Bug 4728
| Summary: | second (ultrabay) battery not detected if inserted after boot | ||
|---|---|---|---|
| Product: | ACPI | Reporter: | Eugene Pavlovsky (heilong) |
| Component: | Config-Hotplug | Assignee: | acpi_config-hotplug |
| Status: | REJECTED DUPLICATE | ||
| Severity: | normal | ||
| Priority: | P2 | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.12-rc5 | Subsystem: | |
| Regression: | --- | Bisected commit-id: | |
I might should have replied your email in acpi-dev, but I can't access bugzilla till today, so I can't point out the exact tract for this issue. Ok, Could you please test the patch in bug 2884? *** This bug has been marked as a duplicate of 2884 *** Yes, these kernel parameters make it work for me. Will the battery driver be fixed so that there would be no need for them? |
I've got an ultrabay battery for my thinkpad r51. If I boot my kernel with it, i can see both batteries in /proc/acpi/battery/BAT{0,1}. Then, I can hotswap ultrabay battery for a cdrom (hdc=cdrom boot param makes it work, though it's a kluge too, it'd be good to have some way for re-scanning IDE devices) then back and everything still works (CD works, /proac/acpi/battery/BAT1 works when battery is swapped back). But if I load my kernel without the extra battery plugged in, then add the battery, it won't show up in /proc/acpi/battery/BAT1, only the main (BAT0) will be there. Any way to tell the kernel to rescan the ACPI stuff? The process if plugging the battery itself generates an ACPI event which I can handle and execute smth. It'd be also great to know if there's any way of telling the kernel to rescan IDE devices - the "hdc=cdrom" way is not very pretty (there's idectl 1 rescan (uses hdparm, actually), but it needs the /dev/hdc device, and udev won't create it if there was no cd drive on kernel boot).