For now the driver does not probe if usbkbd was not present in stdin.
This presents two issues, we can not probe the driver before setting stdin
and we can not use this driver in other manner than stdin console.
This patch fixes this by adding an else statement. It simply probes the
driver without console management in the case "usbkbd" is not in stdin.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Using the XHCI driver, the function `usb_kbd_poll_for_event` takes
30-40ms to run. The exact time is dependent on the polling interval the
keyboard requests in its descriptor, and likely cannot be significantly
reduced without major rework to the XHCI driver.
The U-Boot EFI console service sets a timer to poll the keyboard every 5
microseconds, and this timer is checked every time a block is read off
disk. The net effect is that, on my system, loading a ~40MiB kernel and
initrd takes about 62 seconds with a slower keyboard and 53 seconds
with a faster one, with the vast majority of the time spent polling the
keyboard.
To solve this problem, this patch adds a 20ms delay between consecutive
calls to `usb_kbd_poll_for_event`. This is sufficient to reduce the
total loading time to under half a second for both keyboards, and does
not impact the perceived keystroke latency.
Signed-off-by: Thomas Watson <twatson52@icloud.com>
The OUT endpoint can just be ignored as it is not used, just as the
corresponding Set_Report request for IN-only interfaces. E.g. the
Linux gadget hid keyboard also provides an interrupt endpoint.
Also cleanup confusing debug messages like "found set protocol", which
is printed when a keyboard device is found, while the Set_Protocol request
is issued quite some time later.
Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
In case of IOMUX enabled it assumes that console devices in the list
are available to get them stopped properly via ->stop() callback.
However, the USB keyboard driver violates this assumption and tries
to play tricks so the device get destroyed while being listed as
an active console.
Swap the order of device deregistration and IOMUX update along with
converting to use iomux_replace_device() jelper to avoid the use-after-free.
Fixes: 3cbcb28928 ("usb: Fix usb_kbd_deregister when console-muxing is used")
Fixes: 8a83487030 ("dm: usb: Add a remove() method for USB keyboards")
Reported-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Move constant USB_KBD_BOOT_REPORT_SIZE. This allows us to reuse it.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Provide support for F1-F12, Insert, Delete, Home, End, Page Up, Page Down.
As this leads to a size increase provide a customizing setting
CONFIG_USB_KEYBOARD_FN_KEYS.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Avoid duplicate translation of arrow key codes.
Reduce code size by avoiding strings and eliminating
usb_kbd_put_sequence().
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
usb_kbd_buffer is defined as u8[]. So let usb_kbd_put_queue() use u8 as
type of the parameter for the new byte.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
So far arrows key pressed on an USB keyboard got translated to some
low ASCII control sequences (Ctrl+N, Ctrl+P). Some programs understand
these codes, but the standard for those keys is to use ANSI control
sequences for cursor movement (ESC [ A).
Our own boot menu is a victim of this, currently we cannot change the
selection with an USB keyboard due to this.
Since we already implement a queue for USB key codes, we can just insert
the three character ANSI sequence into the key buffer. This fixes the
bootmenu, and is more universal for other users (UEFI) as well.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This allows to disable the USB driver model in SPL because it checks
the CONFIG_SPL_DM_USB variable for SPL builds. Nothing changes for
regular non-SPL builds.
Signed-off-by: Sven Schwermer <sven@svenschwermer.de>
Driver supports only one instance of usb keyboard.
Remove the first dependency on generic usbkbd DEVNAME.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
When U-Boot started using SPDX tags we were among the early adopters and
there weren't a lot of other examples to borrow from. So we picked the
area of the file that usually had a full license text and replaced it
with an appropriate SPDX-License-Identifier: entry. Since then, the
Linux Kernel has adopted SPDX tags and they place it as the very first
line in a file (except where shebangs are used, then it's second line)
and with slightly different comment styles than us.
In part due to community overlap, in part due to better tag visibility
and in part for other minor reasons, switch over to that style.
This commit changes all instances where we have a single declared
license in the tag as both the before and after are identical in tag
contents. There's also a few places where I found we did not have a tag
and have introduced one.
Signed-off-by: Tom Rini <trini@konsulko.com>
We are now using an env_ prefix for environment functions. Rename these
two functions for consistency. Also add function comments in common.h.
Quite a few places use getenv() in a condition context, provoking a
warning from checkpatch. These are fixed up in this patch also.
Suggested-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Simon Glass <sjg@chromium.org>
CONFIG_CONSOLE_MUX and CONFIG_SYS_CONSOLE_IS_IN_ENV are not applicable
for SPL. Update the console code to use CONFIG_IS_ENABLED(), so that these
options will be inactive in SPL.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This converts the following to Kconfig:
CONFIG_SYS_STDIO_DEREGISTER
This option should never be enabled in SPL, so use
CONFIG_IS_ENABLED(SYS_STDIO_DEREGISTER) when checking the option.
Signed-off-by: Simon Glass <sjg@chromium.org>
[trini: Re-sync]
Signed-off-by: Tom Rini <trini@konsulko.com>
The dm usb_kbd_remove function() will deregister the usb keyboard for
us on a "usb reset" / "usb stop" so there is no need to manually call
usb_kbd_deregister() in the dm case.
This commit removes usb_kbd_deregister() in the dm case fixing the
following "usb reset" errors:
usb_kbd_remove: warning, ret=-6
device_remove: Device 'usb_kbd' failed to remove, but children are gone
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Scan code 0x39 is CapsLock, which is not a printable character and thus
is not covered by either usb_kbd_numkey_shifted[] or usb_kbd_numkey[].
Fix the scan code check to avoid looking it up in either of the arrays.
Signed-off-by: Marek Vasut <marex@denx.de>
At present USB keyboards are not properly removed with driver model. Add the
code to handle this.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Marek Vasut <marex@denx.de>
When CONFIG_SYS_USB_EVENT_POLL_VIA_CONTROL_EP is enabled, use a
GET_REPORT control transfer to retrieve the initial state of the
keyboard. This matches the technique used to poll the keyboard state.
This is useful since it eliminates the remaining use of interrupt
transfers from the USB keyboard driver, which allows it to work with
USB HCD that don't support interrupt transfers.
Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
The console includes a global variable and several functions that are only
used by a small subset of U-Boot files. Before adding more functions, move
the definitions into their own header file.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a uclass for keyboard input, mirroring the existing stdio methods.
This is enabled by a new CONFIG_DM_KEYBOARD option.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Switch USB keyboards over to use driver model instead of scanning with the
horrible usb_get_dev_index() function. This involves creating a new uclass
for keyboards, although so far there is no API.
Signed-off-by: Simon Glass <sjg@chromium.org>
Now that we have a new header file for cache-aligned allocation, we should
move the stack-based allocation macro there also.
Signed-off-by: Simon Glass <sjg@chromium.org>
In Linux USB_DEVICE() is used to declare a USB device by vendor/device ID.
We should follow the same convention in U-Boot. Rename the existing
USB_DEVICE() macro to U_BOOT_USB_DEVICE() and bring in the USB_DEVICE()
macro from Linux for use in U-Boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
When we're polling and thus handling key-repeat in software, make sure
to disable idle reports, some keyboards may have these enabled by default
messing up our software keyrepeat.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The usb-kbd key repeat code assumes that reports get repeated every 40 ms,
this is never true when using CONFIG_SYS_USB_EVENT_POLL_VIA_CONTROL_EP, and
does not always works for CONFIG_SYS_USB_EVENT_POLL and
CONFIG_SYS_USB_EVENT_POLL_VIA_INT_QUEUE since not all usb keyboards honor
the usb_set_idle() command.
For CONFIG_SYS_USB_EVENT_POLL we must use usb_set_idle() since we do a
blocking wait for the hid report, so if we do not tell the keyboard to send
a hid report every 40ms even if nothing changes then we will block u-boot
for 1s (the default u-boot usb interrupt packet timeout). Note that in this
case on keyboards which do not support usb_set_idle() we loose and we actually
get 1s latencies on other u-boot activities.
For the other poll-methods this commit stops using usb_set_idle() and instead
repeats the last received hid-report every 40 ms as long as no new hid-report
is received. This fixes key-repeat not working at all with
CONFIG_SYS_USB_EVENT_POLL_VIA_CONTROL_EP and fixes it not working with
keyboards which do not implement usb_set_idle() when using
CONFIG_SYS_USB_EVENT_POLL_VIA_INT_QUEUE.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Allow USB keyboards to work with driver model. The main difference is that
we can have multiple buses (each with its own device numbering) and each
bus must be scanned.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Marek Vasut <marex@denx.de>
Before adding driver model support, split out code from this over-long
function.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Marek Vasut <marex@denx.de>
When iomuxing is used we must not only deregister the device with stdio.c,
but also remove the reference to the device in the console_devices array
used by console-muxing. Add a call to iomux_doenv to usb_kbd_deregister to
update console_devices, which will drop the reference.
This fixes the console filling with "Failed to enqueue URB to controller"
messages after a "usb stop force", or when the USB keyboard is gone after a
"usb reset".
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Currently create_int_queue is only implemented by the ehci code, and that
does not honor interrupt intervals, but other drivers which might also want
to implement create_int_queue may honor intervals, so add an interval param.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Waiting an interrupt packet to complete in usb_kbd_poll_for_event, causes
a 40 ms latency for each call to usb_kbd_testc, which is undesirable.
Using control messages leads to lower (but still not 0) latency, but some
devices do not work well with control messages (e.g. my kvm behaves funny
with them).
This commit adds support for using the int_queue mechanism which at least
the ehci-hcd driver supports. This allows polling with 0 latency, while
using interrupt packets.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Use the new force parameter to make the stdio_deregister succeed, replacing
stdin with a nulldev, and assume that the usb keyboard will come back after
the reset.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
In some cases we really want to move forward with a deregister, add a force
parameter to allow this, and replace the dev with a nulldev in this case.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
We now always properly deregister the keyboard before calling
drv_usb_kbd_init(), so we can drop the check for already being registered.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
