diff options
-rw-r--r-- | CHANGELOG | 1 | ||||
-rw-r--r-- | README.md | 2 | ||||
-rw-r--r-- | scripts/database/database/clblast.py | 2 | ||||
-rw-r--r-- | src/database/kernels/copy.hpp | 18 | ||||
-rw-r--r-- | src/database/kernels/pad.hpp | 12 | ||||
-rw-r--r-- | src/database/kernels/padtranspose.hpp | 8 | ||||
-rw-r--r-- | src/database/kernels/transpose.hpp | 12 | ||||
-rw-r--r-- | src/database/kernels/xaxpy.hpp | 10 | ||||
-rw-r--r-- | src/database/kernels/xdot.hpp | 22 | ||||
-rw-r--r-- | src/database/kernels/xgemm.hpp | 24 | ||||
-rw-r--r-- | src/database/kernels/xgemm_direct.hpp | 34 | ||||
-rw-r--r-- | src/database/kernels/xgemv.hpp | 8 | ||||
-rw-r--r-- | src/database/kernels/xgemv_fast.hpp | 12 | ||||
-rw-r--r-- | src/database/kernels/xgemv_fast_rot.hpp | 18 | ||||
-rw-r--r-- | src/database/kernels/xger.hpp | 14 |
15 files changed, 146 insertions, 51 deletions
@@ -1,6 +1,7 @@ Development version (next release) - Fixed a bug when using offsets in the direct version of the GEMM kernels +- Added tuned parameters for various devices (see README) Version 0.10.0 - Updated to version 8.0 of the CLCudaAPI C++11 OpenCL header @@ -127,6 +127,7 @@ The CLBlast library will be tuned in the future for the most commonly used OpenC - Pitcairn - Tahiti - Tonga + - Turks * Intel GPUs: - HD Graphics 530 - HD Graphics 5500 BroadWell U-Processor GT2 @@ -137,6 +138,7 @@ The CLBlast library will be tuned in the future for the most commonly used OpenC - Iris Pro * Intel CPUs: - Core i5-6200U + - Core i7-2670QM - Core i7-3770K - Core i7-5930K * Other devices: diff --git a/scripts/database/database/clblast.py b/scripts/database/database/clblast.py index d89b6350..8af3ab5b 100644 --- a/scripts/database/database/clblast.py +++ b/scripts/database/database/clblast.py @@ -123,7 +123,7 @@ def print_cpp_database(database, output_dir): devices = sorted(set([s["device"] for s in type_database])) for device_name in devices: device_database = [s for s in type_database if s["device"] == device_name] - device_name_quoted = "\"%s\"," % device_name + device_name_quoted = "\"%s\"," % device_name.strip() device_name_cpp = " { %-50s { " % device_name_quoted f.write(device_name_cpp) diff --git a/src/database/kernels/copy.hpp b/src/database/kernels/copy.hpp index 1bc63691..cf746901 100644 --- a/src/database/kernels/copy.hpp +++ b/src/database/kernels/copy.hpp @@ -44,6 +44,7 @@ const Database::DatabaseEntry CopySingle = { { "Pitcairn", { {"COPY_DIMX",8}, {"COPY_DIMY",16}, {"COPY_VW",4}, {"COPY_WPT",1} } }, { "Tahiti", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, { "Tonga", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",4} } }, + { "Turks", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",2} } }, { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, } }, @@ -55,10 +56,11 @@ const Database::DatabaseEntry CopySingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, - { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, + { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, } }, { // Intel GPUs @@ -99,7 +101,7 @@ const Database::DatabaseEntry CopySingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, + { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",4} } }, } }, } @@ -117,11 +119,13 @@ const Database::DatabaseEntry CopyComplexSingle = { { "Pitcairn", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",2} } }, { "Tahiti", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, { "Tonga", { {"COPY_DIMX",8}, {"COPY_DIMY",32}, {"COPY_VW",1}, {"COPY_WPT",2} } }, - { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } }, + { "Turks", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",1} } }, + { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",4} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",4}, {"COPY_WPT",2} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, @@ -164,7 +168,7 @@ const Database::DatabaseEntry CopyComplexSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } }, + { "default", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",2} } }, } }, } @@ -193,10 +197,11 @@ const Database::DatabaseEntry CopyDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",32}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",32}, {"COPY_VW",2}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"COPY_DIMX",16}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, - { "default", { {"COPY_DIMX",16}, {"COPY_DIMY",16}, {"COPY_VW",8}, {"COPY_WPT",1} } }, + { "default", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, } }, { // Intel accelerators @@ -225,7 +230,7 @@ const Database::DatabaseEntry CopyDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",2} } }, + { "default", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } }, } }, } @@ -254,6 +259,7 @@ const Database::DatabaseEntry CopyComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"COPY_DIMX",32}, {"COPY_DIMY",32}, {"COPY_VW",8}, {"COPY_WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",8}, {"COPY_WPT",1} } }, diff --git a/src/database/kernels/pad.hpp b/src/database/kernels/pad.hpp index 310d3a78..5d594046 100644 --- a/src/database/kernels/pad.hpp +++ b/src/database/kernels/pad.hpp @@ -44,7 +44,8 @@ const Database::DatabaseEntry PadSingle = { { "Pitcairn", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } }, { "Tahiti", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } }, { "Tonga", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",2}, {"PAD_WPTY",2} } }, - { "default", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",4} } }, + { "Turks", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } }, + { "default", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } }, } }, { // ARM GPUs @@ -55,6 +56,7 @@ const Database::DatabaseEntry PadSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",32}, {"PAD_WPTX",4}, {"PAD_WPTY",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",16}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PAD_DIMX",16}, {"PAD_DIMY",32}, {"PAD_WPTX",4}, {"PAD_WPTY",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, @@ -117,7 +119,8 @@ const Database::DatabaseEntry PadComplexSingle = { { "Pitcairn", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } }, { "Tahiti", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } }, { "Tonga", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } }, - { "default", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } }, + { "Turks", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",4} } }, + { "default", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } }, } }, { // ARM GPUs @@ -128,10 +131,11 @@ const Database::DatabaseEntry PadComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",2} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",32}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",16}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, - { "default", { {"PAD_DIMX",32}, {"PAD_DIMY",16}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, + { "default", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",2} } }, } }, { // Intel GPUs @@ -201,6 +205,7 @@ const Database::DatabaseEntry PadDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",2} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",32}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } }, @@ -262,6 +267,7 @@ const Database::DatabaseEntry PadComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PAD_DIMX",16}, {"PAD_DIMY",32}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } }, diff --git a/src/database/kernels/padtranspose.hpp b/src/database/kernels/padtranspose.hpp index 8ef09e85..c009425e 100644 --- a/src/database/kernels/padtranspose.hpp +++ b/src/database/kernels/padtranspose.hpp @@ -44,6 +44,7 @@ const Database::DatabaseEntry PadtransposeSingle = { { "Pitcairn", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",4} } }, { "Tahiti", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",4} } }, { "Tonga", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } }, + { "Turks", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } }, { "default", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",4} } }, } }, @@ -55,6 +56,7 @@ const Database::DatabaseEntry PadtransposeSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",32}, {"PADTRA_WPT",1} } }, @@ -117,6 +119,7 @@ const Database::DatabaseEntry PadtransposeComplexSingle = { { "Pitcairn", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, { "Tahiti", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } }, { "Tonga", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } }, + { "Turks", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",4} } }, { "default", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, } }, @@ -128,6 +131,7 @@ const Database::DatabaseEntry PadtransposeComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, @@ -201,6 +205,7 @@ const Database::DatabaseEntry PadtransposeDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",1} } }, @@ -262,10 +267,11 @@ const Database::DatabaseEntry PadtransposeComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, - { "default", { {"PADTRA_PAD",1}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, + { "default", { {"PADTRA_PAD",0}, {"PADTRA_TILE",8}, {"PADTRA_WPT",4} } }, } }, { // Intel accelerators diff --git a/src/database/kernels/transpose.hpp b/src/database/kernels/transpose.hpp index 23fecb49..56ce93a9 100644 --- a/src/database/kernels/transpose.hpp +++ b/src/database/kernels/transpose.hpp @@ -44,7 +44,8 @@ const Database::DatabaseEntry TransposeSingle = { { "Pitcairn", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } }, { "Tahiti", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",4} } }, { "Tonga", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, - { "default", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, + { "Turks", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, + { "default", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, } }, { // ARM GPUs @@ -55,6 +56,7 @@ const Database::DatabaseEntry TransposeSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",16} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, @@ -117,6 +119,7 @@ const Database::DatabaseEntry TransposeComplexSingle = { { "Pitcairn", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } }, { "Tahiti", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } }, { "Tonga", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } }, + { "Turks", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",4} } }, { "default", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, } }, @@ -128,6 +131,7 @@ const Database::DatabaseEntry TransposeComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } }, @@ -166,7 +170,7 @@ const Database::DatabaseEntry TransposeComplexSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, + { "default", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } }, } }, } @@ -195,10 +199,11 @@ const Database::DatabaseEntry TransposeDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, - { "default", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, + { "default", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } }, } }, { // Intel accelerators @@ -256,6 +261,7 @@ const Database::DatabaseEntry TransposeComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"TRA_DIM",4}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } }, diff --git a/src/database/kernels/xaxpy.hpp b/src/database/kernels/xaxpy.hpp index 52845e96..a852a895 100644 --- a/src/database/kernels/xaxpy.hpp +++ b/src/database/kernels/xaxpy.hpp @@ -44,6 +44,7 @@ const Database::DatabaseEntry XaxpySingle = { { "Pitcairn", { {"VW",2}, {"WGS",128}, {"WPT",1} } }, { "Tahiti", { {"VW",2}, {"WGS",64}, {"WPT",1} } }, { "Tonga", { {"VW",1}, {"WGS",256}, {"WPT",8} } }, + { "Turks", { {"VW",2}, {"WGS",256}, {"WPT",1} } }, { "default", { {"VW",2}, {"WGS",64}, {"WPT",2} } }, } }, @@ -55,10 +56,11 @@ const Database::DatabaseEntry XaxpySingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW",8}, {"WGS",512}, {"WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW",1}, {"WGS",512}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW",4}, {"WGS",256}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW",1}, {"WGS",128}, {"WPT",1} } }, - { "default", { {"VW",2}, {"WGS",256}, {"WPT",1} } }, + { "default", { {"VW",8}, {"WGS",1024}, {"WPT",2} } }, } }, { // Intel GPUs @@ -117,6 +119,7 @@ const Database::DatabaseEntry XaxpyComplexSingle = { { "Pitcairn", { {"VW",1}, {"WGS",64}, {"WPT",1} } }, { "Tahiti", { {"VW",1}, {"WGS",64}, {"WPT",1} } }, { "Tonga", { {"VW",1}, {"WGS",256}, {"WPT",8} } }, + { "Turks", { {"VW",2}, {"WGS",256}, {"WPT",1} } }, { "default", { {"VW",1}, {"WGS",128}, {"WPT",1} } }, } }, @@ -128,6 +131,7 @@ const Database::DatabaseEntry XaxpyComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW",4}, {"WGS",1024}, {"WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW",4}, {"WGS",256}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW",1}, {"WGS",1024}, {"WPT",2} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW",2}, {"WGS",1024}, {"WPT",1} } }, @@ -201,6 +205,7 @@ const Database::DatabaseEntry XaxpyDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW",4}, {"WGS",64}, {"WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW",1}, {"WGS",1024}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW",8}, {"WGS",64}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW",8}, {"WGS",2048}, {"WPT",1} } }, @@ -262,6 +267,7 @@ const Database::DatabaseEntry XaxpyComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW",4}, {"WGS",1024}, {"WPT",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW",8}, {"WGS",128}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW",8}, {"WGS",512}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW",1}, {"WGS",256}, {"WPT",1} } }, @@ -294,7 +300,7 @@ const Database::DatabaseEntry XaxpyComplexDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"VW",1}, {"WGS",128}, {"WPT",1} } }, + { "default", { {"VW",1}, {"WGS",256}, {"WPT",1} } }, } }, } diff --git a/src/database/kernels/xdot.hpp b/src/database/kernels/xdot.hpp index 8b07c539..a7a4e105 100644 --- a/src/database/kernels/xdot.hpp +++ b/src/database/kernels/xdot.hpp @@ -43,13 +43,15 @@ const Database::DatabaseEntry XdotSingle = { { "Pitcairn", { {"WGS1",128}, {"WGS2",32} } }, { "Tahiti", { {"WGS1",128}, {"WGS2",32} } }, { "Tonga", { {"WGS1",64}, {"WGS2",32} } }, - { "default", { {"WGS1",128}, {"WGS2",32} } }, + { "Turks", { {"WGS1",128}, {"WGS2",64} } }, + { "default", { {"WGS1",128}, {"WGS2",64} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",32}, {"WGS2",32} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",1024}, {"WGS2",32} } }, - { "default", { {"WGS1",1024}, {"WGS2",32} } }, + { "default", { {"WGS1",256}, {"WGS2",32} } }, } }, { // Intel GPUs @@ -81,7 +83,7 @@ const Database::DatabaseEntry XdotSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"WGS1",256}, {"WGS2",32} } }, + { "default", { {"WGS1",128}, {"WGS2",32} } }, } }, } @@ -98,13 +100,15 @@ const Database::DatabaseEntry XdotComplexSingle = { { "Pitcairn", { {"WGS1",256}, {"WGS2",32} } }, { "Tahiti", { {"WGS1",64}, {"WGS2",32} } }, { "Tonga", { {"WGS1",256}, {"WGS2",64} } }, - { "default", { {"WGS1",256}, {"WGS2",64} } }, + { "Turks", { {"WGS1",128}, {"WGS2",32} } }, + { "default", { {"WGS1",256}, {"WGS2",32} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",128}, {"WGS2",64} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",1024}, {"WGS2",32} } }, - { "default", { {"WGS1",1024}, {"WGS2",32} } }, + { "default", { {"WGS1",128}, {"WGS2",64} } }, } }, { // Intel GPUs @@ -136,7 +140,7 @@ const Database::DatabaseEntry XdotComplexSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"WGS1",256}, {"WGS2",64} } }, + { "default", { {"WGS1",256}, {"WGS2",32} } }, } }, } @@ -158,8 +162,9 @@ const Database::DatabaseEntry XdotDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",64}, {"WGS2",128} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",512}, {"WGS2",64} } }, - { "default", { {"WGS1",512}, {"WGS2",64} } }, + { "default", { {"WGS1",128}, {"WGS2",64} } }, } }, { // NVIDIA GPUs @@ -202,8 +207,9 @@ const Database::DatabaseEntry XdotComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",32}, {"WGS2",128} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",1024}, {"WGS2",32} } }, - { "default", { {"WGS1",1024}, {"WGS2",32} } }, + { "default", { {"WGS1",128}, {"WGS2",32} } }, } }, { // NVIDIA GPUs diff --git a/src/database/kernels/xgemm.hpp b/src/database/kernels/xgemm.hpp index 66ac8a9f..06f0f26a 100644 --- a/src/database/kernels/xgemm.hpp +++ b/src/database/kernels/xgemm.hpp @@ -43,7 +43,8 @@ const Database::DatabaseEntry XgemmSingle = { { "Pitcairn", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } }, { "Tahiti", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",32}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",1} } }, { "Tonga", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",8} } }, - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",4} } }, + { "Turks", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",4} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",4} } }, } }, { // ARM GPUs @@ -54,10 +55,11 @@ const Database::DatabaseEntry XgemmSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"KWG",32}, {"KWI",8}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",32}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, + { "default", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, } }, { // Intel GPUs @@ -116,7 +118,8 @@ const Database::DatabaseEntry XgemmComplexSingle = { { "Pitcairn", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",4}, {"VWN",2} } }, { "Tahiti", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",16}, {"NWG",32}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",1} } }, { "Tonga", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",1} } }, - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",32}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",2} } }, + { "Turks", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",1} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, } }, { // ARM GPUs @@ -127,10 +130,11 @@ const Database::DatabaseEntry XgemmComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",1}, {"VWM",1}, {"VWN",2} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",16}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",4} } }, - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, + { "default", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, } }, { // Intel GPUs @@ -171,7 +175,7 @@ const Database::DatabaseEntry XgemmComplexSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",1} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",1} } }, } }, } @@ -200,10 +204,11 @@ const Database::DatabaseEntry XgemmDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",8} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",8} } }, - { "default", { {"KWG",32}, {"KWI",8}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",1} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",4} } }, } }, { // Intel accelerators @@ -232,7 +237,7 @@ const Database::DatabaseEntry XgemmDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",4} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",4} } }, } }, } @@ -261,10 +266,11 @@ const Database::DatabaseEntry XgemmComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",0}, {"SB",1}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",1} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",32}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",0}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",8}, {"NDIMC",16}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",1}, {"VWN",8} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"KWG",32}, {"KWI",8}, {"MDIMA",8}, {"MDIMC",32}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, + { "default", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",1} } }, } }, { // Intel accelerators @@ -292,7 +298,7 @@ const Database::DatabaseEntry XgemmComplexDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",16}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",2} } }, + { "default", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",2} } }, } }, } diff --git a/src/database/kernels/xgemm_direct.hpp b/src/database/kernels/xgemm_direct.hpp index 4413cf1b..d7aacfa3 100644 --- a/src/database/kernels/xgemm_direct.hpp +++ b/src/database/kernels/xgemm_direct.hpp @@ -39,9 +39,16 @@ const Database::DatabaseEntry XgemmDirectSingle = { kDeviceTypeGPU, "AMD", { { "AMD Radeon R9 M370X Compute Engine", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",32} } }, { "Tonga", { {"KWID",16}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",32}, {"NDIMCD",8}, {"PADA",0}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",32} } }, + { "Turks", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, } }, + { // Intel CPUs + kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",1}, {"VWND",8}, {"WGD",64} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",1}, {"VWND",8}, {"WGD",64} } }, + } + }, { // Intel GPUs kDeviceTypeGPU, "Intel", { { "Intel(R) HD Graphics Skylake ULT GT2", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",8} } }, @@ -58,7 +65,7 @@ const Database::DatabaseEntry XgemmDirectSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",4}, {"VWND",2}, {"WGD",32} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",32} } }, } }, } @@ -72,9 +79,16 @@ const Database::DatabaseEntry XgemmDirectComplexSingle = { kDeviceTypeGPU, "AMD", { { "AMD Radeon R9 M370X Compute Engine", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, { "Tonga", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",32} } }, + { "Turks", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",16} } }, { "default", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",32} } }, } }, + { // Intel CPUs + kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",4}, {"VWND",4}, {"WGD",32} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",4}, {"VWND",4}, {"WGD",32} } }, + } + }, { // Intel GPUs kDeviceTypeGPU, "Intel", { { "Intel(R) HD Graphics Skylake ULT GT2", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, @@ -91,7 +105,7 @@ const Database::DatabaseEntry XgemmDirectComplexSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",16} } }, } }, } @@ -108,6 +122,12 @@ const Database::DatabaseEntry XgemmDirectDouble = { { "default", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, } }, + { // Intel CPUs + kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",4}, {"VWND",4}, {"WGD",32} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",4}, {"VWND",4}, {"WGD",32} } }, + } + }, { // NVIDIA GPUs kDeviceTypeGPU, "NVIDIA", { { "GeForce GTX 750 Ti", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",4}, {"WGD",32} } }, @@ -117,7 +137,7 @@ const Database::DatabaseEntry XgemmDirectDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",16} } }, } }, } @@ -134,6 +154,12 @@ const Database::DatabaseEntry XgemmDirectComplexDouble = { { "default", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, } }, + { // Intel CPUs + kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",32}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",1}, {"VWND",1}, {"WGD",32} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",32}, {"NDIMCD",8}, {"PADA",0}, {"PADB",0}, {"VWMD",1}, {"VWND",1}, {"WGD",32} } }, + } + }, { // NVIDIA GPUs kDeviceTypeGPU, "NVIDIA", { { "GeForce GTX 750 Ti", { {"KWID",2}, {"MDIMAD",32}, {"MDIMCD",32}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",32} } }, @@ -143,7 +169,7 @@ const Database::DatabaseEntry XgemmDirectComplexDouble = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"KWID",2}, {"MDIMAD",16}, {"MDIMCD",16}, {"NDIMBD",16}, {"NDIMCD",16}, {"PADA",1}, {"PADB",1}, {"VWMD",1}, {"VWND",1}, {"WGD",16} } }, + { "default", { {"KWID",2}, {"MDIMAD",8}, {"MDIMCD",8}, {"NDIMBD",8}, {"NDIMCD",8}, {"PADA",1}, {"PADB",1}, {"VWMD",2}, {"VWND",2}, {"WGD",16} } }, } }, } diff --git a/src/database/kernels/xgemv.hpp b/src/database/kernels/xgemv.hpp index 5f25f210..7d1781f9 100644 --- a/src/database/kernels/xgemv.hpp +++ b/src/database/kernels/xgemv.hpp @@ -44,11 +44,13 @@ const Database::DatabaseEntry XgemvSingle = { { "Pitcairn", { {"WGS1",256}, {"WPT1",1} } }, { "Tahiti", { {"WGS1",256}, {"WPT1",1} } }, { "Tonga", { {"WGS1",128}, {"WPT1",2} } }, + { "Turks", { {"WGS1",32}, {"WPT1",1} } }, { "default", { {"WGS1",128}, {"WPT1",1} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",128}, {"WPT1",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",64}, {"WPT1",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"WGS1",64}, {"WPT1",4} } }, { "default", { {"WGS1",64}, {"WPT1",4} } }, @@ -110,14 +112,16 @@ const Database::DatabaseEntry XgemvComplexSingle = { { "Pitcairn", { {"WGS1",64}, {"WPT1",1} } }, { "Tahiti", { {"WGS1",64}, {"WPT1",1} } }, { "Tonga", { {"WGS1",32}, {"WPT1",1} } }, + { "Turks", { {"WGS1",64}, {"WPT1",1} } }, { "default", { {"WGS1",64}, {"WPT1",1} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",32}, {"WPT1",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",128}, {"WPT1",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"WGS1",64}, {"WPT1",4} } }, - { "default", { {"WGS1",64}, {"WPT1",1} } }, + { "default", { {"WGS1",64}, {"WPT1",2} } }, } }, { // Intel GPUs @@ -177,6 +181,7 @@ const Database::DatabaseEntry XgemvDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",64}, {"WPT1",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",64}, {"WPT1",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"WGS1",64}, {"WPT1",4} } }, { "default", { {"WGS1",64}, {"WPT1",4} } }, @@ -231,6 +236,7 @@ const Database::DatabaseEntry XgemvComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",64}, {"WPT1",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",64}, {"WPT1",1} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"WGS1",64}, {"WPT1",4} } }, { "default", { {"WGS1",64}, {"WPT1",4} } }, diff --git a/src/database/kernels/xgemv_fast.hpp b/src/database/kernels/xgemv_fast.hpp index 994a220c..988687d5 100644 --- a/src/database/kernels/xgemv_fast.hpp +++ b/src/database/kernels/xgemv_fast.hpp @@ -44,14 +44,16 @@ const Database::DatabaseEntry XgemvFastSingle = { { "Pitcairn", { {"VW2",1}, {"WGS2",64}, {"WPT2",1} } }, { "Tahiti", { {"VW2",1}, {"WGS2",64}, {"WPT2",1} } }, { "Tonga", { {"VW2",1}, {"WGS2",16}, {"WPT2",4} } }, + { "Turks", { {"VW2",1}, {"WGS2",256}, {"WPT2",1} } }, { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",1} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW2",1}, {"WGS2",32}, {"WPT2",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW2",4}, {"WGS2",128}, {"WPT2",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, - { "default", { {"VW2",4}, {"WGS2",64}, {"WPT2",4} } }, + { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, } }, { // Intel GPUs @@ -110,14 +112,16 @@ const Database::DatabaseEntry XgemvFastComplexSingle = { { "Pitcairn", { {"VW2",1}, {"WGS2",64}, {"WPT2",1} } }, { "Tahiti", { {"VW2",1}, {"WGS2",128}, {"WPT2",1} } }, { "Tonga", { {"VW2",2}, {"WGS2",32}, {"WPT2",2} } }, + { "Turks", { {"VW2",1}, {"WGS2",16}, {"WPT2",1} } }, { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",1} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW2",2}, {"WGS2",64}, {"WPT2",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW2",1}, {"WGS2",128}, {"WPT2",2} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW2",4}, {"WGS2",64}, {"WPT2",4} } }, - { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",2} } }, + { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, } }, { // Intel GPUs @@ -173,6 +177,7 @@ const Database::DatabaseEntry XgemvFastDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW2",4}, {"WGS2",128}, {"WPT2",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, { "default", { {"VW2",1}, {"WGS2",64}, {"WPT2",4} } }, @@ -227,9 +232,10 @@ const Database::DatabaseEntry XgemvFastComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW2",4}, {"WGS2",32}, {"WPT2",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"VW2",2}, {"WGS2",64}, {"WPT2",4} } }, { "Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz", { {"VW2",4}, {"WGS2",64}, {"WPT2",4} } }, - { "default", { {"VW2",2}, {"WGS2",64}, {"WPT2",4} } }, + { "default", { {"VW2",4}, {"WGS2",64}, {"WPT2",4} } }, } }, { // Intel accelerators diff --git a/src/database/kernels/xgemv_fast_rot.hpp b/src/database/kernels/xgemv_fast_rot.hpp index da8bcfeb..6fbb4619 100644 --- a/src/database/kernels/xgemv_fast_rot.hpp +++ b/src/database/kernels/xgemv_fast_rot.hpp @@ -39,11 +39,13 @@ const Database::DatabaseEntry XgemvFastRotSingle = { kDeviceTypeGPU, "AMD", { { "AMD Radeon R9 M370X Compute Engine", { {"VW3",8}, {"WGS3",64}, {"WPT3",32} } }, { "Tonga", { {"VW3",8}, {"WGS3",128}, {"WPT3",32} } }, + { "Turks", { {"VW3",8}, {"WGS3",128}, {"WPT3",16} } }, { "default", { {"VW3",8}, {"WGS3",128}, {"WPT3",32} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW3",8}, {"WGS3",16}, {"WPT3",8} } }, { "default", { {"VW3",8}, {"WGS3",16}, {"WPT3",8} } }, } @@ -68,7 +70,7 @@ const Database::DatabaseEntry XgemvFastRotSingle = { }, { // Default kDeviceTypeAll, "default", { - { "default", { {"VW3",8}, {"WGS3",128}, {"WPT3",32} } }, + { "default", { {"VW3",8}, {"WGS3",32}, {"WPT3",32} } }, } }, } @@ -82,13 +84,15 @@ const Database::DatabaseEntry XgemvFastRotComplexSingle = { kDeviceTypeGPU, "AMD", { { "AMD Radeon R9 M370X Compute Engine", { {"VW3",8}, {"WGS3",16}, {"WPT3",16} } }, { "Tonga", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, - { "default", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, + { "Turks", { {"VW3",4}, {"WGS3",32}, {"WPT3",8} } }, + { "default", { {"VW3",8}, {"WGS3",32}, {"WPT3",32} } }, } }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW3",4}, {"WGS3",16}, {"WPT3",16} } }, - { "default", { {"VW3",4}, {"WGS3",16}, {"WPT3",16} } }, + { "default", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, } }, { // Intel GPUs @@ -122,8 +126,9 @@ const Database::DatabaseEntry XgemvFastRotDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW3",4}, {"WGS3",32}, {"WPT3",32} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW3",8}, {"WGS3",16}, {"WPT3",8} } }, - { "default", { {"VW3",8}, {"WGS3",16}, {"WPT3",8} } }, + { "default", { {"VW3",8}, {"WGS3",32}, {"WPT3",32} } }, } }, { // NVIDIA GPUs @@ -155,13 +160,14 @@ const Database::DatabaseEntry XgemvFastRotComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"VW3",2}, {"WGS3",16}, {"WPT3",16} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"VW3",8}, {"WGS3",16}, {"WPT3",16} } }, - { "default", { {"VW3",8}, {"WGS3",16}, {"WPT3",16} } }, + { "default", { {"VW3",4}, {"WGS3",16}, {"WPT3",16} } }, } }, { // Default kDeviceTypeAll, "default", { - { "default", { {"VW3",8}, {"WGS3",32}, {"WPT3",16} } }, + { "default", { {"VW3",4}, {"WGS3",16}, {"WPT3",16} } }, } }, } diff --git a/src/database/kernels/xger.hpp b/src/database/kernels/xger.hpp index 5e2be6a9..f7295ff4 100644 --- a/src/database/kernels/xger.hpp +++ b/src/database/kernels/xger.hpp @@ -44,6 +44,7 @@ const Database::DatabaseEntry XgerSingle = { { "Pitcairn", { {"WGS1",64}, {"WGS2",1}, {"WPT",1} } }, { "Tahiti", { {"WGS1",256}, {"WGS2",1}, {"WPT",1} } }, { "Tonga", { {"WGS1",256}, {"WGS2",1}, {"WPT",2} } }, + { "Turks", { {"WGS1",64}, {"WGS2",4}, {"WPT",2} } }, { "default", { {"WGS1",32}, {"WGS2",8}, {"WPT",1} } }, } }, @@ -55,9 +56,10 @@ const Database::DatabaseEntry XgerSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",32}, {"WGS2",4}, {"WPT",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",128}, {"WGS2",2}, {"WPT",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"WGS1",128}, {"WGS2",1}, {"WPT",4} } }, - { "default", { {"WGS1",128}, {"WGS2",8}, {"WPT",4} } }, + { "default", { {"WGS1",64}, {"WGS2",1}, {"WPT",4} } }, } }, { // Intel GPUs @@ -105,7 +107,8 @@ const Database::DatabaseEntry XgerComplexSingle = { { "Pitcairn", { {"WGS1",128}, {"WGS2",2}, {"WPT",1} } }, { "Tahiti", { {"WGS1",64}, {"WGS2",2}, {"WPT",1} } }, { "Tonga", { {"WGS1",64}, {"WGS2",1}, {"WPT",1} } }, - { "default", { {"WGS1",64}, {"WGS2",1}, {"WPT",1} } }, + { "Turks", { {"WGS1",128}, {"WGS2",2}, {"WPT",1} } }, + { "default", { {"WGS1",128}, {"WGS2",2}, {"WPT",1} } }, } }, { // ARM GPUs @@ -116,9 +119,10 @@ const Database::DatabaseEntry XgerComplexSingle = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",128}, {"WGS2",2}, {"WPT",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",256}, {"WGS2",1}, {"WPT",4} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"WGS1",512}, {"WGS2",4}, {"WPT",2} } }, - { "default", { {"WGS1",512}, {"WGS2",4}, {"WPT",2} } }, + { "default", { {"WGS1",256}, {"WGS2",1}, {"WPT",4} } }, } }, { // Intel GPUs @@ -177,9 +181,10 @@ const Database::DatabaseEntry XgerDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",256}, {"WGS2",1}, {"WPT",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",512}, {"WGS2",16}, {"WPT",1} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"WGS1",512}, {"WGS2",8}, {"WPT",2} } }, - { "default", { {"WGS1",512}, {"WGS2",8}, {"WPT",2} } }, + { "default", { {"WGS1",256}, {"WGS2",4}, {"WPT",2} } }, } }, { // NVIDIA GPUs @@ -227,6 +232,7 @@ const Database::DatabaseEntry XgerComplexDouble = { }, { // Intel CPUs kDeviceTypeCPU, "Intel", { + { "Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz", { {"WGS1",128}, {"WGS2",4}, {"WPT",4} } }, { "Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz", { {"WGS1",512}, {"WGS2",4}, {"WPT",2} } }, { "Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz", { {"WGS1",256}, {"WGS2",1}, {"WPT",2} } }, { "default", { {"WGS1",256}, {"WGS2",1}, {"WPT",2} } }, |