Messages in this thread | | | Date | Tue, 14 Aug 2012 14:21:44 +0530 | From | Laxman Dewangan <> | Subject | Re: "regulator: Enable supply regulator if child rail is enabled" broken? |
| |
On Tuesday 14 August 2012 02:08 PM, Rabin Vincent wrote: > It appears that this patch causes incorrect reference counting on supply > regulators if the child regulator is enabled with for example boot_on=true: > > commit b2296bd43e781976743354c668a356b0df98e1da > Author: Laxman Dewangan<ldewangan@nvidia.com> > Date: Mon Jan 2 13:08:45 2012 +0530 > > regulator: Enable supply regulator if child rail is enabled. > > If a regulator is enabled at boot (by constraints), then this patch calls > regulator_enable() on the supply regulator, leading to a use_count increase on > the supply. However, this patch adds no code to perform the corresponding > regulator_disable(), leaving the supply regulator on even when all consumers > have disabled it. > > AFAICS, this patch would have only worked for always_on regulators.
Yaah, It is missing the regulator_disable() in regulator_init_complete() for some cases. Please let me know if I need to send patch for fixing this issue.
| |