On Wed, 18 Apr 2012, Stephen Warren wrote:
> From: Stephen Warren <swarren@xxxxxxxxxx>
>
> In the ClearPortFeature/USB_PORT_FEAT_ENABLE case, ehci_hub_control()
> would read from status_reg, clear PORT_PE, and write the result back to
> status_reg. This would clear any bits in PORT_RWC_BITS that were set in
> the registers. Fix this by masking these bits off before the write.
>
> Since this masking is common across all ClearPortFeature cases, move it
> into a single early location to avoid duplicating it.
>
> Remove the same bugfix from ehci-tegra.c's tegra_ehci_hub_control(), now
> that this case is correctly handled by the core.
>
> Signed-off-by: Stephen Warren <swarren@xxxxxxxxxx>
Acked-by: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
> ---
> drivers/usb/host/ehci-hub.c | 17 +++++++----------
> drivers/usb/host/ehci-tegra.c | 13 +------------
> 2 files changed, 8 insertions(+), 22 deletions(-)
>
> diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c
> index 38fe076..3ff35b9 100644
> --- a/drivers/usb/host/ehci-hub.c
> +++ b/drivers/usb/host/ehci-hub.c
> @@ -699,6 +699,7 @@ static int ehci_hub_control (
> goto error;
> wIndex--;
> temp = ehci_readl(ehci, status_reg);
> + temp &= ~PORT_RWC_BITS;
>
> /*
> * Even if OWNER is set, so the port is owned by the
> @@ -712,8 +713,7 @@ static int ehci_hub_control (
> ehci_writel(ehci, temp & ~PORT_PE, status_reg);
> break;
> case USB_PORT_FEAT_C_ENABLE:
> - ehci_writel(ehci, (temp & ~PORT_RWC_BITS) | PORT_PEC,
> - status_reg);
> + ehci_writel(ehci, temp | PORT_PEC, status_reg);
> break;
> case USB_PORT_FEAT_SUSPEND:
> if (temp & PORT_RESET)
> @@ -742,7 +742,7 @@ static int ehci_hub_control (
> spin_lock_irqsave(&ehci->lock, flags);
> }
> /* resume signaling for 20 msec */
> - temp &= ~(PORT_RWC_BITS | PORT_WAKE_BITS);
> + temp &= ~PORT_WAKE_BITS;
> ehci_writel(ehci, temp | PORT_RESUME, status_reg);
> ehci->reset_done[wIndex] = jiffies
> + msecs_to_jiffies(20);
> @@ -752,9 +752,8 @@ static int ehci_hub_control (
> break;
> case USB_PORT_FEAT_POWER:
> if (HCS_PPC (ehci->hcs_params))
> - ehci_writel(ehci,
> - temp & ~(PORT_RWC_BITS | PORT_POWER),
> - status_reg);
> + ehci_writel(ehci, temp & ~PORT_POWER,
> + status_reg);
> break;
> case USB_PORT_FEAT_C_CONNECTION:
> if (ehci->has_lpm) {
> @@ -762,12 +761,10 @@ static int ehci_hub_control (
> temp &= ~PORT_LPM;
> temp &= ~PORT_DEV_ADDR;
> }
> - ehci_writel(ehci, (temp & ~PORT_RWC_BITS) | PORT_CSC,
> - status_reg);
> + ehci_writel(ehci, temp | PORT_CSC, status_reg);
> break;
> case USB_PORT_FEAT_C_OVER_CURRENT:
> - ehci_writel(ehci, (temp & ~PORT_RWC_BITS) | PORT_OCC,
> - status_reg);
> + ehci_writel(ehci, temp | PORT_OCC, status_reg);
> break;
> case USB_PORT_FEAT_C_RESET:
> /* GetPortStatus clears reset */
> diff --git a/drivers/usb/host/ehci-tegra.c b/drivers/usb/host/ehci-tegra.c
> index 10bb3ea..0300137 100644
> --- a/drivers/usb/host/ehci-tegra.c
> +++ b/drivers/usb/host/ehci-tegra.c
> @@ -148,18 +148,7 @@ static int tegra_ehci_hub_control(
>
> spin_lock_irqsave(&ehci->lock, flags);
>
> - /*
> - * In ehci_hub_control() for USB_PORT_FEAT_ENABLE clears the other bits
> - * that are write on clear, by writing back the register read value, so
> - * USB_PORT_FEAT_ENABLE is handled by masking the set on clear bits
> - */
> - if (typeReq == ClearPortFeature && wValue == USB_PORT_FEAT_ENABLE) {
> - temp = ehci_readl(ehci, status_reg) & ~PORT_RWC_BITS;
> - ehci_writel(ehci, temp & ~PORT_PE, status_reg);
> - goto done;
> - }
> -
> - else if (typeReq == GetPortStatus) {
> + if (typeReq == GetPortStatus) {
> temp = ehci_readl(ehci, status_reg);
> if (tegra->port_resuming && !(temp & PORT_SUSPEND)) {
> /* Resume completed, re-enable disconnect detection */
>
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
