| Message ID | 20191203190716.18917-2-yuri.benditovich@daynix.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Remove 'remote wakeup' flag from USB config descriptor | expand |
On Tue, Dec 03, 2019 at 09:07:15PM +0200, Yuri Benditovich wrote: > If the redirected device has this capability, Windows guest may > place the device into D2 and expect it to wake when the device > becomes active, but this will never happen. For example, when > internal Bluetooth adapter is redirected, keyboards and mice > connected to it do not work. Current commit removes this > capability (starting from machine 4.2) > Set 'usb-host.suppress-remote-wake' property to 'off' to keep > 'remote wake' as is or to 'on' to remove 'remote wake' on > 4.1 or earlier. > > Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com> > --- > hw/core/machine.c | 1 + > hw/usb/host-libusb.c | 20 ++++++++++++++++++++ > hw/usb/trace-events | 1 + > 3 files changed, 22 insertions(+) > > diff --git a/hw/core/machine.c b/hw/core/machine.c > index 1689ad3bf8..8c0eaad091 100644 > --- a/hw/core/machine.c > +++ b/hw/core/machine.c > @@ -29,6 +29,7 @@ > > GlobalProperty hw_compat_4_1[] = { > { "virtio-pci", "x-pcie-flr-init", "off" }, > + { "usb-host", "suppress-remote-wake", "off" }, > }; > const size_t hw_compat_4_1_len = G_N_ELEMENTS(hw_compat_4_1); In case this doesn't get merged in QEMU 4.2.0, the patch needs to be updated to change hw_compat_4_2 instead (after applying the 5.0 machine type patch[1]). [1] https://patchew.org/QEMU/20191112104811.30323-1-cohuck@redhat.com/
On Tue, Dec 10, 2019 at 12:32 AM Eduardo Habkost <ehabkost@redhat.com> wrote: > > On Tue, Dec 03, 2019 at 09:07:15PM +0200, Yuri Benditovich wrote: > > If the redirected device has this capability, Windows guest may > > place the device into D2 and expect it to wake when the device > > becomes active, but this will never happen. For example, when > > internal Bluetooth adapter is redirected, keyboards and mice > > connected to it do not work. Current commit removes this > > capability (starting from machine 4.2) > > Set 'usb-host.suppress-remote-wake' property to 'off' to keep > > 'remote wake' as is or to 'on' to remove 'remote wake' on > > 4.1 or earlier. > > > > Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com> > > --- > > hw/core/machine.c | 1 + > > hw/usb/host-libusb.c | 20 ++++++++++++++++++++ > > hw/usb/trace-events | 1 + > > 3 files changed, 22 insertions(+) > > > > diff --git a/hw/core/machine.c b/hw/core/machine.c > > index 1689ad3bf8..8c0eaad091 100644 > > --- a/hw/core/machine.c > > +++ b/hw/core/machine.c > > @@ -29,6 +29,7 @@ > > > > GlobalProperty hw_compat_4_1[] = { > > { "virtio-pci", "x-pcie-flr-init", "off" }, > > + { "usb-host", "suppress-remote-wake", "off" }, > > }; > > const size_t hw_compat_4_1_len = G_N_ELEMENTS(hw_compat_4_1); > > In case this doesn't get merged in QEMU 4.2.0, the patch needs to > be updated to change hw_compat_4_2 instead (after applying the > 5.0 machine type patch[1]). Of course I will resubmit if needed. > > [1] https://patchew.org/QEMU/20191112104811.30323-1-cohuck@redhat.com/ > > -- > Eduardo >
> > > GlobalProperty hw_compat_4_1[] = { > > > { "virtio-pci", "x-pcie-flr-init", "off" }, > > > + { "usb-host", "suppress-remote-wake", "off" }, > > > }; > > > const size_t hw_compat_4_1_len = G_N_ELEMENTS(hw_compat_4_1); > > > > In case this doesn't get merged in QEMU 4.2.0, the patch needs to > > be updated to change hw_compat_4_2 instead (after applying the > > 5.0 machine type patch[1]). > > Of course I will resubmit if needed. Ping. Submission was too close to the release, so it'll go into 5.0. Please resend with compat properties updated accordingly (otherwise the patches are fine). thanks, Gerd
diff --git a/hw/core/machine.c b/hw/core/machine.c index 1689ad3bf8..8c0eaad091 100644 --- a/hw/core/machine.c +++ b/hw/core/machine.c @@ -29,6 +29,7 @@ GlobalProperty hw_compat_4_1[] = { { "virtio-pci", "x-pcie-flr-init", "off" }, + { "usb-host", "suppress-remote-wake", "off" }, }; const size_t hw_compat_4_1_len = G_N_ELEMENTS(hw_compat_4_1); diff --git a/hw/usb/host-libusb.c b/hw/usb/host-libusb.c index fcf48c0193..00e0e36369 100644 --- a/hw/usb/host-libusb.c +++ b/hw/usb/host-libusb.c @@ -88,6 +88,7 @@ struct USBHostDevice { bool needs_autoscan; bool allow_one_guest_reset; bool allow_all_guest_resets; + bool suppress_remote_wake; /* state */ QTAILQ_ENTRY(USBHostDevice) next; @@ -386,6 +387,8 @@ static void LIBUSB_CALL usb_host_req_complete_ctrl(struct libusb_transfer *xfer) r->p->status = status_map[xfer->status]; r->p->actual_length = xfer->actual_length; if (r->in && xfer->actual_length) { + USBDevice *udev = USB_DEVICE(s); + struct libusb_config_descriptor *conf = (void *)r->cbuf; memcpy(r->cbuf, r->buffer + 8, xfer->actual_length); /* Fix up USB-3 ep0 maxpacket size to allow superspeed connected devices @@ -394,6 +397,21 @@ static void LIBUSB_CALL usb_host_req_complete_ctrl(struct libusb_transfer *xfer) r->cbuf[7] == 9) { r->cbuf[7] = 64; } + /* + *If this is GET_DESCRIPTOR request for configuration descriptor, + * remove 'remote wakeup' flag from it to prevent idle power down + * in Windows guest + */ + if (s->suppress_remote_wake && + udev->setup_buf[0] == USB_DIR_IN && + udev->setup_buf[1] == USB_REQ_GET_DESCRIPTOR && + udev->setup_buf[3] == USB_DT_CONFIG && udev->setup_buf[2] == 0 && + xfer->actual_length > + offsetof(struct libusb_config_descriptor, bmAttributes) && + (conf->bmAttributes & USB_CFG_ATT_WAKEUP)) { + trace_usb_host_remote_wakeup_removed(s->bus_num, s->addr); + conf->bmAttributes &= ~USB_CFG_ATT_WAKEUP; + } } trace_usb_host_req_complete(s->bus_num, s->addr, r->p, r->p->status, r->p->actual_length); @@ -1596,6 +1614,8 @@ static Property usb_host_dev_properties[] = { LIBUSB_LOG_LEVEL_WARNING), DEFINE_PROP_BIT("pipeline", USBHostDevice, options, USB_HOST_OPT_PIPELINE, true), + DEFINE_PROP_BOOL("suppress-remote-wake", USBHostDevice, + suppress_remote_wake, true), DEFINE_PROP_END_OF_LIST(), }; diff --git a/hw/usb/trace-events b/hw/usb/trace-events index 2d3713351c..1c24d82c09 100644 --- a/hw/usb/trace-events +++ b/hw/usb/trace-events @@ -266,3 +266,4 @@ usb_host_parse_config(int bus, int addr, int value, int active) "dev %d:%d, valu usb_host_parse_interface(int bus, int addr, int num, int alt, int active) "dev %d:%d, num %d, alt %d, active %d" usb_host_parse_endpoint(int bus, int addr, int ep, const char *dir, const char *type, int active) "dev %d:%d, ep %d, %s, %s, active %d" usb_host_parse_error(int bus, int addr, const char *errmsg) "dev %d:%d, msg %s" +usb_host_remote_wakeup_removed(int bus, int addr) "dev %d:%d"
If the redirected device has this capability, Windows guest may place the device into D2 and expect it to wake when the device becomes active, but this will never happen. For example, when internal Bluetooth adapter is redirected, keyboards and mice connected to it do not work. Current commit removes this capability (starting from machine 4.2) Set 'usb-host.suppress-remote-wake' property to 'off' to keep 'remote wake' as is or to 'on' to remove 'remote wake' on 4.1 or earlier. Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com> --- hw/core/machine.c | 1 + hw/usb/host-libusb.c | 20 ++++++++++++++++++++ hw/usb/trace-events | 1 + 3 files changed, 22 insertions(+)