Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
tsoc
openmm
Commits
81cd9f5f
Commit
81cd9f5f
authored
Jan 18, 2011
by
Peter Eastman
Browse files
Fixed problems on compute level 1.2 GPUs
parent
ff5f5d5a
Changes
44
Show whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
38 additions
and
38 deletions
+38
-38
platforms/cuda/src/kernels/kSettle.cu
platforms/cuda/src/kernels/kSettle.cu
+2
-2
platforms/cuda/src/kernels/kShakeH.cu
platforms/cuda/src/kernels/kShakeH.cu
+3
-3
platforms/cuda/src/kernels/kVerletUpdate.cu
platforms/cuda/src/kernels/kVerletUpdate.cu
+1
-1
platforms/cuda/src/kernels/kVerletUpdate.h
platforms/cuda/src/kernels/kVerletUpdate.h
+2
-2
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaElectrostatic.h
...orms/cuda/src/kernels/kCalculateAmoebaCudaElectrostatic.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaFixedEAndGkFields.h
.../cuda/src/kernels/kCalculateAmoebaCudaFixedEAndGkFields.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaFixedEField.h
...tforms/cuda/src/kernels/kCalculateAmoebaCudaFixedEField.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.cu
...latforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.cu
+2
-2
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.h
...platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwoodEDiff.h
...orms/cuda/src/kernels/kCalculateAmoebaCudaKirkwoodEDiff.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaLocalForces.cu
...forms/cuda/src/kernels/kCalculateAmoebaCudaLocalForces.cu
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMapTorques.cu
...tforms/cuda/src/kernels/kCalculateAmoebaCudaMapTorques.cu
+5
-5
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.cu
...c/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.cu
+4
-4
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.h
...rc/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedField.cu
...uda/src/kernels/kCalculateAmoebaCudaMutualInducedField.cu
+3
-3
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedField.h
...cuda/src/kernels/kCalculateAmoebaCudaMutualInducedField.h
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPME.cu
...eba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPME.cu
+1
-1
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPmeFixedEField.cu
...ms/cuda/src/kernels/kCalculateAmoebaCudaPmeFixedEField.cu
+2
-2
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPmeMutualInducedField.cu
.../src/kernels/kCalculateAmoebaCudaPmeMutualInducedField.cu
+3
-3
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaReduce.h
...a/platforms/cuda/src/kernels/kCalculateAmoebaCudaReduce.h
+2
-2
No files found.
platforms/cuda/src/kernels/kSettle.cu
View file @
81cd9f5f
...
...
@@ -60,7 +60,7 @@ void GetSettleSim(gpuContext gpu)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_SHAKE_THREADS_PER_BLOCK
,
1
)
...
...
@@ -242,7 +242,7 @@ void kApplyFirstSettle(gpuContext gpu)
__global__
void
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_SHAKE_THREADS_PER_BLOCK
,
1
)
...
...
platforms/cuda/src/kernels/kShakeH.cu
View file @
81cd9f5f
...
...
@@ -67,7 +67,7 @@ void GetShakeHSim(gpuContext gpu)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_SHAKE_THREADS_PER_BLOCK
,
1
)
...
...
@@ -238,7 +238,7 @@ void kApplyFirstShake(gpuContext gpu)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_SHAKE_THREADS_PER_BLOCK
,
1
)
...
...
@@ -430,7 +430,7 @@ void kApplySecondShake_kernel()
__global__
void
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_SHAKE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_SHAKE_THREADS_PER_BLOCK
,
1
)
...
...
platforms/cuda/src/kernels/kVerletUpdate.cu
View file @
81cd9f5f
...
...
@@ -92,7 +92,7 @@ void kVerletUpdatePart2(gpuContext gpu)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_UPDATE_THREADS_PER_BLOCK
,
1
)
...
...
platforms/cuda/src/kernels/kVerletUpdate.h
View file @
81cd9f5f
...
...
@@ -33,7 +33,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_UPDATE_THREADS_PER_BLOCK
,
1
)
...
...
@@ -122,7 +122,7 @@ void kVerletUpdatePart1_kernel()
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_UPDATE_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_UPDATE_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaElectrostatic.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
384
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
128
,
1
)
#else
__launch_bounds__
(
64
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaFixedEAndGkFields.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
256
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
128
,
1
)
#else
__launch_bounds__
(
64
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaFixedEField.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_NONBOND_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_NONBOND_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_NONBOND_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.cu
View file @
81cd9f5f
...
...
@@ -1576,7 +1576,7 @@ static void kReduce_dBorn(amoebaGpuContext amoebaGpu )
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -1633,7 +1633,7 @@ __launch_bounds__(G8X_THREADS_PER_BLOCK, 1)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwood.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
512
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
128
,
1
)
#else
__launch_bounds__
(
64
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaKirkwoodEDiff.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
384
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
96
,
1
)
#else
__launch_bounds__
(
32
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaLocalForces.cu
View file @
81cd9f5f
...
...
@@ -298,7 +298,7 @@ __device__ void bicubic( float4 y, float4 y1i, float4 y2i, float4 y12i, float x1
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_LOCALFORCES_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_LOCALFORCES_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_LOCALFORCES_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMapTorques.cu
View file @
81cd9f5f
...
...
@@ -54,7 +54,7 @@ __device__ static void crossVector3( float* vector1, float* vector2, float* vect
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -345,7 +345,7 @@ void amoebaMapTorqueToForce_kernel( float* torque, int maxDiff, float* tempElecF
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -513,7 +513,7 @@ void amoebaMapTorqueToForceOld_kernel( float* torque, int maxDiff, float* tempEl
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -566,7 +566,7 @@ void amoebaMapTorqueReduce_kernel(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -620,7 +620,7 @@ void amoebaMapTorqueReduce_kernel2(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.cu
View file @
81cd9f5f
...
...
@@ -239,7 +239,7 @@ __device__ static int debugAccumulate( int index, float4* debugArray, float* fie
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -273,7 +273,7 @@ void kInitializeMutualInducedAndGkField_kernel(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -342,7 +342,7 @@ void kReduceMutualInducedAndGkFieldDelta_kernel( float* arrayOfDeltas1, float* a
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -375,7 +375,7 @@ void kSorUpdateMutualInducedAndGkField_kernel(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedAndGkFields.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
384
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
128
,
1
)
#else
__launch_bounds__
(
64
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedField.cu
View file @
81cd9f5f
...
...
@@ -106,7 +106,7 @@ __device__ void calculateMutualInducedFieldPairIxn_kernel( MutualInducedParticle
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -134,7 +134,7 @@ void kInitializeMutualInducedField_kernel(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -182,7 +182,7 @@ void kReduceMutualInducedFieldDelta_kernel(int numberOfEntries, float* arrayOfDe
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaMutualInducedField.h
View file @
81cd9f5f
...
...
@@ -29,7 +29,7 @@
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_NONBOND_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_NONBOND_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_NONBOND_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPME.cu
View file @
81cd9f5f
...
...
@@ -163,7 +163,7 @@ void kComputeAmoebaBsplines_kernel()
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
1024
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
512
,
1
)
#else
__launch_bounds__
(
256
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPmeFixedEField.cu
View file @
81cd9f5f
...
...
@@ -34,7 +34,7 @@ void GetCalculateAmoebaCudaPmeFixedEFieldSim(amoebaGpuContext amoebaGpu)
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -83,7 +83,7 @@ static void kReducePmeEFieldPolar_kernel( unsigned int fieldComponents, unsigned
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaPmeMutualInducedField.cu
View file @
81cd9f5f
...
...
@@ -228,7 +228,7 @@ __device__ void calculatePmeDirectMutualInducedFieldPairIxn_kernel( MutualInduce
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -256,7 +256,7 @@ static void kInitializeMutualInducedField_kernel(
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -313,7 +313,7 @@ static void kReduceMutualInducedFieldDelta_kernel(int numberOfEntries, float* ar
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
plugins/amoeba/platforms/cuda/src/kernels/kCalculateAmoebaCudaReduce.h
View file @
81cd9f5f
...
...
@@ -8,7 +8,7 @@ typedef unsigned int uint;
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
@@ -73,7 +73,7 @@ void METHOD_NAME(kCalculateAmoebaReduce, N2ToNBlockLevel)( float *N2Array, float
__global__
#if (__CUDA_ARCH__ >= 200)
__launch_bounds__
(
GF1XX_THREADS_PER_BLOCK
,
1
)
#elif (__CUDA_ARCH__ >= 1
3
0)
#elif (__CUDA_ARCH__ >= 1
2
0)
__launch_bounds__
(
GT2XX_THREADS_PER_BLOCK
,
1
)
#else
__launch_bounds__
(
G8X_THREADS_PER_BLOCK
,
1
)
...
...
Prev
1
2
3
Next
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment
