Index of /wiki/ParallelComputingSpring2015/cuda/stanford/assignments/mp3/
Name | Last Modified | Size |
---|---|---|
Parent Directory | ||
black_scholes.cu | 2022-06-01 22:52 | 7k |
black_scholes.h | 2022-06-01 22:52 | 4k |
compact.cu | 2022-06-01 22:52 | 3k |
compact.h | 2022-06-01 22:52 | 1k |
Makefile | 2022-06-01 22:52 | 2k |
mp3-part1.cu | 2022-06-01 22:52 | 13k |
mp3-part2.cu | 2022-06-01 22:52 | 7k |
mp3-part3.cu | 2022-06-01 22:52 | 14k |
mp3-questions.txt | 2022-06-01 22:52 | 3k |
mp3-util.h | 2022-06-01 22:52 | 3k |
scan.cu | 2022-06-01 22:52 | 5k |
scan.h | 2022-06-01 22:52 | 1k |
scatter.cu | 2022-06-01 22:52 | 5k |
scatter.h | 2022-06-01 22:52 | 1k |
Welcome to the second Machine Problem!
DUE DATE: Thu April 22, 11:59pm
MP3 is a three-part assignment. Detailed instructions for each of the three
parts of MP3 are included in commentary at the top of the files mp3-part1.cu,
mp3-part2.cu, and mp3-part3.cu. When compiled out of the box, each of
these parts should report an error allocating device memory and exit
immediately. Your task is to correctly implement sections marked TODO: in
each part, causing the main() functions of each to return normally.
SUBMISSION INSTRUCTIONS
=========================
You can submit an assignment from any of the cluster machines by using
our submit script. Th submit script bundles the entire current directory into
a submission. Thus, you use it by CDing to a the directory for your assignment,
and running:
> cd *some directory*
> /usr/class/cs193g/bin/submit mp3
This will submit the current directory as your assignment. You can submit
as many times as you want, and we will use your last submission.
1) RUNNING ON PUPS USING OUR MAKEFILE:
======================================
To compile the three problems (part1, part2, and part3), run:
> make all
To run any of the problems, you can also use the makefile, like so:
> make run1
> make run2
> make run3
2) RUNNING ON YOUR OWN MACHINE USING OUR MAKEFILE:
==================================================
part1 and part3 of MP3 require GPU support for double precision (64b) floating point
arithmetic. The GPUs in the pups cluster support double precision, but your development
machine may not. This is fine -- you can develop in single precision, but you should
verify that your code compiles and runs in double precision on the pups cluster machines.
In order to develop in single precision, you should modify the Makefile by following the
instructions therein.
Proudly Served by LiteSpeed Web Server at wrfranklin.org Port 443