Messages in this thread | | | Date | Sat, 11 Dec 2021 11:47:04 -0800 | Subject | Re: rockchip_i2s_tdm.c:undefined reference to `clk_set_parent' | From | Randy Dunlap <> |
| |
Hi--
On 12/7/21 03:32, Nicolas Frattaroli wrote: > On Montag, 6. Dezember 2021 23:46:38 CET kernel test robot wrote: >> [...] >> >> All errors (new ones prefixed by >>): >> >> mips-linux-ld: sound/soc/jz4740/jz4740-i2s.o: in function `jz4740_i2s_set_sysclk': >> jz4740-i2s.c:(.text+0x3ec): undefined reference to `clk_set_parent' >> mips-linux-ld: jz4740-i2s.c:(.text+0x44c): undefined reference to `clk_set_parent' >> mips-linux-ld: sound/soc/rockchip/rockchip_i2s_tdm.o: in function `rockchip_i2s_tdm_calibrate_mclk.isra.0': >>>> rockchip_i2s_tdm.c:(.text+0x10d4): undefined reference to `clk_set_parent' >>>> mips-linux-ld: rockchip_i2s_tdm.c:(.text+0x1180): undefined reference to `clk_set_parent' >> > > According to some previous conversations I've stumbled upon[1], > this appears to be due to certain MIPS configurations not > implementing the clock API properly, so they don't provide a > clk_set_parent despite advertising that they have support for > clocks. > > So my question is: do I need to care about this? This hardware > will never be used on MIPS, and a lot of other drivers (as seen in > the errors snippet from the test robot) have the same issue, and > the problem is most likely not in my driver but in that specific > configuration's clock API implementation. > > > [1]: https://lore.kernel.org/lkml/8a41b718-a6f6-6b7f-1699-18ab619884c3@infradead.org/
Yes, AFAIK this is fixed by:
commit fc1aabb08886 Author: Randy Dunlap <rdunlap@infradead.org> Date: Sun Nov 14 17:20:51 2021 -0800
mips: lantiq: add support for clk_get_parent()
I can't reproduce this build error with the provided config file in a current kernel tree.
-- ~Randy
| |