I used clock_gettime(CLOCK_THREAD_CPUTIME_ID, &ts) to get the CPU time of the thread but found that the time fetched through it on M1 was very small. The iPhone and non-M1 devices are fine.
Here is the code:
double CPUTime()
{
struct timespec ts;
int ret = clock_gettime(CLOCK_THREAD_CPUTIME_ID, &ts);
return ts.tv_sec + ts.tv_nsec/1000.0/1000.0/1000.0;
}
double RealTime()
{
struct timespec ts;
int ret = clock_gettime(CLOCK_MONOTONIC, &ts);
return ts.tv_sec + ts.tv_nsec/1000.0/1000.0/1000.0;
}
void longTask()
{
double begin = RealTime();
double limit = 1.0;
static double g_result = 0;
while (1) {
for (int i=0; i<1000000; i++) {
g_result = sin(g_result+5);
}
double now = RealTime();
if (now - begin >= limit) break;
}
return;
}
void testClocks()
{
double t1, t0, cpuTimeCost, realTimeCost;
t0 = CPUTime();
longTask();
t1 = CPUTime();
cpuTimeCost = t1 - t0;
double t10 = RealTime();
longTask();
double t11 = RealTime();
realTimeCost = t11 - t10;
NSLog(#"\nthread cpu: %lfs\nreal: %lfs", cpuTimeCost, realTimeCost);
return;
}
the output is:
thread cpu: 0.024395s
real: 1.006627s
By contrast, on iPhone or Mac with an Intel chip, these two values are almost the same.
Any help here? I've tried the Rosetta and it has the same problem.
Related
We are developing a POS APP using xamarin.forms, in that we need to print the receipt to an esc/pos thermal printer connected via LAN.
We have multi language support with the App, printing multiple language with the esc/pos commands by changing code page works perfectly.
But its working for some supported language only, for other language its printing garbage characters(unreadable ones).
so we thought of creating a pdf for the receipt and print that one. we tried to create the pdf and then convert to bitmap and then send to the printer by using esc pos commands, but its not printing anything.
public BitImage(String filename)
{
Java.IO.File file = new Java.IO.File(filename);
var pdfRenderer = new PdfRenderer(ParcelFileDescriptor.Open(file, ParcelFileMode.ReadOnly));
PdfRenderer.Page page = pdfRenderer.OpenPage(0);
Bitmap bmp = Bitmap.CreateBitmap(page.Width, page.Height, Bitmap.Config.Argb8888);
page.Render(bmp, null, null, PdfRenderMode.ForPrint);
load(bmp);
}
private void load(Bitmap bmp)
{
int w = bmp.Width;
int h = bmp.Height;
int bw = (w + 7) / 8;
if (bw > 255)
bw = 255;
int bh = h / 8;
if (bh > 24)
{
bh = 24;
}
initData(bw * 8, bh * 8);
for (int y = 0; y < h; y++)
{
for (int x = 0; x < w; x++)
{
if (bmp.GetPixel(x, y) == Color.Black)
setPixel(x, y);
}
}
}
private void initData(int w, int h)
{
width = w;
height = h;
pitch = h / 8;
data = new byte[w * pitch];
}
private void setPixel(int x, int y)
{
if (x >= width || y >= height)
{
return;
}
int mask = (0x0080 >> (y % 8));
data[(x * pitch) + (y / 8)] |= (byte)mask;
}
public void PrintData()
{
byte[] CMD_INIT = { 0x1B, 0x40 };
byte[] CMD_UPLOAD_IMAGE = { 0x1D, 0x2A, 0, 0 };
byte[] CMD_PRINT_IMAGE = { 0x1D, 0x2F, 0 };
byte[] CMD_CUT = { 0x1D, 0x56, 0x01 };
CMD_UPLOAD_IMAGE[2] = (byte)(width / 8);
CMD_UPLOAD_IMAGE[3] = (byte)(height / 8);
#region Print Via Lan
Socket pSocket = new Socket(SocketType.Stream, ProtocolType.IP);
pSocket.SendTimeout = 1500;
pSocket.Connect("192.168.15.168", 9100);
pSocket.Send(CMD_INIT);
pSocket.Send(CMD_UPLOAD_IMAGE);
pSocket.Send(data);
pSocket.Send(CMD_PRINT_IMAGE);
pSocket.Send(CMD_CUT);
pSocket.Close();
#endregion
}
Please help me, whether i am doing it in correct way?
or is there any better way to do the same?
You can use libraries like SkiaSharp to make Image/PDF from your data in any language and print them properly using any printer.
I've created a sample to demonstrate how to print images properly with ESC\POS printers in C#: GitHub code repo
everyone.
I got this kernel:
__kernel void FuncionCL(__global char* in, __global char* out, __global int* S2)
{
__private int op1, op2, op3;
__private int C;
__private uint WorkDim, C2;
op1 = 1;
op2 = 2;
WorkDim = get_global_size(0);
__private int ID;
ID = get_global_id(0);
for(C = 0; C < 1000000; C++)
{
for(C2 = ID; C2 < 1000; C2 += WorkDim)
{
op3 = op1 + op2;
}
}
out[0] = 90;
out[1] = 89;
*S2 = (int) WorkDim;
}
It crashes not only the application, the graphic controller too. I i change the for increment for the constant value '16' (the get_global_size() function returns) then the code runs fine. What's the problem?
If i run the code with:
WorkDim = 16;
in the line 8 instead of:
WorkDim = get_global_size(0);
The code runs 400 times faster, that's the problem. Why if the value is the same?
**EDIT: ** Well, now i know why, the code is so slow and there are multiply reasons:
1.- The occupancy.
2.- All the threads do the same iterations in the first loop, the right code looks like this:
__kernel void FuncionCL(__global char* in, __global char* out, __global int* S2)
{
__private int op1, op2, op3;
__private int C;
__private uint WorkDim, C2;
op1 = 1;
op2 = 2;
WorkDim = get_global_size(0);
__private int ID;
ID = get_global_id(0);
for(C = ID; C < 1000000; C += WorkDim)
{
for(C2 = ID; C2 < 1000; C2 += WorkDim)
{
op3 = op1 + op2;
}
}
out[0] = 90;
out[1] = 89;
*S2 = (int) WorkDim;
}
Now my code runs 6.1 times faster on the GPU than CPU.
Each item there is doing 1000000*1000 = 1Gop. Just too much, takes too long to do that and the driver restarts the GPU. (I am guessing global size is 1 in your example)
It is a total waste of resources to run a CL kernel with so little work items, it will make the GPU do almost-serial computation and take too long.
At least 1024 global items are needed in new GPUs to fully use their resources.
EDIT: The loop is probably optimized by the compiler when it has a static value. Therefore giving an "amazing" speedup.
I'm a beginner in OpenCL. I'm trying to implement an OpenCL application.I have a doubt that how to write opencl kernel code . i have given a original c code.
Question :- help me to change that given c code into opencl kernel code?.
ORIGINAL C CODE:
int i, j;
// initialization of indexes
for (i = 0; i<n; i++)
Index[i] = i;
// Bubble sort
for (i = 0; i<n - 1; i++)
{
for (j = i + 1; j<n; j++)
{
if (I[i] > I[j])
{
double z = I[i]; // exchange attractiveness
I[i] = I[j];
I[j] = z;
z = f[i]; // exchange fitness
f[i] = f[j];
f[j] = z;
int k = Index[i]; // exchange indexes
Index[i] = Index[j];
Index[j] = k;
}
}
}
Example for 4096 element arrays(alternate bubble1 and bubble2 at least 2048 times--->4096(N) kernel executions ):
index init on host side since its just assignment.
Auxiliary functions:
void swap2p(__private int * I,int i,int j)
{
int tmp=I[i];
I[i]=I[j];
I[j]=tmp;
}
void swap2g(__global int * I,int i,int j)
{
int tmp=I[i];
I[i]=I[j];
I[j]=tmp;
}
Alternating kernel-1:
__kernel void bubble1(__global int * I, __global int * f, __global int * Index){
int threadId=get_global_id(0);
__private int vals[2];
if(threadId*2+1<4096)
{
vals[0]=I[threadId*2];
vals[1]=I[threadId*2+1];
if(vals[0]>vals[1])
{
swap2p(vals,threadId*2,threadId*2+1);
swap2g(f,threadId*2,threadId*2+1);
swap2g(Index,threadId*2,threadId*2+1);
I[threadId*2]=vals[0];
I[threadId*2+1]=vals[1];
}
}
}
alternating kernel-2:
__kernel void bubble2(__global int * I){
int threadId=get_global_id(0);
__private int vals[2];
if(threadId*2+2<4096)
{
vals[0]=I[threadId*2+1];
vals[1]=I[threadId*2+2];
if(vals[0]>vals[1])
{
swap2p(vals,threadId*2+1,threadId*2+2);
swap2g(f,threadId*2+1,threadId*2+2);
swap2g(Index,threadId*2+1,threadId*2+2);
I[threadId*2+1]=vals[0];
I[threadId*2+2]=vals[1];
}
}
}
Global thread number: N/2 (2048)
Edit: Solved
SOLUTION:
I'm running Arduino 1.0.5
I fixed the problem by changing /Sketchbook/library/ros_lib/ros/node_handle.h line 260 from
}else if (topic_ == TopicInfo::ID_TX_STOP){
to
}else if (topic_ == ID_TX_STOP){
However, this gave me a new error message:
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp: In member function 'size_t Print::print(const __FlashStringHelper*)':
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp:44:9: error: 'prog_char' does not name a type
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp:47:23: error: 'p' was not declared in this scope
So to fix this I edited usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp, line 44 from
const prog_char *p = (const prog_char *)ifsh;
to
const char PROGMEM *p = (const char PROGMEM *)ifsh;
Now it compiles!
Original Problem:
I've been using this tutorial to get everything set up: (http://wiki.ros.org/rosserial_arduino/Tutorials/Arduino%20IDE%20Setup) I can install everything without an issue I think, the ros_lib folder is placed in my sketchbook libraries. But when I do the next tutorial with the helloworld example, the code does not verify correctly. When I try to verify with the checkmark in the Arduino IDE, I get the following set of error codes:
In file included from /home/user/sketchbook/libraries/ros_lib/ros.h:38:0,
from HelloWorld.cpp:6:
/home/user/sketchbook/libraries/ros_lib/ros/node_handle.h: In member function 'virtual int ros::NodeHandle_::spinOnce()':
/home/user/sketchbook/libraries/ros_lib/ros/node_handle.h:260:45: error: expected unqualified-id before numeric constant
/home/user/sketchbook/libraries/ros_lib/ros/node_handle.h:260:45: error: expected ')' before numeric constant
I've reinstalled the ros_lib as well as rosserial and I keep getting this error, so I don't know what the problem is. I looked around line 260 of the node_handle.h file but I didn't notice anything out of place.
Here's node_handle.h: (spacing might be a little off)
/*
* Software License Agreement (BSD License)
*
* Copyright (c) 2011, Willow Garage, Inc.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials provided
* with the distribution.
* * Neither the name of Willow Garage, Inc. nor the names of its
* contributors may be used to endorse or promote prducts derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
* ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef ROS_NODE_HANDLE_H_
#define ROS_NODE_HANDLE_H_
#include "std_msgs/Time.h"
#include "rosserial_msgs/TopicInfo.h"
#include "rosserial_msgs/Log.h"
#include "rosserial_msgs/RequestParam.h"
#define SYNC_SECONDS 5
#define MODE_FIRST_FF 0
/*
* The second sync byte is a protocol version. It's value is 0xff for the first
* version of the rosserial protocol (used up to hydro), 0xfe for the second version
* (introduced in hydro), 0xfd for the next, and so on. Its purpose is to enable
* detection of mismatched protocol versions (e.g. hydro rosserial_python with groovy
* rosserial_arduino. It must be changed in both this file and in
* rosserial_python/src/rosserial_python/SerialClient.py
*/
#define MODE_PROTOCOL_VER 1
#define PROTOCOL_VER1 0xff // through groovy
#define PROTOCOL_VER2 0xfe // in hydro
#define PROTOCOL_VER PROTOCOL_VER2
#define MODE_SIZE_L 2
#define MODE_SIZE_H 3
#define MODE_SIZE_CHECKSUM 4 // checksum for msg size received from size L and H
#define MODE_TOPIC_L 5 // waiting for topic id
#define MODE_TOPIC_H 6
#define MODE_MESSAGE 7
#define MODE_MSG_CHECKSUM 8 // checksum for msg and topic id
#define MSG_TIMEOUT 20 //20 milliseconds to recieve all of message data
#define ID_TX_STOP 11 //hardcode for hydro version
#include "msg.h"
namespace ros {
class NodeHandleBase_{
public:
virtual int publish(int id, const Msg* msg)=0;
virtual int spinOnce()=0;
virtual bool connected()=0;
};
}
#include "publisher.h"
#include "subscriber.h"
#include "service_server.h"
#include "service_client.h"
namespace ros {
using rosserial_msgs::TopicInfo;
/* Node Handle */
template<class Hardware,
int MAX_SUBSCRIBERS=25,
int MAX_PUBLISHERS=25,
int INPUT_SIZE=512,
int OUTPUT_SIZE=512>
class NodeHandle_ : public NodeHandleBase_
{
protected:
Hardware hardware_;
/* time used for syncing */
unsigned long rt_time;
/* used for computing current time */
unsigned long sec_offset, nsec_offset;
unsigned char message_in[INPUT_SIZE];
unsigned char message_out[OUTPUT_SIZE];
Publisher * publishers[MAX_PUBLISHERS];
Subscriber_ * subscribers[MAX_SUBSCRIBERS];
/*
* Setup Functions
*/
public:
NodeHandle_() : configured_(false) {
for(unsigned int i=0; i< MAX_PUBLISHERS; i++)
publishers[i] = 0;
for(unsigned int i=0; i< MAX_SUBSCRIBERS; i++)
subscribers[i] = 0;
for(unsigned int i=0; i< INPUT_SIZE; i++)
message_in[i] = 0;
for(unsigned int i=0; i< OUTPUT_SIZE; i++)
message_out[i] = 0;
req_param_resp.ints_length = 0;
req_param_resp.ints = NULL;
req_param_resp.floats_length = 0;
req_param_resp.floats = NULL;
req_param_resp.ints_length = 0;
req_param_resp.ints = NULL;
}
Hardware* getHardware(){
return &hardware_;
}
/* Start serial, initialize buffers */
void initNode(){
hardware_.init();
mode_ = 0;
bytes_ = 0;
index_ = 0;
topic_ = 0;
};
/* Start a named port, which may be network server IP, initialize buffers */
void initNode(char *portName){
hardware_.init(portName);
mode_ = 0;
bytes_ = 0;
index_ = 0;
topic_ = 0;
};
protected:
//State machine variables for spinOnce
int mode_;
int bytes_;
int topic_;
int index_;
int checksum_;
bool configured_;
/* used for syncing the time */
unsigned long last_sync_time;
unsigned long last_sync_receive_time;
unsigned long last_msg_timeout_time;
public:
/* This function goes in your loop() function, it handles
* serial input and callbacks for subscribers.
*/
virtual int spinOnce(){
/* restart if timed out */
unsigned long c_time = hardware_.time();
if( (c_time - last_sync_receive_time) > (SYNC_SECONDS*2200) ){
configured_ = false;
}
/* reset if message has timed out */
if ( mode_ != MODE_FIRST_FF){
if (c_time > last_msg_timeout_time){
mode_ = MODE_FIRST_FF;
}
}
/* while available buffer, read data */
while( true )
{
int data = hardware_.read();
if( data < 0 )
break;
checksum_ += data;
if( mode_ == MODE_MESSAGE ){ /* message data being recieved */
message_in[index_++] = data;
bytes_--;
if(bytes_ == 0) /* is message complete? if so, checksum */
mode_ = MODE_MSG_CHECKSUM;
}else if( mode_ == MODE_FIRST_FF ){
if(data == 0xff){
mode_++;
last_msg_timeout_time = c_time + MSG_TIMEOUT;
}
}else if( mode_ == MODE_PROTOCOL_VER ){
if(data == PROTOCOL_VER){
mode_++;
}else{
mode_ = MODE_FIRST_FF;
if (configured_ == false)
requestSyncTime(); /* send a msg back showing our protocol version */
}
}else if( mode_ == MODE_SIZE_L ){ /* bottom half of message size */
bytes_ = data;
index_ = 0;
mode_++;
checksum_ = data; /* first byte for calculating size checksum */
}else if( mode_ == MODE_SIZE_H ){ /* top half of message size */
bytes_ += data<<8;
mode_++;
}else if( mode_ == MODE_SIZE_CHECKSUM ){
if( (checksum_%256) == 255)
mode_++;
else
mode_ = MODE_FIRST_FF; /* Abandon the frame if the msg len is wrong */
}else if( mode_ == MODE_TOPIC_L ){ /* bottom half of topic id */
topic_ = data;
mode_++;
checksum_ = data; /* first byte included in checksum */
}else if( mode_ == MODE_TOPIC_H ){ /* top half of topic id */
topic_ += data<<8;
mode_ = MODE_MESSAGE;
if(bytes_ == 0)
mode_ = MODE_MSG_CHECKSUM;
}else if( mode_ == MODE_MSG_CHECKSUM ){ /* do checksum */
mode_ = MODE_FIRST_FF;
if( (checksum_%256) == 255){
if(topic_ == TopicInfo::ID_PUBLISHER){
requestSyncTime();
negotiateTopics();
last_sync_time = c_time;
last_sync_receive_time = c_time;
return -1;
}else if(topic_ == TopicInfo::ID_TIME){
syncTime(message_in);
}else if (topic_ == TopicInfo::ID_PARAMETER_REQUEST){
req_param_resp.deserialize(message_in);
param_recieved= true;
}else if(topic_ == TopicInfo::ID_TX_STOP){
configured_ = false;
}else{
if(subscribers[topic_-100])
subscribers[topic_-100]->callback( message_in );
}
}
}
}
/* occasionally sync time */
if( configured_ && ((c_time-last_sync_time) > (SYNC_SECONDS*500) )){
requestSyncTime();
last_sync_time = c_time;
}
return 0;
}
/* Are we connected to the PC? */
virtual bool connected() {
return configured_;
};
/********************************************************************
* Time functions
*/
void requestSyncTime()
{
std_msgs::Time t;
publish(TopicInfo::ID_TIME, &t);
rt_time = hardware_.time();
}
void syncTime( unsigned char * data )
{
std_msgs::Time t;
unsigned long offset = hardware_.time() - rt_time;
t.deserialize(data);
t.data.sec += offset/1000;
t.data.nsec += (offset%1000)*1000000UL;
this->setNow(t.data);
last_sync_receive_time = hardware_.time();
}
Time now(){
unsigned long ms = hardware_.time();
Time current_time;
current_time.sec = ms/1000 + sec_offset;
current_time.nsec = (ms%1000)*1000000UL + nsec_offset;
normalizeSecNSec(current_time.sec, current_time.nsec);
return current_time;
}
void setNow( Time & new_now )
{
unsigned long ms = hardware_.time();
sec_offset = new_now.sec - ms/1000 - 1;
nsec_offset = new_now.nsec - (ms%1000)*1000000UL + 1000000000UL;
normalizeSecNSec(sec_offset, nsec_offset);
}
/********************************************************************
* Topic Management
*/
/* Register a new publisher */
bool advertise(Publisher & p)
{
for(int i = 0; i < MAX_PUBLISHERS; i++){
if(publishers[i] == 0){ // empty slot
publishers[i] = &p;
p.id_ = i+100+MAX_SUBSCRIBERS;
p.nh_ = this;
return true;
}
}
return false;
}
/* Register a new subscriber */
template<typename MsgT>
bool subscribe(Subscriber< MsgT> & s){
for(int i = 0; i < MAX_SUBSCRIBERS; i++){
if(subscribers[i] == 0){ // empty slot
subscribers[i] = (Subscriber_*) &s;
s.id_ = i+100;
return true;
}
}
return false;
}
/* Register a new Service Server */
template<typename MReq, typename MRes>
bool advertiseService(ServiceServer<MReq,MRes>& srv){
bool v = advertise(srv.pub);
for(int i = 0; i < MAX_SUBSCRIBERS; i++){
if(subscribers[i] == 0){ // empty slot
subscribers[i] = (Subscriber_*) &srv;
srv.id_ = i+100;
return v;
}
}
return false;
}
/* Register a new Service Client */
template<typename MReq, typename MRes>
bool serviceClient(ServiceClient<MReq, MRes>& srv){
bool v = advertise(srv.pub);
for(int i = 0; i < MAX_SUBSCRIBERS; i++){
if(subscribers[i] == 0){ // empty slot
subscribers[i] = (Subscriber_*) &srv;
srv.id_ = i+100;
return v;
}
}
return false;
}
void negotiateTopics()
{
rosserial_msgs::TopicInfo ti;
int i;
for(i = 0; i < MAX_PUBLISHERS; i++)
{
if(publishers[i] != 0) // non-empty slot
{
ti.topic_id = publishers[i]->id_;
ti.topic_name = (char *) publishers[i]->topic_;
ti.message_type = (char *) publishers[i]->msg_->getType();
ti.md5sum = (char *) publishers[i]->msg_->getMD5();
ti.buffer_size = OUTPUT_SIZE;
publish( publishers[i]->getEndpointType(), &ti );
}
}
for(i = 0; i < MAX_SUBSCRIBERS; i++)
{
if(subscribers[i] != 0) // non-empty slot
{
ti.topic_id = subscribers[i]->id_;
ti.topic_name = (char *) subscribers[i]->topic_;
ti.message_type = (char *) subscribers[i]->getMsgType();
ti.md5sum = (char *) subscribers[i]->getMsgMD5();
ti.buffer_size = INPUT_SIZE;
publish( subscribers[i]->getEndpointType(), &ti );
}
}
configured_ = true;
}
virtual int publish(int id, const Msg * msg)
{
if(id >= 100 && !configured_)
return 0;
/* serialize message */
unsigned int l = msg->serialize(message_out+7);
/* setup the header */
message_out[0] = 0xff;
message_out[1] = PROTOCOL_VER;
message_out[2] = (unsigned char) ((unsigned int)l&255);
message_out[3] = (unsigned char) ((unsigned int)l>>8);
message_out[4] = 255 - ((message_out[2] + message_out[3])%256);
message_out[5] = (unsigned char) ((int)id&255);
message_out[6] = (unsigned char) ((int)id>>8);
/* calculate checksum */
int chk = 0;
for(int i =5; i<l+7; i++)
chk += message_out[i];
l += 7;
message_out[l++] = 255 - (chk%256);
if( l <= OUTPUT_SIZE ){
hardware_.write(message_out, l);
return l;
}else{
logerror("Message from device dropped: message larger than buffer.");
return -1;
}
}
/********************************************************************
* Logging
*/
private:
void log(char byte, const char * msg){
rosserial_msgs::Log l;
l.level= byte;
l.msg = (char*)msg;
publish(rosserial_msgs::TopicInfo::ID_LOG, &l);
}
public:
void logdebug(const char* msg){
log(rosserial_msgs::Log::ROSDEBUG, msg);
}
void loginfo(const char * msg){
log(rosserial_msgs::Log::INFO, msg);
}
void logwarn(const char *msg){
log(rosserial_msgs::Log::WARN, msg);
}
void logerror(const char*msg){
log(rosserial_msgs::Log::ERROR, msg);
}
void logfatal(const char*msg){
log(rosserial_msgs::Log::FATAL, msg);
}
/********************************************************************
* Parameters
*/
private:
bool param_recieved;
rosserial_msgs::RequestParamResponse req_param_resp;
bool requestParam(const char * name, int time_out = 1000){
param_recieved = false;
rosserial_msgs::RequestParamRequest req;
req.name = (char*)name;
publish(TopicInfo::ID_PARAMETER_REQUEST, &req);
unsigned int end_time = hardware_.time() + time_out;
while(!param_recieved ){
spinOnce();
if (hardware_.time() > end_time) return false;
}
return true;
}
public:
bool getParam(const char* name, int* param, int length =1){
if (requestParam(name) ){
if (length == req_param_resp.ints_length){
//copy it over
for(int i=0; i<length; i++)
param[i] = req_param_resp.ints[i];
return true;
}
}
return false;
}
bool getParam(const char* name, float* param, int length=1){
if (requestParam(name) ){
if (length == req_param_resp.floats_length){
//copy it over
for(int i=0; i<length; i++)
param[i] = req_param_resp.floats[i];
return true;
}
}
return false;
}
bool getParam(const char* name, char** param, int length=1){
if (requestParam(name) ){
if (length == req_param_resp.strings_length){
//copy it over
for(int i=0; i<length; i++)
strcpy(param[i],req_param_resp.strings[i]);
return true;
}
}
return false;
}
};
}
#endif
I tried commenting out line 160 and 161:
//}else if (topic_ == TopicInfo::ID_TX_STOP){
//configured_ = false;
This gave me a different error message:
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp: In member function 'size_t Print::print(const __FlashStringHelper*)':
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp:44:9: error: 'prog_char' does not name a type
/usr/share/arduino/hardware/arduino/cores/arduino/Print.cpp:47:23: error: 'p' was not declared in this scope
So here's Print.cpp
/*
Print.cpp - Base class that provides print() and println()
Copyright (c) 2008 David A. Mellis. All right reserved.
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
Modified 23 November 2006 by David A. Mellis
*/
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <math.h>
#include "Arduino.h"
#include "Print.h"
// Public Methods //////////////////////////////////////////////////////////////
/* default implementation: may be overridden */
size_t Print::write(const uint8_t *buffer, size_t size)
{
size_t n = 0;
while (size--) {
n += write(*buffer++);
}
return n;
}
size_t Print::print(const __FlashStringHelper *ifsh)
{
const prog_char *p = (const prog_char *)ifsh;
size_t n = 0;
while (1) {
unsigned char c = pgm_read_byte(p++);
if (c == 0) break;
n += write(c);
}
return n;
}
size_t Print::print(const String &s)
{
size_t n = 0;
for (uint16_t i = 0; i < s.length(); i++) {
n += write(s[i]);
}
return n;
}
size_t Print::print(const char str[])
{
return write(str);
}
size_t Print::print(char c)
{
return write(c);
}
size_t Print::print(unsigned char b, int base)
{
return print((unsigned long) b, base);
}
size_t Print::print(int n, int base)
{
return print((long) n, base);
}
size_t Print::print(unsigned int n, int base)
{
return print((unsigned long) n, base);
}
size_t Print::print(long n, int base)
{
if (base == 0) {
return write(n);
} else if (base == 10) {
if (n < 0) {
int t = print('-');
n = -n;
return printNumber(n, 10) + t;
}
return printNumber(n, 10);
} else {
return printNumber(n, base);
}
}
size_t Print::print(unsigned long n, int base)
{
if (base == 0) return write(n);
else return printNumber(n, base);
}
size_t Print::print(double n, int digits)
{
return printFloat(n, digits);
}
size_t Print::println(const __FlashStringHelper *ifsh)
{
size_t n = print(ifsh);
n += println();
return n;
}
size_t Print::print(const Printable& x)
{
return x.printTo(*this);
}
size_t Print::println(void)
{
size_t n = print('\r');
n += print('\n');
return n;
}
size_t Print::println(const String &s)
{
size_t n = print(s);
n += println();
return n;
}
size_t Print::println(const char c[])
{
size_t n = print(c);
n += println();
return n;
}
size_t Print::println(char c)
{
size_t n = print(c);
n += println();
return n;
}
size_t Print::println(unsigned char b, int base)
{
size_t n = print(b, base);
n += println();
return n;
}
size_t Print::println(int num, int base)
{
size_t n = print(num, base);
n += println();
return n;
}
size_t Print::println(unsigned int num, int base)
{
size_t n = print(num, base);
n += println();
return n;
}
size_t Print::println(long num, int base)
{
size_t n = print(num, base);
n += println();
return n;
}
size_t Print::println(unsigned long num, int base)
{
size_t n = print(num, base);
n += println();
return n;
}
size_t Print::println(double num, int digits)
{
size_t n = print(num, digits);
n += println();
return n;
}
size_t Print::println(const Printable& x)
{
size_t n = print(x);
n += println();
return n;
}
// Private Methods /////////////////////////////////////////////////////////////
size_t Print::printNumber(unsigned long n, uint8_t base) {
char buf[8 * sizeof(long) + 1]; // Assumes 8-bit chars plus zero byte.
char *str = &buf[sizeof(buf) - 1];
*str = '\0';
// prevent crash if called with base == 1
if (base < 2) base = 10;
do {
unsigned long m = n;
n /= base;
char c = m - base * n;
*--str = c < 10 ? c + '0' : c + 'A' - 10;
} while(n);
return write(str);
}
size_t Print::printFloat(double number, uint8_t digits)
{
size_t n = 0;
// Handle negative numbers
if (number < 0.0)
{
n += print('-');
number = -number;
}
// Round correctly so that print(1.999, 2) prints as "2.00"
double rounding = 0.5;
for (uint8_t i=0; i<digits; ++i)
rounding /= 10.0;
number += rounding;
// Extract the integer part of the number and print it
unsigned long int_part = (unsigned long)number;
double remainder = number - (double)int_part;
n += print(int_part);
// Print the decimal point, but only if there are digits beyond
if (digits > 0) {
n += print(".");
}
// Extract digits from the remainder one at a time
while (digits-- > 0)
{
remainder *= 10.0;
int toPrint = int(remainder);
n += print(toPrint);
remainder -= toPrint;
}
return n;
}
Trying to use the same code (sort of) as what I have used when running using TBB (threading building blocks).
I don't have a great deal of experience with OpenCL, but I think most of the main code is correct. I believe the errors are in the .cl file, where it does the math.
Here is my mandelbrot code in TBB:
Mandelbrot TBB
Here is my code in OpenCL
Mandelbrot OpenCL
Any help would be greatly appreciated.
I changed the code in the kernel, and it ran fine. My new kernel code is the following:
// voronoi kernels
//
// local memory version
//
kernel void voronoiL(write_only image2d_t outputImage)
{
// get id of element in array
int x = get_global_id(0);
int y = get_global_id(1);
int w = get_global_size(0);
int h = get_global_size(1);
float4 result = (float4)(0.0f,0.0f,0.0f,1.0f);
float MinRe = -2.0f;
float MaxRe = 1.0f;
float MinIm = -1.5f;
float MaxIm = MinIm+(MaxRe-MinRe)*h/w;
float Re_factor = (MaxRe-MinRe)/(w-1);
float Im_factor = (MaxIm-MinIm)/(h-1);
float MaxIterations = 50;
//C imaginary
float c_im = MaxIm - y*Im_factor;
//C real
float c_re = MinRe + x*Re_factor;
//Z real
float Z_re = c_re, Z_im = c_im;
bool isInside = true;
bool col2 = false;
bool col3 = false;
int iteration =0;
for(int n=0; n<MaxIterations; n++)
{
// Z - real and imaginary
float Z_re2 = Z_re*Z_re, Z_im2 = Z_im*Z_im;
//if Z real squared plus Z imaginary squared is greater than c squared
if(Z_re2 + Z_im2 > 4)
{
if(n >= 0 && n <= (MaxIterations/2-1))
{
col2 = true;
isInside = false;
break;
}
else if(n >= MaxIterations/2 && n <= MaxIterations-1)
{
col3 = true;
isInside = false;
break;
}
}
Z_im = 2*Z_re*Z_im + c_im;
Z_re = Z_re2 - Z_im2 + c_re;
iteration++;
}
if(col2)
{
result = (float4)(iteration*0.05f,0.0f, 0.0f, 1.0f);
}
else if(col3)
{
result = (float4)(255, iteration*0.05f, iteration*0.05f, 1.0f);
}
else if(isInside)
{
result = (float4)(0.0f, 0.0f, 0.0f, 1.0f);
}
write_imagef(outputImage, (int2)(x, y), result);
}
You can also find it here:
https://docs.google.com/file/d/0B6DBARvnB__iUjNSTWJubFhUSDA/edit
See this link. It's developed by #eric-bainville. The CPU code both native and with OpenCL is not optimal (it does not use SSE/AVX) but I think the GPU code may be good. For the CPU you can speed up the code quite a bit by using AVX and operating on eight pixels at once.
http://www.bealto.com/mp-mandelbrot.html