Merge pull request #1360 from peastman/debugcl

Debugging OpenCL errors on travis

.travis.yml

@@ -17,6 +17,7 @@ env:
 matrix:
   include:
     - sudo: required
+      dist: trusty
       env: ==CPU_OPENCL==
            OPENCL=true
            CUDA=false
@@ -33,7 +34,9 @@ matrix:
            -DOPENMM_BUILD_AMOEBA_PLUGIN=OFF
            -DOPENMM_BUILD_PYTHON_WRAPPERS=OFF
            -DOPENMM_BUILD_C_AND_FORTRAN_WRAPPERS=OFF
-           -DOPENMM_BUILD_EXAMPLES=OFF"
+           -DOPENMM_BUILD_EXAMPLES=OFF
+           -DOPENCL_INCLUDE_DIR=$HOME/AMDAPPSDK/include
+           -DOPENCL_LIBRARY=$HOME/AMDAPPSDK/lib/x86_64/libOpenCL.so"
       addons: {apt: {packages: []}}
 
     - sudo: required
@@ -107,9 +110,16 @@ before_install:
       sudo easy_install pytest;
     fi
   - if [[ "$OPENCL" == "true" ]]; then
-      sudo add-apt-repository "deb http://archive.ubuntu.com/ubuntu $(lsb_release -sc) main universe restricted multiverse";
-      sudo apt-get -yq update > /dev/null 2>&1 ;
-      sudo apt-get install -qq fglrx=2:8.960-0ubuntu1 opencl-headers;
+      wget https://jenkins.choderalab.org/userContent/AMD-APP-SDKInstaller-v3.0.130.135-GA-linux64.tar.bz2;
+      tar -xjf AMD-APP-SDK*.tar.bz2;
+      AMDAPPSDK=${HOME}/AMDAPPSDK;
+      export OPENCL_VENDOR_PATH=${AMDAPPSDK}/etc/OpenCL/vendors;
+      mkdir -p ${OPENCL_VENDOR_PATH};
+      sh AMD-APP-SDK*.sh --tar -xf -C ${AMDAPPSDK};
+      echo libamdocl64.so > ${OPENCL_VENDOR_PATH}/amdocl64.icd;
+      export LD_LIBRARY_PATH=${AMDAPPSDK}/lib/x86_64:${LD_LIBRARY_PATH};
+      chmod +x ${AMDAPPSDK}/bin/x86_64/clinfo;
+      ${AMDAPPSDK}/bin/x86_64/clinfo;
     fi
   # Install swig for Python wrappers. However, testing CUDA and OpenCL, we
   # skip the Python wrapper for speed. We're not using anaconda python,

platforms/cuda/src/kernels/customManyParticle.cu

@@ -59,7 +59,7 @@ inline __device__ real4 computeCross(real4 vec1, real4 vec2) {
 /**
  * Determine whether a particular interaction is in the list of exclusions.
  */
-inline __device__ bool isInteractionExcluded(int atom1, int atom2, int* __restrict__ exclusions, int* __restrict__ exclusionStartIndex) {
+inline __device__ bool isInteractionExcluded(int atom1, int atom2, const int* __restrict__ exclusions, const int* __restrict__ exclusionStartIndex) {
     int first = exclusionStartIndex[atom1];
     int last = exclusionStartIndex[atom1+1];
     for (int i = last-1; i >= first; i--) {
@@ -180,7 +180,7 @@ extern "C" __global__ void findNeighbors(real4 periodicBoxSize, real4 invPeriodi
         const real4* __restrict__ posq, const real4* __restrict__ blockCenter, const real4* __restrict__ blockBoundingBox, int2* __restrict__ neighborPairs,
         int* __restrict__ numNeighborPairs, int* __restrict__ numNeighborsForAtom, int maxNeighborPairs
 #ifdef USE_EXCLUSIONS
-        , int* __restrict__ exclusions, int* __restrict__ exclusionStartIndex
+        , const int* __restrict__ exclusions, const int* __restrict__ exclusionStartIndex
 #endif
         ) {
     __shared__ real3 positionCache[FIND_NEIGHBORS_WORKGROUP_SIZE];
@@ -265,7 +265,8 @@ extern "C" __global__ void findNeighbors(real4 periodicBoxSize, real4 invPeriodi
                 }
             }
         }
-        numNeighborsForAtom[atom1] = totalNeighborsForAtom1;
+        if (atom1 < NUM_ATOMS)
+            numNeighborsForAtom[atom1] = totalNeighborsForAtom1;
     }
 }
 
@@ -308,6 +309,7 @@ extern "C" __global__ void computeNeighborStartIndices(int* __restrict__ numNeig
             numNeighborsForAtom[globalIndex] = 0; // Clear this so the next kernel can use it as a counter
         }
         globalOffset += posBuffer[blockDim.x-1];
+        __syncthreads();
     }
     if (threadIdx.x == 0)
         neighborStartIndex[0] = 0;

platforms/opencl/src/kernels/customManyParticle.cl

@@ -55,7 +55,7 @@ inline real4 computeCross(real4 vec1, real4 vec2) {
 /**
  * Determine whether a particular interaction is in the list of exclusions.
  */
-inline bool isInteractionExcluded(int atom1, int atom2, __global int* restrict exclusions, __global int* restrict exclusionStartIndex) {
+inline bool isInteractionExcluded(int atom1, int atom2, __global const int* restrict exclusions, __global const int* restrict exclusionStartIndex) {
     int first = exclusionStartIndex[atom1];
     int last = exclusionStartIndex[atom1+1];
     for (int i = last-1; i >= first; i--) {
@@ -174,7 +174,7 @@ __kernel void findNeighbors(real4 periodicBoxSize, real4 invPeriodicBoxSize, rea
         __global const real4* restrict posq, __global const real4* restrict blockCenter, __global const real4* restrict blockBoundingBox, __global int2* restrict neighborPairs,
         __global int* restrict numNeighborPairs, __global int* restrict numNeighborsForAtom, int maxNeighborPairs
 #ifdef USE_EXCLUSIONS
-        , __global int* restrict exclusions, __global int* restrict exclusionStartIndex
+        , __global const int* restrict exclusions, __global const int* restrict exclusionStartIndex
 #endif
         ) {
     __local real4 positionCache[FIND_NEIGHBORS_WORKGROUP_SIZE];
@@ -264,7 +264,9 @@ __kernel void findNeighbors(real4 periodicBoxSize, real4 invPeriodicBoxSize, rea
                 }
             }
         }
-        numNeighborsForAtom[atom1] = totalNeighborsForAtom1;
+        if (atom1 < NUM_ATOMS)
+            numNeighborsForAtom[atom1] = totalNeighborsForAtom1;
+        SYNC_WARPS;
     }
 }
 
@@ -307,6 +309,7 @@ __kernel void computeNeighborStartIndices(__global int* restrict numNeighborsFor
             numNeighborsForAtom[globalIndex] = 0; // Clear this so the next kernel can use it as a counter
         }
         globalOffset += posBuffer[get_local_size(0)-1];
+        barrier(CLK_LOCAL_MEM_FENCE);
     }
     if (get_local_id(0) == 0)
         neighborStartIndex[0] = 0;