Hi Peter, On Mon, 2025-01-20 at 22:34 +0000, Peter Griffin wrote: > gs101 needs it's own suspend/resume callbacks to use the newly > added eint_fltcon_offset for saving & restoring fltcon0 & fltcon1 > registers. It also differs to previous SoCs in that fltcon1 > register doesn't always exist for each bank. > > exynosautov920 also has dedicated logic for using eint_con_offset > and eint_mask_offset for saving & restoring it's registers. > > Refactor the existing platform specific suspend/resume callback > so that each SoC variant has their own callback containing the > SoC specific logic. > > Additionally we now call drvdata->suspend() & drvdata->resume() > from within the loop that iterates the banks in > samsung_pinctrl_suspend() and samsung_pinctrl_resume(). Maybe split this patch in two: * first to do the refactoring plus adding exynosautov920_pinctrl_suspend() and exynosautov920_pinctrl_resume() * second to add gs101_pinctrl_suspend() / gs101_pinctrl_resume() This way, it's obvious which part is the bugfix and which part is the preparation and I believe it'd be easier to read. Cheers, Andre'
