|
Bugzilla – Full Text Bug Listing |
| Summary: | rtc_cmos: probe of 00:02 failed with error -16 | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE 10.3 | Reporter: | Marko Schugardt <marko.schugardt> |
| Component: | Kernel | Assignee: | E-mail List <kernel-maintainers> |
| Status: | RESOLVED WONTFIX | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | forgotten_06k4h7fA91, jengelh, marko.schugardt |
| Version: | Final | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
|
Description
Marko Schugardt
2007-08-11 18:04:52 UTC
I suppose it's because the old-style rtc module (CONFIG_RTC=y) blocks rtc_cmos. Yeah, IMHO it should be removed and rtc_cmos be made the proper one. Note that udev also needs an extra rule to symlink rtc to rtc0 because programs like vmware and hwclock do not know of /dev/rtc0 yet. outdated I still do see this issue (http://lkml.org/lkml/2007/8/25/143) on my VIA EPIA MII-6000E, on which I have installed openSUSE 10.3 today. epia:~ # rpm -q kernel-default kernel-default-2.6.22.9-0.4 epia:~ # zcat /proc/config.gz | grep CONFIG_RTC CONFIG_RTC=y CONFIG_RTC_LIB=m CONFIG_RTC_CLASS=m CONFIG_RTC_INTF_SYSFS=y CONFIG_RTC_INTF_PROC=y CONFIG_RTC_INTF_DEV=y CONFIG_RTC_INTF_DEV_UIE_EMUL=y CONFIG_RTC_DRV_TEST=m CONFIG_RTC_DRV_DS1307=m CONFIG_RTC_DRV_DS1672=m CONFIG_RTC_DRV_MAX6900=m CONFIG_RTC_DRV_RS5C372=m CONFIG_RTC_DRV_ISL1208=m CONFIG_RTC_DRV_X1205=m CONFIG_RTC_DRV_PCF8563=m CONFIG_RTC_DRV_PCF8583=m CONFIG_RTC_DRV_RS5C348=m CONFIG_RTC_DRV_MAX6902=m CONFIG_RTC_DRV_CMOS=m CONFIG_RTC_DRV_DS1553=m CONFIG_RTC_DRV_DS1742=m CONFIG_RTC_DRV_M48T86=m CONFIG_RTC_DRV_V3020=m epia:~ # dmesg | grep rtc_cmos rtc_cmos 00:05: rtc core: registered rtc_cmos as rtc0 rtc_cmos: probe of 00:05 failed with error -16 It's just a warning, /dev/rtc is still usable, and driven by the legacy rtc driver. If you want to get rid of that, I can suggest you the (non-novell alert!) -ccj kernel and udev where I have disabled CONFIG_RTC for good and added a symlink into udev. (-16 = EBUSY, I suspect) But there must be something wrong. Although rtc_cmos is successfully loaded, I neither get /proc/acpi/alarm nor /sys/class/rtc/rtc0/wakealarm. I need at least one of them to let my VDR machine wake up at a specified time. >But there must be something wrong. Although rtc_cmos is successfully loaded, I >neither get /proc/acpi/alarm nor /sys/class/rtc/rtc0/wakealarm. The CMOS clock is already taken by rtc.ko (compiled in) when rtc_cmos.ko is about to be loaded! >I need at least one of them to let my VDR machine wake up at a specified time. I do have a /sys/class/rtc/rtc0/wakealarm, if rtc_cmos.ko manages to get hold of the CMOS RTC first. (Which is unlikely to happen with the stock SUSE kernel atm.) I understand your explanation of this error, but I do not understand how this bug could make it into the final release. According to https://bugzilla.novell.com/show_bug.cgi?id=287539, it looks like the bug will stay "forever". So I'll have to use your kernel. I'm disappointed of openSUSE 10.3! Reinhard - The reality is that sometimes bugs make it into released products. The problem was that fixing this bug ended up causing quite a few more elsewhere, and we were too close to the scheduled release date. Apologies, but RTC wakeup wasn't a big enough feature to delay shipping 10.3 Closing as WONTFIX. |