One specific USB 3.0 device behaves strangely when reset by
usb_new_device()'s call to hub_port_reset(). For some reason, the device
appears to briefly drop off the bus when this second bus reset is
executed, yet if we retry this loop, it'll eventually come back after
another two resets.
If USB bus reset is executed over and over within usb_new_device()'s call
to hub_port_reset(), I see the following sequence of results, which
repeats as long as you want:
1) STAT_C_CONNECTION = 1 STAT_CONNECTION = 0 USB_PORT_STAT_ENABLE 0
2) STAT_C_CONNECTION = 1 STAT_CONNECTION = 1 USB_PORT_STAT_ENABLE 0
3) STAT_C_CONNECTION = 1 STAT_CONNECTION = 1 USB_PORT_STAT_ENABLE 1
The device in question is a SanDisk Ultra USB 3.0 16GB memory stick with
USB VID/PID 0x0781/0x5581.
In order to allow this device to work with U-Boot, ignore the
{C_,}CONNECTION bits in the status/change registers, and only use the
ENABLE bit to determine if the reset was successful.
To be honest, extensive investigation has failed to determine why this
problem occurs. I'd love to know! I don't know if it's caused by:
* A HW bug in the device
* A HW bug in the Tegra USB controller
* A SW bug in the U-Boot Tegra USB driver
* A SW bug in the U-Boot USB core
This issue only occurs when the device's USB3 pins are attached to the
host; if only the USB2 pins are connected the issue does not occur. The
USB3 controller on Tegra is in reset, so is not actively communicating
with the device at all - a USB3 analyzer confirms this. Slightly
unplugging the device (so the USB3 pins don't contact) or using a USB2
cable or hub as an intermediary avoids the problem. For some reason,
the Linux kernel (either on the same Tegra board, or on an x86 host)
has no issue with the device, and I observe no disconnections during
reset.
This change won't affect any USB device that already works, since such
devices could not currently be triggering the error return this patch
removes, or they wouldn't be working currently.
However, this patch is quite reliable in practice, hence I hope it's
acceptable to solve the problem.
The only potential fallout I can see from this patch is:
* A broken device that triggers C_CONNECTION/!CONNECTION now causes the
loop in hub_port_reset() to run multiple times. If it never succeeds,
this will cause "usb start" to take roughly 1s extra to execute.
* If the user unplugs a device while hub_port_reset() is executing, and
very quickly swaps in a new device, hub_port_reset() might succeed on
the new device. This would mean that any information cached about the
original device (from the descriptor read in usb_new_device(), which
simply caches the max packet size) might be invalid, which would cause
problems talking to the new device. However, without this change, the
new device wouldn't work anyway, so this is probably not much of a
loss.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Now that we wait the correct specification-mandated time at the end of
usb_hub_power_on(), I suspect that CONFIG_USB_HUB_MIN_POWER_ON_DELAY has
no purpose.
For cm_t35.h, we already wait longer than the original MIN_POWER_ON_DELAY,
so this change is safe.
For gw_ventana.h, we will wait as long as the original MIN_POWER_ON_DELAY
iff pgood_delay was at least 200ms. I'm not sure if this is the case or
not, hence I've CC'd relevant people to test this change.
Cc: Igor Grinberg <grinberg@compulab.co.il>
Cc: Tim Harvey <tharvey@gateworks.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
usb_hub_power_on() currently waits for the maximum of (a) the hub port's
power output to become good, (b) the max time the USB specification
allows a device to take to connect.
However, these two operations must occur in series rather than in
parallel. First, the power supply ramps up to the level required to
power the USB device, and then the device may take a certain amount of
time to connect (assert D+/D- pullups).
Related, the maximum time that a device has to assert pullups is 1s not
100ms.
This is explained in "Connect Timing ECN.pdf", itself part of
usb_20_042814.zip from www.usb.org.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
The existing USB configuration parsing code relies on the descriptors'
own length values when reading through the configuration blob. Since the
size of those descriptors is always well-defined, we should rather use
the known sizes instead of trusting device-provided values to be
correct. Also adds some safety to potential out-of-order descriptors.
Change-Id: I16f69dfdd6793aa0fe930b5148d4521f3e5c3090
Signed-off-by: Julius Werner <jwerner@chromium.org>
Add a __weak function that can be overridden to reset devices
attached to an ehci devices after the FEAT_POWER has been submitted
Signed-off-by: Dan Murphy <dmurphy@ti.com>
When power cycling the hub ports, a misbehaving port will prevent all ports
from being powered on because we quit at the first sign of trouble.
Skip problematic ports instead of failing the entire power on.
Cc: Marek Vasut <marex@denx.de>
Cc: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
This patch adds support to both Faraday FUSBH200 and FOTG210,
the differences between Faraday EHCI and standard EHCI are
listed bellow:
1. The PORTSC starts at 0x30 instead of 0x44.
2. The CONFIGFLAG(0x40) is not only un-implemented, and
also has its address space removed.
3. Faraday EHCI is a TDI design, but it doesn't
compatible with the general TDI implementation
found at both U-Boot and Linux.
4. The ISOC descriptors differ from standard EHCI in
several ways. But since U-boot doesn't support ISOC,
we don't have to worry about that.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
CC: Marek Vasut <marex@denx.de>
This patch makes the minimum power-on delay for USB HUB
become configurable. The original design waits at least
100 msec here, but some EHCI controlers(e.g. Faraday EHCI)
are known to require much longer delay interval.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
CC: Marek Vasut <marex@denx.de>
Fix the Port status bit constants and Port feature number
constants as a part of USB 2.0 and USB 3.0 Hub class.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Untill now we power-cycle (aka: disable power on a port
and re-enabling again) one port at a time.
Delay of 20ms for Port-power to change multiplies with
number of ports in this case.
So better we parallelize this process:
disable power on all ports, wait for port-power to stabilize
and then re-enable the power subsequently.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
This adds usb framework support for super-speed usb, which will
further facilitate to add stack support for xHCI.
Signed-off-by: Vikas C Sajjan <vikas.sajjan@samsung.com>
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Patch b6d7852c increases timeout for enumeration, taking
worst case to be 10 sec.
get_timer() api returns timestamp in milliseconds, which is
what we are checking in the do-while() loop in usb_hub_configure()
(get_timer(start) < CONFIG_SYS_HZ * 10).
This should give us a required check for 10 seconds, and thereby
we don't need to add additional mdelay of 100 microseconds in
each cycle.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
Reviewed-by: Vipin Kumar <vipin.kumar@st.com>
XHCI ports are powered on after a H/W reset, however
EHCI ports are not. So disabling and re-enabling power
on all ports invariably.
Signed-off-by: Amar <amarendra.xt@samsung.com>
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
USB_PRINTF, USB_HUB_PRINTF, USB_STOR_PRINTF, USB_KBD_PRINTF
are nothing but conditional debug prints, depending on DEBUG.
So better remove them and use debug() simply.
Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com>
If probe of a newly connected device fails for some reason, clean up
the allocated entry in usb_dev array.
Signed-off-by: Milind Choudhary <milindc@codeaurora.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
The current logic reads the port status just once after usb_hub_power_on and
expects the portstatus and portchange to report the connection status
immediately and correctly.
Few pen drives are not able to report both of them immediately ie. those pens
report the connection change but not the connected state after the first read.
This opportunity once lost is gone for ever because the u-boot, unlike linux or
any other OS, works in polling mode.
This patch modifies the logic to read the port status continuously until the
portstatus and portchange both report a connection change as well as a connected
state or no connection change and no connection. This logic is placed in a
timeout of 10 sec. At the end of it, the pen drive would have either reported a
ONE or a ZERO in bit 1 of portstatus as well as portchange.
It enhances the set of pen drives which can eventually be detected by u-boot
Note: This 10 second timeout is based purely on several experiments done with
the broken pen drives
Signed-off-by: Vipin Kumar <vipin.kumar@st.com>
Acked-by: Igor Grinberg <grinberg@compulab.co.il>
Carry an index in the lowlevel usb functions to make specify the
respective usb controller.
Also pass through an controller struct from lowlevel_init to the
creation of the root usb device of this controller.
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Marek Vasut <marex@denx.de>
usb_hub_descriptor has to be packed as it's used for
communication with the device. Member wHubCharacteristics
violates the natural alignment rules.
Use explicit unaligned access functions for this member.
Fixes ARMv7 traping while using USB.
v2: fix typo found by Thomas Langer
v3: rebased on top of u-boot-usb/master
Signed-off-by: Lucas Stach <dev@lynxeye.de>
This avoids cache-alignment warnings shown in console
when a usb command is entered.
Whenever X bytes of unaligned buffer is invalidated, arm core
invalidates X + Y bytes as per the cache line size and throws
these warnings.
Signed-off-by: Puneet Saxena <puneets@nvidia.com>
Signed-off-by: Marek Vasut <marex@denx.de>
Common code has a mdelay() func, so use that instead of the usb-specific
wait_ms() func. This also fixes the build errors:
ohci-hcd.c: In function 'submit_common_msg':
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1519:9: sorry, unimplemented: called from here
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1816:10: sorry, unimplemented: called from here
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1827:10: sorry, unimplemented: called from here
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1844:10: sorry, unimplemented: called from here
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1563:11: sorry, unimplemented: called from here
/usr/local/src/u-boot/blackfin/include/usb.h:202:44: sorry, unimplemented: inlining failed in call to 'wait_ms': function body not available
ohci-hcd.c:1583:9: sorry, unimplemented: called from here
make[1]: *** [ohci-hcd.o] Error 1
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Marek Vasut <marex@denx.de>
Following the corresponding Linux code, this patch relaxes reset timings
waiting at least 100ms after power to the ports. There are some reports
that it helps make enumeration work better on some high speed devices.
Furthermore, the wait is only done once after power has been enabled
on all ports.
CC: Remy Bohmer <linux@bohmer.net>
CC: Vincent Palatin <vpalatin@chromium.org>
Signed-off-by: Wolfgang Grandegger <wg@denx.de>
