On Thursday, 29 August 2024 05:36:47 EDT Heiko Stuebner wrote: > On Wed, 28 Aug 2024 15:42:50 +0000, Detlev Casanova wrote: > > Add clock and reset ID defines for rk3576. > > > > Compared to the downstream bindings written by Elaine, this uses > > continous gapless IDs starting at 0. Thus all numbers are > > different between downstream and upstream, but names are kept > > exactly the same. > > > > [...] > > Applied, thanks! > > [1/3] dt-bindings: clock, reset: Add support for rk3576 > commit: 49c04453db81fc806906e26ef9fc53bdb635ff39 > [2/3] clk: rockchip: Add new pll type pll_rk3588_ddr > commit: e781bffc296766b55dbd048890d558655031e8d1 > [3/3] clk: rockchip: Add clock controller for the RK3576 > commit: cc40f5baa91bb7b031f5622e11a4e443cb771527 Awesome ! > general remark, please take a look at your mail setup. > Amazon seems to break the generated message-ids. > > Your cover-letter is > > 0102019199a75f9b-aab57db6-806a-474b-8295-e5be5a99d424-000000@xxxxxxxxxxxxxx > onses.com > > while the patch (1-3) mails say > In-Reply-To: <20240828154243.57286-1-detlev.casanova@xxxxxxxxxxxxx> > > So that amazon thing somehow broke the message-ids in your mails. Yes, sorry, we are working on that. Recent switch in email setup and still figuring out all the issues :/ > I've also dropped the whole module part. > > As always that Android GKI madness was cause for issues. > The driver claims to be buildable as module, but it looks like nobody > tried that: > > First build-failure: > -------------------- > > ../drivers/clk/rockchip/clk-rk3576.c:1800:36: warning: ‘struct > platform_device’ declared inside parameter list will not be visible outside > of this definition or declaration 1800 | static int clk_rk3576_probe(struct > platform_device *pdev) > > | ^~~~~~~~~~~~~~~ > > ../drivers/clk/rockchip/clk-rk3576.c: In function ‘clk_rk3576_probe’: > ../drivers/clk/rockchip/clk-rk3576.c:1802:38: error: invalid use of > undefined type ‘struct platform_device’ 1802 | struct device_node > *np = pdev->dev.of_node; > > | ^~ > > ... > > missing platform_device header > > Second build-failure, after fixing the whole module madnes: > ----------------------------------------------------------- > > MODPOST Module.symvers > ERROR: modpost: missing MODULE_LICENSE() in > drivers/clk/rockchip/rst-rk3576.o ERROR: modpost: "rk3576_rst_init" > [drivers/clk/rockchip/clk-rk3576.ko] undefined! make[3]: *** > [../scripts/Makefile.modpost:145: Module.symvers] Fehler 1 > > > > So when applying the series, I simply removed the whole module-part and made > the init look like rk3588 for now. > > > Somehow I always get the impression the whole "clock as a module" thing > is just there so Rockchip can ship something completely out of tree on > Android devices. Mmh, I'll more careful with that, but indeed, I did not try building the clock driver as a module. > Best regards,
