aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Changed from size_t to uint64_tSebastiano Tronto2024-10-121-5/+5
|
* Added shelltestSebastiano Tronto2024-10-1210-7/+44
| | | | Will add more tests when I feel like it
* Added buffer sizes in API argsSebastiano Tronto2024-10-119-93/+180
|
* Improved commentSebastiano Tronto2024-10-111-1/+1
|
* Error codes, documentation, change nisstype from string to flagsSebastiano Tronto2024-10-113-94/+460
|
* Improved error messages and add some comments in nissy.hSebastiano Tronto2024-10-109-119/+228
|
* added solve_scramble to shellSebastiano Tronto2024-10-101-1/+20
|
* Removed "options" from solver selectionSebastiano Tronto2024-10-1014-183/+161
| | | | | | | | Now the solver name fully determines the solver (and options) to be used. For example, now one must specify "h48h0k4" as the name of the solver. This PR also fixes a couple of things in tools.
* Small warning fixes to gendataSebastiano Tronto2024-10-101-3/+10
|
* Fixed tableinfo alignment issuesSebastiano Tronto2024-10-101-22/+50
|
* Reintroduce -Werror, but exclude unused optionsSebastiano Tronto2024-10-101-1/+2
|
* Removed un-common function from common.hSebastiano Tronto2024-10-101-1/+0
|
* Added cachegrind output to .gitignoreSebastiano Tronto2024-10-101-0/+1
|
* Revert useless optimization (fix portable and neon)Sebastiano Tronto2024-10-071-12/+10
|
* Remove -Werror compiler optionSebastiano Tronto2024-10-071-1/+1
|
* Remove double returnSebastiano Tronto2024-10-071-1/+0
|
* Tiny optimization for moves (avoid unnecessary CO)Sebastiano Tronto2024-10-062-11/+13
| | | | This is like 1% improvement to solver, but it is easy enough so I'll keep it
* Removed some useless outputSebastiano Tronto2024-10-062-7/+7
|
* Fix concurrency issuesSebastiano Tronto2024-10-063-44/+73
| | | | | So I did not have any misconception about Atomics, I just could not apply it where I did because the logic doesn't allow for it.
* Remove unnecessary _AtomicsSebastiano Tronto2024-10-054-27/+34
|
* Merge branch 'master' of tronto.net:h48Sebastiano Tronto2024-10-052-1/+97
|\
| * Added expected distributionsSebastiano Tronto2024-10-052-1/+97
| |
* | Fix rare data raceSebastiano Tronto2024-10-055-30/+30
|/ | | | | | I had some misconception about how _Atomic and const worked. This means the previous commit (expected distribution) could be incorrect.
* Added distribution check to toolSebastiano Tronto2024-10-054-5/+97
|
* Add transformations.mdSebastiano Tronto2024-10-041-0/+9
|
* Removed excessive LOG from gendata cocsepSebastiano Tronto2024-10-041-2/+0
|
* Avoid repeating typedefs in testsSebastiano Tronto2024-10-0420-261/+170
|
* Removed slow testSebastiano Tronto2024-10-043-108/+0
| | | | | | | | This test generated a pruning table and it was too slow (around a minute). Now the whole test suite runs in 49s on my laptop, which is more reasonable. Pruning table generation should be tested with tools. They run in release mode and are much faster, but can be still run in debug mode optionally.
* Revert table-based transformationsSebastiano Tronto2024-10-0411-863/+600
| | | | Long live the giant switch!
* Table-based transformations (will revert immediately)Sebastiano Tronto2024-10-0410-383/+467
| | | | | | | | This commit changes transformations to being based on a lookup table intead of a big switch. This is something I had planned for a long time and the code is nicer with this change. But unfortunately the performance is noticeably worse, so I'll revert all these changes in the next commit. I wanted to push this anyway so we can keep track of this.
* Added documentation on how to run derivedataSebastiano Tronto2024-10-041-0/+11
|
* Clean up tool outputSebastiano Tronto2024-10-047-29/+34
|
* Remove more locksSebastiano Tronto2024-10-031-23/+19
|
* Removed useless lockSebastiano Tronto2024-10-031-9/+1
|
* Improve gendata performance by replacing locks with atomic operationsSebastiano Tronto2024-10-036-63/+54
|
* Made test 120 (gendata h0k4) fasterSebastiano Tronto2024-10-033-3/+2
| | | | | | The test is still slow (around 25 seconds on gazelle). This is probably because of how sanitizers interact with multi-threading, and the fact that there is a lot of locking going on.
* Do not run make clean automatically on every build and other Makefile changesSebastiano Tronto2024-10-034-24/+23
|
* Updated h48 docSebastiano Tronto2024-10-011-22/+81
|
* Renamed macros in shell.cSebastiano Tronto2024-10-011-56/+56
|
* Merge pull request #3 from enricotenuti/masterSebastiano Tronto2024-09-296-7/+315
|\ | | | | Threaded solver
| * Merge remote-tracking branch 'upstream/master'enricotenuti2024-09-2933-347/+538
| |\ | |/ |/| | | Merge upstream
* | Use address and UB sanitizers by default, alwaysSebastiano Tronto2024-09-281-12/+5
| |
* | Solved alignment UB issueSebastiano Tronto2024-09-284-32/+36
| |
* | oopsieSebastiano Tronto2024-09-271-1/+1
| |
* | Made table derivation tool more flexibleSebastiano Tronto2024-09-275-41/+73
| | | | | | | | | | | | | | | | | | | | So apparently my RAM is broken. That took me a while to figure out. While I get a replacement, I have to restrict myself to a weaker test for the intermediate tables: instead of deriving them from the huge table and checking that they are the same, I have to derive a small h0k2 table from the intermediate ones and check that it is correct. This is not a 100% proof of correctness, but it is good enough (and much faster).
* | Merge branch 'master' of tronto.net:h48Sebastiano Tronto2024-09-271-1/+4
|\ \
| * | Fixed checkdataSebastiano Tronto2024-09-271-1/+4
| | |
* | | Use constant instead of magic numberSebastiano Tronto2024-09-274-18/+14
| | |
* | | Merge branch 'master' of tronto.net:h48Sebastiano Tronto2024-09-2728-295/+279
|\| |
| * | small fixSebastiano Tronto2024-09-261-1/+1
| | |

Generated with cgit - Back to sebastiano.tronto.net