boryiingsu / clspmv Goto Github PK
View Code? Open in Web Editor NEWThe repository of the clSpMV code.
clspmv's People
Contributors
Stargazers
Watchers
clspmv's Issues
cocktail_example error
I use cuda-7.0 and I edit the makefile according to my system configuration.
When I run the benchmarking, I get the error message at the end.
------------------------------------------------------------------------
COO Dim 1024 BN 4 GN 15 opttime 0.018090 ms optflop 0.452847 optmethod 0
------------------------------------------------------------------------
Program File Name: /home/aoyulong/spmv/clSpMV/clSpMV_NV/kernels/spmv_coo.cl
---
>>> Build Log:
<<< End of Build Log
process size 32 nnz 4096 gsize 4352 active_warp 129
Error at line 156 code -9999 message Unknown
Then I run the cocktail_example, I get the following messages:
MatInfo: Width 121192 Height 121192 NNZ 2624331
Max Offset 121040 Min Offset -121040
Histogram: <-15: 1021338 -15~-1 240906 < 0-15 340749 > 16 1021338
Cacheline usage per row: max 40 min 0 avg 9.237219
SpMV serial COO time 0.007290 s
Read Overhead
filename /home/aoyulong/spmv/clSpMV/clSpMV_NV/benchmark/overhead.ben
filename /home/aoyulong/spmv/clSpMV/clSpMV_NV/benchmark/overhead.ben
42 Entries read
Read bdia
240 Entries read
Read dia
240 Entries read
Read sbell
1112 Entries read
Read bell
1112 Entries read
Read bcsr
576 Entries read
Read sell
93 Entries read
Read ell
93 Entries read
Read csr
588 Entries read
Read coo
13 Entries read
bdia Max method 1 value 68.679888
dia Max method 2 value 69.579421
sbell Max method 0 value 53.363765
sbell Max method 0 value 63.799897
sbell Max method 0 value 64.794302
sbell Max method 0 value 66.808057
sbell Max method 0 value 59.341946
sbell Max method 0 value 69.086807
sbell Max method 0 value 70.590549
sbell Max method 0 value 66.819063
bell Max method 0 value 53.405211
bell Max method 1 value 65.800716
bell Max method 0 value 65.175433
bell Max method 0 value 66.559819
bell Max method 0 value 60.487999
bell Max method 1 value 71.004429
bell Max method 0 value 70.273990
bell Max method 0 value 66.281354
bcsr Max method 0 value 3.900219
bcsr Max method 0 value 3.707462
bcsr Max method 0 value 5.044556
bcsr Max method 0 value 3.548505
bcsr Max method 0 value 2.829009
bcsr Max method 0 value 3.394457
bcsr Max method 0 value 4.113381
bcsr Max method 0 value 3.896984
sell Max method 0 value 37.735076
ell Max method 0 value 38.411910
csr Max method 0 gnum 180 value 11.540831
coo Max method 0 gnum 15 value 0.452847
dia flop 69.579421 block flop 71.004429 flat flop 38.411910
---------------------------------------------
Extract Blocks
---------------------------------------------
Matrix nnz 2624331
sbell Max method 0 value 53.363765
sbell Max method 0 value 63.799897
sbell Max method 0 value 64.794302
sbell Max method 0 value 66.808057
sbell Max method 0 value 59.341946
sbell Max method 0 value 69.086807
sbell Max method 0 value 70.590549
sbell Max method 0 value 66.819063
bell Max method 0 value 53.405211
bell Max method 1 value 65.800716
bell Max method 0 value 65.175433
bell Max method 0 value 66.559819
bell Max method 0 value 60.487999
bell Max method 1 value 71.004429
bell Max method 0 value 70.273990
bell Max method 0 value 66.281354
bcsr Max method 0 value 3.900219
bcsr Max method 0 value 3.707462
bcsr Max method 0 value 5.044556
bcsr Max method 0 value 3.548505
bcsr Max method 0 value 2.829009
bcsr Max method 0 value 3.394457
bcsr Max method 0 value 4.113381
bcsr Max method 0 value 3.896984
sell Max method 0 value 37.735076
ell Max method 0 value 38.411910
csr Max method 0 gnum 180 value 11.540831
coo Max method 0 gnum 15 value 0.452847
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 1816670
sell Max method 0 value 36.249341
ell Max method 0 value 36.694613
csr Max method 0 gnum 180 value 7.928531
coo Max method 0 gnum 15 value 0.452847
bestnnz 1629411 tmpnnz 1629411
slice partial nnz 1816197 full nnz 1816670
Block size (1, 4) full block num 1377747 nnz 2624331 partial num 211620 nnz 807661 full time 0.109473 partial time 0.108955 block over 0.006281 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 1987756
sell Max method 0 value 36.813969
ell Max method 0 value 37.304945
csr Max method 0 gnum 180 value 8.444574
coo Max method 0 gnum 15 value 0.452847
bestnnz 1671646 tmpnnz 1671646
slice partial nnz 1987252 full nnz 1987756
Block size (2, 4) full block num 847603 nnz 2624331 partial num 85128 nnz 636575 full time 0.108414 partial time 0.104752 block over 0.005363 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2193413
sell Max method 0 value 37.121005
ell Max method 0 value 37.673933
csr Max method 0 gnum 180 value 9.476659
coo Max method 0 gnum 15 value 0.452847
bestnnz 1852739 tmpnnz 1852739
slice partial nnz 2192835 full nnz 2193413
Block size (4, 4) full block num 698452 nnz 2624331 partial num 30097 nnz 430918 full time 0.176119 partial time 0.102659 block over 0.004656 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2370929
sell Max method 0 value 37.428040
ell Max method 0 value 38.042922
csr Max method 0 gnum 180 value 10.508745
coo Max method 0 gnum 15 value 0.452847
bestnnz 1958355 tmpnnz 1958355
slice partial nnz 2370381 full nnz 2370929
Block size (8, 4) full block num 611372 nnz 2624331 partial num 10323 nnz 253402 full time 0.296667 partial time 0.103412 block over 0.003830 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2174079
sell Max method 0 value 37.121005
ell Max method 0 value 37.673933
csr Max method 0 gnum 180 value 9.476659
coo Max method 0 gnum 15 value 0.452847
bestnnz 1910967 tmpnnz 1910967
slice partial nnz 2172271 full nnz 2174079
Block size (1, 8) full block num 1222760 nnz 2624331 partial num 70098 nnz 450252 full time 0.168000 partial time 0.109075 block over 0.006281 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2228571
sell Max method 0 value 37.121005
ell Max method 0 value 37.673933
csr Max method 0 gnum 180 value 9.476659
coo Max method 0 gnum 15 value 0.452847
bestnnz 1889902 tmpnnz 1889902
slice partial nnz 2226931 full nnz 2228571
Block size (2, 8) full block num 746709 nnz 2624331 partial num 32172 nnz 395760 full time 0.173625 partial time 0.104961 block over 0.005363 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2347727
sell Max method 0 value 37.274523
ell Max method 0 value 37.858428
csr Max method 0 gnum 180 value 9.992702
coo Max method 0 gnum 15 value 0.452847
bestnnz 1975867 tmpnnz 1975867
slice partial nnz 2347171 full nnz 2347727
Block size (4, 8) full block num 611372 nnz 2624331 partial num 11612 nnz 276604 full time 0.281802 partial time 0.102032 block over 0.004656 flat over 0.004946
---------------------------------------------
Estimate Flat Time
---------------------------------------------
Matrix nnz 2364479
sell Max method 0 value 37.428040
ell Max method 0 value 38.042922
csr Max method 0 gnum 180 value 10.508745
coo Max method 0 gnum 15 value 0.452847
bestnnz 1936616 tmpnnz 1936616
slice partial nnz 2363891 full nnz 2364479
Block size (8, 8) full block num 530327 nnz 2624331 partial num 5901 nnz 259852 full time 0.511783 partial time 0.104002 block over 0.003830 flat over 0.004946
No block extracted
Block category analysis time 1.226713 s
---------------------------------------------
Extract Diagonals
---------------------------------------------
Matrix nnz 2624331
bdia Max method 1 value 68.679888
dia Max method 2 value 69.579421
sell Max method 0 value 37.735076
ell Max method 0 value 38.411910
csr Max method 0 gnum 180 value 11.540831
coo Max method 0 gnum 15 value 0.452847
max flat 38.411910 max dia 69.579421 overhead 0.004946 cur nnz 251737 total nnz 2624331
Extract 0 diagonals
Dia category analysis time 0.005381 s
---------------------------------------------
Extract Flat Elements
---------------------------------------------
Matrix nnz 2624331
sell Max method 0 value 37.735076
ell Max method 0 value 38.411910
csr Max method 0 gnum 180 value 11.540831
coo Max method 0 gnum 15 value 0.452847
---------------------------------------------
Choose between sell and ell
---------------------------------------------
Matrix nnz 2624331
bestnnz 2171679 tmpnnz 2171679
slice partial nnz 2623613 full nnz 2624331
---------------------------------------------
Extract sell
---------------------------------------------
Matrix nnz 2624331
---------------------------------------------
Choose between csr and coo
---------------------------------------------
Matrix nnz 0
No csr or coo extracted
Flat category analysis time 0.024781 s
*****************************
Cocktail matrix info
*****************************
Original Matrix Width 121192 Height 121192 nnz 2624331
Use sell
Implementation no. 0
NNZ 2624331
slice height 32
------------------------------
Program File Name: /home/aoyulong/spmv/clSpMV/clSpMV_NV/kernels/kernel_all.cl
---
Error: clBuildProgram() returned -11.
>>> Build Log:
In file included from <kernel>:10:
./spmv_csr_scalar.cl:113:101: error: illegal address space on image or sampler type
__kernel void gpu_csr_sc_pm_u4_tx(__global int* rowptr, __global int* colid, __global float* data, __global __read_only image2d_t vec, __global float* result, int row_num)
^
<<< End of Build Log
Error at line 710 code -45 message Invalid program executable
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.