Bugzilla – Bug 113264
SuSEconfig.fonts using excessive time
Last modified: 2005-09-05 15:59:22 UTC
I timed the usage of SuSEconfig modules on a multiple run (taking a bit of time on a firest run is tolerated, a rerun should better not be above 1 second). Executing /sbin/conf.d/SuSEconfig.fonts... Creating cache files for Xft .............................. Creating fonts.{scale,dir} files ........... real 0m7.643s user 0m4.596s sys 0m0.292s it's ok to generate caches, but over 7 seconds for direct re-runs should be optimized a bit more.
Less than 4 seconds for direct reruns for me: mfabian@magellan:/var/tmp/abuild$ time sudo /sbin/conf.d/SuSEconfig.fonts Creating cache files for Xft ................................................................................... Configure X11 to use available CID-keyed fonts ... Creating fonts.{scale,dir} files ......................... real 0m3.745s user 0m1.312s sys 0m0.116s mfabian@magellan:/var/tmp/abuild$ time sudo /sbin/conf.d/SuSEconfig.fonts Creating cache files for Xft ................................................................................... Configure X11 to use available CID-keyed fonts ... Creating fonts.{scale,dir} files ......................... real 0m3.614s user 0m1.356s sys 0m0.084s mfabian@magellan:/var/tmp/abuild$ time sudo /sbin/conf.d/SuSEconfig.fonts Creating cache files for Xft ................................................................................... Configure X11 to use available CID-keyed fonts ... Creating fonts.{scale,dir} files ......................... real 0m3.654s user 0m1.356s sys 0m0.080s mfabian@magellan:/var/tmp/abuild$ And I have probably more fonts installed than anybody else: mfabian@magellan:/var/tmp/abuild$ fc-list : | wc 1409 2571 49953 mfabian@magellan:/var/tmp/abuild$ mfabian@fatou:~$ fc-list : | wc 681 1520 25260 mfabian@fatou:~$
I'll retry after updating to beta4 (the numbers were done on fatou).
Rudi, is this fixed?
I didn't do anything to make it faster yet, i.e. there should be no change so far. It is not as slow on my machine magellan though, although magellan is slower than fatou, it did run faster for me on magellan than in Rudi's original comment. First I'll fix bug #79259 now which will make it unfortately even slower (although not much I hope), then I'll see whether I can optimize something.
I cannot see why this should be `critical', reduced to `minor' and closed the bug with `nofix' due to the comment from Mike.