22278999e7
Fixed algorithm problem.
2017-02-02 10:35:43 +01:00
a5b68282b3
Cancel if `primes.list[i] >= numRoot'.
...
It has been solved!!!
2017-02-01 23:12:23 +01:00
b9cafadf8e
Parallelized, but needs algorithm optimization.
2017-02-01 19:36:05 +01:00
32076a67cd
Remove parallelization code to start anew.
2017-02-01 17:18:03 +01:00
9673a4da57
Use shared as default.
2017-01-31 22:24:06 +01:00
76936644cd
Almost completely parallelized.
2017-01-31 19:39:35 +01:00
a6ecea514f
Restrict will no longer apply to `addToList'
2017-01-17 00:11:51 +01:00
22c7702cf0
We only need to test up to the sqrt(num).
2017-01-01 17:05:25 +01:00
f8726497a4
Don't test for divisibility by 2.
...
We skip all even numbers, so we can skip 2, this also means giving 3 as
a given.
2016-12-30 17:57:28 +01:00
3e902efd18
Small cleanup.
2016-12-30 17:51:48 +01:00
3246870b75
Better help information.
2016-12-28 14:43:17 +01:00
09166efe89
Small improvements and optimizations.
...
All error catching is now in `main.c' so we can neatly release memory as
well as adding the `restrict' optimizer to the `addToList()' function.
2016-12-28 00:30:31 +01:00
8b35c5aea3
Fixed title for v0.7
2016-12-27 22:53:22 +01:00
1264edc8c8
We're going to release v0.7 first.
2016-12-27 22:52:06 +01:00
63aa8e14cb
Fixed the damn exporting for once.
...
Now I'm checking to see if it was actually worth a damn to have raw
files in the first place (-_-;).
2016-12-27 22:34:34 +01:00
f2eb3e869e
Fixed the file issue...
...
but the export issue still remains -_- I hate seg faults.
2016-12-27 21:47:16 +01:00
a695cce709
Moving shit to files.c
2016-12-27 21:31:24 +01:00
096cb2eb16
Moving file I/O to own file. Everything works except export.
2016-12-26 18:50:51 +01:00
1783b16024
Use raw I/O for file saving.
2016-12-26 16:51:51 +01:00
c602b5fe1b
Added OpenMP to CI.
2016-12-26 16:35:52 +01:00
6b2411e860
Now using OpenMP for parallel programming and next version is v1.0
2016-12-26 15:57:28 +01:00
6c2f96416b
Read and write primes to and from file.
2016-12-26 15:56:56 +01:00
fa3f2dd2b1
New goal before paralellization
2016-12-22 16:31:24 +01:00
591ee92971
Cast is unnecessary in C.
2016-12-15 15:16:04 +01:00
803c6f9e06
`size_t' makes this safeguard unnecessary.
2016-12-14 23:42:20 +01:00
4390fca3ef
Added comments.
2016-12-14 23:38:22 +01:00
2629c12f1a
Set goal for v1.0
2016-12-14 22:26:43 +01:00
af79d206d3
Added entry for v0.6
2016-12-14 20:12:46 +01:00
4b034ce5e3
Print more info about the file writing.
2016-12-14 19:34:50 +01:00
f146dbf11c
Able to save found primes to a file afterwards.
2016-12-14 19:28:28 +01:00
656fee720e
Allow user to choose base.
2016-12-14 19:06:33 +01:00
5e45656e1a
Forgot that from the switch to C.
2016-12-14 17:58:12 +01:00
c522196d66
Argument parsing!
2016-12-14 17:52:52 +01:00
b5dcadce19
Added new goal for v0.5
2016-12-14 17:31:36 +01:00
003b94dcdb
Added some more useful docs and stuff.
2016-12-14 16:14:58 +01:00
fcee95da17
Fixed something from the description
2016-12-14 16:14:37 +01:00
43620ba2d3
Enable documentation.
2016-12-14 16:08:11 +01:00
9f1160242a
Removed warning (no longer necessary).
2016-12-14 15:56:28 +01:00
b414bff9dc
Minor optimization.
2016-12-14 15:55:39 +01:00
d8c81b172b
Print some fun information at the end.
2016-12-14 15:29:06 +01:00
c2f0fb0ffd
Defining changes for v0.5
2016-12-14 15:05:24 +01:00
5bbac132bc
Use size_t, which is better for arrays of very large sizes.
2016-12-14 14:13:32 +01:00
2a3e97f4bc
Prepare version number so I don't forget later.
2016-12-13 18:09:05 +01:00
06d5ddb0cc
Fixed problem with the algorithm.
...
If it's equal to half then we want to check if it's divisible, since if
it's half then it is NOT prime.
2016-12-13 18:05:02 +01:00
a9019291c2
New goal for v0.4
2016-12-13 17:51:48 +01:00
5aa0b333c0
Made a fix, because previously it was not measuring primes.
2016-12-13 16:32:10 +01:00
3110c74174
Added entry about `restrict' keyword.
2016-12-13 16:12:31 +01:00
a5ce845c68
`restrict' keyword must be in declaration.
2016-12-13 16:12:03 +01:00
449fef2994
Add restrict
...
Add restrict for better pointer optimizations. This is not being applied
to `addToList()' because that function we want to thread later on.
2016-12-13 11:26:12 +01:00
2e9326b5fb
Preparing for v0.4
...
I'm going to be learning OpenCL, after which I will decide whether to
use OpenCL or OpenMP for the development of Indivisible. This mostly
depends on the compatibility of each library with GMP, especially since
I believe GMP already does some of its own threading.
2016-12-12 23:21:22 +01:00
