Restore the old behavior of ehci_setup_phy() and ohci_setup_phy() to
return success when generic_phy_get_by_index() return -ENOENT.
Fixes: 84e561407a ("phy: Add generic_{setup,shutdown}_phy() helpers")
Fixes: 10005004db ("usb: ohci: Make usage of generic_{setup,shutdown}_phy() helpers")
Fixes: 083f8aa978 ("usb: ehci: Make usage of generic_{setup,shutdown}_phy() helpers")
Fixes: 75341e9c16 ("usb: ehci: Remove unused ehci_{setup,shutdown}_phy() helpers")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
generic_phy_exit() typically return 0 for a struct phy that has been
initialized with a generic_phy_init() call.
generic_setup_phy() returns the value from a generic_phy_exit() call
when generic_phy_power_on() fails. This hides the failed state of the
power_on ops from the caller of generic_setup_phy().
Fix this by ignoring the return value of the generic_phy_exit() call and
return the value from the generic_phy_power_on() call.
Fixes: 84e561407a ("phy: Add generic_{setup,shutdown}_phy() helpers")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Generic phy helpers typically use generic_phy_valid() to determine if
the helper should perform its function on a passed struct phy.
generic_phy_valid() treat any struct phy having phy->dev set as valid.
With generic_phy_get_by_index_nodev() setting phy->dev to a valid struct
udevice early, there can be situations where the struct phy is returned
as valid when initialization in fact failed and returned an error.
Fix this by setting phy->dev back to NULL when any of the calls to
of_xlate ops, device_get_supply_regulator or phy_alloc_counts fail. Also
extend the dm_test_phy_base test with a test where of_xlate ops fail.
Fixes: 72e5016f87 ("drivers: phy: add generic PHY framework")
Fixes: b9688df3cb ("drivers: phy: Set phy->dev to NULL when generic_phy_get_by_index() fails")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
generic_phy_get_by_name() does not initialize phy->dev to NULL before
returning when dev_read_stringlist_search() fails. This can lead to an
uninitialized or reused struct phy erroneously be report as valid by
generic_phy_valid().
Fix this issue by initializing phy->dev to NULL, also extend the
dm_test_phy_base test with calls to generic_phy_valid().
Fixes: b9688df3cb ("drivers: phy: Set phy->dev to NULL when generic_phy_get_by_index() fails")
Fixes: 868d58f69c ("usb: dwc3: Fix non-usb3 configurations")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Wrap the assert macros in ({ ... }) so they can be safely used both as
right side argument as well as in conditionals without curly brackets
around them. In the process, find a bunch of missing semicolons, fix
them.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
On boards using the RK3399 SoC, the USB OHCI and EHCI controllers share
the same PHY device instance. While these controllers are being stopped
they both attempt to power-off and deinitialize it, but trying to
power-off the deinitialized PHY device results in a hang. This usually
happens just before booting an OS, and can be explicitly triggered by
running "usb start; usb stop" in the U-Boot shell.
Implement a uclass-wide counting mechanism for PHY initialization and
power state change requests, so that we don't power-off/deinitialize a
PHY instance until all of its users want it done. The Allwinner A10 USB
PHY driver does this counting in-driver, remove those parts in favour of
this in-uclass implementation.
The sandbox PHY operations test needs some changes since the uclass will
no longer call into the drivers for actions matching its tracked state
(e.g. powering-off a powered-off PHY). Update that test, and add a new
one which simulates multiple users of a single PHY.
The major complication here is that PHY handles aren't deduplicated per
instance, so the obvious idea of putting the counts in the PHY handles
don't immediately work. It seems possible to bind a child udevice per
PHY instance to the PHY provider and deduplicate the handles in each
child's uclass-private areas, like in the CLK framework. An alternative
approach could be to use those bound child udevices themselves as the
PHY handles. Instead, to avoid the architectural changes those would
require, this patch solves things by dynamically allocating a list of
structs (one per instance) in the provider's uclass-private area.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Peter Robinson <pbrobinson@gmail.com> - Rock960
Usage of lists_bind_fdt() in bind command imposes to add
a compatible string for bind-test node.
Others impacts are:
- bind-test node is binded at sandbox start, so no need to bind it
in test_bind_unbind_with_node() test.
- As explained just above, after sandbox start, now a phy exist.
In test/dm/phy.c, it was verified that a third phy didn't exist,
now we must verified that a fourth phy doesn't exist.
Signed-off-by: Patrice Chotard <patrice.chotard@st.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The test flags used by driver model are currently not available to other
tests. Rather than creating two sets of flags, make these flags generic
by changing the DM_ prefix to UT_ and moving them to the test.h header.
This will allow adding other test flags without confusion.
Signed-off-by: Simon Glass <sjg@chromium.org>
These header file should not be included in other header files. Remove
them and add to each individual file. Add test/test.h to test/ui.h since
that is a reasonable place.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a test item for the phy_bulk API
Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Weijie Gao <weijie.gao@mediatek.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.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 have a large number of places where while we historically referenced
gd in the code we no longer do, as well as cases where the code added
that line "just in case" during development and never dropped it.
Signed-off-by: Tom Rini <trini@konsulko.com>
Use ut_asserteq() to test equality since this gives a better error message
on failure. Also make a few of the tests more specific.
Signed-off-by: Simon Glass <sjg@chromium.org>
Those tests check:
- the ability for a phy-user to get a phy based on its name or its index
- the ability of a phy device (provider) to manage multiple ports
- the ability to perform operations on the phy (init,deinit,on,off)
- the behavior of the uclass when optional operations are not implemented
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
