summaryrefslogtreecommitdiff
path: root/src
diff options
context:
space:
mode:
authorCedric Nugteren <web@cedricnugteren.nl>2016-07-03 20:30:47 +0200
committerCedric Nugteren <web@cedricnugteren.nl>2016-07-03 20:30:47 +0200
commit9683b50c557c7e415389b321d437be39ce0e519d (patch)
tree1bbf1b7033ac8ee1496873348e5e8a11de4b0e1a /src
parent4105a7959861597b97c1d72a64fff773a426a940 (diff)
Added tuning results for GTX670, GTX750, and GTX1070 (thanks to gcp)
Diffstat (limited to 'src')
-rw-r--r--src/database/kernels/copy.hpp12
-rw-r--r--src/database/kernels/pad.hpp14
-rw-r--r--src/database/kernels/padtranspose.hpp12
-rw-r--r--src/database/kernels/transpose.hpp16
-rw-r--r--src/database/kernels/xaxpy.hpp12
-rw-r--r--src/database/kernels/xdot.hpp14
-rw-r--r--src/database/kernels/xgemm.hpp14
-rw-r--r--src/database/kernels/xgemv.hpp10
-rw-r--r--src/database/kernels/xger.hpp16
9 files changed, 113 insertions, 7 deletions
diff --git a/src/database/kernels/copy.hpp b/src/database/kernels/copy.hpp
index 14946af4..a76a08e7 100644
--- a/src/database/kernels/copy.hpp
+++ b/src/database/kernels/copy.hpp
@@ -75,8 +75,11 @@ const Database::DatabaseEntry Database::CopySingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",1} } },
+ { "GeForce GTX 1070", { {"COPY_DIMX",8}, {"COPY_DIMY",16}, {"COPY_VW",4}, {"COPY_WPT",1} } },
{ "GeForce GTX 480", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",1} } },
+ { "GeForce GTX 670", { {"COPY_DIMX",16}, {"COPY_DIMY",32}, {"COPY_VW",4}, {"COPY_WPT",1} } },
{ "GeForce GTX 680", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",4}, {"COPY_WPT",1} } },
+ { "GeForce GTX 750", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",2} } },
{ "GeForce GTX 750 Ti", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",1} } },
{ "GeForce GTX 980", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX TITAN", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",4} } },
@@ -133,7 +136,10 @@ const Database::DatabaseEntry Database::CopyComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 1070", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX 480", { {"COPY_DIMX",16}, {"COPY_DIMY",16}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 670", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 750", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",2} } },
{ "GeForce GTX 750 Ti", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX 980", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX TITAN X", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
@@ -186,8 +192,11 @@ const Database::DatabaseEntry Database::CopyDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",2}, {"COPY_WPT",1} } },
+ { "GeForce GTX 1070", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",4}, {"COPY_WPT",1} } },
{ "GeForce GTX 480", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",1} } },
+ { "GeForce GTX 670", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",1} } },
{ "GeForce GTX 680", { {"COPY_DIMX",16}, {"COPY_DIMY",32}, {"COPY_VW",2}, {"COPY_WPT",1} } },
+ { "GeForce GTX 750", { {"COPY_DIMX",8}, {"COPY_DIMY",16}, {"COPY_VW",2}, {"COPY_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",1} } },
{ "GeForce GTX 980", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",2}, {"COPY_WPT",1} } },
{ "GeForce GTX TITAN", { {"COPY_DIMX",16}, {"COPY_DIMY",32}, {"COPY_VW",2}, {"COPY_WPT",2} } },
@@ -241,8 +250,11 @@ const Database::DatabaseEntry Database::CopyComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 1070", { {"COPY_DIMX",8}, {"COPY_DIMY",32}, {"COPY_VW",1}, {"COPY_WPT",4} } },
{ "GeForce GTX 480", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 670", { {"COPY_DIMX",16}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX 680", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
+ { "GeForce GTX 750", { {"COPY_DIMX",32}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"COPY_DIMX",32}, {"COPY_DIMY",16}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX 980", { {"COPY_DIMX",8}, {"COPY_DIMY",8}, {"COPY_VW",1}, {"COPY_WPT",1} } },
{ "GeForce GTX TITAN", { {"COPY_DIMX",16}, {"COPY_DIMY",16}, {"COPY_VW",1}, {"COPY_WPT",1} } },
diff --git a/src/database/kernels/pad.hpp b/src/database/kernels/pad.hpp
index db4df9f0..6bb0bc86 100644
--- a/src/database/kernels/pad.hpp
+++ b/src/database/kernels/pad.hpp
@@ -75,8 +75,11 @@ const Database::DatabaseEntry Database::PadSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 1070", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 480", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",4} } },
+ { "GeForce GTX 670", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",2} } },
{ "GeForce GTX 680", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 750", { {"PAD_DIMX",32}, {"PAD_DIMY",16}, {"PAD_WPTX",4}, {"PAD_WPTY",2} } },
{ "GeForce GTX 750 Ti", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",4}, {"PAD_WPTY",1} } },
{ "GeForce GTX 980", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX TITAN", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
@@ -139,15 +142,18 @@ const Database::DatabaseEntry Database::PadComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 1070", { {"PAD_DIMX",8}, {"PAD_DIMY",32}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 480", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 670", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } },
{ "GeForce GTX 680", { {"PAD_DIMX",16}, {"PAD_DIMY",32}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } },
+ { "GeForce GTX 750", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
{ "GeForce GTX 750 Ti", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 980", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX TITAN", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
{ "GeForce GTX TITAN X", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "Tesla K20m", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } },
{ "Tesla K40m", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
- { "default", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "default", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
}
},
{ // Default
@@ -194,8 +200,11 @@ const Database::DatabaseEntry Database::PadDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 1070", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 480", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 670", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",2}, {"PAD_WPTY",1} } },
{ "GeForce GTX 680", { {"PAD_DIMX",32}, {"PAD_DIMY",32}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } },
+ { "GeForce GTX 750", { {"PAD_DIMX",32}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 750 Ti", { {"PAD_DIMX",8}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 980", { {"PAD_DIMX",8}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX TITAN", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
@@ -249,8 +258,11 @@ const Database::DatabaseEntry Database::PadComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 1070", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",2}, {"PAD_WPTY",2} } },
{ "GeForce GTX 480", { {"PAD_DIMX",16}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 670", { {"PAD_DIMX",32}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 680", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
+ { "GeForce GTX 750", { {"PAD_DIMX",8}, {"PAD_DIMY",8}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 750 Ti", { {"PAD_DIMX",32}, {"PAD_DIMY",32}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX 980", { {"PAD_DIMX",16}, {"PAD_DIMY",16}, {"PAD_WPTX",1}, {"PAD_WPTY",1} } },
{ "GeForce GTX TITAN", { {"PAD_DIMX",8}, {"PAD_DIMY",32}, {"PAD_WPTX",1}, {"PAD_WPTY",2} } },
diff --git a/src/database/kernels/padtranspose.hpp b/src/database/kernels/padtranspose.hpp
index 7fedd15a..b117f9fe 100644
--- a/src/database/kernels/padtranspose.hpp
+++ b/src/database/kernels/padtranspose.hpp
@@ -75,8 +75,11 @@ const Database::DatabaseEntry Database::PadtransposeSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",2} } },
+ { "GeForce GTX 1070", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 480", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
+ { "GeForce GTX 670", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 680", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
+ { "GeForce GTX 750", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 750 Ti", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 980", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
@@ -139,8 +142,11 @@ const Database::DatabaseEntry Database::PadtransposeComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 1070", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 480", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 670", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 680", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 750", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 750 Ti", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 980", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
@@ -194,8 +200,11 @@ const Database::DatabaseEntry Database::PadtransposeDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 1070", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 480", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 670", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 680", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 750", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 750 Ti", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 980", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
@@ -249,8 +258,11 @@ const Database::DatabaseEntry Database::PadtransposeComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 1070", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 480", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 670", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 680", { {"PADTRA_PAD",1}, {"PADTRA_TILE",32}, {"PADTRA_WPT",1} } },
+ { "GeForce GTX 750", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",2} } },
{ "GeForce GTX 980", { {"PADTRA_PAD",0}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"PADTRA_PAD",1}, {"PADTRA_TILE",16}, {"PADTRA_WPT",1} } },
diff --git a/src/database/kernels/transpose.hpp b/src/database/kernels/transpose.hpp
index 4229e39f..8faa53f4 100644
--- a/src/database/kernels/transpose.hpp
+++ b/src/database/kernels/transpose.hpp
@@ -75,15 +75,18 @@ const Database::DatabaseEntry Database::TransposeSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"TRA_DIM",32}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
+ { "GeForce GTX 1070", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",4} } },
{ "GeForce GTX 480", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } },
+ { "GeForce GTX 670", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
{ "GeForce GTX 680", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
+ { "GeForce GTX 750", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",8} } },
{ "GeForce GTX 750 Ti", { {"TRA_DIM",32}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } },
{ "GeForce GTX 980", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } },
{ "GeForce GTX TITAN X", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } },
{ "Tesla K20m", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } },
{ "Tesla K40m", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",4} } },
- { "default", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
+ { "default", { {"TRA_DIM",4}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
}
},
{ // Default
@@ -133,8 +136,11 @@ const Database::DatabaseEntry Database::TransposeComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
+ { "GeForce GTX 1070", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
{ "GeForce GTX 480", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
+ { "GeForce GTX 670", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
{ "GeForce GTX 680", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
+ { "GeForce GTX 750", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 980", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
@@ -188,8 +194,11 @@ const Database::DatabaseEntry Database::TransposeDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
+ { "GeForce GTX 1070", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
{ "GeForce GTX 480", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } },
+ { "GeForce GTX 670", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
{ "GeForce GTX 680", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",2} } },
+ { "GeForce GTX 750", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"TRA_DIM",32}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 980", { {"TRA_DIM",16}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } },
{ "GeForce GTX TITAN", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",2} } },
@@ -237,15 +246,18 @@ const Database::DatabaseEntry Database::TransposeComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
+ { "GeForce GTX 1070", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 480", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
+ { "GeForce GTX 670", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
{ "GeForce GTX 680", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",1}, {"TRA_WPT",1} } },
+ { "GeForce GTX 750", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 750 Ti", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX 980", { {"TRA_DIM",32}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX TITAN", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "GeForce GTX TITAN X", { {"TRA_DIM",32}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "Tesla K20m", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
{ "Tesla K40m", { {"TRA_DIM",16}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
- { "default", { {"TRA_DIM",8}, {"TRA_PAD",1}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
+ { "default", { {"TRA_DIM",8}, {"TRA_PAD",0}, {"TRA_SHUFFLE",0}, {"TRA_WPT",1} } },
}
},
{ // Default
diff --git a/src/database/kernels/xaxpy.hpp b/src/database/kernels/xaxpy.hpp
index d8088ca2..b6ea1d7e 100644
--- a/src/database/kernels/xaxpy.hpp
+++ b/src/database/kernels/xaxpy.hpp
@@ -75,8 +75,11 @@ const Database::DatabaseEntry Database::XaxpySingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"VW",2}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 1070", { {"VW",2}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX 480", { {"VW",4}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 670", { {"VW",2}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX 680", { {"VW",2}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 750", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX 750 Ti", { {"VW",1}, {"WGS",1024}, {"WPT",1} } },
{ "GeForce GTX 980", { {"VW",1}, {"WGS",1024}, {"WPT",1} } },
{ "GeForce GTX TITAN", { {"VW",4}, {"WGS",256}, {"WPT",1} } },
@@ -139,8 +142,11 @@ const Database::DatabaseEntry Database::XaxpyComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"VW",1}, {"WGS",512}, {"WPT",1} } },
+ { "GeForce GTX 1070", { {"VW",1}, {"WGS",64}, {"WPT",2} } },
{ "GeForce GTX 480", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
+ { "GeForce GTX 670", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
{ "GeForce GTX 680", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
+ { "GeForce GTX 750", { {"VW",1}, {"WGS",512}, {"WPT",1} } },
{ "GeForce GTX 750 Ti", { {"VW",1}, {"WGS",512}, {"WPT",1} } },
{ "GeForce GTX 980", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX TITAN", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
@@ -194,8 +200,11 @@ const Database::DatabaseEntry Database::XaxpyDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 1070", { {"VW",1}, {"WGS",64}, {"WPT",8} } },
{ "GeForce GTX 480", { {"VW",2}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 670", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX 680", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 750", { {"VW",1}, {"WGS",128}, {"WPT",1} } },
{ "GeForce GTX 750 Ti", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
{ "GeForce GTX 980", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
{ "GeForce GTX TITAN", { {"VW",2}, {"WGS",1024}, {"WPT",1} } },
@@ -249,8 +258,11 @@ const Database::DatabaseEntry Database::XaxpyComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 1070", { {"VW",1}, {"WGS",64}, {"WPT",2} } },
{ "GeForce GTX 480", { {"VW",1}, {"WGS",128}, {"WPT",1} } },
+ { "GeForce GTX 670", { {"VW",1}, {"WGS",256}, {"WPT",1} } },
{ "GeForce GTX 680", { {"VW",1}, {"WGS",64}, {"WPT",1} } },
+ { "GeForce GTX 750", { {"VW",1}, {"WGS",1024}, {"WPT",1} } },
{ "GeForce GTX 750 Ti", { {"VW",1}, {"WGS",256}, {"WPT",2} } },
{ "GeForce GTX 980", { {"VW",1}, {"WGS",1024}, {"WPT",1} } },
{ "GeForce GTX TITAN", { {"VW",1}, {"WGS",64}, {"WPT",4} } },
diff --git a/src/database/kernels/xdot.hpp b/src/database/kernels/xdot.hpp
index 48288f95..8670c50b 100644
--- a/src/database/kernels/xdot.hpp
+++ b/src/database/kernels/xdot.hpp
@@ -60,8 +60,11 @@ const Database::DatabaseEntry Database::XdotSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WGS2",32} } },
+ { "GeForce GTX 1070", { {"WGS1",128}, {"WGS2",1024} } },
{ "GeForce GTX 480", { {"WGS1",512}, {"WGS2",32} } },
+ { "GeForce GTX 670", { {"WGS1",512}, {"WGS2",1024} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WGS2",128} } },
+ { "GeForce GTX 750", { {"WGS1",128}, {"WGS2",32} } },
{ "GeForce GTX 980", { {"WGS1",256}, {"WGS2",32} } },
{ "GeForce GTX TITAN X", { {"WGS1",256}, {"WGS2",32} } },
{ "Tesla K20m", { {"WGS1",1024}, {"WGS2",32} } },
@@ -106,8 +109,11 @@ const Database::DatabaseEntry Database::XdotComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",64}, {"WGS2",32} } },
+ { "GeForce GTX 1070", { {"WGS1",128}, {"WGS2",32} } },
{ "GeForce GTX 480", { {"WGS1",512}, {"WGS2",32} } },
+ { "GeForce GTX 670", { {"WGS1",256}, {"WGS2",32} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WGS2",64} } },
+ { "GeForce GTX 750", { {"WGS1",64}, {"WGS2",32} } },
{ "GeForce GTX 980", { {"WGS1",256}, {"WGS2",64} } },
{ "GeForce GTX TITAN X", { {"WGS1",256}, {"WGS2",32} } },
{ "Tesla K20m", { {"WGS1",512}, {"WGS2",32} } },
@@ -144,12 +150,15 @@ const Database::DatabaseEntry Database::XdotDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WGS2",32} } },
+ { "GeForce GTX 1070", { {"WGS1",128}, {"WGS2",512} } },
{ "GeForce GTX 480", { {"WGS1",512}, {"WGS2",32} } },
+ { "GeForce GTX 670", { {"WGS1",256}, {"WGS2",32} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WGS2",64} } },
+ { "GeForce GTX 750", { {"WGS1",64}, {"WGS2",256} } },
{ "GeForce GTX 980", { {"WGS1",128}, {"WGS2",32} } },
{ "GeForce GTX TITAN X", { {"WGS1",256}, {"WGS2",32} } },
{ "Tesla K20m", { {"WGS1",512}, {"WGS2",32} } },
- { "default", { {"WGS1",128}, {"WGS2",32} } },
+ { "default", { {"WGS1",64}, {"WGS2",32} } },
}
},
{ // Default
@@ -182,8 +191,11 @@ const Database::DatabaseEntry Database::XdotComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",64}, {"WGS2",32} } },
+ { "GeForce GTX 1070", { {"WGS1",128}, {"WGS2",64} } },
{ "GeForce GTX 480", { {"WGS1",512}, {"WGS2",32} } },
+ { "GeForce GTX 670", { {"WGS1",512}, {"WGS2",128} } },
{ "GeForce GTX 680", { {"WGS1",256}, {"WGS2",64} } },
+ { "GeForce GTX 750", { {"WGS1",256}, {"WGS2",32} } },
{ "GeForce GTX 980", { {"WGS1",64}, {"WGS2",32} } },
{ "GeForce GTX TITAN X", { {"WGS1",128}, {"WGS2",32} } },
{ "Tesla K20m", { {"WGS1",128}, {"WGS2",32} } },
diff --git a/src/database/kernels/xgemm.hpp b/src/database/kernels/xgemm.hpp
index 27cebc8a..f2cd46d1 100644
--- a/src/database/kernels/xgemm.hpp
+++ b/src/database/kernels/xgemm.hpp
@@ -69,15 +69,18 @@ const Database::DatabaseEntry Database::XgemmSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",4} } },
+ { "GeForce GTX 1070", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",4}, {"VWN",1} } },
{ "GeForce GTX 480", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",2} } },
+ { "GeForce GTX 670", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",4} } },
{ "GeForce GTX 680", { {"KWG",32}, {"KWI",8}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",4}, {"VWN",2} } },
+ { "GeForce GTX 750", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",2} } },
{ "GeForce GTX 750 Ti", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",4}, {"VWN",4} } },
{ "GeForce GTX 980", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",4}, {"VWN",8} } },
{ "GeForce GTX TITAN", { {"KWG",16}, {"KWI",8}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",2} } },
{ "GeForce GTX TITAN X", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",4}, {"VWN",8} } },
{ "Tesla K20m", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",4} } },
{ "Tesla K40m", { {"KWG",16}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",4} } },
- { "default", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",2} } },
+ { "default", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
}
},
{ // Default
@@ -133,8 +136,11 @@ const Database::DatabaseEntry Database::XgemmComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"KWG",16}, {"KWI",8}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",128}, {"SA",1}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",4} } },
+ { "GeForce GTX 1070", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",4} } },
{ "GeForce GTX 480", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",128}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",2} } },
+ { "GeForce GTX 670", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",32}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",1} } },
{ "GeForce GTX 680", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",32}, {"NWG",128}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",2}, {"VWN",2} } },
+ { "GeForce GTX 750", { {"KWG",16}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",2} } },
{ "GeForce GTX 750 Ti", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",8}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",4} } },
{ "GeForce GTX 980", { {"KWG",32}, {"KWI",8}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",1} } },
{ "GeForce GTX TITAN", { {"KWG",16}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
@@ -188,8 +194,11 @@ const Database::DatabaseEntry Database::XgemmDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",2} } },
+ { "GeForce GTX 1070", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",8} } },
{ "GeForce GTX 480", { {"KWG",16}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",64}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",2} } },
+ { "GeForce GTX 670", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",32}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
{ "GeForce GTX 680", { {"KWG",32}, {"KWI",8}, {"MDIMA",8}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",128}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",4} } },
+ { "GeForce GTX 750", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",32}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",8}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",2}, {"VWN",1} } },
{ "GeForce GTX 750 Ti", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",1} } },
{ "GeForce GTX 980", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",32}, {"NDIMC",32}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",2}, {"VWN",4} } },
{ "GeForce GTX TITAN", { {"KWG",16}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",128}, {"SA",1}, {"SB",1}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",2} } },
@@ -243,8 +252,11 @@ const Database::DatabaseEntry Database::XgemmComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",16}, {"NWG",64}, {"SA",1}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",1}, {"VWN",1} } },
+ { "GeForce GTX 1070", { {"KWG",32}, {"KWI",8}, {"MDIMA",32}, {"MDIMC",16}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",1}, {"VWN",4} } },
{ "GeForce GTX 480", { {"KWG",16}, {"KWI",2}, {"MDIMA",32}, {"MDIMC",32}, {"MWG",32}, {"NDIMB",32}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
+ { "GeForce GTX 670", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",16}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",64}, {"SA",1}, {"SB",0}, {"STRM",0}, {"STRN",1}, {"VWM",1}, {"VWN",2} } },
{ "GeForce GTX 680", { {"KWG",16}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",64}, {"NDIMB",16}, {"NDIMC",32}, {"NWG",32}, {"SA",0}, {"SB",1}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
+ { "GeForce GTX 750", { {"KWG",32}, {"KWI",2}, {"MDIMA",8}, {"MDIMC",32}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",64}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",4} } },
{ "GeForce GTX 750 Ti", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",16}, {"NDIMB",8}, {"NDIMC",8}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",0}, {"STRN",0}, {"VWM",1}, {"VWN",4} } },
{ "GeForce GTX 980", { {"KWG",16}, {"KWI",2}, {"MDIMA",16}, {"MDIMC",8}, {"MWG",32}, {"NDIMB",8}, {"NDIMC",16}, {"NWG",128}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",1}, {"VWM",2}, {"VWN",2} } },
{ "GeForce GTX TITAN X", { {"KWG",32}, {"KWI",8}, {"MDIMA",16}, {"MDIMC",16}, {"MWG",128}, {"NDIMB",16}, {"NDIMC",16}, {"NWG",32}, {"SA",0}, {"SB",0}, {"STRM",1}, {"STRN",0}, {"VWM",1}, {"VWN",1} } },
diff --git a/src/database/kernels/xgemv.hpp b/src/database/kernels/xgemv.hpp
index ce258f2f..0f66e497 100644
--- a/src/database/kernels/xgemv.hpp
+++ b/src/database/kernels/xgemv.hpp
@@ -68,8 +68,11 @@ const Database::DatabaseEntry Database::XgemvSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",256}, {"WPT1",1}, {"VW2",2}, {"WGS2",256}, {"WPT2",2}, {"VW3",2}, {"WGS3",64}, {"WPT3",2} } },
+ { "GeForce GTX 1070", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",2}, {"WGS3",64}, {"WPT3",2} } },
{ "GeForce GTX 480", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",128}, {"WPT2",1}, {"VW3",2}, {"WGS3",64}, {"WPT3",2} } },
+ { "GeForce GTX 670", { {"WGS1",64}, {"WPT1",1}, {"VW2",2}, {"WGS2",256}, {"WPT2",2}, {"VW3",2}, {"WGS3",64}, {"WPT3",2} } },
{ "GeForce GTX 680", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",128}, {"WPT2",1}, {"VW3",2}, {"WGS3",128}, {"WPT3",2} } },
+ { "GeForce GTX 750", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",2}, {"WGS3",128}, {"WPT3",2} } },
{ "GeForce GTX 750 Ti", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",4}, {"WGS3",128}, {"WPT3",4} } },
{ "GeForce GTX 980", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",128}, {"WPT3",1} } },
{ "GeForce GTX TITAN", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",256}, {"WPT3",1} } },
@@ -125,8 +128,11 @@ const Database::DatabaseEntry Database::XgemvComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 1070", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1} } },
{ "GeForce GTX 480", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 670", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
{ "GeForce GTX 680", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 750", { {"WGS1",128}, {"WPT1",1} } },
{ "GeForce GTX 750 Ti", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1} } },
{ "GeForce GTX TITAN", { {"WGS1",256}, {"WPT1",1} } },
{ "default", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
@@ -169,8 +175,11 @@ const Database::DatabaseEntry Database::XgemvDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 1070", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",128}, {"WPT3",1} } },
{ "GeForce GTX 480", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 670", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",128}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",128}, {"WPT2",1}, {"VW3",2}, {"WGS3",128}, {"WPT3",2} } },
+ { "GeForce GTX 750", { {"WGS1",64}, {"WPT1",1}, {"VW2",2}, {"WGS2",256}, {"WPT2",2}, {"VW3",2}, {"WGS3",64}, {"WPT3",2} } },
{ "GeForce GTX 750 Ti", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",2}, {"WGS3",256}, {"WPT3",2} } },
{ "GeForce GTX 980", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
{ "GeForce GTX TITAN", { {"WGS1",256}, {"WPT1",1}, {"VW2",1}, {"WGS2",256}, {"WPT2",1}, {"VW3",1}, {"WGS3",256}, {"WPT3",1} } },
@@ -218,6 +227,7 @@ const Database::DatabaseEntry Database::XgemvComplexDouble = {
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",128}, {"WPT2",1}, {"VW3",1}, {"WGS3",256}, {"WPT3",1} } },
{ "GeForce GTX 480", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
+ { "GeForce GTX 670", { {"WGS1",128}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
{ "default", { {"WGS1",64}, {"WPT1",1}, {"VW2",1}, {"WGS2",64}, {"WPT2",1}, {"VW3",1}, {"WGS3",64}, {"WPT3",1} } },
}
},
diff --git a/src/database/kernels/xger.hpp b/src/database/kernels/xger.hpp
index 3727cc57..f02482a0 100644
--- a/src/database/kernels/xger.hpp
+++ b/src/database/kernels/xger.hpp
@@ -67,10 +67,13 @@ const Database::DatabaseEntry Database::XgerSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WGS2",1}, {"WPT",2} } },
+ { "GeForce GTX 1070", { {"WGS1",512}, {"WGS2",1}, {"WPT",1} } },
{ "GeForce GTX 480", { {"WGS1",256}, {"WGS2",1}, {"WPT",4} } },
+ { "GeForce GTX 670", { {"WGS1",32}, {"WGS2",8}, {"WPT",2} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WGS2",1}, {"WPT",4} } },
+ { "GeForce GTX 750", { {"WGS1",64}, {"WGS2",16}, {"WPT",4} } },
{ "GeForce GTX TITAN", { {"WGS1",32}, {"WGS2",4}, {"WPT",2} } },
- { "default", { {"WGS1",32}, {"WGS2",1}, {"WPT",2} } },
+ { "default", { {"WGS1",32}, {"WGS2",1}, {"WPT",1} } },
}
},
{ // Default
@@ -118,8 +121,11 @@ const Database::DatabaseEntry Database::XgerComplexSingle = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",64}, {"WGS2",4}, {"WPT",2} } },
+ { "GeForce GTX 1070", { {"WGS1",16}, {"WGS2",64}, {"WPT",2} } },
{ "GeForce GTX 480", { {"WGS1",128}, {"WGS2",2}, {"WPT",2} } },
+ { "GeForce GTX 670", { {"WGS1",16}, {"WGS2",32}, {"WPT",2} } },
{ "GeForce GTX 680", { {"WGS1",32}, {"WGS2",4}, {"WPT",2} } },
+ { "GeForce GTX 750", { {"WGS1",32}, {"WGS2",16}, {"WPT",4} } },
{ "GeForce GTX TITAN", { {"WGS1",16}, {"WGS2",16}, {"WPT",2} } },
{ "default", { {"WGS1",16}, {"WGS2",2}, {"WPT",2} } },
}
@@ -161,10 +167,13 @@ const Database::DatabaseEntry Database::XgerDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",128}, {"WGS2",8}, {"WPT",2} } },
+ { "GeForce GTX 1070", { {"WGS1",32}, {"WGS2",8}, {"WPT",1} } },
{ "GeForce GTX 480", { {"WGS1",32}, {"WGS2",4}, {"WPT",2} } },
+ { "GeForce GTX 670", { {"WGS1",32}, {"WGS2",32}, {"WPT",2} } },
{ "GeForce GTX 680", { {"WGS1",128}, {"WGS2",4}, {"WPT",2} } },
+ { "GeForce GTX 750", { {"WGS1",256}, {"WGS2",2}, {"WPT",2} } },
{ "GeForce GTX TITAN", { {"WGS1",16}, {"WGS2",8}, {"WPT",2} } },
- { "default", { {"WGS1",16}, {"WGS2",4}, {"WPT",2} } },
+ { "default", { {"WGS1",16}, {"WGS2",2}, {"WPT",1} } },
}
},
{ // Default
@@ -204,8 +213,11 @@ const Database::DatabaseEntry Database::XgerComplexDouble = {
{ // NVIDIA GPUs
kDeviceTypeGPU, "NVIDIA", {
{ "GRID K520", { {"WGS1",16}, {"WGS2",8}, {"WPT",2} } },
+ { "GeForce GTX 1070", { {"WGS1",8}, {"WGS2",128}, {"WPT",1} } },
{ "GeForce GTX 480", { {"WGS1",64}, {"WGS2",2}, {"WPT",2} } },
+ { "GeForce GTX 670", { {"WGS1",8}, {"WGS2",16}, {"WPT",2} } },
{ "GeForce GTX 680", { {"WGS1",8}, {"WGS2",16}, {"WPT",1} } },
+ { "GeForce GTX 750", { {"WGS1",8}, {"WGS2",32}, {"WPT",4} } },
{ "GeForce GTX TITAN", { {"WGS1",32}, {"WGS2",4}, {"WPT",2} } },
{ "default", { {"WGS1",8}, {"WGS2",2}, {"WPT",1} } },
}