4979 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Peter Boyle	78ebd93281	Cuda 9.1 happy	2019-07-12 17:11:00 +01:00
Peter Boyle	3d58daf70f	Safety check	2019-07-12 17:10:35 +01:00
Peter Boyle	6e3c3214a3	Offload loops	2019-07-02 17:25:40 +01:00
Peter Boyle	d6ffadb33b	Coalesced write	2019-07-02 17:25:13 +01:00
Peter Boyle	4c3225412b	Drop 5dVEC	2019-07-01 07:31:26 +01:00
Peter Boyle	b8f7bfbb26	Dont stream as poor perf in some cases	2019-07-01 07:30:25 +01:00
Peter Boyle	7b7c470917	Accelerator loop	2019-07-01 07:29:51 +01:00
Peter Boyle	532e226b22	cuda 9.1 fixes	2019-07-01 07:29:22 +01:00
Peter Boyle	6a13731818	Move GPU cuda call earlier	2019-07-01 07:28:41 +01:00
Peter Boyle	1cd4ee0706	Thrust used on GPU builds	2019-06-18 12:50:35 +01:00
Peter Boyle	b8f71b6777	Fix NVCC warning unused variable	2019-06-17 13:58:45 +01:00
Peter Boyle	703dc20377	Compile tests fix	2019-06-16 13:59:29 +01:00
Peter Boyle	d976e5c514	Pow is being awkward in thrust for reasons I don't understand. Possible thrust bug.	2019-06-16 12:05:11 +01:00
Peter Boyle	d7b3efe893	Compile fix	2019-06-15 17:03:15 +01:00
Peter Boyle	f710d7bd45	TODO list update	2019-06-15 12:54:27 +01:00
Peter Boyle	cb336aa8f8	Thread loop constructs changing a little	2019-06-15 12:54:11 +01:00
Peter Boyle	462900b48d	Modified entire test directory to suit new GPU constructs for looping	2019-06-15 12:53:27 +01:00
Peter Boyle	0561c2edeb	Benchmarks modified for new GPU constructs	2019-06-15 12:52:56 +01:00
Peter Boyle	0184719216	Change to predicate type	2019-06-15 12:52:26 +01:00
Peter Boyle	24202dbc51	Thread loop construct change	2019-06-15 12:52:07 +01:00
Peter Boyle	d763c303c5	Clean acceleerator barrier	2019-06-15 12:51:45 +01:00
Peter Boyle	8e394d3bf9	New loop construct	2019-06-15 12:51:15 +01:00
Peter Boyle	b881d5489b	Move SchurDiagTwoKappa to Algorithms	2019-06-15 12:50:45 +01:00
Peter Boyle	82306913a8	Move Schur operator into correct place	2019-06-15 12:49:22 +01:00
Peter Boyle	49f90cc7eb	use pragma once	2019-06-15 12:45:22 +01:00
Peter Boyle	b77af0210b	Thread loop. Probably deprecate this impl	2019-06-15 12:44:56 +01:00
Peter Boyle	5254ede2d8	New loops. Revisit as accelerator loop in future audit	2019-06-15 12:44:29 +01:00
Peter Boyle	16e5d7945e	Hard to make 5D vec work with GPU code	2019-06-15 12:43:43 +01:00
Peter Boyle	decc99ca76	Accelerator version	2019-06-15 12:43:00 +01:00
Peter Boyle	464cd65931	Still to test this fully	2019-06-15 12:35:14 +01:00
Peter Boyle	a1ec2f4723	Still to test this routine fully	2019-06-15 12:33:55 +01:00
Peter Boyle	ea9662ec85	Thread loop changes	2019-06-15 09:09:57 +01:00
Peter Boyle	52c74f1cac	Thread loop changes	2019-06-15 09:08:16 +01:00
Peter Boyle	9a13d2992c	lean up	2019-06-15 09:05:16 +01:00
Peter Boyle	b0449ae270	Thread loop changes	2019-06-15 09:04:19 +01:00
Peter Boyle	1299225105	Accelerator loop changes	2019-06-15 09:03:46 +01:00
Peter Boyle	5925e7f405	Thread for changes	2019-06-15 09:01:30 +01:00
Peter Boyle	be1fd4930f	Template instantiation make happy changes	2019-06-15 08:37:34 +01:00
Peter Boyle	377fa5dec1	looping construct	2019-06-15 08:36:48 +01:00
Peter Boyle	e8b78f596e	Looping construct changes	2019-06-15 08:35:57 +01:00
Peter Boyle	09720c40cd	Coalesced loops	2019-06-15 08:35:26 +01:00
Peter Boyle	bb024dd114	Loop construct changed	2019-06-15 08:30:05 +01:00
Peter Boyle	52456b9ec7	New loop construct	2019-06-15 08:28:45 +01:00
Peter Boyle	b285138be4	Better checking on types	2019-06-15 08:27:48 +01:00
Peter Boyle	c7dbf4c87e	Scalar support for GPU threads	2019-06-15 08:25:43 +01:00
Peter Boyle	1e889c93b8	Insert a GPU synchronise	2019-06-15 08:23:26 +01:00
Peter Boyle	7379047482	Threading and acceleration primitives further changes. accelerator_barrier() needed and used	2019-06-15 08:22:48 +01:00
Peter Boyle	d836ce3b78	Clean up of acceleration and threading primitives	2019-06-15 08:14:21 +01:00
Peter Boyle	cefaacbc07	Changing accelerator loop. Still have work to do for multi-GPU code	2019-06-15 08:10:24 +01:00
Peter Boyle	0074ef7f69	thread loops	2019-06-15 08:04:29 +01:00