Messages in this thread | | | Subject | Re: [PATCH] x86: new Intel Atom SoC power management controller driver | From | Joe Perches <> | Date | Thu, 06 Mar 2014 16:41:11 -0800 |
| |
On Fri, 2014-03-07 at 08:26 +0800, Li, Aubrey wrote:
> my patch is word wrapped while I was not aware, hope this one work.
trivial notes:
> diff --git a/arch/x86/include/asm/pmc_atom.h b/arch/x86/include/asm/pmc_atom.h [] > +#define BIT_LPSS1_F0_DMA (1 << 0) > +#define BIT_LPSS1_F1_PWM1 (1 << 1)
These could use BIT(foo)
#define BIT_LPSS1_F0_DMA BIT(0) #define BIT_LPSS1_F1_PWM1 BIT(1)
etc..
> +#define BIT_SMB (1 << 0) > +#define BIT_USH_SS_PHY (1 << 1) > +#define BIT_OTG_SS_PHY (1 << 2) > +#define BIT_DFX (1 << 3)
here too.
> diff --git a/arch/x86/kernel/pmc_atom.c b/arch/x86/kernel/pmc_atom.c [] > @@ -0,0 +1,301 @@ > +/* > + * Intel Atom SOC Power Management Controller Driver [] > + */
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> +static void pmc_power_off(void) > +{ [] > + printk(KERN_INFO, "PMC: Preparing to enter system sleep state S5\n");
pr_info("Preparing to enter system sleep state S5\n");
> + if (func_dis_index & dev_map[dev_index].bit_mask) { > + seq_printf(s, "Dev: %-32s\tState: %s [%s]\n", > + dev_map[dev_index].name, "Disabled", > + (d3_sts_index & dev_map[dev_index].bit_mask) ? > + "D3":"D0"); > + } else { > + seq_printf(s, "Dev: %-32s\tState: %s [%s]\n", > + dev_map[dev_index].name, "Enabled ", > + (d3_sts_index & dev_map[dev_index].bit_mask) ? > + "D3":"D0"); > + }
Maybe better with a single seq_printf
seq_printf(s, "Dev: %-32s\tState: %s [%s]\n", dev_map[dev_index].name, dev_map[dev_index].bit_mask & func_dis_index ? "Disabled" : " Enabled ", dev_map[dev_index].bit_mask & d3_sts_index ? "D3" : "D0");
> +static int pmc_probe(struct pci_dev *pdev, > + const struct pci_device_id *unused) > +{
> + pmc = devm_kzalloc(&pdev->dev, sizeof(struct pmc_dev), GFP_KERNEL); > + if (!pmc) { > + dev_err(&pdev->dev, "error: could not allocate memory\n");
Unnecessary OOM message, kzalloc will dump_stack()
| |