Hi Simon,
On Tue, Dec 8, 2015 at 11:38 AM, Simon Glass sjg@chromium.org wrote:
Set up a PCH init() method where we will put init code. Rename the existing bd82x6x_init() to bd82x6x_init_extra().
Signed-off-by: Simon Glass sjg@chromium.org
arch/x86/cpu/ivybridge/bd82x6x.c | 8 +++++++- arch/x86/cpu/ivybridge/cpu.c | 11 +++++++++++ arch/x86/cpu/ivybridge/pci.c | 2 +- arch/x86/include/asm/arch-ivybridge/bd82x6x.h | 2 +- 4 files changed, 20 insertions(+), 3 deletions(-)
diff --git a/arch/x86/cpu/ivybridge/bd82x6x.c b/arch/x86/cpu/ivybridge/bd82x6x.c index be39bcd..fce6622 100644 --- a/arch/x86/cpu/ivybridge/bd82x6x.c +++ b/arch/x86/cpu/ivybridge/bd82x6x.c @@ -100,7 +100,7 @@ static int bd82x6x_probe(struct udevice *dev) return 0; }
-int bd82x6x_init(void) +int bd82x6x_init_extra(void) { const void *blob = gd->fdt_blob; int sata_node; @@ -125,7 +125,13 @@ static int bd82x6x_pch_get_version(struct udevice *dev) return 9; }
+static int bd82x6x_init(struct udevice *dev) +{
return 0;+}
static const struct pch_ops bd82x6x_pch_ops = {
.init = bd82x6x_init, .get_version = bd82x6x_pch_get_version,};
diff --git a/arch/x86/cpu/ivybridge/cpu.c b/arch/x86/cpu/ivybridge/cpu.c index 5f4683c..c1f5ec9 100644 --- a/arch/x86/cpu/ivybridge/cpu.c +++ b/arch/x86/cpu/ivybridge/cpu.c @@ -15,6 +15,7 @@ #include <dm.h> #include <errno.h> #include <fdtdec.h> +#include <pch.h> #include <asm/cpu.h> #include <asm/io.h> #include <asm/lapic.h> @@ -211,6 +212,7 @@ int print_cpuinfo(void) { enum pei_boot_mode_t boot_mode = PEI_BOOT_NONE; char processor_name[CPU_MAX_NAME_LEN];
struct udevice *dev; const char *name; uint32_t pm1_cnt; uint16_t pm1_sts;@@ -241,6 +243,15 @@ int print_cpuinfo(void) }
/* Early chipset init required before RAM init can work */
ret = uclass_first_device(UCLASS_PCH, &dev);
Previous patch mentioned that probing LPC driver will automatically probe its parent PCH. Is this not the case?
if (ret)return ret;if (!dev)return -ENODEV;ret = pch_init(dev);if (ret)return ret;sandybridge_early_init(SANDYBRIDGE_MOBILE); /* Check PM1_STS[15] to see if we are waking from Sx */diff --git a/arch/x86/cpu/ivybridge/pci.c b/arch/x86/cpu/ivybridge/pci.c index 5e90f30..8af99b4 100644 --- a/arch/x86/cpu/ivybridge/pci.c +++ b/arch/x86/cpu/ivybridge/pci.c @@ -26,7 +26,7 @@ static int pci_ivybridge_probe(struct udevice *bus) if (!(gd->flags & GD_FLG_RELOC)) return 0; post_code(0x50);
bd82x6x_init();
bd82x6x_init_extra(); post_code(0x51); reg16 = 0xff;diff --git a/arch/x86/include/asm/arch-ivybridge/bd82x6x.h b/arch/x86/include/asm/arch-ivybridge/bd82x6x.h index fcdf6e2..d76cb8d 100644 --- a/arch/x86/include/asm/arch-ivybridge/bd82x6x.h +++ b/arch/x86/include/asm/arch-ivybridge/bd82x6x.h @@ -13,7 +13,7 @@ void bd82x6x_pci_init(pci_dev_t dev); void bd82x6x_usb_ehci_init(pci_dev_t dev); void bd82x6x_usb_xhci_init(pci_dev_t dev); int gma_func0_init(struct udevice *dev, const void *blob, int node); -int bd82x6x_init(void); +int bd82x6x_init_extra(void);
/**
- struct x86_cpu_priv - Information about a single CPU
--
As I mentioned previously, I don't understand why we create the init op for PCH uclass, instead we cannot put the init into PCH driver's probe routine?
Regards, Bin