Commits · cc645538cbafacbcda37807c130af1bf9541938b · Christoph Alt / pystencils

13 May, 2021 1 commit

Merge branch 'arm64' into 'master' · cc645538

Markus Holzer authored May 13, 2021

Cherry-pick non-sizeless SVE things from !234

See merge request pycodegen/pystencils!245

cc645538

11 May, 2021 2 commits
- Merge branch 'Manifest' into 'master' · 2cf3eec1
  Michael Kuron authored May 11, 2021
```
ship C-file

See merge request pycodegen/pystencils!246
```
  2cf3eec1
- ship C-file · 3a4a4d75
  Markus Holzer authored May 11, 2021
  
  3a4a4d75
07 May, 2021 4 commits
- Cherry-pick updated SVE Philox · 896b4192
  Michael Kuron authored May 07, 2021
  
  896b4192
- Add ARM SVE CI job · 7e6be86e
  Michael Kuron authored May 06, 2021
  
  7e6be86e
- don't zero cachelines beyond the end of a field · 5cf6febc
  Michael Kuron authored May 06, 2021
  
  5cf6febc
- move NontemporalFence and CachelineSize to backend · 2d5fcf53
  Michael Kuron authored Apr 24, 2021
  
  2d5fcf53
06 May, 2021 2 commits
- Merge branch 'qemu' into 'master' · 269c0bfa
  Markus Holzer authored May 06, 2021
```
Add CI job for non-x86 vectorization

See merge request pycodegen/pystencils!244
```
  269c0bfa
- Add CI job for non-x86 vectorization · 78c2cbc0
  Michael Kuron authored May 04, 2021
  
  78c2cbc0
03 May, 2021 1 commit
- Merge branch 'hash' into 'master' · 0b740d1f
  Markus Holzer authored May 03, 2021
```
atomically write cache status file

See merge request pycodegen/pystencils!243
```
  0b740d1f
02 May, 2021 2 commits
- Merge branch 'scattergather' into 'master' · 8f72741d
  Markus Holzer authored May 02, 2021
```
Vector scatter/gather support

Closes #34

See merge request pycodegen/pystencils!241
```
  8f72741d
- Vector scatter/gather support · d314ef10
  Michael Kuron authored May 02, 2021 and Markus Holzer committed May 02, 2021
  
  d314ef10
29 Apr, 2021 3 commits
- atomically write cache status file · 2b466db5
  Michael Kuron authored Apr 29, 2021
  
  2b466db5
- Merge branch 'cachelinezero' into 'master' · 63536d22
  Markus Holzer authored Apr 29, 2021
```
Disallow OpenMP + blocking + cacheline-zero

See merge request pycodegen/pystencils!242
```
  63536d22
- Disallow OpenMP + blocking + cacheline-zero · 0fe7727d
  Michael Kuron authored Apr 29, 2021
  
  0fe7727d
28 Apr, 2021 4 commits
- Merge branch 'hash' into 'master' · e9bd89c8
  Markus Holzer authored Apr 28, 2021
```
Incorporate header files and compiler flags into object cache hash

See merge request pycodegen/pystencils!240
```
  e9bd89c8
- Incorporate header files and compiler flags into object cache hash · 0738e9e0
  Michael Kuron authored Apr 28, 2021 and Markus Holzer committed Apr 28, 2021
  
  0738e9e0
- Merge branch 'vec_tests' into 'master' · 10d3503d
  Markus Holzer authored Apr 28, 2021
```
maskStore improvements

See merge request pycodegen/pystencils!235
```
  10d3503d
- Merge branch 'amd' into 'master' · e2fbba20
  Markus Holzer authored Apr 28, 2021
```
Adapt an ifdef for AMD Epyc 7003

See merge request pycodegen/pystencils!236
```
  e2fbba20
27 Apr, 2021 2 commits
- Merge branch 'Versioneer' into 'master' · bb667c68
  Michael Kuron authored Apr 27, 2021
```
Versioneer

See merge request pycodegen/pystencils!238
```
  bb667c68
- Versioneer · 1d261175
  Markus Holzer authored Apr 27, 2021 and Michael Kuron committed Apr 27, 2021
  
  1d261175
26 Apr, 2021 2 commits
- Merge branch 'pickle' into 'master' · 17ea6bd2
  Markus Holzer authored Apr 26, 2021
```
Sympy 1.9 support

Closes #35

See merge request pycodegen/pystencils!239
```
  17ea6bd2
- Sympy 1.9 support · c1bb1ec2
  Michael Kuron authored Apr 26, 2021 and Markus Holzer committed Apr 26, 2021
  
  c1bb1ec2
21 Apr, 2021 2 commits

Fix an ifdef for AMD Epyc 7003 · 4812968d
Michael Kuron authored Apr 21, 2021

4812968d

maskStore improvements · c331d24f

Michael Kuron authored Apr 21, 2021

- fix the aligned version
- make sure the test case is incommensurate with the vector width
- implement a fallback for instruction sets that don't support it natively

c331d24f

20 Apr, 2021 4 commits
- Merge branch 'vec_tests' into 'master' · 059de5fb
  Markus Holzer authored Apr 20, 2021
```
Vectorization: improve test coverage

See merge request pycodegen/pystencils!233
```
  059de5fb
- Vectorization tests: run with all available instruction sets, add test for maskStore · 686a3ad8
  Michael Kuron authored Apr 20, 2021
  
  686a3ad8
- Merge branch 'sve' into 'master' · b1522533
  Markus Holzer authored Apr 20, 2021
```
SVE vectorization

See merge request pycodegen/pystencils!232
```
  b1522533
- SVE vectorization · 7238f9bf
  Michael Kuron authored Apr 20, 2021 and Markus Holzer committed Apr 20, 2021
  
  7238f9bf
14 Apr, 2021 1 commit
- Merge branch 'nontemporal' into 'master' · 589ca872
  Markus Holzer authored Apr 14, 2021
```
fix error introduced in !230

See merge request pycodegen/pystencils!231
```
  589ca872
13 Apr, 2021 1 commit
- fix error introduced in !230 · c5712bcb
  Michael Kuron authored Apr 13, 2021
  
  c5712bcb
12 Apr, 2021 1 commit
- Merge branch 'nontemporal' into 'master' · 147f6901
  Markus Holzer authored Apr 12, 2021
```
Improve non-temporal stores

Closes #25

See merge request pycodegen/pystencils!230
```
  147f6901
03 Apr, 2021 1 commit
- logic fix · 1d39f5f5
  Michael Kuron authored Apr 03, 2021
  
  1d39f5f5
02 Apr, 2021 1 commit
- remove stream from instruction sets that don't have it · 8f92d147
  Michael Kuron authored Apr 02, 2021
  
  8f92d147
01 Apr, 2021 1 commit
- Allow fields to be aligned to cacheline size · bcd2d628
  Michael Kuron authored Apr 01, 2021
  
  bcd2d628
31 Mar, 2021 5 commits
- cache line zero: require alignment to cache line size when OpenMP enabled · a100b5b0
  Michael Kuron authored Mar 31, 2021
```
otherwise we can delete stuff from the next outer loop, which in 2D would be the one OpenMP-parallelized
```
  a100b5b0
- nontemporal cleanup · 31be359b
  Michael Kuron authored Mar 31, 2021
  
  31be359b
- nontemporal stores: flush cacheline if available · 7e6d2a21
  Michael Kuron authored Mar 31, 2021
  
  7e6d2a21
- improve TemporayMemoryAllocation: don't align more than necessary, behave correctly on MSVC · 5a88c472
  Michael Kuron authored Mar 31, 2021
  
  5a88c472
- use cache line zero on architectures without nontemporal store · 9c784763
  Michael Kuron authored Mar 31, 2021
  
  9c784763