| Message ID | 1465138776-6003-4-git-send-email-hdegoede@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi, On Sun, Jun 05, 2016 at 04:59:36PM +0200, Hans de Goede wrote: > phy-sun4i-usb now has proper dr_mode handling, it always registers an > extcon, and sends a notify with the mode (even when in peripheral- / > host-only mode) at least once. > > So we can simply the sunxi musb glue by always registering its extcon > notifier and relying on sunxi_musb_work() to enable vbus when in > host-only mode. > > This also enables host- and peripheral-only mode with vbus monitoring. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> It's been a bit painful to track all the patches needed so that it applies properly, but I've finally been able to test it on a Sinlinx SinA33 with peripheral-only mUSB, and it works like a charm. You can add my Tested-by. Thanks! Maxime
Hi, On 08-06-16 12:23, Maxime Ripard wrote: > Hi, > > On Sun, Jun 05, 2016 at 04:59:36PM +0200, Hans de Goede wrote: >> phy-sun4i-usb now has proper dr_mode handling, it always registers an >> extcon, and sends a notify with the mode (even when in peripheral- / >> host-only mode) at least once. >> >> So we can simply the sunxi musb glue by always registering its extcon >> notifier and relying on sunxi_musb_work() to enable vbus when in >> host-only mode. >> >> This also enables host- and peripheral-only mode with vbus monitoring. >> >> Signed-off-by: Hans de Goede <hdegoede@redhat.com> > > It's been a bit painful to track all the patches needed so that it > applies properly, but I've finally been able to test it on a Sinlinx > SinA33 with peripheral-only mUSB, and it works like a charm. > > You can add my Tested-by. Great, thanks for testing. This is the board which has an otg connector with vbus not connected, right? Yet it does have a functional id-pin, right ? In that case you should be able to put it in dual-role mode (only specify the id-pin in the phy dts node, no vbus / vbus-monitoring) and then it _should_ work in host mode if you use a powered hub. I'm fine with putting in peripheral-only mode, but as said dual-role might work with a powered hub. Regards, Hans
Hi, On Sun, Jun 05, 2016 at 04:59:36PM +0200, Hans de Goede wrote: > phy-sun4i-usb now has proper dr_mode handling, it always registers an > extcon, and sends a notify with the mode (even when in peripheral- / > host-only mode) at least once. > > So we can simply the sunxi musb glue by always registering its extcon > notifier and relying on sunxi_musb_work() to enable vbus when in > host-only mode. > > This also enables host- and peripheral-only mode with vbus monitoring. > > Signed-off-by: Hans de Goede <hdegoede@redhat.com> > --- > Changes in v2: > -No changes > Changes in v3: > -No changes Acked-by: Bin Liu <b-liu@ti.com> Regards, -Bin.
Hi, On Wed, Jun 08, 2016 at 12:30:20PM +0200, Hans de Goede wrote: > Hi, > > On 08-06-16 12:23, Maxime Ripard wrote: > >Hi, > > > >On Sun, Jun 05, 2016 at 04:59:36PM +0200, Hans de Goede wrote: > >>phy-sun4i-usb now has proper dr_mode handling, it always registers an > >>extcon, and sends a notify with the mode (even when in peripheral- / > >>host-only mode) at least once. > >> > >>So we can simply the sunxi musb glue by always registering its extcon > >>notifier and relying on sunxi_musb_work() to enable vbus when in > >>host-only mode. > >> > >>This also enables host- and peripheral-only mode with vbus monitoring. > >> > >>Signed-off-by: Hans de Goede <hdegoede@redhat.com> > > > >It's been a bit painful to track all the patches needed so that it > >applies properly, but I've finally been able to test it on a Sinlinx > >SinA33 with peripheral-only mUSB, and it works like a charm. > > > >You can add my Tested-by. > > Great, thanks for testing. > > This is the board which has an otg connector with vbus not connected, > right? Yet it does have a functional id-pin, right ? It's that one, yes. > In that case you should be able to put it in dual-role mode (only > specify the id-pin in the phy dts node, no vbus / vbus-monitoring) > and then it _should_ work in host mode if you use a powered hub. > > I'm fine with putting in peripheral-only mode, but as said > dual-role might work with a powered hub. Good point, I'll test that. Thanks! Maxime
diff --git a/drivers/usb/musb/sunxi.c b/drivers/usb/musb/sunxi.c index e7d4617..b88a2f6 100644 --- a/drivers/usb/musb/sunxi.c +++ b/drivers/usb/musb/sunxi.c @@ -255,12 +255,10 @@ static int sunxi_musb_init(struct musb *musb) writeb(SUNXI_MUSB_VEND0_PIO_MODE, musb->mregs + SUNXI_MUSB_VEND0); /* Register notifier before calling phy_init() */ - if (musb->port_mode == MUSB_PORT_MODE_DUAL_ROLE) { - ret = extcon_register_notifier(glue->extcon, EXTCON_USB_HOST, - &glue->host_nb); - if (ret) - goto error_reset_assert; - } + ret = extcon_register_notifier(glue->extcon, EXTCON_USB_HOST, + &glue->host_nb); + if (ret) + goto error_reset_assert; ret = phy_init(glue->phy); if (ret) @@ -274,9 +272,8 @@ static int sunxi_musb_init(struct musb *musb) return 0; error_unregister_notifier: - if (musb->port_mode == MUSB_PORT_MODE_DUAL_ROLE) - extcon_unregister_notifier(glue->extcon, EXTCON_USB_HOST, - &glue->host_nb); + extcon_unregister_notifier(glue->extcon, EXTCON_USB_HOST, + &glue->host_nb); error_reset_assert: if (test_bit(SUNXI_MUSB_FL_HAS_RESET, &glue->flags)) reset_control_assert(glue->rst); @@ -300,9 +297,8 @@ static int sunxi_musb_exit(struct musb *musb) phy_exit(glue->phy); - if (musb->port_mode == MUSB_PORT_MODE_DUAL_ROLE) - extcon_unregister_notifier(glue->extcon, EXTCON_USB_HOST, - &glue->host_nb); + extcon_unregister_notifier(glue->extcon, EXTCON_USB_HOST, + &glue->host_nb); if (test_bit(SUNXI_MUSB_FL_HAS_RESET, &glue->flags)) reset_control_assert(glue->rst); @@ -314,25 +310,6 @@ static int sunxi_musb_exit(struct musb *musb) return 0; } -static int sunxi_set_mode(struct musb *musb, u8 mode) -{ - struct sunxi_glue *glue = dev_get_drvdata(musb->controller->parent); - int ret; - - if (mode == MUSB_HOST) { - ret = phy_power_on(glue->phy); - if (ret) - return ret; - - set_bit(SUNXI_MUSB_FL_PHY_ON, &glue->flags); - /* Stop musb work from turning vbus off again */ - set_bit(SUNXI_MUSB_FL_VBUS_ON, &glue->flags); - musb->xceiv->otg->state = OTG_STATE_A_WAIT_VRISE; - } - - return 0; -} - static void sunxi_musb_enable(struct musb *musb) { struct sunxi_glue *glue = dev_get_drvdata(musb->controller->parent); @@ -579,7 +556,6 @@ static const struct musb_platform_ops sunxi_musb_ops = { .exit = sunxi_musb_exit, .enable = sunxi_musb_enable, .disable = sunxi_musb_disable, - .set_mode = sunxi_set_mode, .fifo_offset = sunxi_musb_fifo_offset, .ep_offset = sunxi_musb_ep_offset, .busctl_offset = sunxi_musb_busctl_offset, @@ -635,10 +611,6 @@ static int sunxi_musb_probe(struct platform_device *pdev) return -EINVAL; } - glue = devm_kzalloc(&pdev->dev, sizeof(*glue), GFP_KERNEL); - if (!glue) - return -ENOMEM; - memset(&pdata, 0, sizeof(pdata)); switch (usb_get_dr_mode(&pdev->dev)) { #if defined CONFIG_USB_MUSB_DUAL_ROLE || defined CONFIG_USB_MUSB_HOST @@ -646,15 +618,13 @@ static int sunxi_musb_probe(struct platform_device *pdev) pdata.mode = MUSB_PORT_MODE_HOST; break; #endif +#if defined CONFIG_USB_MUSB_DUAL_ROLE || defined CONFIG_USB_MUSB_GADGET + case USB_DR_MODE_PERIPHERAL: + pdata.mode = MUSB_PORT_MODE_GADGET; + break; +#endif #ifdef CONFIG_USB_MUSB_DUAL_ROLE case USB_DR_MODE_OTG: - glue->extcon = extcon_get_edev_by_phandle(&pdev->dev, 0); - if (IS_ERR(glue->extcon)) { - if (PTR_ERR(glue->extcon) == -EPROBE_DEFER) - return -EPROBE_DEFER; - dev_err(&pdev->dev, "Invalid or missing extcon\n"); - return PTR_ERR(glue->extcon); - } pdata.mode = MUSB_PORT_MODE_DUAL_ROLE; break; #endif @@ -665,6 +635,10 @@ static int sunxi_musb_probe(struct platform_device *pdev) pdata.platform_ops = &sunxi_musb_ops; pdata.config = &sunxi_musb_hdrc_config; + glue = devm_kzalloc(&pdev->dev, sizeof(*glue), GFP_KERNEL); + if (!glue) + return -ENOMEM; + glue->dev = &pdev->dev; INIT_WORK(&glue->work, sunxi_musb_work); glue->host_nb.notifier_call = sunxi_musb_host_notifier; @@ -698,6 +672,14 @@ static int sunxi_musb_probe(struct platform_device *pdev) } } + glue->extcon = extcon_get_edev_by_phandle(&pdev->dev, 0); + if (IS_ERR(glue->extcon)) { + if (PTR_ERR(glue->extcon) == -EPROBE_DEFER) + return -EPROBE_DEFER; + dev_err(&pdev->dev, "Invalid or missing extcon\n"); + return PTR_ERR(glue->extcon); + } + glue->phy = devm_phy_get(&pdev->dev, "usb"); if (IS_ERR(glue->phy)) { if (PTR_ERR(glue->phy) == -EPROBE_DEFER)
phy-sun4i-usb now has proper dr_mode handling, it always registers an extcon, and sends a notify with the mode (even when in peripheral- / host-only mode) at least once. So we can simply the sunxi musb glue by always registering its extcon notifier and relying on sunxi_musb_work() to enable vbus when in host-only mode. This also enables host- and peripheral-only mode with vbus monitoring. Signed-off-by: Hans de Goede <hdegoede@redhat.com> --- Changes in v2: -No changes Changes in v3: -No changes --- drivers/usb/musb/sunxi.c | 68 ++++++++++++++++++------------------------------ 1 file changed, 25 insertions(+), 43 deletions(-)