On Fri, Feb 14, 2025 at 09:39:09PM +0300, Fedor Pchelkin wrote:
> On Fri, 14. Feb 22:44, Manivannan Sadhasivam wrote:
> > On Thu, Feb 13, 2025 at 02:24:16PM +0300, Vitalii Mordan wrote:
> > > If owl_port->clk was enabled in owl_uart_probe(), it must be disabled in
> > > all error paths to ensure proper cleanup. However, if uart_add_one_port()
> > > returns an error in owl_uart_probe(), the owl_port->clk clock will not be
> > > disabled.
> > >
> > > Use the devm_clk_get_enabled() helper function to ensure proper call
> > > balance for owl_port->clk.
> > >
> >
> > Do not use newly introduced APIs to fix old bugs. The bug should be fixed
> > separately to allow backporting and the conversion should be done on top.
>
> These relatively new helpers are already available in all currently
> supported stable kernels including 5.4.y.
>
> Commit 7ef9651e9792 ("clk: Provide new devm_clk helpers for prepared and
> enabled clocks") was conveniently backported there as a dependency for
> the similar bug fixes.
Ah, then fine with me.
- Mani
--
மணிவண்ணன் சதாசிவம்
