comp files
This commit is contained in:
Aryan Singh committed
1 parent
fa3efcaf07
commit
1f77d0131d
27 files changed
+1238
-798
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, -43.363079
|
||||
-39.836346, -39.774330
|
||||
-43.065910, -43.004066
|
||||
-42.828045, -42.763325
|
||||
-43.529541, -43.461735
|
||||
-44.318398, -44.251251
|
||||
-41.227425, -41.161076
|
||||
-41.270760, -41.205547
|
||||
-42.541393, -42.472733
|
||||
-42.394997, -42.328682
|
||||
---------------- Logits OK! (max err 0.036241, RMSE 0.001194) ----------------
|
||||
_________________________ All tests passed! _________________________
|
||||
Collecting data...
|
||||
Generating '/tmp/nsys-report-8038.qdstrm'
|
||||
␍[1/8] [0% ] report10.nsys-rep␍[1/8] [0% ] report10.nsys-rep␍[1/8] [====26% ] report10.nsys-rep␍[1/8] [==========48% ] report10.nsys-rep␍[1/8] [==========49% ] report10.nsys-rep␍[1/8] [===========50% ] report10.nsys-rep␍[1/8] [=====32% ] report10.nsys-rep␍[1/8] [===========50% ] report10.nsys-rep␍[1/8] [===========51% ] report10.nsys-rep␍[1/8] [===========52% ] report10.nsys-rep␍[1/8] [=============58% ] report10.nsys-rep␍[1/8] [==================75% ] report10.nsys-rep␍[1/8] [======================91% ] report10.nsys-rep␍[1/8] [========================100%] report10.nsys-rep␍[1/8] [========================100%] report10.nsys-rep
|
||||
[2/8] [0% ] report10.sqlite
[2/8] [1% ] report10.sqlite
[2/8] [2% ] report10.sqlite
[2/8] [3% ] report10.sqlite
[2/8] [4% ] report10.sqlite
[2/8] [5% ] report10.sqlite
[2/8] [6% ] report10.sqlite
[2/8] [7% ] report10.sqlite
[2/8] [8% ] report10.sqlite
[2/8] [9% ] report10.sqlite
[2/8] [10% ] report10.sqlite
[2/8] [11% ] report10.sqlite
[2/8] [12% ] report10.sqlite
[2/8] [13% ] report10.sqlite
[2/8] [14% ] report10.sqlite
[2/8] [=15% ] report10.sqlite
[2/8] [=16% ] report10.sqlite
[2/8] [=17% ] report10.sqlite
[2/8] [==18% ] report10.sqlite
[2/8] [==19% ] report10.sqlite
[2/8] [==20% ] report10.sqlite
[2/8] [==21% ] report10.sqlite
[2/8] [===22% ] report10.sqlite
[2/8] [===23% ] report10.sqlite
[2/8] [===24% ] report10.sqlite
[2/8] [====25% ] report10.sqlite
[2/8] [====26% ] report10.sqlite
[2/8] [====27% ] report10.sqlite
[2/8] [====28% ] report10.sqlite
[2/8] [=====29% ] report10.sqlite
[2/8] [=====30% ] report10.sqlite
[2/8] [=====31% ] report10.sqlite
[2/8] [=====32% ] report10.sqlite
[2/8] [======33% ] report10.sqlite
[2/8] [======34% ] report10.sqlite
[2/8] [======35% ] report10.sqlite
[2/8] [=======36% ] report10.sqlite
[2/8] [=======37% ] report10.sqlite
[2/8] [=======38% ] report10.sqlite
[2/8] [=======39% ] report10.sqlite
[2/8] [========40% ] report10.sqlite
[2/8] [========41% ] report10.sqlite
[2/8] [========42% ] report10.sqlite
[2/8] [=========43% ] report10.sqlite
[2/8] [=========44% ] report10.sqlite
[2/8] [=========45% ] report10.sqlite
[2/8] [=========46% ] report10.sqlite
[2/8] [==========47% ] report10.sqlite
[2/8] [==========48% ] report10.sqlite
[2/8] [==========49% ] report10.sqlite
[2/8] [===========50% ] report10.sqlite
[2/8] [===========51% ] report10.sqlite
[2/8] [===========52% ] report10.sqlite
[2/8] [===========53% ] report10.sqlite
[2/8] [============54% ] report10.sqlite
[2/8] [============55% ] report10.sqlite
[2/8] [============56% ] report10.sqlite
[2/8] [============57% ] report10.sqlite
[2/8] [=============58% ] report10.sqlite
[2/8] [=============59% ] report10.sqlite
[2/8] [=============60% ] report10.sqlite
[2/8] [==============61% ] report10.sqlite
[2/8] [==============62% ] report10.sqlite
[2/8] [==============63% ] report10.sqlite
[2/8] [==============64% ] report10.sqlite
[2/8] [===============65% ] report10.sqlite
[2/8] [===============66% ] report10.sqlite
[2/8] [===============67% ] report10.sqlite
[2/8] [================68% ] report10.sqlite
[2/8] [================69% ] report10.sqlite
[2/8] [================70% ] report10.sqlite
[2/8] [================71% ] report10.sqlite
[2/8] [=================72% ] report10.sqlite
[2/8] [=================73% ] report10.sqlite
[2/8] [=================74% ] report10.sqlite
[2/8] [==================75% ] report10.sqlite
[2/8] [==================76% ] report10.sqlite
[2/8] [==================77% ] report10.sqlite
[2/8] [==================78% ] report10.sqlite
[2/8] [===================79% ] report10.sqlite
[2/8] [===================80% ] report10.sqlite
[2/8] [===================81% ] report10.sqlite
[2/8] [===================82% ] report10.sqlite
[2/8] [====================83% ] report10.sqlite
[2/8] [====================84% ] report10.sqlite
[2/8] [====================85% ] report10.sqlite
[2/8] [=====================86% ] report10.sqlite
[2/8] [=====================87% ] report10.sqlite
[2/8] [=====================88% ] report10.sqlite
[2/8] [=====================89% ] report10.sqlite
[2/8] [======================90% ] report10.sqlite
[2/8] [======================91% ] report10.sqlite
[2/8] [======================92% ] report10.sqlite
[2/8] [=======================93% ] report10.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
|
||||
-------- --------------- --------- ----------- ---------- -------- --------- ----------- ----------------------
|
||||
56.0 996338383 20 49816919.1 37144026.0 1233 100147022 47273324.4 poll
|
||||
34.2 608280586 6 101380097.7 818540.0 7524 540651394 216790051.0 fread
|
||||
7.5 133295089 592 225160.6 22788.0 1002 20761770 945697.8 ioctl
|
||||
1.3 22334199 38 587742.1 6277.0 3637 21155518 3428230.9 fopen
|
||||
0.4 7061795 1 7061795.0 7061795.0 7061795 7061795 0.0 pthread_cond_broadcast
|
||||
0.4 6959863 1 6959863.0 6959863.0 6959863 6959863 0.0 pthread_cond_wait
|
||||
0.1 2388902 25 95556.1 9268.0 6172 1630658 322960.7 mmap64
|
||||
0.1 1595422 10 159542.2 32852.0 22562 552156 214109.1 write
|
||||
0.0 738626 32 23082.1 2129.0 1352 379712 82280.0 fclose
|
||||
0.0 572887 17 33699.2 5110.0 1984 378079 90516.0 mmap
|
||||
0.0 359054 2 179527.0 179527.0 164950 194104 20615.0 pthread_create
|
||||
0.0 331892 55 6034.4 5961.0 2354 12935 2361.7 open64
|
||||
0.0 76394 1 76394.0 76394.0 76394 76394 0.0 fgets
|
||||
0.0 65062 1 65062.0 65062.0 65062 65062 0.0 fflush
|
||||
0.0 56836 9 6315.1 4839.0 2995 18494 4843.8 munmap
|
||||
0.0 36999 6 6166.5 6011.5 3125 9508 2180.5 open
|
||||
0.0 20970 3 6990.0 8316.0 3286 9368 3250.6 pipe2
|
||||
0.0 18394 10 1839.4 1353.0 1042 3456 943.7 read
|
||||
0.0 16671 2 8335.5 8335.5 5571 11100 3909.6 socket
|
||||
0.0 8135 1 8135.0 8135.0 8135 8135 0.0 connect
|
||||
0.0 6442 1 6442.0 6442.0 6442 6442 0.0 fopen64
|
||||
0.0 3367 2 1683.5 1683.5 1423 1944 368.4 fwrite
|
||||
0.0 1984 1 1984.0 1984.0 1984 1984 0.0 bind
|
||||
0.0 1182 1 1182.0 1182.0 1182 1182 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
|
||||
-------- --------------- --------- ---------- ---------- -------- -------- ----------- ---------------------------------
|
||||
80.8 58664668 3 19554889.3 21881269.0 17532 36765867 18484292.3 cudaMemcpy
|
||||
8.6 6233820 12 519485.0 7173.0 190 2255630 790172.6 cudaFree
|
||||
3.4 2475383 1 2475383.0 2475383.0 2475383 2475383 0.0 cudaFreeHost
|
||||
2.5 1802090 1 1802090.0 1802090.0 1802090 1802090 0.0 cudaGetDeviceProperties_v2_v12000
|
||||
1.8 1334142 1 1334142.0 1334142.0 1334142 1334142 0.0 cudaMallocHost
|
||||
1.4 1040310 171 6083.7 2826.0 2675 390603 30047.0 cudaLaunchKernel
|
||||
1.1 769711 6 128285.2 124803.5 5030 312856 117219.1 cudaMalloc
|
||||
0.2 153352 838 183.0 150.0 90 762 90.7 cuGetProcAddress_v2
|
||||
0.1 41117 1 41117.0 41117.0 41117 41117 0.0 cudaMemset
|
||||
0.0 20561 18 1142.3 270.5 250 5761 1731.3 cudaEventCreateWithFlags
|
||||
0.0 13585 4 3396.3 2785.0 1002 7013 2566.9 cudaDeviceSynchronize
|
||||
0.0 8154 18 453.0 296.0 230 2504 527.8 cudaEventDestroy
|
||||
0.0 5150 4 1287.5 1227.5 1082 1613 233.6 cuInit
|
||||
0.0 1423 3 474.3 170.0 91 1162 596.8 cuModuleGetLoadingMode
|
||||
0.0 801 2 400.5 400.5 260 541 198.7 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
|
||||
-------- --------------- --------- -------- -------- -------- -------- ----------- ----------------------------------------------------------------------------------------------------
|
||||
90.6 13694036 49 279470.1 217344.0 97088 3138461 429723.5 matmul_forward_kernel(float *, const float *, const float *, const float *, int, int, int, int)
|
||||
2.0 302144 12 25178.7 25184.0 24960 25248 72.0 gelu_forward_kernel(float *, const float *, int)
|
||||
1.3 189280 24 7886.7 7888.0 7488 8288 357.1 residual_forward_kernel(float *, float *, float *, int)
|
||||
1.2 186336 12 15528.0 15536.0 15392 15744 101.3 softmax_forward_kernel(float *, float, const float *, int, int)
|
||||
1.1 170272 25 6810.9 6528.0 6112 7904 602.7 layernorm_forward_kernel(float *, float *, float *, const float *, const float *, const float *, in…
|
||||
0.8 126272 12 10522.7 10528.0 10400 10656 75.8 preattention_scores_kernel(float *, float *, int, int, int, int, float *)
|
||||
0.8 125728 12 10477.3 10464.0 10368 10688 98.8 attention_v_kernel(float *, float *, float *, int, int, int, int)
|
||||
0.8 119232 12 9936.0 9920.0 9760 10112 144.7 permute_kernel(float *, float *, float *, const float *, int, int, int, int)
|
||||
0.7 113183 1 113183.0 113183.0 113183 113183 0.0 encoder_forward_kernel(float *, const int *, const float *, const float *, int, int, int)
|
||||
0.6 94496 12 7874.7 7872.0 7840 7936 21.4 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
|
||||
-------- --------------- ----- ---------- ---------- -------- -------- ----------- ----------------------------
|
||||
82.9 36637635 2 18318817.5 18318817.5 1088 36636547 25905181.5 [CUDA memcpy Host-to-Device]
|
||||
16.2 7173307 1 7173307.0 7173307.0 7173307 7173307 0.0 [CUDA memcpy Device-to-Host]
|
||||
0.9 409248 1 409248.0 409248.0 409248 409248 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/report10.nsys-rep
|
||||
/u/asingh41/sp26_ece408_Temperature1.5/report10.sqlite
|
||||
Reference in new issue
Block a user