gmac0 is always connected to switch phy4 and mdio1 is always needed.
So add phy handle for eth0 and enable mdio1 by default.
Move fixed-link for gmac1 from device dts to ar9331.dtsi because gmac1 is always connected to builtin switch.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Enable mdio1 by default because mdio1 is needed when eth1 is enabled.
PS: If a ar9341 device has only one port and you only want to use gmac0,
change compatible of gmac1 to "syscon", "simple-mfd" in dts.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
This patch did several things:
1. Probe the builtin switch as a separated mdio device.
2. Register a separated mdio bus for builtin switch.
3. Use generic mdio read/write function instead of calling ag71xx_mdio_mii_read/write directly.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
We need to have mdio1 belonging to gmac1 initialized before gmac0.
Split it into a separated mdio device to get both mdios ready before probing gmac.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
The builtin switch has it's initial valid mac address(00:00:01:00:00:00).
Since the builtin switch is an independent device, setting mac address of gmac1 to builtin switch isn't a good idea and this makes it impossilbe to split builtin switch apart as an independent platform device.
Remove these functions and apply default VLAN during initialization as a preparation for further driver splitting.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Only build images for straight OpenWrt (using all flash; wipes out
partitions that contain information only important for accessing a
now defunct cloud service with the stock firmware) since the stock
firmware is now irrelevant.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The wrong MAC addresses (from the point of view of the physical device
label) were being assigned to the wrong interfaces. Fix that.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
While the stock firmware and previous ar71xx versions of openwrt used the
single ethernet port as a DHCP client, for unmodified openwrt usage it
makes more sense to do the standard openwrt thing and make the ethernet
port a static lan with known address so that users can find the device on
the network more easily.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The Skydog cloud service no longer exists hence supporting going back
to stock firmware with cloud support is no longer applicable.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The reset button was incorrectly returning KEY_WPS_BUTTON as the key
code. We want KEY_RESTART., so make that fix.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The PCIe wireless MAC address address is better labelled as WMAC
than MAC to emphasize that it is for a wireless interface.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
The CAP324 was an AP for a NaaS offering that is now defunct. While
previously supported in the ar71xx arch, there were some errata (to
be fixed shortly).
Notable differences from ar71xx support:
1) The method of getting the ath9k firmware for the PCIe 2ghz wifi has
changed (due to changes in how the arch handles this), since this device
doesn't use the EEPROM except to get the MAC address of the wifi.
2) /etc/config/wireless will need to be regenerated as the path(s) to
the wireless device(s) have changed.
3) ath79 OpenWrt firmware no longer supports build an image that allows
reverting to stock firmware (as the cloud service no longer exists, the
stock firmware is useless), instead using all of the flash for image and
overlay (less u-boot/env and art).
4) Initial network config treats the ethernet port as a Lan port with
the standard default address (192.168.1.1 unless changed in .config
--e.g. via menuconfig) instead of using DHCP (this was the default for
the stock firmware, however for openwrt use this is rather confusion and
counter-productive as the user has a harder time finding the device on
the network.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
Add ath79 arch support for PowerCloud Systems CR5000. Previously
supported under ar71xx (however there are some errors in that support;
to be fixed shortly).
Info:
* This board is based on the Atheros DB120 reference design, but doesn't
use the on-board switch. Instead it attachs GMAC0 to an AR8327 switch.
* It only uses GMAC0 and the WAN is simply a VLAN in the stock firmware.
* It has 64MB RAM and 8MB flash.
* In the dts version we get rid of using 'open-drain' for the AR8327
LED controls.
* As with the platform data version we disable JTAG as this conflicts
with one of the pair of GPIO's required for the power/status LED
(GPIO2 and GPIO4 are used for this LED).
* The pcie card wifi has an EEPROM but gets it's MAC address from
the ART partition.
* The SoC wifi (2.4 GHz) is all from the ART.
* The USB is support comes from the SoC.
NB. This is actually an AR9342 rather than AR9344 but we use the 9344
definitions because there are no relevant differences for this board.
NB: Building only images that don't support reverting to the old
cloud-based firmware as the Skydog cloud service for the CR5000 no
longer exists.
Signed-off-by: Daniel F. Dickinson <cshored@thecshore.com>
81d446b045 introduced incomplete
support for this device.
This patch attempts to correct the situation based on OEM source
code.
LED1-3 are GSM mode on OFW (2G/3G/4G) hence unassigned here.
Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
Tested-by: David Ehrmann <ehrmann@gmail.com>
The active_low flag was missing for the user LED. This LED is open drain
(confirmed in OEM source) and open drain only makes sense for active low
GPIOs.
The two wireless LEDs mentioned in the comments are also #defined for
future reference.
Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
Tested-by: Ryan Mounce <ryan@mounce.com.au>
e15c63a375 introduced code that was trying
to register GPIO 1 as both an LED and a button. The OEM source makes it
clear that LED1 is not wired to the SoC GPIOs. GPIO 1 is the reset button.
Furthermore the (green) power led default state should also be defined,
(matching OEM source), and it should be used by diag.sh since it's
currently the only software-controllable LED.
This patch fixes these issues and renames the corresponding #defines for
clarity
Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
The gpios that control power toggle for USB on the RouterBOARD devices
are active low _off_ switches.
When they are active (low), power is off. When they are inactive
(high), power is on.
Rename GPIO defines, set gpios to GPIOF_ACTIVE_LOW for consistency and
reflect their true action in the display name. This brings openwrt code
in line with OEM.
Signed-off-by: Thibaut VARÈNE <hacks@slashdirt.org>
Tested-by: Ryan Mounce <ryan@mounce.com.au>
convert the usb and both sata port power related gpio-hogs to
what they really are: fixed-regulators.
The ethernet phy-reset gpio-hog is replaced by a proper
upstream (4.15+) reset-gpios property in the mdio-node.
So this will work eventually.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
@vahid-dan reported a issue with extracting the rpi images with
Gnome's Archive Manager:
"Ubuntu Archive Manager cannot extract the file and it just
throws a general error message: "An error occurred while
extracting files".
<https://forum.lede-project.org/t/corrupted-pre-built-v18-06-0-rc2-image-for-rpi>
The MBL's rootfs.img.gz image is generated in much the same way.
Hence this patch preemptively splits the rootfs.img.gz image into
a sysupgrade and a factory image.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
This patch adds support for squashfs as the root filesystem.
advantages:
- migrate from a existing -ext4 installation and back
with the sysupgrade utility
- existing partition layout will not be lost during switch
- slightly smaller image size as compared to the -ext4 image.
disadvantages:
- needs f2fs + tools. This is because fstools rootdisk.c decides based
on the partition size (currently root partitions > 100 MiB) f2fs is
used as the rootfs_data filesystem.
- rootfs_data is placed into the rootfs partition after the squashfs.
This makes it difficult for tools that expect a /dev/sda${X} device.
It also makes it difficult for data recovery tools as they might not
expect to find a embedded partition or will be slightly confused.
... or will not support f2fs.
For people with existing build configurations: make sure to include mkf2fs
and f2fsck packages into the image. Otherwise the new -squashfs image will
only boot from the ram-overlay.
Note:
All overlay data (configurations/all installed packages/...) will be
placed in inside the rootfs partition (i.e. /dev/sda2) just after the
squashfs image.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
@vahid-dan reported a issue with extracting the rpi images with
Gnome's Archive Manager:
"Ubuntu Archive Manager cannot extract the file and it just
throws a general error message: "An error occurred while
extracting files".
<https://forum.lede-project.org/t/corrupted-pre-built-v18-06-0-rc2-image-for-rpi>
@blogic told me to split the single sdcard.img.gz for the RPi
into a sysupgrade and a factory image for all brcm2708 targets.
The factory images will have no metadata attached, this way
these utilities that can't deal with the attached metadata will
not fail for no reason.
Cc: John Crispin <john@phrozen.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
It was present as 4.4 compatibility, but since we now use 4.9 or later
with the new upstream solution, we don't need it anymore.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
That upstream commit caused instability in flash reads. It was reported
but there isn't any proper fix as for now.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This includes Linksys EA9500 support, BCM53573 timer fix and
upstream-ready partitions patch that replaces two downstream hacks.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
USB 3.0 PHY is attached to the MDIO bus and should be supported
(accessed) as a MDIO device. This wasn't known initially which resulted
in writing driver that was working with MDIO bus (using some magic
values) without knowing it.
This commit updates DT to properly describe MDIO & USB 3.0 PHY and
enables required kernel drivers.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This is a new & warm feature that allows nesting partiitons in DT and
mixing their types (e.g. static vs. dynamic). It's very useful for
boards that have most partitions static but some of them require extra
parsing (e.g. a "firmware" partition).
It's required to successfully backport support for new devices using
that new syntax in their DT files.
Since brcm63xx has a custom alternative patch the upstream one is being
reverted for it. The plan is to make brcm63xx use the upstream
implementation.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
The OCEDO Raccoon only has one ethernet port, but currently uci sections
for WAN and LAN are created.
Additionally, newer versions of the devices U-Boot (units with SteelWRT)
set the kernel-cmdline and therefore overwrite the partition-layout.
We fix this by overwriting the cmdline supplied by the bootloader.
Signed-off-by: David Bauer <mail@david-bauer.net>
The get_status_led() function was removed due to the convertion to dts
alias based status led.
Since we don't need the boardname any longer, the functions.sh include
isn't required any more.
Fixes: c9c4b2116c ("ramips: Use dts alias based status led")
Signed-off-by: Mathias Kresin <dev@kresin.me>
With a10a204aab ("kernel: make ubi auto-attach check for a tar file
magic") the check for the magic was added without considering a failing
mtd_read(). If the read fails, no check is done and the mount code is
called straight away.
Failing with an error message for such cases seems to me the cleaner way,
as it would allow to spot hidden/workaround issues.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The first block(s) of the ubi mtd device might be bad. We need to take
care on our own to skip the bad block(s) and read the next one(s).
Don't treat recoverable read errors as fatal and check for the UBI magic
if the data of a block could be recovered using ECC or similar.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Move the put_mtd_device() called on multiple error conditions to a goto
label to use it later for more error conditions.
The early return on failed open of the mtd device and mismatching mtd
type allows to get rid of one level of indentation. By jumping to the
cleanup code, a refcount bug is fixed for the wrong flash type condition.
While at it, make clear that we only check for the UBI magic if the read
from flash was successful.
Signed-off-by: Mathias Kresin <dev@kresin.me>
