Add parsec_advise_data_on_device for zpotrf_L

ICLDisco · Jun 14, 2024 · a4b81a9 · a4b81a9
1 parent 36d60da
commit a4b81a9
Show file tree

Hide file tree

Showing 4 changed files with 101 additions and 11 deletions.
diff --git a/parsec b/parsec
diff --git a/src/zpotrf_L.jdf b/src/zpotrf_L.jdf
@@ -87,6 +87,49 @@ smallnb       [type = "int" hidden = on default = "descA->mb" ]
 CuHandlesID   [type = "int" hidden = on default = -1 ]
 POWorkspaceID [type = "int" hidden = on default = -1 ]
 
+nb_gpu_devices      [ type = "int"   hidden = on default = 0 ]
+gpu_device_index    [ type = "int *" hidden = on default = "NULL"]
+gpu_rows            [ type = "int" hidden = on default = 1]
+gpu_cols            [ type = "int" hidden = on default = 1]
+grid_rows           [ type = "int" hidden = on default = 1]
+grid_cols           [ type = "int" hidden = on default = 1]
+
+
+/**************************************************
+ *               potrf_bind_A                     *
+ **************************************************/
+potrf_bind_A(m, n)
+
+// Execution space
+m = 0 .. descA->mt-1
+n = 0 .. m
+
+loc_A = %{ return LOC(descA, m, n); %}
+
+// Parallel partitioning
+:descA(m, n)
+
+READ A <- ddescA(m, n)                                        [ type        = %{ return ADTT_READ(ddescA, loc_A, DEFAULT, TILED); %}
+                                                                type_data   = %{ return ADTT_READ(ddescA, loc_A, DEFAULT, LAPACK); %} ]
+       -> (m == 0 && n == 0) ? T potrf_zpotrf(0)                      
+       -> (n == 0)? C potrf_ztrsm(m, n)                               
+       -> (m == n && n > 0) ? T potrf_zherk(0, m)                     
+       -> (m != n && n > 0) ? C potrf_zgemm(m, n, 0)                  
+
+BODY
+{
+#if defined(PARSEC_HAVE_DEV_CUDA_SUPPORT)  || defined(PARSEC_HAVE_DEV_HIP_SUPPORT)
+    if( nb_gpu_devices > 0 ) {
+        int g = (m / grid_rows % gpu_rows) * gpu_cols + n / grid_cols % gpu_cols; 
+        parsec_advise_data_on_device( _f_A->original,
+                                    gpu_device_index[g],
+                                    PARSEC_DEV_DATA_ADVICE_PREFERRED_DEVICE );
+    }
+#endif
+}
+END
+
+
 /**************************************************
  *               potrf_zpotrf                     *
  **************************************************/
@@ -106,8 +149,7 @@ loc_T = %{ return LOC(descA, k, k); %}
 
 // Parameters
 
-RW T <- (k == 0) ? ddescA(k, k)            [ type        = %{ return ADTT_READ(ddescA, loc_T, DEFAULT, TILED); %}
-                                             type_data   = %{ return ADTT_READ(ddescA, loc_T, DEFAULT, LAPACK); %} ]
+RW T <- (k == 0) ? A potrf_bind_A(k, k)    [ type_remote = %{ return ADTT_DC(ddescA, loc_T, DEFAULT, TILED); %} ] 
      <- (k != 0) ? T potrf_zherk(k-1, k)   [ type_remote = %{ return ADTT_DC(ddescA, loc_T, DEFAULT, TILED); %} ]
      -> T potrf_ztrsm(k+1..descA->mt-1, k) /* dep OUT: rely on datacopy dtt for sending */
      -> ddescA(k, k)                       [ type        = %{ return ADTT_CP(_f_T, ddescA, loc_T, DEFAULT); %}
@@ -207,8 +249,7 @@ loc_C = %{ return LOC(descA, m, k); %}
 
 // Parameters
 READ  T <- T potrf_zpotrf(k)                     [ type_remote = %{ return ADTT_DC(ddescA, loc_T, DEFAULT, TILED); %} ]
-RW    C <- (k == 0) ? ddescA(m, k)               [ type        = %{ return ADTT_READ(ddescA, loc_C, DEFAULT, TILED); %}
-                                                   type_data   = %{ return ADTT_READ(ddescA, loc_C, DEFAULT, LAPACK); %} ]
+RW    C <- (k == 0) ? A potrf_bind_A(m, k)       [ type_remote = %{ return ADTT_DC(ddescA, loc_C, DEFAULT, TILED); %} ] 
         <- (k != 0) ? C potrf_zgemm(m, k, k-1)   [ type_remote = %{ return ADTT_DC(ddescA, loc_C, DEFAULT, TILED); %} ]
         -> A potrf_zherk(k, m)                   /* dep OUT: rely on datacopy dtt for sending */
         -> A potrf_zgemm(m, k+1..m-1, k)         /* dep OUT: rely on datacopy dtt for sending */
@@ -319,9 +360,8 @@ loc_T = %{ return LOC(descA, m, m); %}
 
 //Parameters
 READ  A <- C potrf_ztrsm(m, k)                [ type_remote = %{ return ADTT_DC(ddescA, loc_A, DEFAULT, TILED); %} ]
-RW    T <- (k == 0)   ? ddescA(m, m)          [ type        = %{ return ADTT_READ(ddescA, loc_T, DEFAULT, TILED); %}
-                                                type_data   = %{ return ADTT_READ(ddescA, loc_T, DEFAULT, LAPACK); %} ]
-        <- (k != 0)   ? T potrf_zherk(k-1, m) [ type_remote = %{ return ADTT_DC(ddescA, loc_T, DEFAULT, TILED); %} ]
+RW    T <- (k == 0)   ? A potrf_bind_A(m, m)  [ type_remote = %{ return ADTT_DC(ddescA, loc_T, DEFAULT, TILED); %} ] 
+        <- (k != 0)   ? T potrf_zherk(k-1, m) /* dep OUT: rely on datacopy dtt for sending */ 
         -> (m == k+1) ? T potrf_zpotrf(m)  : T potrf_zherk(k+1, m) /* dep OUT: rely on datacopy dtt for sending */
 
 ; (m >= (descA->mt - PRI_CHANGE)) ? (descA->mt - m) * (descA->mt - m) * (descA->mt - m) + 3 * (m - k) : PRI_MAX
@@ -422,8 +462,7 @@ loc_C = %{ return LOC(descA, m, n); %}
 // Parameters
 READ  A <- C potrf_ztrsm(m, k)                   [ type_remote = %{ return ADTT_DC(ddescA, loc_A, DEFAULT, TILED); %} ]
 READ  B <- C potrf_ztrsm(n, k)                   [ type_remote = %{ return ADTT_DC(ddescA, loc_B, DEFAULT, TILED); %} ]
-RW    C <- (k == 0)   ? ddescA(m, n)             [ type        = %{ return ADTT_READ(ddescA, loc_C, DEFAULT, TILED); %}
-                                                   type_data   = %{ return ADTT_READ(ddescA, loc_C, DEFAULT, LAPACK); %} ]
+RW    C <- (k == 0)   ? A potrf_bind_A(m, n)     [ type_remote = %{ return ADTT_DC(ddescA, loc_C, DEFAULT, TILED); %} ] 
         <- (k != 0)   ? C potrf_zgemm(m, n, k-1) [ type_remote = %{ return ADTT_DC(ddescA, loc_C, DEFAULT, TILED); %} ]
         -> (n == k+1) ? C potrf_ztrsm(m, n) : C potrf_zgemm(m, n, k+1) /* dep OUT: rely on datacopy dtt for sending */
 

diff --git a/src/zpotrf_wrapper.c b/src/zpotrf_wrapper.c
@@ -22,6 +22,7 @@
 #include "zpotrf_U.h"
 #include "zpotrf_L.h"
 #include "cores/dplasma_plasmatypes.h"
+#include "parsec/data_dist/matrix/sym_two_dim_rectangle_cyclic.h"
 
 #define MAX_SHAPES 1
 
@@ -104,6 +105,42 @@ static void destroy_workspace(void *_ws, void *_n)
     free(ws);
     (void)_n;
 }
+
+/* Find all devices */
+static void parsec_find_nb_devices(int **dev_index, int *nb) {
+    for(int i = 0; i < (int)parsec_nb_devices; i++) {
+        parsec_device_module_t *device = parsec_mca_device_get(i); 
+        if( PARSEC_DEV_CUDA == device->type || PARSEC_DEV_HIP == device->type ) {
+            (*nb)++;
+        }
+    }
+#if defined(DPLASMA_DEBUG)
+    if((*nb) == 0) { 
+        char hostname[256];
+        gethostname(hostname, 256);
+        fprintf(stderr, "No CUDA device found on rank %d on %s\n",
+                parsec->my_rank, hostname);
+    }
+#endif
+    *dev_index = (int *)malloc((*nb) * sizeof(int));
+    *nb = 0;
+    for(int i = 0; i < (int)parsec_nb_devices; i++) {
+        parsec_device_module_t *device = parsec_mca_device_get(i);
+        if( PARSEC_DEV_CUDA == device->type || PARSEC_DEV_HIP == device->type ) {
+            (*dev_index)[(*nb)++] = device->device_index;
+        }
+    }
+}
+
+/* Get the most suitable process grid */
+static int parsec_grid_calculation( int nb_process ) {
+    int P;
+    for( P = (int)(sqrt(nb_process + 1.0)); P > 0; P-- ) {
+        if( 0 == nb_process % P ) break;
+    }
+    return P;
+}
+
 #endif
 
 /**
@@ -210,6 +247,19 @@ dplasma_zpotrf_New( dplasma_enum_t uplo,
                                                            destroy_workspace, NULL,
                                                            zpotrf_create_workspace, parsec_zpotrf,
                                                            NULL);
+    int nb = 0, *dev_index;
+    parsec_find_nb_devices(&dev_index, &nb);
+    parsec_zpotrf->_g_nb_gpu_devices = nb;
+    parsec_zpotrf->_g_gpu_device_index = dev_index;
+    parsec_zpotrf->_g_gpu_cols = parsec_grid_calculation(nb); 
+    parsec_zpotrf->_g_gpu_rows = nb/parsec_zpotrf->_g_gpu_cols; 
+    parsec_zpotrf->_g_grid_rows = ((parsec_matrix_sym_block_cyclic_t *)A)->grid.rows;
+    parsec_zpotrf->_g_grid_cols = ((parsec_matrix_sym_block_cyclic_t *)A)->grid.cols;
+#if defined(DPLASMA_DEBUG)
+    printf("nb_gpu_devices %d gpu_rows %d gpu_cols %d grid_rows %d grid_cols %d\n",
+            parsec_zpotrf->_g_nb_gpu_devices, parsec_zpotrf->_g_gpu_rows,
+            parsec_zpotrf->_g_gpu_cols, parsec_zpotrf->_g_grid_rows, parsec_zpotrf->_g_grid_cols);
+#endif
 #else
     parsec_zpotrf->_g_CuHandlesID = PARSEC_INFO_ID_UNDEFINED;
     parsec_zpotrf->_g_POWorkspaceID = PARSEC_INFO_ID_UNDEFINED;

diff --git a/tests/testing_zpotrf.c b/tests/testing_zpotrf.c
@@ -18,7 +18,8 @@ int main(int argc, char ** argv)
 {
     parsec_context_t* parsec;
     int iparam[IPARAM_SIZEOF];
-    dplasma_enum_t uplo = dplasmaUpper;
+    //dplasma_enum_t uplo = dplasmaUpper;
+    dplasma_enum_t uplo = dplasmaLower;
     int info = 0;
     int ret = 0;