diff options
author | Sarah Sharp <sarah.a.sharp@linux.intel.com> | 2009-04-28 04:58:26 +0200 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@suse.de> | 2009-06-16 06:44:49 +0200 |
commit | 79abb1ab13cee5ba488210798b6e7bbae0b391ac (patch) | |
tree | 9bb008003d30146a092609a59882925640a5c93c /drivers/usb | |
parent | USB: Parse and store the SuperSpeed endpoint companion descriptors. (diff) | |
download | linux-79abb1ab13cee5ba488210798b6e7bbae0b391ac.tar.xz linux-79abb1ab13cee5ba488210798b6e7bbae0b391ac.zip |
USB: Support for bandwidth allocation.
Originally, the USB core had no support for allocating bandwidth when a
particular configuration or alternate setting for an interface was
selected. Instead, the device driver's URB submission would fail if
there was not enough bandwidth for a periodic endpoint. Drivers could
work around this, by using the scatter-gather list API to guarantee
bandwidth.
This patch adds host controller API to allow the USB core to allocate or
deallocate bandwidth for an endpoint. Endpoints are added to or dropped
from a copy of the current schedule by calling add_endpoint() or
drop_endpoint(), and then the schedule is atomically evaluated with a
call to check_bandwidth(). This allows all the endpoints for a new
configuration or alternate setting to be added at the same time that the
endpoints from the old configuration or alt setting are dropped.
Endpoints must be added to the schedule before any URBs are submitted to
them. The HCD must be allowed to reject a new configuration or alt
setting before the control transfer is sent to the device requesting the
change. It may reject the change because there is not enough bandwidth,
not enough internal resources (such as memory on an embedded host
controller), or perhaps even for security reasons in a virtualized
environment.
If the call to check_bandwidth() fails, the USB core must call
reset_bandwidth(). This causes the schedule to be reverted back to the
state it was in just after the last successful check_bandwidth() call.
If the call succeeds, the host controller driver (and hardware) will have
changed its internal state to match the new configuration or alternate
setting. The USB core can then issue a control transfer to the device to
change the configuration or alt setting. This allows the core to test new
configurations or alternate settings before unbinding drivers bound to
interfaces in the old configuration.
WIP:
The USB core must add endpoints from all interfaces in a configuration
to the schedule, because a driver may claim that interface at any time.
A slight optimization might be to add the endpoints to the schedule once
a driver claims that interface. FIXME
This patch does not cover changing alternate settings, but it does
handle a configuration change or de-configuration. FIXME
The code for managing the schedule is currently HCD specific. A generic
scheduling algorithm could be added for host controllers without
built-in scheduling support. For now, if a host controller does not
define the check_bandwidth() function, the call to
usb_hcd_check_bandwidth() will always succeed.
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'drivers/usb')
-rw-r--r-- | drivers/usb/core/hcd.c | 86 | ||||
-rw-r--r-- | drivers/usb/core/hcd.h | 32 | ||||
-rw-r--r-- | drivers/usb/core/message.c | 20 | ||||
-rw-r--r-- | drivers/usb/core/urb.c | 6 |
4 files changed, 144 insertions, 0 deletions
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index 823744d80cb9..b2da4753b12e 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -1560,6 +1560,92 @@ rescan: } } +/* Check whether a new configuration or alt setting for an interface + * will exceed the bandwidth for the bus (or the host controller resources). + * Only pass in a non-NULL config or interface, not both! + * Passing NULL for both new_config and new_intf means the device will be + * de-configured by issuing a set configuration 0 command. + */ +int usb_hcd_check_bandwidth(struct usb_device *udev, + struct usb_host_config *new_config, + struct usb_interface *new_intf) +{ + int num_intfs, i, j; + struct usb_interface_cache *intf_cache; + struct usb_host_interface *alt = 0; + int ret = 0; + struct usb_hcd *hcd; + struct usb_host_endpoint *ep; + + hcd = bus_to_hcd(udev->bus); + if (!hcd->driver->check_bandwidth) + return 0; + + /* Configuration is being removed - set configuration 0 */ + if (!new_config && !new_intf) { + for (i = 1; i < 16; ++i) { + ep = udev->ep_out[i]; + if (ep) + hcd->driver->drop_endpoint(hcd, udev, ep); + ep = udev->ep_in[i]; + if (ep) + hcd->driver->drop_endpoint(hcd, udev, ep); + } + hcd->driver->check_bandwidth(hcd, udev); + return 0; + } + /* Check if the HCD says there's enough bandwidth. Enable all endpoints + * each interface's alt setting 0 and ask the HCD to check the bandwidth + * of the bus. There will always be bandwidth for endpoint 0, so it's + * ok to exclude it. + */ + if (new_config) { + num_intfs = new_config->desc.bNumInterfaces; + /* Remove endpoints (except endpoint 0, which is always on the + * schedule) from the old config from the schedule + */ + for (i = 1; i < 16; ++i) { + ep = udev->ep_out[i]; + if (ep) { + ret = hcd->driver->drop_endpoint(hcd, udev, ep); + if (ret < 0) + goto reset; + } + ep = udev->ep_in[i]; + if (ep) { + ret = hcd->driver->drop_endpoint(hcd, udev, ep); + if (ret < 0) + goto reset; + } + } + for (i = 0; i < num_intfs; ++i) { + + /* Dig the endpoints for alt setting 0 out of the + * interface cache for this interface + */ + intf_cache = new_config->intf_cache[i]; + for (j = 0; j < intf_cache->num_altsetting; j++) { + if (intf_cache->altsetting[j].desc.bAlternateSetting == 0) + alt = &intf_cache->altsetting[j]; + } + if (!alt) { + printk(KERN_DEBUG "Did not find alt setting 0 for intf %d\n", i); + continue; + } + for (j = 0; j < alt->desc.bNumEndpoints; j++) { + ret = hcd->driver->add_endpoint(hcd, udev, &alt->endpoint[j]); + if (ret < 0) + goto reset; + } + } + } + ret = hcd->driver->check_bandwidth(hcd, udev); +reset: + if (ret < 0) + hcd->driver->reset_bandwidth(hcd, udev); + return ret; +} + /* Disables the endpoint: synchronizes with the hcd to make sure all * endpoint state is gone from hardware. usb_hcd_flush_endpoint() must * have been called previously. Use for set_configuration, set_interface, diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h index ae6d9db41ca9..d397ecfd5b17 100644 --- a/drivers/usb/core/hcd.h +++ b/drivers/usb/core/hcd.h @@ -232,6 +232,35 @@ struct hc_driver { int (*alloc_dev)(struct usb_hcd *, struct usb_device *); /* Called by usb_release_dev to free HC device structures */ void (*free_dev)(struct usb_hcd *, struct usb_device *); + + /* Bandwidth computation functions */ + /* Note that add_endpoint() can only be called once per endpoint before + * check_bandwidth() or reset_bandwidth() must be called. + * drop_endpoint() can only be called once per endpoint also. + * A call to xhci_drop_endpoint() followed by a call to xhci_add_endpoint() will + * add the endpoint to the schedule with possibly new parameters denoted by a + * different endpoint descriptor in usb_host_endpoint. + * A call to xhci_add_endpoint() followed by a call to xhci_drop_endpoint() is + * not allowed. + */ + /* Allocate endpoint resources and add them to a new schedule */ + int (*add_endpoint)(struct usb_hcd *, struct usb_device *, struct usb_host_endpoint *); + /* Drop an endpoint from a new schedule */ + int (*drop_endpoint)(struct usb_hcd *, struct usb_device *, struct usb_host_endpoint *); + /* Check that a new hardware configuration, set using + * endpoint_enable and endpoint_disable, does not exceed bus + * bandwidth. This must be called before any set configuration + * or set interface requests are sent to the device. + */ + int (*check_bandwidth)(struct usb_hcd *, struct usb_device *); + /* Reset the device schedule to the last known good schedule, + * which was set from a previous successful call to + * check_bandwidth(). This reverts any add_endpoint() and + * drop_endpoint() calls since that last successful call. + * Used for when a check_bandwidth() call fails due to resource + * or bandwidth constraints. + */ + void (*reset_bandwidth)(struct usb_hcd *, struct usb_device *); /* Returns the hardware-chosen device address */ int (*address_device)(struct usb_hcd *, struct usb_device *udev); }; @@ -252,6 +281,9 @@ extern void usb_hcd_disable_endpoint(struct usb_device *udev, extern void usb_hcd_reset_endpoint(struct usb_device *udev, struct usb_host_endpoint *ep); extern void usb_hcd_synchronize_unlinks(struct usb_device *udev); +extern int usb_hcd_check_bandwidth(struct usb_device *udev, + struct usb_host_config *new_config, + struct usb_interface *new_intf); extern int usb_hcd_get_frame_number(struct usb_device *udev); extern struct usb_hcd *usb_create_hcd(const struct hc_driver *driver, diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 9bd26dec7599..3a2e69ec2f29 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -510,6 +510,10 @@ EXPORT_SYMBOL_GPL(usb_sg_init); * could be transferred. That capability is less useful for low or full * speed interrupt endpoints, which allow at most one packet per millisecond, * of at most 8 or 64 bytes (respectively). + * + * It is not necessary to call this function to reserve bandwidth for devices + * under an xHCI host controller, as the bandwidth is reserved when the + * configuration or interface alt setting is selected. */ void usb_sg_wait(struct usb_sg_request *io) { @@ -1653,6 +1657,21 @@ free_interfaces: if (ret) goto free_interfaces; + /* Make sure we have bandwidth (and available HCD resources) for this + * configuration. Remove endpoints from the schedule if we're dropping + * this configuration to set configuration 0. After this point, the + * host controller will not allow submissions to dropped endpoints. If + * this call fails, the device state is unchanged. + */ + if (cp) + ret = usb_hcd_check_bandwidth(dev, cp, NULL); + else + ret = usb_hcd_check_bandwidth(dev, NULL, NULL); + if (ret < 0) { + usb_autosuspend_device(dev); + goto free_interfaces; + } + /* if it's already configured, clear out old state first. * getting rid of old interfaces means unbinding their drivers. */ @@ -1675,6 +1694,7 @@ free_interfaces: dev->actconfig = cp; if (!cp) { usb_set_device_state(dev, USB_STATE_ADDRESS); + usb_hcd_check_bandwidth(dev, NULL, NULL); usb_autosuspend_device(dev); goto free_interfaces; } diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c index 02eb0ef7a4c3..0885d4abdc62 100644 --- a/drivers/usb/core/urb.c +++ b/drivers/usb/core/urb.c @@ -241,6 +241,12 @@ EXPORT_SYMBOL_GPL(usb_unanchor_urb); * If the USB subsystem can't allocate sufficient bandwidth to perform * the periodic request, submitting such a periodic request should fail. * + * For devices under xHCI, the bandwidth is reserved at configuration time, or + * when the alt setting is selected. If there is not enough bus bandwidth, the + * configuration/alt setting request will fail. Therefore, submissions to + * periodic endpoints on devices under xHCI should never fail due to bandwidth + * constraints. + * * Device drivers must explicitly request that repetition, by ensuring that * some URB is always on the endpoint's queue (except possibly for short * periods during completion callacks). When there is no longer an urb |