On 26/06/2019 11:53:59-0700, Stephen Boyd wrote: > Quoting Alexandre Belloni (2019-06-21 02:33:02) > > On 20/06/2019 10:30:42+0000, [email protected] wrote: > > > Hi, > > > > > > On 18.06.2019 12:55, Alexandre Belloni wrote: > > > > On 13/06/2019 15:37:06+0000, [email protected] wrote: > > > >> From: Claudiu Beznea <[email protected]> > > > >> > > > >> Hi, > > > >> > > > >> This series tries to improve error path for slow clock registrations > > > >> by adding functions to free resources and using them on failures. > > > >> > > > > > > > > Does the platform even boot when the slow clock is not available? > > > > > > > > The TCB clocksource would fail at: > > > > > > > > tc.slow_clk = of_clk_get_by_name(node->parent, "slow_clk"); > > > > if (IS_ERR(tc.slow_clk)) > > > > return PTR_ERR(tc.slow_clk); > > > > > > > > > > In case of using TC as clocksource, yes, the platform wouldn't boot if > > > slow > > > clock is not available, because, anyway the TC needs it. PIT may work > > > without it (if slow clock is not used to drive the PIT). > > > > > > For sure there are other IPs (which may be or are driven by slow clock) > > > which may not work if slow clock is driven them. > > > > > > Anyway, please let me know if you feel this series has no meaning. > > > > > > > Well, I'm not sure it is worth it but at the same time, it is not adding > > many lines and you already developed it... > > > > Is that a Reviewed-by or a Rejected-by tag? >
Reviewed-by: Alexandre Belloni <[email protected]> -- Alexandre Belloni, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
