [====26% ] report4.sqlite [2/8] [====27% ] report4.sqlite [2/8] [====28% ] report4.sqlite [2/8] [=====29% ] report4.sqlite [2/8] [=====30% ] report4.sqlite [2/8] [=====31% ] report4.sqlite [2/8] [=====32% ] report4.sqlite [2/8] [======33% ] report4.sqlite [2/8] [======34% ] report4.sqlite [2/8] [======35% ] report4.sqlite [2/8] [=======36% ] report4.sqlite [2/8] [=======37% ] report4.sqlite [2/8] [=======38% ] report4.sqlite [2/8] [=======39% ] report4.sqlite [2/8] [========40% ] report4.sqlite [2/8] [========41% ] report4.sqlite [2/8] [========42% ] report4.sqlite [2/8] [=========43% ] report4.sqlite [2/8] [=========44% ] report4.sqlite [2/8] [=========45% ] report4.sqlite [2/8] [=========46% ] report4.sqlite [2/8] [==========47% ] report4.sqlite [2/8] [==========48% ] report4.sqlite [2/8] [==========49% ] report4.sqlite [2/8] [===========50% ] report4.sqlite [2/8] [===========51% ] report4.sqlite [2/8] [===========52% ] report4.sqlite [2/8] [===========53% ] report4.sqlite [2/8] [============54% ] report4.sqlite [2/8] [============55% ] report4.sqlite [2/8] [============56% ] report4.sqlite [2/8] [============57% ] report4.sqlite [2/8] [=============58% ] report4.sqlite [2/8] [=============59% ] report4.sqlite [2/8] [=============60% ] report4.sqlite [2/8] [==============61% ] report4.sqlite [2/8] [==============62% ] report4.sqlite [2/8] [==============63% ] report4.sqlite [2/8] [==============64% ] report4.sqlite [2/8] [===============65% ] report4.sqlite [2/8] [===============66% ] report4.sqlite [2/8] [===============67% ] report4.sqlite [2/8] [================68% ] report4.sqlite [2/8] [================69% ] report4.sqlite [2/8] [================70% ] report4.sqlite [2/8] [================71% ] report4.sqlite [2/8] [=================72% ] report4.sqlite [2/8] [=================73% ] report4.sqlite [2/8] [=================74% ] report4.sqlite [2/8] [==================75% ] report4.sqlite [2/8] [==================76% ] report4.sqlite [2/8] [==================77% ] report4.sqlite [2/8] [==================78% ] report4.sqlite [2/8] [===================79% ] report4.sqlite [2/8] [===================80% ] report4.sqlite [2/8] [===================81% ] report4.sqlite [2/8] [===================82% ] report4.sqlite [2/8] [====================83% ] report4.sqlite [2/8] [====================84% ] report4.sqlite [2/8] [====================85% ] report4.sqlite [2/8] [=====================86% ] report4.sqlite [2/8] [=====================87% ] report4.sqlite [2/8] [=====================88% ] report4.sqlite [2/8] [=====================89% ] report4.sqlite [2/8] [======================90% ] report4.sqlite [2/8] [======================91% ] report4.sqlite [2/8] [======================92% ] report4.sqlite [2/8] [=======================93% ] report4.sqlite [2/8] [=======================94% ] report4.sqlite [2/8] [=======================95% ] report4.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
-------- --------------- --------- ---------- ---------- -------- --------- ----------- ----------------------
63.4 596797662 16 37299853.9 11463331.5 1312 100128054 44505075.2 poll
12.3 115657312 38 3043613.5 6326.5 3507 114185851 18517552.8 fopen
12.0 112644556 591 190599.9 23554.0 1092 20946538 916359.7 ioctl
9.5 89859962 6 14976660.3 107220.5 7223 81799661 32883386.3 fread
0.9 8564355 1 8564355.0 8564355.0 8564355 8564355 0.0 pthread_cond_wait
0.8 7100598 1 7100598.0 7100598.0 7100598 7100598 0.0 pthread_cond_broadcast
0.5 4587263 32 143352.0 2204.5 1333 3746586 671302.8 fclose
0.2 2342308 25 93692.3 9026.0 6772 1633214 323364.9 mmap64
0.2 2014448 10 201444.8 31454.0 22051 884024 304027.7 write
0.1 593932 17 34937.2 8366.0 1873 379470 90751.4 mmap
0.0 372607 2 186303.5 186303.5 168375 204232 25354.7 pthread_create
0.0 320273 55 5823.1 6021.0 2334 10550 2072.2 open64
0.0 170338 1 170338.0 170338.0 170338 170338 0.0 fflush
0.0 77535 1 77535.0 77535.0 77535 77535 0.0 fgets
0.0 60311 8 7538.9 4633.5 2735 23083 6839.3 munmap
0.0 37641 6 6273.5 6372.0 3216 9387 2126.0 open
0.0 22875 3 7625.0 9138.0 3808 9929 3329.2 pipe2
0.0 17262 9 1918.0 1423.0 1022 3747 993.8 read
0.0 16491 2 8245.5 8245.5 5891 10600 3329.8 socket
0.0 10971 1 10971.0 10971.0 10971 10971 0.0 connect
0.0 6632 1 6632.0 6632.0 6632 6632 0.0 fopen64
0.0 6061 1 6061.0 6061.0 6061 6061 0.0 pthread_mutex_trylock
0.0 3176 2 1588.0 1588.0 1453 1723 190.9 fwrite
0.0 2365 1 2365.0 2365.0 2365 2365 0.0 bind
0.0 1042 1 1042.0 1042.0 1042 1042 0.0 fcntl
[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
-------- --------------- --------- ---------- ---------- -------- -------- ----------- ---------------------------------
87.1 91151498 3 30383832.7 36093633.0 21130 55036735 27948713.1 cudaMemcpy
5.1 5373556 12 447796.3 6998.0 200 1853236 632488.6 cudaFree
2.3 2415988 1 2415988.0 2415988.0 2415988 2415988 0.0 cudaFreeHost
1.7 1811777 1 1811777.0 1811777.0 1811777 1811777 0.0 cudaMallocHost
1.4 1447858 1 1447858.0 1447858.0 1447858 1447858 0.0 cudaGetDeviceProperties_v2_v12000
1.1 1146304 171 6703.5 2825.0 2715 449320 34539.6 cudaLaunchKernel
1.0 1047369 6 174561.5 124498.0 3266 462335 182577.2 cudaMalloc
0.1 147859 838 176.4 150.0 80 721 81.0 cuGetProcAddress_v2
0.0 46527 1 46527.0 46527.0 46527 46527 0.0 cudaMemset
0.0 15610 18 867.2 301.0 250 5870 1417.2 cudaEventCreateWithFlags
0.0 13797 4 3449.3 2790.5 1002 7214 2662.6 cudaDeviceSynchronize
0.0 7324 18 406.9 270.5 201 2194 468.7 cudaEventDestroy
0.0 5211 4 1302.8 1253.0 1242 1463 107.0 cuInit
0.0 1252 3 417.3 130.0 110 1012 515.1 cuModuleGetLoadingMode
0.0 862 2 431.0 431.0 281 581 212.1 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
-------- --------------- --------- -------- -------- -------- -------- ----------- ----------------------------------------------------------------------------------------------------
88.5 42371434 49 864723.1 794053.0 249378 11507721 1574324.4 matmul_forward_kernel(float *, const float *, const float *, const float *, int, int, int, int)
8.5 4090620 25 163624.8 164129.0 158689 166433 2048.2 layernorm_forward_kernel(float *, float *, float *, const float *, const float *, const float *, in…
0.6 282529 12 23544.1 23536.0 22944 23968 352.1 attention_v_kernel(float *, float *, float *, int, int, int, int)
0.6 278114 12 23176.2 23136.0 22848 23680 232.5 preattention_scores_kernel(float *, float *, int, int, int, int, float *)
0.5 226402 12 18866.8 18864.0 18784 18945 50.2 gelu_forward_kernel(float *, const float *, int)
0.5 215713 12 17976.1 17968.5 17856 18144 94.6 softmax_forward_kernel(float *, float, const float *, int, int)
0.3 145216 24 6050.7 6064.0 5632 6496 364.6 residual_forward_kernel(float *, float *, float *, int)
0.2 110113 12 9176.1 9200.0 8928 9472 142.4 permute_kernel(float *, float *, float *, const float *, int, int, int, int)
0.2 85280 1 85280.0 85280.0 85280 85280 0.0 encoder_forward_kernel(float *, const int *, const float *, const float *, int, int, int)
0.1 70369 12 5864.1 5856.0 5824 5920 27.6 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
-------- --------------- ----- ---------- ---------- -------- -------- ----------- ----------------------------
81.7 35990437 2 17995218.5 17995218.5 864 35989573 25447860.2 [CUDA memcpy Host-to-Device]
17.4 7673713 1 7673713.0 7673713.0 7673713 7673713 0.0 [CUDA memcpy Device-to-Host]
0.9 407011 1 407011.0 407011.0 407011 407011 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/report4.nsys-rep
/u/asingh41/sp26_ece408_Temperature1.5/report4.sqlite