script to return info from /proc/ - linux

Question

I am trying to write a script that will return info from the /proc/cpuinfo, /proc/meminfo and /proc/version files.
From the cpuinfo file, I want to return the cpu Mhz and model name.
I can get these via these commands
more /proc/cpuinfo | grep "model name" | head -n 1
more /proc/cpuinfo | grep "cpu MHz"
for the meminfo file, I want to get total memory, memory free and total used. I can get the first 2 via these commands:
more /proc/meminfo | grep MemTotal
more /proc/meminfo | grep MemFree
and I can get the linux version # with this:
more /proc/version
I can then have this saved as a file via redirecting the first output into a file and then append the next info items with using a >> instead of >.
My problem is this - how do I write a script that will take the info from the above and place it into this format:
/proc/cpuinfo, Model name: (result of first command above)
/proc/cpuinfo, cpu Mhz: (result of 2nd)
/proc/meminfo, MemTotal: (result of 3rd)
/proc/meminfo, MemFree: (result of 4th)
/proc/meminfo, MemUsed: (calculate it based off memtotal and memfree)
/proc/version, Linux version #:
I know how to use cut, awk and more, etc but do not know how to set this up. I do not know how to force the calculation of the mem used either.
Any help you can give would be appreciated.
EDIT: I use the more because I am not too familiar with Linux.
I am getting closer and closer to what I want to do with a combination of what is posted here and what I need to come up with.
MATH function -
I just want to take the memtotal and subtract memfree from it.
Could I just create a variable such as
memused=$(bc $memtotal - memfree)
and then echo it out?

Answer 1

With a simple shell function like:
filedata() {
grep -H "$#" | sed -e 's/:/, /'
}
You can get most of the data you need by calling
filedata 'model name' /proc/cpuinfo
filedata -E 'Mem(Total|Free)' /proc/meminfo
filedata . /proc/version
To get MemUsed you could use something like:
awk '/MemFree/ {free=$2} /MemTotal/ {total=$2} END {print FILENAME",","MemUsed:", total-free}' /proc/meminfo
Alternatively the following awk script will do it all for you (though not in exactly the order of your example output):
awk '/model name|cpu MHz|MemTotal|MemFree|^Linux/ {
print FILENAME",",$0
}
/MemTotal|MemFree/ {
v=$1
gsub(/^Mem/, "", v)
gsub(/:$/, "", v)
mem[v]=$2
}
END {
print "/proc/meminfo, MemUsed:", mem["Total"] - mem["Free"]
}' /proc/cpuinfo /proc/meminfo /proc/version

Answer 2

In a simplified way you can do the following:
EXAMPLE
#!/bin/sh
LOCATION=$1
if [ "$#" -ne "1" ]
then
echo "Usage: ./$0 <FILE>"
else
model=$(cat /proc/cpuinfo |grep -m 1 "model name"|cut -d' ' -f 4-);
mhz=$(cat /proc/cpuinfo |grep -m 1 "cpu MHz"|cut -d' ' -f 3-);
mem=$(cat /proc/meminfo | grep MemTotal|cut -d' ' -f 2-);
free=$(cat /proc/meminfo | grep MemFree|cut -d' ' -f 2-);
ver=$(cat /proc/version|cut -d' ' -f 3);
fi
echo -e \
"/proc/cpuinfo, Model Name: $model
/proc/cpuinfo, CPU MHz: $mhz
/proc/meminfo, MemTotal: $mem
/proc/meminfo, MemFree: $free
/proc/version, Linux Verion #: $ver" > $LOCATION
That will place each result in a variable so you can echo it into a file that you declare when you call the script like sh test.sh mynewfile.txt.
As for "I do not know how to force the calculation of the mem used either." please update you question to include how you expect those values to be present (kb, MB, GB) and a sample output you are looking for.

Answer 3

Here is how to get value using awk only:
model=$(awk -F: '/model name/ {print $2;exit}' /proc/cpuinfo)
mhz=$(awk -F: '/cpu MHz/ {print $2;exit}' /proc/cpuinfo)
mem=$(awk -F"[: ]+" '/MemTotal/ {print $2;exit}' /proc/meminfo)
etc