Commits · master · Christoph Klein / srcsrmv

Dec 12, 2023
- Remove test/csr-convert.cu · 42aa2da7
  Ferdinand-Joseph Vanmaele authored 1 year ago
```
Disabled test starting from commit c402d011
```
  42aa2da7
Sep 05, 2023
- Fix missing includes · 492e4803
  Paul Große-Bley authored 1 year ago
  
  492e4803
Nov 04, 2022
- Add config init · b94d5af2
  dontspoilerme authored 2 years ago
  
  b94d5af2
- Add csrmv balanced without atomics · 16d207e9
  dontspoilerme authored 2 years ago
  
  16d207e9
Jun 14, 2022
- Add work-balanced scheme in benchmark · 021dead7
  dontspoilerme authored 3 years ago
  
  021dead7
- Fix: move config init functions to separate header · 21bfafcf
  dontspoilerme authored 3 years ago
  
  21bfafcf
Mar 29, 2022
- Add autotuning for srcsrmv spmv in bench-find-max2 · fbb54bd2
  dontspoilerme authored 3 years ago
  
  fbb54bd2
Mar 19, 2022
- add make_row_block_starts · 32046180
  dontspoilerme authored 3 years ago
  
  32046180
Mar 15, 2022
- Fix: start profiling after warmup · eb2bcdef
  dontspoilerme authored 3 years ago
  
  eb2bcdef
- Add find-max-with-col and max1 to bench-find-max1 · ffd4f260
  dontspoilerme authored 3 years ago
  
  ffd4f260
Mar 14, 2022
- Fix: config serialization for profiling · 4636e3f4
  dontspoilerme authored 3 years ago
  
  4636e3f4
Mar 13, 2022
- update bench-find-max2 for paper · 19d6314c
  dontspoilerme authored 3 years ago
  
  19d6314c
Mar 12, 2022
- update apps for hand-shaking paper · b32e8e1d
  dontspoilerme authored 3 years ago
  
  b32e8e1d
- Update CMakeLists of apps · 97a9a0e7
  dontspoilerme authored 3 years ago
  
  97a9a0e7
Feb 10, 2022
- Fix: stream.id() -> stream.handle() due to cuda-api-wrappers upgrade · 445621dc
  dontspoilerme authored 3 years ago
  
  445621dc
- Merge branch '12-incompatibility-with-cuda-api-wrappers-0-4-4' into 'master' · e4325fa2
  Christoph Klein authored 3 years ago
```
Upgrade to cuda-api-wrappers v0.4.5

Closes #12

See merge request !2
```
  e4325fa2
- Upgrade to cuda-api-wrappers v0.4.5 · 501b7449
  Paul Große-Bley authored 3 years ago and Christoph Klein committed 3 years ago
  
  501b7449
- Merge branch '11-add-cmake-options-to-install-apps-or-tests-for-spack-builds' into 'master' · 95fa8416
  Christoph Klein authored 3 years ago
```
Resolve "Add CMake options to install apps or tests for spack builds"

Closes #11

See merge request !1
```
  95fa8416
Feb 08, 2022
- Add options to install tests and benchmarks · 95f0b953
  Paul Große-Bley authored 3 years ago and Paul Große-Bley committed 3 years ago
  
  95f0b953
- Add bench-with-nvbench · b234d149
  dontspoilerme authored 3 years ago
  
  b234d149
Feb 07, 2022
- Rework bench.cu · 729c20ae
  dontspoilerme authored 3 years ago
  
  729c20ae
- Add config_t based interface for dynamic launch of work adaptive csrmv with get_row_block_start · 7171663d
  dontspoilerme authored 3 years ago
  
  7171663d
- Add segmented reduction from CUB in find-max2 bench · a57b8618
  dontspoilerme authored 3 years ago
  
  a57b8618
Feb 02, 2022
- Calc row_indices on-chip in case of binary coeff_multiplication_operator · f6a6f0df
  dontspoilerme authored 3 years ago
  
  f6a6f0df
Jan 23, 2022
- Fix: invalid memory accesses in gsl_ExpectsAudit in work balanced srcsrmv · 44357c6d
  dontspoilerme authored 3 years ago
  
  44357c6d
Jan 21, 2022
- trivial writeout option + remove additional reg_vals and row_ends · 0f963032
  dontspoilerme authored 3 years ago
```
register arrays for CUB's scan.
Better unified interface for the register values.
```
  0f963032
Jan 20, 2022
- Consistent setting of shared memory size · 95f0569a
  dontspoilerme authored 3 years ago
  
  95f0569a
- Remove unused constant::null_edge_id from bench-find-max2 · 43d9c4d7
  dontspoilerme authored 3 years ago
  
  43d9c4d7
- Less shared memory consumption for CUB scan in work adaptive csrmv kernel · ed0f6628
  dontspoilerme authored 3 years ago
  
  ed0f6628
Jan 19, 2022
- Only save type TD in registers and not TM before writing the result to shared memory · 60e4b0f6
  dontspoilerme authored 3 years ago
  
  60e4b0f6
Jan 18, 2022
- Fix: compilation + header for find-max2.h · 16deee3a
  dontspoilerme authored 3 years ago
  
  16deee3a
- Add cli/options.h · d180331f
  dontspoilerme authored 3 years ago
  
  d180331f
- Change sh mem iterator for user defined shared memory layouts in work adaptive csrmv · a3a8e530
  dontspoilerme authored 3 years ago
  
  a3a8e530
Jan 17, 2022
- Cosmetic changes + remove a comment. · 4b552ad7
  dontspoilerme authored 3 years ago
  
  4b552ad7
- more benchmarks regarding the find-max problem · b97715cb
  dontspoilerme authored 3 years ago
  
  b97715cb
Jan 16, 2022
- Add get_row_block_start_block_dim to config_t · bf73b214
  dontspoilerme authored 3 years ago
  
  bf73b214
- Add restricting result parameter for work adaptive csrmv · 7ccd31d0
  dontspoilerme authored 3 years ago
  
  7ccd31d0
Jan 14, 2022
- Add config_t · 7848227a
  dontspoilerme authored 3 years ago
  
  7848227a
Jan 11, 2022
- Fix: compilation · ad88259b
  dontspoilerme authored 3 years ago
  
  ad88259b
- Support for generic types in warp scan/reduce · c708acd2
  dontspoilerme authored 3 years ago
```
+ benchmark for find-max2
```
  c708acd2

Admin message