Hi!
> +sub()
> +{
> + minuend=$(echo $1 | sed -r 's/([[:digit:]]{1,}).*/\1/')
> + if echo $1 | grep -q -E "m|M"; then
> + minuend=$((minuend*1024*1024))
> + flag=m
> + elif echo $1 | grep -q -E "k|K"; then
> + minuend=$((minuend*1024))
> + flag=k
> + else
> + flag=0
> + fi
> +
> + subtrahend=$(echo $2 | sed -r 's/([[:digit:]]{1,}).*/\1/')
> + if echo $2 | grep -q -E "m|M"; then
> + subtrahend=$((subtrahend*1024*1024))
> + elif echo $2 | grep -q -E "k|K"; then
> + subtrahend=$((subtrahend*1024))
> + fi
> +
> + rest=$((minuend - subtrahend))
> + if [ $flag == m ]; then
> + echo $((rest/1024/1024))
> + elif [ $flag == k ]; then
> + echo $((rest/1024))
> + else
> + echo $rest
> + fi
I don't get why we multiply the numbers, substract them and then divide
them again. Can't we just substract them?
Also does this work for other filesystems than ext2? It may be a good
idea to run the test for a few more filesystems as well. We may add a
parameter which fs to use and add entries as:
df01 -f ext2
df01 -f ext3
...
to the runtest file.
Otherwise it looks good.
--
Cyril Hrubis
[email protected]
------------------------------------------------------------------------------
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list