clocks | analog clock made with many different gui toolkits | Emulator library

I am getting started with C programming and I am trying to solve a practice lesson in the book I am learning from. The task I need to do is to print a line to the terminal then wait 2 seconds and after 2 seconds return the cursor to the beginning of the line in the terminal and replace, overwrite the line with something else.

The example code in the book contain lots of boilerplate code for other tasks I also need to do, for now I am trying to isolate and solve this first problem on its own, but it does not work and I don't understand why.

This is the code I'm trying:

I'm experimenting with threads. My program is supposed to take a vector and sum it by breaking it down into different sections and creating a thread to sum each section. Currently, my vector has 5 * 10^8 elements, which should be easily handled by my pc. However, the creation of each thread (4 threads in my case) takes an insanely long time. I'm wondering why...?

I'm trying to learn about multithreaded algorithms so I've implemented a simple find max number function of an array. I've made a baseline program (findMax1.c) which loads from a file about 263 million int numbers into memory. Then I simply use a for loop to find the max number. Then I've made another program (findMax2.c) which uses 4 threads. I chose 4 threads because the CPU (intel i5 4460) I'm using has 4 cores and 1 thread per core. So my guess is that if I assign each core a chunk of the array to process it would be more efficient because that way I'll have fewer cache misses. Now, each thread finds the max number from each chunk, then I join all threads to finally find the max number from all those chunks. The baseline program findMax1.c takes about 660ms to complete the task, so my initial thought was that findMax2.c (which uses 4 threads) would take about 165ms (660ms / 4) to complete since now I have 4 threads running all in parallel to do the same task, but findMax2.c takes about 610ms. Only 50ms less than findMax1.c. What am I missing? is there something wrong with the implementation of the threaded program?

findMax1.c

I know that this 'eigen speed-up' questions arise regularly but after reading many of them and trying several flags I cannot get a better time with c++ eigen comparing with the traditional way of performing a transpose. Actually using blocking is much more efficient. The following is the code

I've installed Windows 10 21H2 on both my desktop (AMD 5950X system with RTX3080) and my laptop (Dell XPS 9560 with i7-7700HQ and GTX1050) following the instructions on https://docs.nvidia.com/cuda/wsl-user-guide/index.html:

1. Install CUDA-capable driver in Windows
2. Update WSL2 kernel in PowerShell: wsl --update
3. Install CUDA toolkit in Ubuntu 20.04 in WSL2 (Note that you don't install a CUDA driver in WSL2, the instructions explicitly tell that the CUDA driver should not be installed.):

I have a crontab scheduling many things on an Ubuntu server which runs on UTC time. Some of the scripts being run are time sensitive in relation to the web applications they are providing functionality for.

e.g. Something has to be updated at 4pm every day, as the end users see it by their clock.

This is not an issue during the winter when the UK is on GMT, but from April to October, the clocks go forward an hour for British Summer Time (BST). The scripts running on the server then update the web application at what appears to be 5pm for the users, instead of 4pm.

Is there a way to conditionally adjust the crontab's scheduled times for this time zone change?

The intention below is to run on the first day of every month at 00:00. During BST, my understanding is that this will run at 01:00 BST during daylight savings:

I'm trying to compute the intersection between 3 geometries using VTK.

I've started with the BooleanOperationPolyDataFilter example and made minor tweaks such as:

Adding a third sphere:

This is the code that shows the access behavior of std::vector slows down when std::vector is sorted by std::sort().

In an attempt to profile & optimize a caching algorithm, I got stuck at something I don't understand.

Here is the hot-spot of the perf's report (in Ubuntu 18.04 LTS and g++ 7.5):

How does just a "mov" operatiaon between rax and rdx registers cause ~30% of total run-time of program? It's not a register-heavy program (an LRU-cache approximation algorithm that is doing ~50million lookups per second at max and this is around 400MB/s throughput(and certainly not faster than RAM for bigger key-value pairs) which should not be related to register bandwidth at all)

Test system's CPU is FX8150 and has 1 channel memory attached with 9GB/s bandwidth which is way higher than this single-thread cache can achieve for just "int" key + "int" value pairs. So I guess it should be safe to leave RAM out of this problem. Also the mov instruction looks like a part of std::unordered_map's lookup operations. I know this CPU is old but not really ancient-old so perhaps compiler is not using the right instruction here due to some "support for old CPU" issue?

Source code to reproduce the hot-spot: