109 Commits

Author	SHA1	Message	Date
Daniel	c62aff806a	small updates and notes for further writing	2025-04-15 19:32:39 +02:00
Daniel	a5c34a53b7	benchmarking: reverted previous; made interpreter use fast math	2025-04-13 13:26:35 +02:00
Daniel	6d6874c7ba	benchmarking: added results for test 4 on uni-pc	2025-04-13 11:45:52 +02:00
Daniel	af3b72f196	benchmarking: used int32 wherever possible; resulted in noticeable performance drop	2025-04-13 11:32:54 +02:00
Wiplinger Daniel - s2310454043	101ccef67b	benchmarking: tuned blocksizes; slightly improved performance; mostly improved standard deviation	2025-04-12 13:20:50 +02:00
Wiplinger Daniel - s2310454043	c6e2ce47aa	benchmarking: redid performance tests on uni pc	2025-04-12 12:13:28 +02:00
Daniel	9746db56c0	benchmarking: added tests for performance tuning with Nsight compute and systems	2025-04-12 10:24:45 +02:00
Daniel	258d33c338	benchmarking: redid inbounds tests because first test run did not use inbounds everywhere	2025-04-06 14:29:07 +02:00
Daniel	20fcbab4ca	concept and design: added transpiler section	2025-04-06 13:59:14 +02:00
Daniel	9e1094ac43	benchmarking: added reminder to redo one of the performance tests	2025-04-05 12:56:06 +02:00
Daniel	2a8de064a6	concept & design: improved existing sections	2025-04-05 12:54:26 +02:00
Daniel	d8f5454e9c	concept and design: started writing this chapter	2025-04-03 13:43:23 +02:00
Daniel	2b9c394f1b	benchmarking: improved performance with @inbounds. still slower in most cases	2025-04-01 21:48:59 +02:00
Daniel	d9c83caad9	benchmarking: updated blocksize to 256 with moderate improvements	2025-03-30 13:56:25 +02:00
Daniel	1dc0c1898d	benchmarking: fixed bugs; took initial_benchmark	2025-03-30 12:54:50 +02:00
Daniel	ad175abac0	benchmarking: added gpu evaluators to performance testing. getting execution errors still	2025-03-29 13:35:59 +01:00
Daniel	690ee33db1	benchmarks: started preparing benchmarks	2025-03-29 12:01:06 +01:00
Daniel	effd477558	transpiler: generates valid PTX and evaluates expressions correctly	2025-03-28 19:32:48 +01:00
Daniel	9df78ca72e	transpiler: invalid memory access error finally fixed	2025-03-27 22:32:24 +01:00
Daniel	561b37160b	transpiler: trying to fix problem with writing to global memory; not yet fixed	2025-03-27 18:00:31 +01:00
Daniel	eaee21ca75	transpiler: results are now written in results array; preperation for performance testing	2025-03-27 09:55:29 +01:00
Daniel	baa37ea183	code: started finalising transpilation process and preparing for performance testing and tuning	2025-03-23 13:38:22 +01:00
Gabriel Kronberger	f4f39ec47c	Improvements / fixes.	2025-02-19 17:18:43 +01:00
Gabriel Kronberger	942adb8612	Add CPU Interpreter and a test case.	2025-02-19 16:38:11 +01:00
Daniel	4afc15a737	small cleanup	2025-01-26 10:16:23 +01:00
Daniel	7598c51df8	finished generating parameter loading code	2025-01-25 11:15:54 +01:00
Daniel	b2774322a1	added loading of variables from memory into registers. Note: Needed to leave, so code currently not compiling	2025-01-24 13:13:53 +01:00
Daniel	85464083c3	aded symtable for loading vars and params to local memory	2025-01-19 11:00:29 +01:00
Daniel	219c0bb14e	started implementing parameter loading	2025-01-06 14:01:55 +01:00
Daniel	f7926c3438	finished implementing operators	2025-01-05 11:19:03 +01:00
Daniel Wiplinger	a97b804530	testing interpreter because loop seems to do everything. Can't test right now because I don't have a nvidia gpu in my laptop	2024-12-11 11:29:54 +01:00
Daniel	8d129dbfcc	rewrote function for generating code for operators. now the entire operation will be returned and not just the operator	2024-12-10 22:58:18 +01:00
Daniel	67ef9a5139	reworked code to use new 'register manager'	2024-12-08 10:50:09 +01:00
Daniel	203900bb26	nevermind, improved register management worked	2024-12-07 10:23:04 +01:00
Daniel	1e7f6e9010	tried streamlining register management	2024-12-07 10:12:53 +01:00
Daniel	68cedd75fc	updated all to 32-bit to save registers and boost performance	2024-11-01 11:23:58 +01:00
Daniel	9fc55c4c15	started implementing transpilation of expression	2024-10-27 11:48:11 +01:00
Daniel	0e24d74e54	small increment and fixes	2024-10-26 11:41:00 +02:00
Daniel	ee3c5001bd	added information on how to best approach register assignment	2024-10-20 12:24:18 +02:00
Daniel	1f6b40b750	added possibility to create registers	2024-10-19 13:44:38 +02:00
Daniel	7283082699	added guard clause generation	2024-09-28 11:41:13 +02:00
Daniel	d875fc7325	first steps towards ptx generation	2024-09-25 12:45:46 +02:00
Daniel	d60cba7e4d	continued understanding given PTX file and made plan on how to approach the transpiler part	2024-09-22 09:32:39 +02:00
Daniel	44722bfc65	finished interpreter	2024-09-08 11:52:10 +02:00
Daniel	c608abfb17	added more unit tests. Conversion to postfix not working	2024-09-01 12:09:29 +02:00
Daniel	6a1811ae6b	added operator tests. single value operators not working yet	2024-08-25 13:01:58 +02:00
Daniel	929789b6ff	finished interpreter. still need to test it	2024-08-18 11:55:41 +02:00
Daniel	62d10845e9	added operators to the interpreter. storing result still missing	2024-08-08 12:06:42 +02:00
Daniel	f1f8c3b2a4	first simple part of the interpreter finished	2024-08-03 12:11:43 +02:00
Daniel	6b7a4b4bca	added first (wrong) attempt at implementing the interpretation. Added input explanation to the readme for a better understanding	2024-07-24 14:42:26 +02:00