Package Details: modprobed-db 2.44-1

Git Clone URL: https://aur.archlinux.org/modprobed-db.git (read-only, click to copy)
Package Base: modprobed-db
Description: Keeps track of EVERY kernel module ever used - useful for make localmodconfig
Upstream URL: https://wiki.archlinux.org/index.php/Modprobed-db
Licenses: MIT
Conflicts: modprobed_db
Replaces: modprobed_db
Submitter: graysky
Maintainer: graysky
Last Packager: graysky
Votes: 224
Popularity: 1.98
First Submitted: 2013-11-15 22:21
Last Updated: 2020-07-05 20:06

Dependencies (2)

Required by (12)

Sources (1)

Latest Comments

« First ‹ Previous ... 8 9 10 11 12 13 14 Next › Last »

maggie commented on 2011-02-12 13:54

nice tool!

SanskritFritz commented on 2011-01-17 15:48

Just read the contents of the very script, the comments are enough.

SanskritFritz commented on 2011-01-17 15:05

The "database" is a simple textfile, editing it and deleting words is perfectly safe.
What do you mean, by modprobed trying to fetch a module? That is not the function of this package. You have to manually modprobe the modules prior to a kernel compile.

Anonymous comment on 2011-01-17 14:58

How do I remove led_class and wl modules from the db? I stopped using them and I can't compile a kernel because modprobed tries to fetch them

Anonymous comment on 2011-01-11 00:00

@graysky - Been trying to learn gawk and found another way if your interested? Works similar to yours except it uses gawk's regular expression comparison ability to search fields instead of using grep.

Turn the ignore array into something we can use:
var=$(echo ${ignore[*]} | sed 's/\(\<[[:alnum:]]*\>\)/^\1/g; s/ */|/g')

echo $var: (above output)
^nvidia|^nouveau|^vboxdrv|^vboxnetflt|^vboxnetadp

db:
gawk '$1 !~ /'"$var"'/ { print $1 }' /proc/modules | sort > ${db}



graysky commented on 2010-11-04 21:04

@harryNID - thanks for the tips. comm is a new one for me.

Anonymous comment on 2010-11-04 20:56

@graysky - I'm bugging again probably (last time!!) but here is another way to achieve your goal of $db and $dblong without using sed at all. Again this is solely for your amusement nothing more.

$db:
comm -23 <(cut -d\ -f1 /proc/modules | sort) <(echo ${ignore[*]} | tr " " "\n" | sort)

$dblong:
echo $(comm -23 <(cut -d\ -f1 /proc/modules | sort) <(echo ${ignore[*]} | tr " " "\n" | sort))

Anonymous comment on 2010-11-02 22:02

@graysky - First great little script! Here is another way to write your sed statement parsing the ignore array. It's slightly smaller and easier to read if your interested.

cat /proc/modules | cut -d\ -f1 | sort -k1,1 | grep -Ev "$(echo ${ignore[*]} | sed -e "s/^.*/^(&)\$/" -e "s/ /|/g")"

Should be very slightly more efficient by using "cut" instead of "awk" and one less editing statement in the sed command (you might save a billionth of a second perhaps). Strictly speaking the "-k1,1" is really not needed since your just sorting one field but it doesn't hurt.

Hope you don't mind.

graysky commented on 2010-10-25 19:42

@SF - got it this time. Thanks again. Plus, the script is now more robust thanks to some fancy sed usage.

SanskritFritz commented on 2010-10-25 15:01

still
*/5 * * * * /usr/bin/modprobed_db.sh > /dev/null 2>&1
in readme.install ;-)