jefflarkin

#!/bin/bash
# USAGE: Add between aprun options and executable
# For Example: aprun -n 16 -N 1 ./foo arg1 arg2
# Becomes: aprun -n 16 -N 1 ./profile.sh ./foo arg1 arg2

# Enable command-line profiler
export COMPUTE_PROFILE=1

# Set output to CSV (optional)
export COMPUTE_PROFILE_CSV=1

jefflarkin

//Macro for checking cuda errors following a cuda launch or api call
#define cudaCheckError() {                                          \
 cudaError_t e=cudaGetLastError();                                 \
 if(e!=cudaSuccess) {                                              \
   printf("Cuda failure %s:%d: '%s'\n",__FILE__,__LINE__,cudaGetErrorString(e));           \
   exit(0); \
 }                                                                 \
}

jefflarkin

//Macro for checking cuda errors following a cuda launch or api call
#define cudaCheckError() {                                          \
 cudaError_t e=cudaGetLastError();                                 \
 if(e!=cudaSuccess) {                                              \
   printf("Cuda failure %s:%d: '%s'\n",__FILE__,__LINE__,cudaGetErrorString(e));           \
   exit(0); \
 }                                                                 \
}

jefflarkin

#!/bin/bash
# USAGE: Add between aprun options and executable
# For Example: aprun -n 16 -N 1 ./foo arg1 arg2
# Becomes: aprun -n 16 -N 1 ./nvprof.sh ./foo arg1 arg2

# Give each *node* a separate file
LOG=profile_$(hostname).nvp

# Stripe each profile file by 1 to share the load on large runs
lfs setstripe -c 1 $LOG

jefflarkin

/** Charlieplexing 6 LEDs with Arduino Uno     **/
/** Jeff Larkin [http://github.com/jefflarkin] **/
/*
 * Wiring Diagram:
 *
 * pin 2 ===== led0+, led1- ===== led4+, led5-
 * pin 4 ===== led1+, led0- ===== led2+, led3-
 * pin 6 ===== led3+, led2- ===== led5+, led4-
 *
 */

jefflarkin

var Hapi = require('hapi');
var server = Hapi.createServer(process.env.IP,parseInt(process.env.PORT));

// Define the route
var hello = {
    handler: function (request) {
        request.reply({ greeting: 'hello world' });
    }
};

jefflarkin

! Fortran bindings for a small subset of the NVIDIA Tools Extensions library
module nvtx
  use iso_c_binding
  public :: nvtxrangepusha, nvtxrangepop
  public :: nvtxrangepushaargb
  interface
    ! Annotate the timeline with a message
    ! Parameters:
    ! * string : the message in a string format
    subroutine nvtxrangepusha(string) bind(C, name="nvtxRangePushA")

jefflarkin

#include <pthread.h>
#include <nvToolsExt.h>
#include <nvToolsExtCudaRt.h>
// Setup event category name
{{fn name MPI_Init}}
  nvtxNameCategoryA(999, "MPI");
  {{callfn}}
  int rank;
  PMPI_Comm_rank(MPI_COMM_WORLD, &rank);
  char name[256];

jefflarkin

Keybase proof

I hereby claim:

• I am jefflarkin on github.
• I am jefflarkin (https://keybase.io/jefflarkin) on keybase.
• I have a public key whose fingerprint is 13D8 BCCC 1A50 57D3 FB4F E33A 950C 167F 0C1A 041D

To claim this, I am signing this object:

jefflarkin

#!/bin/bash
# USAGE: Add between aprun options and executable
# For Example: aprun -n 16 -N 1 ./foo arg1 arg2
# Becomes: aprun -n 16 -N 1 ./nvprof.sh ./foo arg1 arg2
export PMI_NO_FORK=1

# Give each *rank* a separate file
LOG=timeline_$ALPS_APP_PE.nvprof

# Set the process and context names