Hi Jean, On Wed, Sep 28, 2011 at 05:57:11AM -0400, Jean Delvare wrote: > Hi Jan, Guenter, > > On Fri, 23 Sep 2011 10:37:58 -0700, Guenter Roeck wrote: > > On Fri, 2011-09-23 at 06:35 -0400, Jan Beulich wrote: > > > ... as that has the potential to conflict with (particularly soft) CPU > > > hot removal and re-adding. > > > > > > Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx> > > > > > > --- > > > drivers/hwmon/coretemp.c | 2 +- > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > --- 3.1-rc7-coretemp.orig/drivers/hwmon/coretemp.c > > > +++ 3.1-rc7-coretemp/drivers/hwmon/coretemp.c > > > @@ -724,7 +724,7 @@ static int __cpuinit coretemp_device_add > > > > > > mutex_lock(&pdev_list_mutex); > > > > > > - pdev = platform_device_alloc(DRVNAME, cpu); > > > + pdev = platform_device_alloc(DRVNAME, TO_PHYS_ID(cpu)); > > > > This is incomplete, unfortunately. pdev->id is assumed to match the CPU > > ID, not the physical ID, and the code executes TO_PHYS_ID() again on it. > > I'll update the patch to fix that problem and apply it. > > > > Guenter > > This has been on my to-do list for a long time but I never got around > to doing it. Thanks Jan for stepping in. > > However it seems that the patch isn't OK, not even after Guenter's fix. > chk_ucode_version() assumes that pdev->id is a CPU number, not a > physical CPU ID, so currently the microcode checks are done on random > CPUs. Probably the microcode check should be moved from > coretemp_probe() to get_core_online(), which makes sense anyway, as > there is little point in creating a platform device if it's not going > to work. > This driver is becoming my nemesis :( > I have the following fix: > > From: Jean Delvare <khali@xxxxxxxxxxxx> > Subject: hwmon: (coretemp) Fixup platform device ID change > > With recent change "hwmon: (coretemp) don't use kernel assigned CPU > number as platform device ID", the microcode check is now running on > random CPU. Fix that by checking the microcode before creating the > platform device rather than at probe time. > > Also avoid calling TO_PHYS_ID(cpu) twice in the same function, it's > expensive. > > Signed-off-by: Jean Delvare <khali@xxxxxxxxxxxx> > Cc: Jan Beulich <jbeulich@xxxxxxxx> > Cc: Guenter Roeck <guenter.roeck@xxxxxxxxxxxx> > --- > drivers/hwmon/coretemp.c | 27 ++++++++++++--------------- > 1 file changed, 12 insertions(+), 15 deletions(-) > Thanks a lot for the fix. Applied. Guenter _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
