Wow, thnks for tracking that down.. yeah, my logic for the return code or read_file() was wrong. errno vals are positive, not negative, and i think originally i had read_file returning the number of bytes read or something. Its also why you'd get like 5 error essages all repeating when the cpufreq modules weren't loaded when you should have gotten one. My bad, completely... but at least now I understand what went wrong. Thank you for the help.
No need to include this little patch.. I've fixed this and (hopefully) the centrino problem, and cleaned up the error handling a bit.. so I'm going to release 0.96 on my web site in a a couple of minutes. Bdale should be able to pick it up and work his magic to update the debian package.
sorry for the trouble. john.c
Well, what do you know, your new version works. I downloaded debian's 0.95 source and compared it to your version, and isolated the change that made it work:
--- powernowd_0.95.c 2005-05-08 06:08:58.000000000 +0200 +++ powernowd.c 2005-05-08 06:19:48.000000000 +0200 @@ -416,7 +416,7 @@ strncpy(scratch, cpu->sysfs_dir, 50); strncat(scratch, "scaling_available_frequencies", 50);
- if (((err = read_file(scratch, 0, 1)) < 0) || (step_specified)) { + if (((err = read_file(scratch, 0, 1)) != 0) || (step_specified)) { /* * We don't have scaling_available_frequencies. build the * table from the min, max, and step values. the driver
The whole section calculating the table size was simply skipped.
Sorry to trouble you with this, but i can't seem to figure out what's wrong :/
No trouble, after all, you're taking the time to fix my problem ;-)
For completeness sake, here's the output of your version (0.96) with and without this patch reverted:
I don't know if this little patch is suitable for Debian, because you've fixed the error handling in other places too;
Anyway, thanks for the effort!
-- Mourad
-- John Clemens http://www.deater.net/john john at deater.net "I Hate Quotes" -- Samuel L. Clemens
-- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
