Hi Release Managers, cpufreqd is in a relatively bad shape, unfortunately I haven't been able take care of it as I should have in quite some time. One bug seems particularly bad (severity is important at the moment): http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=619913 The patch is simple, the patch has been in use for some time, debdiff is attached. Can this fix go to wheezy?
On running lintian against the package though there one particular error that I'm wondering if I should fix: E: cpufreqd: non-empty-dependency_libs-in-la-file usr/lib/cpufreqd/cpufreqd_acpi.la E: cpufreqd: non-empty-dependency_libs-in-la-file usr/lib/cpufreqd/cpufreqd_exec.la E: cpufreqd: non-empty-dependency_libs-in-la-file usr/lib/cpufreqd/cpufreqd_governor_parameters.la E: cpufreqd: non-empty-dependency_libs-in-la-file usr/lib/cpufreqd/cpufreqd_sensors.la .la files can be removed in cpufreqd, those are all plugins and it's not going to be any useful to build anything against them. Thanks On Sat, Mar 23, 2013 at 09:41:00AM +0100, Chris wrote: > Hi, > > in my opinion the severity of this bug should be set to grave because > it's not usuable at the moment. This package will be shipped broken in > wheezy. > > I'm still running cpufreqd with the patch above without any problem. -- mattia :wq!
diff -Nru cpufreqd-2.4.2/debian/changelog cpufreqd-2.4.2/debian/changelog --- cpufreqd-2.4.2/debian/changelog 2010-04-18 23:13:27.000000000 +0900 +++ cpufreqd-2.4.2/debian/changelog 2013-03-23 21:51:28.000000000 +0900 @@ -1,3 +1,10 @@ +cpufreqd (2.4.2-2) unstable; urgency=low + + * Fix reading the correct battery attribute + (Closes: #619913 and Closes: #644567). + + -- Mattia Dongili <malat...@debian.org> Sat, 23 Mar 2013 21:42:31 +0900 + cpufreqd (2.4.2-1) unstable; urgency=low * New upstream release: fixes configuration with libsensors4 diff -Nru cpufreqd-2.4.2/debian/patches/619913.patch cpufreqd-2.4.2/debian/patches/619913.patch --- cpufreqd-2.4.2/debian/patches/619913.patch 1970-01-01 09:00:00.000000000 +0900 +++ cpufreqd-2.4.2/debian/patches/619913.patch 2013-03-23 21:51:28.000000000 +0900 @@ -0,0 +1,55 @@ +From: Roland Hieber <roh...@rohieb.name> +Subject: Use power_now instead of current_now to read power from sysfs + +Apparently, current_now was renamed to power_now around kernel 2.6.36. The fix +in this patch first tries to read from the new file name, if this does not +exist, it tries the old name. + +Bug: 644567 +Author: niko2gare <http://sourceforge.net/users/niko2gare/> +Origin: http://sourceforge.net/tracker/?func=detail&aid=3172707&group_id=58904&atid=489239 +Last-Update: 2012-10-12 + +Modification to avoid checking closed battery +From: Bernd Rinn (z-bb) +Origin: https://bugs.launchpad.net/ubuntu/+source/cpufreqd/+bug/733507/+attachment/3022414/+files/cpufreqd-bug%23733507.patch +Index: cpufreqd/src/cpufreqd_acpi_battery.c +=================================================================== +--- cpufreqd.orig/src/cpufreqd_acpi_battery.c 2013-03-23 21:37:28.268379010 +0900 ++++ cpufreqd/src/cpufreqd_acpi_battery.c 2013-03-23 21:38:23.392952250 +0900 +@@ -36,6 +36,7 @@ + #define PRESENT "present" + #define STATUS "status" + #define CURRENT_NOW "current_now" ++#define POWER_NOW "power_now" + + struct battery_info { + int capacity; +@@ -145,9 +146,13 @@ + binfo->status = get_class_device_attribute(binfo->cdev, STATUS); + if (!binfo->status) + return -1; +- binfo->current_now = get_class_device_attribute(binfo->cdev, CURRENT_NOW); +- if (!binfo->current_now) +- return -1; ++ binfo->current_now = get_class_device_attribute(binfo->cdev, POWER_NOW); ++ if (!binfo->current_now) { ++ /* try the "current_now" name */ ++ binfo->current_now = get_class_device_attribute(binfo->cdev, CURRENT_NOW); ++ if (!binfo->current_now) ++ return -1; ++ } + + /* read the last full capacity, this is not going to change + * very often, so no need to poke it later */ +@@ -311,6 +316,10 @@ + + /* Read battery informations */ + for (i = 0; i < bat_dir_num; i++) { ++ if (!info[i].open) { ++ clog(LOG_INFO, "Skipping %s (closed)\n", info[i].cdev->name); ++ continue; ++ } + + if (read_int(info[i].present, &info[i].is_present) != 0) { + clog(LOG_INFO, "Skipping %s\n", info[i].cdev->name); diff -Nru cpufreqd-2.4.2/debian/patches/series cpufreqd-2.4.2/debian/patches/series --- cpufreqd-2.4.2/debian/patches/series 1970-01-01 09:00:00.000000000 +0900 +++ cpufreqd-2.4.2/debian/patches/series 2013-03-23 21:51:28.000000000 +0900 @@ -0,0 +1 @@ +619913.patch
