Currently, we are seeing problem with adc_read_all_channels() for
STM32F, and thus 'adc' console command reports incorrect values. Before
that's fixed, read ADC channels one by one to work around this problem.
BRANCH=Ryu
BUG=chrome-os-partner:33971
TEST='adc' on Ryu
Signed-off-by: Vic Yang <victoryang@chromium.org>
Change-Id: Iae92b82b24f6a843b9d46a8804da1e51d33ed7cb
Reviewed-on: https://chromium-review.googlesource.com/231125
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Check the flash protection at startup, if the RDP is still at level 0
(no read protection) or if the RO partition is not write protected :
- set the write protection on the first 16KB of flash (4 LSB of WRP0)
- push the RDP to level 1, so SWD/serial monitor needs to fully erase
the part before re-writing the code or the write-protection.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=samus
BUG=chrome-os-partner:34935
TEST=dump the content of the option bytes.
Change-Id: I11af64365a6fbc34327b2e463eb8e2d369ffacd2
Reviewed-on: https://chromium-review.googlesource.com/238262
Reviewed-by: Alec Berg <alecaberg@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Trybot-Ready: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
With the type-c connector, it's common for VBUS to make contact before
D+/D-, which can lead to charger misdetection. To work around this, add
a 200ms delay and trigger re-detection when a charger is inserted. This
should fix most misdetects due to unintentional slow plug (though it's
still possible to misdetect if insertion is deliberately very slow).
BUG=chrome-os-partner:34584
TEST=Manual on Samus. Plug in Apple charger 20 times, verify that it is
always detected correctly. Deliberately plug in slowly, verify it is
detected as 500mA SDP port. Repeat tests with various other CDP / SDP
ports to verify detection never over-currents and always indicates a
charger is plugged (or not plugged) correctly.
BRANCH=Samus
Change-Id: I8a776f516d8e7f0cedcb9d8579239eba641cab09
Signed-off-by: Shawn Nematbakhsh <shawnn@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/237241
Reviewed-by: Alec Berg <alecaberg@chromium.org>
STM32 seems to actually measure the rising and falling time of the I2C clock, so
if one uses a really small resistor the timing will go faster than with a bigger
resistor.
This commit makes it so the I2C frequency is limited to max 100kHz (respecting
the spec) no matter what size resistor (essentially we assume 0 rise and fall
times). While this will make stuff slower on boards with big resistors (where
they might have been under 100kHz anyway) this is the best compromise (since the
spec does not specify min frequency) without getting config defines for the
fall/rise times.
The TSCLH of some boards would be too short with the recommended timing
settings from spec, so increases the TSCLH would be better for everyone.
This patch does not touch the higher frequencies since the rise and fall times
do contribute a lot more to clocks, if the same method was used for those
frequencies, the speeds would have to be a lot slower.
BUG=chrome-os-partner:34375
BRANCH=None
TEST=on any EC, note how frequency does not go above 100kHz
TEST=As per tSCL = tSYNC1 + tSYNC2 + { [(SCLH+1) + (SCLL+1)] x
(PRESC+1) x tI2CCLK } from datasheet
Change-Id: Ibbeecac7f3da1b22d2ba3bca29ee3c17bfe997f5
Signed-off-by: Chris Zhong <zyw@rock-chips.com>
Signed-off-by: Alexandru M Stan <amstan@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/234077
Reviewed-by: Randall Spangler <rspangler@chromium.org>
When the EC is busy enough to start missing RX characters an ORECF might come
(Overrun error), we never check for this bit, so that means the interrupt
handler will be called in a loop because it never clears that bit.
This disables the overrun detection feature.
BUG=chrome-os-partner:33451
TEST=get firmware to poll the EC for keyboard presses, type really fast on the
keyboard, note how system does not watchdog reset anymore but loses the
occasional character instead.
BRANCH=None
Change-Id: I711483768e4ba80aaeb4a324c7dee790b3a23682
Signed-off-by: Alexandru M Stan <amstan@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/236088
Reviewed-by: Randall Spangler <rspangler@chromium.org>
According the STM32L RM, when writing to RCC_CFGR to change HSI or MSI
clock source, SWS bits of RCC_CFGR register have to be checked if the
new clock source is taken into account. Also, when writing ACC64 bit and
LATENCY bit to FLASH_ACR register, those bits have to be checked too.
Also changed in this CL is to disable MSI if HSI is enabled, and disable
HSI if MSI is enabled.
BUG=chrome-os-partner:32936
BRANCH=none
TEST=passed suspend_stress_test on big, blaze and nyan
Change-Id: I3ec660d149ecdec3ca3097239612bf2c542d0548
Signed-off-by: Yen Lin <yelin@nvidia.com>
Reviewed-on: https://chromium-review.googlesource.com/234490
Reviewed-by: Kary Jin <karyj@nvidia.com>
Tested-by: Kary Jin <karyj@nvidia.com>
Reviewed-by: Andrew Bresticker <abrestic@chromium.org>
(cherry picked from commit 400d9dca8c41f74cf0c2587e881707b80bb17d3d)
Reviewed-on: https://chromium-review.googlesource.com/236050
Added check for collision just before transmitting on CC line.
To check for collision, RX monitoring is left on all the time
(except when in the act of receiving or transmitting, or in
between receiving and sending a goodCRC), and a
simple check for RX transmission started is used to see if the
CC line is idle or not.
RX monitoring is also changed to only trigger on 3 edges within
20us, as per the PD spec.
When a collision is detected by seeing that CC is not idle, the
transmitting packet is dropped.
BUG=chrome-os-partner:30135
BRANCH=samus
TEST=load onto samus and zinger. make sure we negotiate and make
sure custom VDMs succeed. enabled pings and made sure we stay
alive with pings for a few min.
Also added code to pd_rx_handler to toggle a test point on EVT
board to verify the timing of when we get RX interrupts:
Change-Id: I22d172163319437d3d901e019eda79d4d592f6b8
Signed-off-by: Alec Berg <alecaberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/226118
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Ryu sensor hub has asymectric RO/RW images. The first one is very limited
(not i2c master, no sensor drivers, gesture recognition).
Image size is alter to offer more space for the RW firmware image,
compiled with ryu_sh board.
To write RO image and basic RW image:
flashrom -V -p ec:type=sh,block=0x800 --fast-verify -w /tmp/ryu_sh_loader/ec.bin
To write the expected RW image:
flashrom -V -p ec:type=sh,block=0x800 --fast-verify -w -i EC_RW:/tmp/ryu_sh/ec.bin
BRANCH=ToT
BUG=chrome-os-partner:33908
CQ-DEPEND=CL:231970,CL:233233
TEST=load on Ryu, confirmed limited operation.
Change-Id: Ib976e2b048935adfb9b2b072c071db5be2bc1c09
Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/231984
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
The task based approach made sense when it looked like
there would be a case closed debugging task to handle
multiple bridges (SPI/I2C/USART...). I'm not convinced
anymore that that task will be needed, so this
simplification seems good.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Change-Id: Ic431c287c28d10252246fe9f507d9c5fcc64a077
Reviewed-on: https://chromium-review.googlesource.com/232733
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
This allows the USB SPI bridge to be controlled from the
host at a larger timescale than a single SPI transaction.
This allows the host to signal that many transactions
will take place and that the device should keep the SPI
bridge enabled across them. This allows the device to
hold the AP or other possible user of the SPI bus in
reset while the bridge is enabled.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Change-Id: Ifd6f96b0ff47f35d853735d44e255a205b0e677a
Reviewed-on: https://chromium-review.googlesource.com/232732
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
CL to migrate the flashing VDMs from zinger's custom vdm to
common/usb_pd_flash.c such that other updateable type-C devices can
share.
Additionally adds gaskets to call standard runtime flashing facilities
for USB-PD devices using it.
Signed-off-by: Todd Broch <tbroch@chromium.org>
BRANCH=samus
BUG=chrome-os-partner:31192,chrome-os-partner:31193
TEST=manual,
Try following:
1. From samus_pd console w/ zinger in port 1
pd 1 flash version
pd 1 flash reboot
pd 1 flash info
2. From samus linux prompt w/ zinger in port 1
ectool --name cros_pd flashpd 1 1 <zinger RW payload>
Reading 16384 bytes from
/usr/local/zinger_v1.1.2528-d809e42.ec.RW.bin...
Erasing expected RW hash
Rebooting
Erasing RW flash
Writing RW flash
Rebooting PD into new RW
Complete
3. Repeat 1&2 above on hoho & dingdong.
Change-Id: I018055fa9de128f937c57debdc21dea026137bcf
Reviewed-on: https://chromium-review.googlesource.com/231835
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Alec Berg <alecaberg@chromium.org>
Tested-by: Todd Broch <tbroch@chromium.org>
Commit-Queue: Todd Broch <tbroch@chromium.org>
A non-zero error code returned by the callback causes EP0 to STALL.
This is the common mechanism used in USB to indicate an error while
processing a control request. This simplifies the implementation
of interface callbacks.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Change-Id: I89ceb4892f9f810fcaf6e975e6982fc5b2ae447b
Reviewed-on: https://chromium-review.googlesource.com/232368
Reviewed-by: Anton Staaf <robotboy@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
When using gpio_set_flags_by_mask() to modify a GPIO pin setting, the
former code was not able to revert an open-drain GPIO into push-pull
mode (the other way round was working). Fix it to be able to transition
in each direction.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=all
BUG=none
TEST=on Twinkie, run "tw res none RP3A0" then "adc" and see the RP3A0
GPIO set to 1 and pulling the CC to 3.3V.
Change-Id: Ic2d8ed95ea21b21d261a12ba298931d8bd5e6919
Reviewed-on: https://chromium-review.googlesource.com/231150
Reviewed-by: Vic Yang <victoryang@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Randall Spangler <rspangler@chromium.org>
For USB console, we wait for USB host to grab the data before writing
the next chunk of characters. However, if the USB peripheral is
disabled, the current implementation is confused as to whether the host
has grabbed the previous chunk of data.
Fix this by explicitly checking for USB peripheral enabled.
BRANCH=ryu
BUG=None
TEST=Boot on Ryu and check console isn't slow.
Signed-off-by: Vic Yang <victoryang@chromium.org>
Change-Id: I438c95835e56707b7ca1796734f2e47062799f44
Reviewed-on: https://chromium-review.googlesource.com/231115
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Add commands to send PD packets and to tweak individual parameters (TX
clock frequency, RX detection threshold, resistors on CCx).
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=none
BUG=chrome-os-partner:28337
TEST=verify that the PD sniffing is still working by dumping traffic
between Zinger and Samus.
Connect Twinkie to Zinger, set Rd by using "tw res NONE RD" and see VBUS
going to 5V (reading it using "ina 0").
Send a BIST mode 2 request using the following command :
tw send 2 0x1043 50000000
and see the other end starting sending BIST.
Change-Id: I3c8ddf858435ac1c17a43f59351bbaa69603a209
Reviewed-on: https://chromium-review.googlesource.com/227778
Reviewed-by: Todd Broch <tbroch@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Extend the Zinger runtime to take into account the disable_sleep()
issued by the USB protocol stack and avoid going into deep-sleep while
connected.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=samus
BUG=none
TEST=connect Zinger to a PD power sink (Twinkie) and monitor the stop
mode entry/exit on a GPIO.
Change-Id: I04e35fdd65f3be3da7a4304dc1a92e6268930888
Reviewed-on: https://chromium-review.googlesource.com/230340
Reviewed-by: Alec Berg <alecaberg@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Allow to use larger RSA keys by setting CONFIG_RSA_KEY_SIZE to 4096 or
8192 rather than using the default 2048-bit size.
It's mainly for benchmarking purpose right now as we don't have the RAM
to store the 3x key size buffer and the flash space for the public key
structure.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=samus
BUG=none
TEST=build Zinger with CONFIG_RSA_KEY_SIZE equals to 4096 and run it.
Change-Id: I9839121bf158d0a30dde1e48d875f345191bfec2
Reviewed-on: https://chromium-review.googlesource.com/228925
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
This enables forwarding of the local PD/EC console
over debug USB. It gates the console functionality
based on the CCD mode that is set.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable partial CCD mode on ryu and verify that it is
enumerated by the host correctly, but doesn't respond
to console input, and doesn't generate output.
Enable full CCD mode on ryu and verify that it is
enumerated and that the console works as expected.
Verify that the console still works by default on the
discovery-stm32f072 board.
Change-Id: I0325ce9689486c41387d6075330be1d7d42f1d42
Reviewed-on: https://chromium-review.googlesource.com/229342
Reviewed-by: Anton Staaf <robotboy@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Tested-by: Anton Staaf <robotboy@chromium.org>
Previously enabling USB would automatically (using an
init hook) initialize the USB peripheral. This would
take over the GPIO lines assigned to the USB module.
This is not OK on Ryu for Case Closed Debug because it
interferes with the AP's access to the USB 2.0 lines
even when not in Case Close Debug mode.
This change adds a configuration option to inhibit this
default initialization of the USB peripheral. It also
renames the existing CONFIG_USB_INHIBIT to
CONFIG_USB_INHIBIT_CONNECT now that there are two
possible inhibitions.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable console on ryu_p2 and discovery-stm32f072 board
Verify that it works on both
Change-Id: I6734357131b4356e3d4164349d6c74deac196ce5
Reviewed-on: https://chromium-review.googlesource.com/229138
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Tested-by: Anton Staaf <robotboy@chromium.org>
Previously the usb driver used #ifs to select a mechanism to
enable and disable the DP pullup. This doesn't scale well,
especially when the knowledge of how to do this is not known
to the chip specific code, but instead is board specific (as
it is with the STM32F373).
This change uses the build systems ability to build chip family
specific files per build to select the appropriate behavior.
And on the STM32F3 family of parts it just calls out to a board
specific pair of connect/disconnect functions.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable console on ryu_p2 and discovery-stm32f072 board
Verify that it works on both
Change-Id: I976e02fbc7acbb0f85817d7295b26ee9ecab0711
Reviewed-on: https://chromium-review.googlesource.com/229040
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Previously the USB RAM size was off by a factor of two
for chips that required 32-bit alignment of accesses,
even though the underlying memory was 16-bits in size.
This change adds an additional configuration for the
access size (it still assumes that the underlying memory
is 16-bits in size) and uses that to adjust the USB_RAM
memory section in the linker scripts.
This change also removes the default values for the USB
RAM from stm32/config_chip.h because they mask issues
when new chips are added. It is better for a new chip
to fail to compile until these values are provided.
Finally, this change introduces a common USB API header
so that common code doesn't need to include the STM32
specific header.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable console on ryu_p2 and discovery-stm32f072 board
Verify that it works on both
Change-Id: Id118627f53e9e8ff1bd09fb51f1f9634ff495d19
Reviewed-on: https://chromium-review.googlesource.com/228833
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
The console code never worked with the old style STM32 USB
peripherals because no chips with that version of the peripheral
were being used. The STM32F373 uses the older USB peripheral.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable console on ryu_p2 and discovery-stm32f072 board
Verify that it works on both
Change-Id: I77d36c33712521d7840b4e3ca02ebbea5de3d5df
Reviewed-on: https://chromium-review.googlesource.com/227741
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Previously the version string was special cased in the USB stack
because the build system prevented the inclusion of ec_version.h in
any file other than common/version.c. This lead to common/version.c
being the only place that the USB version string could be computed
and thus the special case of filling in the version string descriptor
at run time. This made the USB stack more complex, and lead to the
common/version.c file including usb.h, which is actually STM32
specific.
Now, the portion of ec_version.h that is deterministic is only
updated when something in the tree actually changes (by way of a
conditional in the makefile), and ec_version.h no longer has to
depend on all object files (other than the special version.o).
This allows anyone to include ec_version.h as needed. In particular,
each board that wants to define a USB version string can directly
include ec_version.h and do so.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
touch files and verify rebuilds happen correctly
Change-Id: Ic84d0b9da90f82ebb4630fb550ec841071e25a49
Reviewed-on: https://chromium-review.googlesource.com/227211
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Per table 6-24 of USB PD spec an alternate mode adapter (AMA) should
include both product & AMA VDOs.
BRANCH=samus
BUG=chrome-os-partner:31192,chrome-os-partner:31193
TEST=manual,
Connect hoho/dingdong to fpie/samus and see product VDO proceed the
AMA VDO in DFP_U console output:
Product VDO -----------------------------v
|------|
SVDM/5 [1] ff008041 340018d1 00000000 50100001 1100000b
Note, hoho's PID == 0x5010
And dingdong (0x5011)
SVDM/5 [1] ff008041 340018d1 00000000 50110001 1100000b
Also see bcdDevice field in descriptor match above data.
$ lsusb -v -d 18d1: | egrep -i "idproduct|bcddev"
idProduct 0x5011
bcdDevice 0.01
Change-Id: I4d898816a45c68c7ff75a54fd348fc11be408ae0
Signed-off-by: Todd Broch <tbroch@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/226125
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
USB Billboard class can be used to advertise an alternate mode capable
device that hasn't entered a mode. Additionally it can remain after
mode entry providing its Billboard capabilities descriptor is
updated.
This CL postpones enumeration which previously occurred after boot
until tAMETimeout has passed and alternate mode has NOT been entered.
Future CL could choose to also (re)enumerate with mode capabilities
although this is not required by the USB PD specification.
BRANCH=none
BUG=chrome-os-partner:31192,chrome-os-partner:31193
TEST=manual,
With DFP_U which does not enter mode see Billboard class enumerate
else it does not.
Change-Id: I59a0815cd0ea551ba9a878907c0184df4ba9480c
Signed-off-by: Todd Broch <tbroch@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224663
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Now that the problem with UART in STOP mode is fixed, let's enable low
power idle for Ryu P2 boards.
BRANCH=None
BUG=chrome-os-partner:33219
TEST=Shut down the AP and unplug AC power. Check the EC goes into
deep-sleep by 'idlestat'.
TEST=Power up the AP and check the EC stays awake.
Change-Id: Ib8ad3763407315ec0f95551fa0fdd258a060e113
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/226312
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Reviewed-by: Alec Berg <alecaberg@chromium.org>
These names can be used to uniquely identify an interface
as being a console forwarded from the AP, or the EC's
console or some other console. This makes it possible to
connect to the correct console from servo without knowing
what board is attached. These strings could also be used
by udev rules to create appropriate symlinks.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Change-Id: I1abd36f952782a03606b09485917be8e154534ce
Reviewed-on: https://chromium-review.googlesource.com/225950
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
This uses the newly allocated simple serial SubClass
and Protocol identifiers to make enumeration of serial
consoles over USB easier. The simple usb serial kernel
driver can select based on vid/class/subclass/protocol,
making it trivial to attach to serial consoles.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable USB console on discovery-stm32f072 and verify
that a new ttyUSB? device is discovered.
Change-Id: I6d3e414a20753573c2ee90a48fb1d5dce2c3972d
Reviewed-on: https://chromium-review.googlesource.com/225869
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Previously the USB console code used two separate
bidirectional endpoints. Using the STM32 USB perihperals
endpoints in this manner means that there are only 8 total
endpoints available. But the STM32 USB peripheral allows
for a single endpoint number to be used for both IN and
OUT transactions, and while this (re-using an endpoint
number for both IN and OUT endpoints) appears to be not
compliant with the letter of the law with respect to the
USB spec, it is supported by USB stacks, and is used in a
number of USB devices.
So this change makes the USB console driver share a single
endpoint number between the IN and OUT bulk endpoints used
to implement the console.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable USB console on discovery board and manually test.
Change-Id: I511e56de2162a6c04ddba80d26b37b4f0cd993fd
Reviewed-on: https://chromium-review.googlesource.com/225868
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
When calling the ep_undefined function, the linker was not considering
it as a Thumb function and not setting the LSB in the address pointer.
This was causing an exception at runtime.
Signed-off-by: Vincent Palatin <vpalatin@chromium.org>
BRANCH=all
BUG=none
TEST=inspect assembly
Change-Id: I8f4d1e351081032e138f593f0b61294031fc09b1
Reviewed-on: https://chromium-review.googlesource.com/226093
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Vincent Palatin <vpalatin@chromium.org>
Tested-by: Vincent Palatin <vpalatin@chromium.org>
Previously the USB console code would wait for up to
30 ms for the USB packet buffer to become available
for transmission, even if it was called from an interrupt
context. This doesn't work because, even if we were OK
with waiting this long in interrupt contexts, which we are
not, we rightly assert that we are not in an interrupt
context in task_wait_event, which usleep calls.
This solution is a quick fix to only wait when not called
from within an interrupt context. The correct solution
is likely to decouple the printf code from directly calling
the console driver code, instead we should place a queue
between the driver and printf logic.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Enable USB console on discovery board and manually test.
Change-Id: I5b6f7bbb77f75132c75935f8fda01e652a236ae0
Reviewed-on: https://chromium-review.googlesource.com/225867
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
By default, UART wakes the chip up on RXNE interrupt. This means when
the chip wakes up, RDR is full and only the shift register is empty, and
this leaves us only the time of a single character to process the
character in RDR. On some system, this is not enough and the first (or
even the second) character is overrun, and thus any multi-character keys
(e.g. arrow keys) break.
To avoid this problem, let's change the wake source to wake on start bit
detection. This gives us the time for one more character to wake up and
process the console input.
BRANCH=None
BUG=chrome-os-partner:33219
TEST=Enable low power mode on Ryu P2. In STOP mode, hit up arrow key and
see the last command show up.
Change-Id: Idce4c0bdfcf3e04ad84152ba525f704a0909f115
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/225771
Reviewed-by: Alec Berg <alecaberg@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@google.com>
Before entering STOP mode, we need to ensure UART Tx has completed.
Otherwise, we may lose some characters or some bits within a character.
For Tx DMA mode, this is already done as we wait until TC (Tx complete)
is set before disabling Tx. However, when not using DMA, we enable sleep
when TXE is set. At this moment, the last character is still in the
shift register and going into sleep causes loss of the whole or part of
the last character.
To avoid this, let's enable TC interrupt and enable sleep only if we
have no more characters to send and TC is set.
BRANCH=None
BUG=chrome-os-partner:33219
TEST=Enable low power mode on Ryu P2. Type when the EC is in STOP mode
and check there is no broken character.
Change-Id: Ife42671882b7f1d1d17734d7d20fb4ba7dffb371
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/225283
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This adds the option to use one 16-bit timer for watchdog helper when
using 32-bit hwtimer. With this, a debug message is dumped before
watchdog fires and we can easily see the task hogging the processor.
BRANCH=None
BUG=None
TEST=On Ryu P2, 'waitms 3000' and see debug message before the EC
reboots.
Change-Id: I498f63a105a0ba1ab7ec7d274dc8b1f16a44140f
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/225253
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This modifies the stray hard-coded DMA channel for UART Rx to the
correct config flag.
BRANCH=None
BUG=chrome-os-partner:32660
TEST=On Ryu, enable UART Rx DMA and check console is still responsive
Change-Id: Icec2de6ad4d34c6e0f8df2a1d51d9fefd982c9f4
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/225239
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
On STM32F0, a single IRQ is used for all RTC events, including RTC_ALARM
and RTC_WAKEUP, and this IRQ is named RTC_WAKEUP. We use RTC_WAKEUP IRQ
in our code while we actually meant to capture RTC alarm event. On
STM32F3, RTC_WAKEUP and RTC_ALARM are separate IRQs and thus we're
having problem. Let's make it explicit which RTC IRQ we want to use and
fix our code to use RTC_ALARM.
BRANCH=None
BUG=chrome-os-partner:33219
TEST=Run 'rtc_alarm' on Ryu and verify the EC gets interrupted after a
second.
Change-Id: Ib1a14a5da49d709a4d071d44fbfa46544cc5929b
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224998
Reviewed-by: Alec Berg <alecaberg@chromium.org>
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Simple control of SPI for flashing over USB.
This driver is working, and using the discovery board
with a W25Q16 flash chip attached flashrom can read,
erase, write, and verify the whole chip in 45 seconds.
Signed-off-by: Anton Staaf <robotboy@chromium.org>
BRANCH=None
BUG=None
TEST=make buildall -j
Change-Id: I224f1f87cd6adc8b64c17de1df98dae0a9cfa6a5
Reviewed-on: https://chromium-review.googlesource.com/218740
Reviewed-by: Anton Staaf <robotboy@chromium.org>
Tested-by: Anton Staaf <robotboy@chromium.org>
Reviewed-by: Randall Spangler <rspangler@chromium.org>
Commit-Queue: Anton Staaf <robotboy@chromium.org>
Unlike STM32F0, we need to configure alternate function for USB module
on STM32F373. Adds the pin configuration for ryu_p2 and also adds the
proper configuration step in USB module.
BRANCH=None
BUG=chrome-os-partner:32660
TEST=With changes to enable USB on ryu_p2, see the device enumerated
Change-Id: I5e2cb7cfc44a1bb88bae69804021c783c8d17968
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224789
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
The new build target ryu_p2 is mostly based on ryu. On ryu_p2, we have a
new EC chip with bigger flash, so make the corresponding changes:
- Pinout changes
- HW Timer: TIM5
- USB PD Tx Timer: TIM3_CH4
- USB PD Rx Timer: TIM2_CH4
- Use UART2 for EC console
- Disable UART Tx DMA as it conflicts with USB PD Tx DMA
- Use 24MHz HSE x2 = 48MHz for SYSCLK
BRANCH=None
BUG=chrome-os-partner:32660
TEST=Sanity check on a new board:
- i2cscan
- PD negotiation
- UART console
- gettime
Change-Id: I4ef6b53a928a2777721e3874032aeb0e6b2b4c92
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/221404
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
This adds the DMA channel definition for UART2 and allows selection of
DMA channel for UART.
BRANCH=None
BUG=chrome-os-partner:32660
TEST=With the CLs to enable the new Ryu boards, check the console is
working.
Change-Id: I964c284899777dda67c264e622aea6aba752ea76
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224176
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
Seems like we were setting outputs too early during boot, sometimes causing a
power leak. SPI should only turn on power levels more active than S3, not on the
S5->S3 transition.
BUG=chrome-os-partner:32824
BRANCH=None
TEST=Pinky powers on, Scope VCC33_PMUIO and VCC33_IO, note that they're smooth
Change-Id: I05c3622d124c2539222b883b895bc9092c5f0b12
Signed-off-by: Alexandru M Stan <amstan@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224508
Reviewed-by: Randall Spangler <rspangler@chromium.org>
We already have interrupt handlers for channel 4 to 7. We need channel 3
for the new Ryu boards. Add the handlers for channel 1 to 3. Also,
instead of copy-pasting interrupt handlers, define a macro and declare
interrupt handlers with it.
BRANCH=None
BUG=chrome-os-partner:32660
TEST=make buildall
TEST=Check PD communication on the new Ryu board (with other CLs to
enable the new boards.)
Change-Id: I51d6bd16739f31a7efbeb4ec19bb91a1546fe21d
Signed-off-by: Vic Yang <victoryang@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/224175
Reviewed-by: Vincent Palatin <vpalatin@chromium.org>
