Re: [PATCH v5] EHCI bus glue for on-chip PMC MSP USB controller

[Dan Carpenter <error27@gmail.com>]

On Tue, Feb 22, 2011 at 09:05:33PM +0530, Anoop P.A wrote:
> From: Anoop <paanoop1@paanoop1-desktop.(none)>
> 
> This patch add bus glue for USB controller commonly found in PMC-Sierra 
> MSP71xx family of SoC's.
> 
> 
> Signed-off-by: Anoop P A <anoop.pa@gmail.com>
> ---
> Changes.
> ehci-pmcmsp.c is based on latest ehci-pci.c.Addressed some stylistic issue 
> pointed by Greg.
> Addressed comments from Matthieu CASTET.

Could you spell that out more completely next time?

> +config USB_EHCI_HCD_PMC_MSP
> +     tristate "EHCI support for on-chip PMC MSP USB controller"

Better to say "EHCI support for on-chip PMC-Sierra MSP71xx USB controllers"

> +     depends on USB_EHCI_HCD && MSP_HAS_USB
> +     default y

New features always default to No.

> +#include <msp_usb.h>

Cannot find the msp_usb.h in linux-next.  Doesn't compile.

> +static void usb_hcd_tdi_set_mode(struct ehci_hcd *ehci)
> +{
> +     u8 *base;
> +     u8 *statreg;
> +     u8 *fiforeg;
> +     u32 val;
> +     struct ehci_regs *reg_base = ehci->regs;
> +
> +     /* get register base */
> +     base = (u8 *)reg_base + USB_EHCI_REG_USB_MODE;
> +     statreg = (u8 *)reg_base + USB_EHCI_REG_USB_STATUS;
> +     fiforeg = (u8 *)reg_base + USB_EHCI_REG_USB_FIFO;
> +
> +     /* Disable controller mode stream */
> +     val = ehci_readl(ehci, (u32 *)base);

It doesn't compile so I can't test this, but I think that this will
cause a sparse warning.  "base" should have an __iomem tag.  Please
run sparse on this driver.

> +/* called after powerup, by probe or system-pm "wakeup" */
> +static int ehci_msp_reinit(struct ehci_hcd *ehci)
> +{
> +     ehci_port_power(ehci, 0);
> +
> +     return 0;

Better to make this function void.

> +}
> +
> +/* called during probe() after chip reset completes */
> +static int ehci_msp_setup(struct usb_hcd *hcd)
> +{
> +     struct ehci_hcd         *ehci = hcd_to_ehci(hcd);
> +     u32                     temp;
> +     int                     retval;

Needs a blank line here to separate declarations from code.

> +     ehci->big_endian_mmio = 1;
> +     ehci->big_endian_desc = 1;
> +
> +     ehci->caps = hcd->regs;
> +     ehci->regs = hcd->regs +
> +                     HC_LENGTH(ehci_readl(ehci, &ehci->caps->hc_capbase));

[snip]

> +     /* data structure init */
> +     retval = ehci_init(hcd);
> +     if (retval)
> +             return retval;
> +
> +     temp = HCS_N_CC(ehci->hcs_params) * HCS_N_PCC(ehci->hcs_params);
> +     temp &= 0x0f;

companion HCs * ports per CC & 0xf?

What's the &= 0x0f for?  It's left out of the printk.

> +     if (temp && HCS_N_PORTS(ehci->hcs_params) > temp) {
> +             ehci_dbg(ehci, "bogus port configuration: "
> +                     "cc=%d x pcc=%d < ports=%d\n",
> +                     HCS_N_CC(ehci->hcs_params),
> +                     HCS_N_PCC(ehci->hcs_params),
> +                     HCS_N_PORTS(ehci->hcs_params));
> +     }

[snip]

> +/*-------------------------------------------------------------------------*/

No need for these blank comments.

> +
> +static void msp_start_hc(struct platform_device *dev)
> +{
> +}
> +
> +static void msp_stop_hc(struct platform_device *dev)
> +{
> +}
> +

I don't understand the point of these empty functions.

> +static int ehci_msp_suspend(struct device *dev)
> +{
> +     struct usb_hcd *hcd = dev_get_drvdata(dev);
> +     struct ehci_hcd *ehci = hcd_to_ehci(hcd);
> +     unsigned long flags;
> +     int rc;
> +
> +     return 0;
> +     rc = 0;
> +
> +     if (time_before(jiffies, ehci->next_statechange))
> +             usleep(10000);

Is there still a usleep() function?  Either way, can you send us
something that compiles on linux-next?

regards,
dan carpenter