Twitter (not set)


Create an array of CPU frequencies in GHz

 $ cpus=($({ echo scale=2; awk '/cpu MHz/ {print $4 " / 1000"}' /proc/cpuinfo; } | bc))

— by openiduser146 on Dec. 28, 2015, 9:02 p.m.


  • The awk command takes the input from /proc/cpuinfo, matches lines containing "cpu MHz", and appends the " / 1000" to the CPU frequency, so it's ready for piping to bc
  • The echo scale=2 is for bc, to get floating point numbers with a precision of maximum two decimal points
  • Group the echo scale=2 and the awk for piping to bc, by enclosing the commands within { ...; }
  • Run the commands in a $(...) subshell
  • Wrap the subshell within (...) to store the output lines as an array

From the cpus array, you can extract the individual CPU values with:


If you don't need the values in GHz, but MHz is enough, then the command is a lot simpler:

cpus=($(awk '/cpu MHz/ {print $4}' /proc/cpuinfo))


Arrays are Bash specific, might not work in older /bin/sh.

/proc/cpuinfo exists only in Linux.


Test git archive before actually creating an archive // fake dry run

 $ git archive master some/project/subdir | tar t

— by openiduser146 on Dec. 22, 2015, 2:29 p.m.


git archive doesn't have a --dry-run flag, and it would be nice to see what files would be in the archive before actually creating it.

  • git archive master some/project/subdir
  • Create an archive from the master branch, with only a specified sub-directory of the project in it (instead of the entire repo)
  • Note: without specifying a file, the archive is dumped to standard output
  • tar t : the t flag of tar is to list the content of an archive. In this example the content comes from standard input (piped from the previous command)

In other words, this command creates an archive without ever saving it in a file, and uses tar t to list the contents. If the output looks good, then you can create the archive with:

git archive master -o file.tar some/project/subdir