Commit Graph

58 Commits

Author SHA1 Message Date
Rafael Ravedutti
9d16bb46c8 Include average neighbors and SIMD iterations per atom on stats
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-20 22:43:08 +02:00
Rafael Ravedutti
557c5e4d3f IUpdate logbook again and do not check for variant on ATOMS_LOOP_RUNS effect
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-14 00:46:30 +02:00
Rafael Ravedutti
40bbaac078 Add hardware information on logbook
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-13 22:27:34 +02:00
Rafael Ravedutti
43ba28e130 Add ATOMS_LOOP_RUNS option and statistics to stub variant
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-12 22:39:54 +02:00
Rafael Ravedutti
55d346510e Compute statistics, useful data volume and cycles per SIMD iteration
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-12 15:04:08 +02:00
Rafael Ravedutti
f295f54fca Add useful data volume calculation for force kernel
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-11 16:57:02 +02:00
Rafael Ravedutti
b1f8e91f0f Add cache measurements for versions with and without force computation
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-04 18:13:17 +02:00
Rafael Ravedutti
749c4d55ab Provide trace output for all reneighboring steps
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-10-04 16:47:07 +02:00
Rafael Ravedutti
2889577a3b Add neighbor lists capacity to index tracer file
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-08-10 17:28:59 +02:00
Rafael Ravedutti
b8d71090aa Add number of atoms in index tracer file
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-08-10 16:43:51 +02:00
Rafael Ravedutti
6b75c158d6 Include atom reference on index tracer and add only_mem ASM version
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-08-10 01:19:26 +02:00
Rafael Ravedutti
57041ceed2 Add likwid-perfctr run with CACHES_PMC for 1 timestep on casclakesp2
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-07-16 00:24:43 +02:00
Rafael Ravedutti
8e1542c87d Sort elements within vector to fix distance calculation on index tracer
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-07-15 23:51:00 +02:00
Rafael Ravedutti
d97fc577b0 Add first version of index and distance tracer
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-07-09 23:49:14 +02:00
Rafael Ravedutti
ecb5ccf6ff Adjust likwid markers for force region
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-06-30 13:44:02 +02:00
Rafael Ravedutti
7ae22a5695 Add forces reading operation on memory tracer and include ampersand in TRACER_PRINT
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-06-21 22:56:44 +02:00
Rafael Ravedutti
0a2ec6376c Add memory tracer and update config.mk with all options
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-06-16 00:56:00 +02:00
Rafael Ravedutti
56ad09156b Fix explicit types for stubbed version
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-20 00:08:10 +02:00
Rafael Ravedutti
4496e91125 Add version with explicit types for atoms
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-19 23:51:02 +02:00
Rafael Ravedutti
6c03ea3f3c Adjust output when computing invalid values
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-07 02:31:53 +02:00
Rafael Ravedutti
e53d9961ef Fix compilation when INTERNAL_LOOP_NTIMES is not set and create avx512 directory
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-06 13:59:02 +02:00
Rafael Ravedutti
15de65303e Add version iterating most internal loop multiple times
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-05 03:04:41 +02:00
Rafael Ravedutti
0a81407948 Fix invalid values for cycles per neighbor
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-29 20:29:00 +02:00
Rafael Ravedutti
11b2d4bcc1 Update results for arch_analysis and stub script
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-29 17:57:37 +02:00
Rafael Ravedutti
c356336dbd Show cycles per atom and neighbor
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-22 16:50:22 +02:00
Rafael Ravedutti
4c53519c73 Add output as CSV
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-22 01:22:18 +02:00
Rafael Ravedutti
fd108d97d8 Fix problem when atoms_per_unit_cell is less or equal than 4
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-22 00:07:42 +02:00
Rafael Ravedutti
3c7dbc833a Allow any values for atoms_per_unit_cell
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-21 13:01:06 +02:00
Rafael Ravedutti
d3121ee08f Adjust computeForce parameters for stub
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-21 11:28:02 +02:00
Rafael Ravedutti
78e6e5c773 Merge master branch into stub
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-15 20:12:36 +02:00
Jan Eitzinger
06ba3b2726 Restructure timing and instrumentation. Add performance metric. 2021-04-15 14:55:02 +02:00
Rafael Ravedutti
a0699dde4c Merge branch 'master' into stub 2021-04-12 23:10:36 +02:00
Jan Eitzinger
35273c491b Introduce generic variant targets
Change Makefiles to allow multiple binary targets specified by VARIANT
environment variabel. For specified variants a matching main-<VARIANT>.c
source file must be present.
2021-04-12 08:24:27 +02:00
Rafael Ravedutti
706f1c38f2 Update stubbed force calculation
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-04-07 00:46:51 +02:00
Rafael Ravedutti
77a0774208 Improve stubbed force calculation version
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-03-30 22:17:30 +02:00
Rafael Ravedutti
7efd30791a Add version with stubbed force calculation
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-03-30 01:54:56 +02:00
Jan Eitzinger
2bc3120112 Put force kernel in separate compilation unit 2021-03-25 06:49:28 +01:00
Jan Eitzinger
d026028439 Streamline build system 2021-03-24 08:43:44 +01:00
Rafael Ravedutti
4b481bb407 Fix macros for AoS
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-03-23 16:21:08 +01:00
Jan Eitzinger
fc1fc9fd45 Port pbc to new macros. Enable debugging flags. 2021-03-23 10:03:55 +01:00
Jan Eitzinger
6679b6c8aa Simplify macros for data structure access.
Still segvaults for AoS.
2021-03-23 09:26:41 +01:00
Rafael Ravedutti
cb0ae2b0bb Replace inline calls by macros
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-03-22 21:51:47 +01:00
Rafael Ravedutti
3f7edb5dbf Add support for AoS data layout
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-03-20 18:32:50 +01:00
Jan Eitzinger
b39957421c Include Likwid Markers. Allow to switch between SP and DP floats. 2020-11-05 12:41:44 +01:00
Jan Eitzinger
0fd51e8a9c Add OpenMP parallelization for computeForce. 2020-08-19 11:19:16 +02:00
Jan Eitzinger
a0acce6cb3 Switch License to LGPL3 2020-08-19 10:47:40 +02:00
Jan Eitzinger
b1f2e23696 Fix more bugs. Enable optimization for CLANG. 2020-08-19 10:00:19 +02:00
Jan Eitzinger
dcee5f97f2 Add timing and result output. 2020-08-19 09:22:43 +02:00
Jan Eitzinger
71ea8dbb0e Refactor. Fix bug in atom initialization. 2020-08-19 09:00:35 +02:00
Jan Eitzinger
e7869286d7 Introduce modular version. 2020-08-18 14:27:28 +02:00