| Message ID | 20180108042235.GA21618@leoy-ThinkPad-T440 (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
On 01/08, Leo Yan wrote: > > I verified Yongjun patch v2 for using devm_ioremap_resource(), just as > Kevin mentioned after using devm_ioremap_resource() the kernel has > memory mapping conflict between stub clock driver and mailbox driver. > > Following Masahiro suggestion, I changed the stub clock driver to use > syscon & regmap to access the memory, I can confirm this can work > well, you could review in below change. > > But this change should note two things, one thing is it is dependent on > mailbox driver upstreaming, so I prefer we can commit one patch for this > change along with mailbox driver patch set; the second thing is we > need change DT binding document. > > If you have any better solution for this, also very welcome. > So I'll apply the original patch it sounds like.
On Wed, Jan 10, 2018 at 02:48:20PM -0800, Stephen Boyd wrote: > On 01/08, Leo Yan wrote: > > > > I verified Yongjun patch v2 for using devm_ioremap_resource(), just as > > Kevin mentioned after using devm_ioremap_resource() the kernel has > > memory mapping conflict between stub clock driver and mailbox driver. > > > > Following Masahiro suggestion, I changed the stub clock driver to use > > syscon & regmap to access the memory, I can confirm this can work > > well, you could review in below change. > > > > But this change should note two things, one thing is it is dependent on > > mailbox driver upstreaming, so I prefer we can commit one patch for this > > change along with mailbox driver patch set; the second thing is we > > need change DT binding document. > > > > If you have any better solution for this, also very welcome. > > > > So I'll apply the original patch it sounds like. Thanks for this and it's good for me, this is simple way to fix the NULL checking issue. Thanks, Leo Yan -- To unsubscribe from this list: send the line "unsubscribe linux-clk" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/drivers/clk/hisilicon/clk-hi3660-stub.c b/drivers/clk/hisilicon/clk-hi3660-stub.c index e8dde4f..40ee63b 100644 --- a/drivers/clk/hisilicon/clk-hi3660-stub.c +++ b/drivers/clk/hisilicon/clk-hi3660-stub.c @@ -25,12 +25,14 @@ #include <linux/err.h> #include <linux/init.h> #include <linux/mailbox_client.h> +#include <linux/mfd/syscon.h> #include <linux/module.h> #include <linux/of.h> #include <linux/platform_device.h> +#include <linux/regmap.h> #include <dt-bindings/clock/hi3660-clock.h> -#define HI3660_STUB_CLOCK_DATA (0x70) +#define HI3660_STUB_CLOCK_DATA (0x570) #define MHZ (1000 * 1000) #define DEFINE_CLK_STUB(_id, _cmd, _name) \ @@ -60,7 +62,7 @@ struct hi3660_stub_clk { unsigned int rate; }; -static void __iomem *freq_reg; +static struct regmap *freq_reg; static struct hi3660_stub_clk_chan stub_clk_chan; static unsigned long hi3660_stub_clk_recalc_rate(struct clk_hw *hw, @@ -72,7 +74,9 @@ static unsigned long hi3660_stub_clk_recalc_rate(struct clk_hw *hw, * LPM3 writes back the CPU frequency in shared SRAM so read * back the frequency. */ - stub_clk->rate = readl(freq_reg + (stub_clk->id << 2)) * MHZ; + regmap_read(freq_reg, HI3660_STUB_CLOCK_DATA + (stub_clk->id << 2), + &stub_clk->rate); + stub_clk->rate *= MHZ; return stub_clk->rate; } @@ -133,7 +137,7 @@ static struct clk_hw *hi3660_stub_clk_hw_get(struct of_phandle_args *clkspec, static int hi3660_stub_clk_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; - struct resource *res; + struct device_node *np = pdev->dev.of_node; unsigned int i; int ret; @@ -148,12 +152,12 @@ static int hi3660_stub_clk_probe(struct platform_device *pdev) if (IS_ERR(stub_clk_chan.mbox)) return PTR_ERR(stub_clk_chan.mbox); - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - freq_reg = devm_ioremap_resource(dev, res); - if (IS_ERR(freq_reg)) + freq_reg = syscon_regmap_lookup_by_phandle(np, + "hisilicon,hi3660-clk-sram"); + if (IS_ERR(freq_reg)) { + dev_err(dev, "failed to get sram regmap\n"); return PTR_ERR(freq_reg); - - freq_reg += HI3660_STUB_CLOCK_DATA; + } for (i = 0; i < HI3660_CLK_STUB_NUM; i++) { ret = devm_clk_hw_register(&pdev->dev, &hi3660_stub_clks[i].hw);