Support for kernel 4.14 was added 2 months ago, make it now the default
kernel to use for the x86 target.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Tested-by: Stijn Segers <francesco.borromini@inventati.org>
The gzip step in the sdcard image build fails because the image filename
already has the gzip extension. This results in an empty image file, to
which the metadata is finally appended.
Remove the .gz extension from the image filename to fix this.
Fixes: e79b096ee1 ("brcm2708: convert to metadata")
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
In the initial patch which adds HSDK board there were no update
of network configuration scripts. Without it by default static IP
is set for br-lan and there is no access to internet.
This patch fixes the issue.
Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com>
CC: Alexey Brodkin <abrodkin@synopsys.com>
CC: Hauke Mehrtens <hauke@hauke-m.de>
CC: John Crispin <john@phrozen.org>
The Orange Pi Zero Plus uses a RTL8211E, add the driver for this PHY.
This is probably used on more boards with a Allwinner SoC.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Calling nand_do_upgrade() from platform_pre_upgrade() was deprecated
with 30f61a34b4 ("base-files: always use staged sysupgrade").
Update the platform upgrade code to use platform_do_upgrade() for NAND
images as well.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The watchdog kill command was meant for busybox watchdog. Busybox watchdog
was replaced by the procd watchdog mid 2013 with commit df7ce9301a
("busybox: disable the watchdog utility by default"), which makes the kill
command obsolete since quite some time.
Signed-off-by: Mathias Kresin <dev@kresin.me>
- removed upstreamed patches
- 0901-spansion_nand_id_fix.patch is disabled, not clear if it's needed
Signed-off-by: Roman Yeryomin <roman@advem.lv>
Signed-off-by: John Crispin <john@phrozen.org>
On MT7621, the REG_ESW_VLAN_VTIM reads are undefined, causing swconfig
to always report `vid: 0` in swconfig show output.
Since a 4K VLAN table is used on this platform, the VLAN ID always
correponds to the actual VLAN table index so provide a specific MT7621
implementation of the get_vid callback which returns the table index
as VLAN ID.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Currently, untagged port primary vlan IDs are set to the VLAN table index,
and not the actual VLAN ID, breaking configurations with IDs deviating from
the VLAN index.
Fix the issue by resolving the per-port pvid property to the target VLAN ID
value before committing to the hardware.
Fixes FS#991, FS#1147, FS#1341
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Avoid overwriting vlan entries with remapped vid in later iterations of
the vlan enumeration loop of mt7530_apply_config().
Fix the problem by refactoring the code to first reset the entire table,
then reprogram only vlans with members to prevent overwriting configured
vlans with unconfigured ones.
Fixes FS#1147, FS#1341
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The current image build code has a number of race conditions and interface
contract violations in the custom image build steps:
- Build/install-zImage, solely used by at91, relies on $(PROFILE_SANITIZED)
which is not available when building with CONFIG_TARGET_MULTI_PROFILE
- Build/at91-sdcard, which may run concurrently, creates scratch files at
fixed locations and manipulates target files directly which can lead
to file corruption and other unexpected failures
Rename the install-zImage macro to at91-install-zImage and move it to the
at91 image Makefile since this target is the sole user. Also utilize "$@"
as output file name and switch the usage of $(PROFILE_SANITIZED) to
$(DEVICE_NAME) in order to fix naming under multi profile builds.
Fix the at91-sdcard macro to construct scratch file paths relative to "$@",
which is guaranteed to be unique and store the final artifact output in "$@"
as well, instead of inside $(BIN_DIR). The generic image build code takes
care of moving a build steps "$@" output to the final destination in a
concurrency-safe manner.
Finally remove the broken install-zImage from the generic image-commands
Makefile.
Fixes: d7a679a036 ("at91: Install zImage.")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This drops support for all the !emmc EVB and adds banannaPi-R2
Also drop mtkhnat until the nftables offoad driver is ready
Signed-off-by: John Crispin <john@phrozen.org>
This patch adds support for GL.iNet GL-B1300
Specification:
- SOC: IPQ4028 / QCA Dakota
- RAM: 256 MiB
- FLASH: 32 MiB
- ETH: Qualcomm Atheros QCA8075 Gigabit Switch (2 x LAN, 1 x WAN)
- USB: 1 x 3.0 (via Synopsys DesignWare DWC3 controller in the SoC)
- WLAN1: Qualcomm Atheros QCA4028 2.4GHz 802.11bgn 2:2x2
- WLAN2: Qualcomm Atheros QCA4028 5GHz 802.11a/n/ac 2:2x2
- INPUT: one reset and one WPS button
- LEDS: 3 leds: Power, WIFI(only for 2.4G currently), and one reserved
- UART: 1 x UART on PCB (3.3V, TX, RX, GND) - 115200 8N1
Installation:
Method 1:
- use serial port to stop uboot
- uboot command: run lf
Method 2:
- push down reset button and power on
- wait until three leds constantly on then release
- upgrade by uboot web at http://192.168.1.1
Note:
- the sysupgrade image need to be renamed to lede-gl-b1300.bin in both method.
- the sysupgrade image can be automatically downloaded if tftp server at
192.168.1.2 have that file.
- the wifi led will be flashing when writing image.
Signed-off-by: Dongming Han <handongming@gl-inet.com>
Synopsys DesignWare HSDK (which stands for ARC HS
Development Kit) is the latest and greatest development
platform that sports quad-core ARC HS38 in real silicon.
Most noticeable features of the board are:
* Quad-core ARC HS38 CPU running at 1GHz
* 4Gb of DDR
* Built-in Vivante GPU (well supported via open source
Etnaviv drivers)
* Built-in Wi-Fi/Bluetooth module (RedPine RS-9113)
And as usual we have:
* [micro] SD-card slot
* 2 USB 2.0 ports
* 1Gbit Ethernet port
* Built-in Digilent JTAG probe
* Serial port accessible via micro-USB port
For more information about HSDK board visit:
https://www.synopsys.com/dw/ipdir.php?ds=arc-hs-development-kit
Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com>
CC: Alexey Brodkin <abrodkin@synopsys.com>
CC: Hauke Mehrtens <hauke@hauke-m.de>
CC: John Crispin <john@phrozen.org>
We have managed to get USB 2.0 working good enough
on all archs38 platforms so we're ready to switch
to much faster USB 2.0.
Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com>
CC: Alexey Brodkin <abrodkin@synopsys.com>
CC: Hauke Mehrtens <hauke@hauke-m.de>
CC: John Crispin <john@phrozen.org>
sysfs attributes 'port_mask' & 'speed_mask' held locks whilst doing
mundane tasks such as sprintf. Refactor code to reduce length of time
locks are held unnecessarily.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Add sysfs 'mode' attribute to swconfig controlled LEDs.
swconfig 'link state' LEDs blink in the presence of port traffic. This
behaviour becomes more obvious as switches start to support
get_port_stats() e.g. commits 0369e35891,
3056d09b40,
4ddbc43cc1,
4d8a66d934.
This blinking can be confusing/distracting if the switch has other LEDs
used to indicate traffic. Provide a 'mode' sysfs attribute that
controls the blink on traffic behaviour.
mode - either "none" (LED is off) or a space separated list of one or more:
link: LED's normal state reflects whether the link is up (has carrier) or not
tx: LED blinks on transmitted data
rx: LED blinks on receive data
Note that 'link' considers any port speed mask that may be applicable.
e.g. if an LED is configured to indicate 1Gbit link speed and mode is
set to 'link rx tx' but the port is connected at 100Mbit then the LED
will not light or blink. A mode of 'tx rx' will blink in the presence of
traffic only if the port matches the rate (if configured)
This maintains compatibility with existing behaviour.
Attribute is 'link tx rx' by default for backwards compatible behaviour.
Many thanks to Thibaut Varene for providing a more sensible led_event
routine after I had mangled the original, and other coding style hints.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Acked-by: Thibaut VARENE <hacks@slashdirt.org>
The name of the PCIe controller node in device tree changed between
kernel 4.9 and kernel 4.14. Migrate the configuration when an update
from kernel 4.9 to 4.14 or back is done to the new name to make
the existing wifi configuration compatible with the new names.
This replaces the "pcie-controller" part with "pcie" on all nodes if the
file exists in sys fs.
This is not done in the uci-defualts, because they are getting executed
to late in the boot process. The kernel module gets loaded before the
uci-defaults scripts are executed. When the mwlwifi driver gets loaded
it will trigger an event via hotplug to detect new devices and as the
paths are not in the uci configuration they will be added again.
When the migration is done before the script will detect that they are
already there.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Add support for kernel 4.14 to the mvebu target.
This also replaces the old sfp and phylink patches with new versions
from Russell's clearfog-4.13 branch
http://git.arm.linux.org.uk/cgit/linux-arm.git/log/?h=clearfog-4.13
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This is needed to prevent copying it into kernel 4.14.
These device tree files are already integrated into kernel 4.14 and we
would like to use the upstream versions only.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Instead of creating an ubifs file system with format version 5 by
default on empty UBI volumes use the older format 4 by default. This
will make it possible to mount these file systems also on older kernel
versions.
When a user wants to do a sysupgrade from kernel 4.14 to kernel 4.9 the
old kernel has to read the file system created by the more recent kernel
which currently does not work for ubifs.
This fixes the problem by creating file systems which are compatible
with older kernel versions by default.
Kernel 4.14 will still be able to read and write UBI FS file system
version 5, it will just not be used when a ubifs partition is created
implicitly on an empty UBI volume.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The root= parameter overrides the automatic rootfs detection by partition
name, so drop it. Fixes boot from flash after renumeration changes of
partitions due to parser changes.
Fixes FS#1350.
Fixes: a27d59bb42 ("brcm63xx: switch to new partition layout specification")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Instead of writing to the currently booted partition set, this
implements full dual-boot support for sysupgrade by always writing to
the other, currently inactive, partition set and toggling the dualflag
after a successful flash.
The currently active/ booted partition set is determined by parsing
/proc/cmdline for its rootfs parameter (supplied by the DTS), instead
of reading from the 0:DUAL_FLAG mtd, in order to prevent (potentially)
bricking both partition sets.
Signed-off-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Instead of hardcoding /dev/loop0, it's safer to query for the first
free loop device and assign that dynamically. While /dev/loop0 is a
reasonable assumption immediately following "losetup --detach-all",
detaching the mounted overlay is no longer strictly necessary once
dual-boot support has been added for the nbg6817.
Signed-off-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Harmonize the variable usage for $kernel and $rootfs to always contain
the full device path, instead of just its basename.
Signed-off-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
PHY drivers were grouped into vendor specific directories
by upstream commit 0b56e9a7e8358e59b21d8a425e463072bfae523c
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Add more registers and flags to ar71x_regs.h for QCA955x and QCA956x
SoCs. Values come from Qualcomm Atheros u-boot code.
Patches can be merged into
622-MIPS-ath79-add-more-register-defines-for-QCA956x-SoC.patch
Signed-off-by: Julien Dusser <julien.dusser@free.fr>
D-Link DAP-1522 is a wireless bridge/access point with 4 LAN
ports and a dual-band wireless chipset.
Specifications:
- Ralink RT2880
- 32 MB of RAM
- 4 MB of Flash
- 4x 10/100/1000 Mbps Ethernet (RTL8366SR)
- 802.11abgn (RT2850)
Flash Instructions:
1. Download lede-ramips-rt288x-dap-1522-a1-squashfs-factory.bin
2. Open the web interface and upload the image
Signed-off-by: George Hopkins <george-hopkins@null.net>
EG-200 is a DIN rail mountable device with one ethernet port, wifi,
an RS-485 port, and an internal USB attached uSD card reader.
Two leds, "modbus" and "etactica" are managed by userspace applications
in factory firmware.
Flash instruction:
Original firmware is based on OpenWrt.
Use sysupgrade image directly in vendor GUI.
Signed-off-by: Karl Palsson <karlp@etactica.com>
The tftp.bin image for Buffalo WHR-G300N was not built, so I was fixed
it after rewriting to new image build code. And the code for
factory-EU.bin was broken, so I deleted it.
Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
- Remove kernel 4.9 support
- Apply specific 4.14 changes directly to source
- Refreshed all
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
The Cavium OCTEON TX is an ARM 64-bit SoC leveraging CPU cores and
periperhals from the Cavium ThunderX SoC.
This initial support provides a 4.14 kernel and kernel+initramfs that is
bootable on the Gateworks Newport GW630x as well as the Cavium sff8104
reference board.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
The Device Tree file for I2SE Duckbill boards was updated upstream.
Let's use the upstream version for upcoming kernel 4.14 by keeping
our current version for v4.9 still around.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
The GnuBee Personal Cloud Two crowdfunded on https://www.crowdsupply.com
It is a low-cost, low-power, network-attached storage device.
Specifications:
- SoC: MediaTek MT7621AT
- RAM: DDR3 512 MB
- Flash: 32 MB
- Six SATA ports for 3.5" Drives
- One SDcard
- One USB 3.0
- Two USB 2.0
- Gigabit Ethernet: Three Ports
- UART 3.5mm Audio Jack or 3 pin header - 57600 8N1
- Three GPIOs available on a pin header
Flash instructions:
The GnuBee Personal Cloud Two ships with libreCMC installed.
libreCMC is a Free Software Foundation approved fork of LEDE/OpenWrt.
As such one can upgrade using the webinterface or sysupgrade.
Das U-Boot has multiple options for recovery or updates including :
- USB
- http
- tftp
Errata:
- While there are three ethernet ports, the third requires support for
the second GMAC. This will come in kernel 4.14.
- The first hard drive slot has a clearance issue with the two fan
headers. Workaround is to pull the headers out and connect the pins to
jumper wires.
- Using this device as a NAS is problematic with the 4.9 kernel as many
/dev/sdX reads throw silent errors. The current theory behind this is
some kind of unhandled DMA mapping error in the kernel. This is not an
issue with kernel 4.4.
Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Updated pfe kernel patch to clean up iounmap(pfe->ddr_baseaddr).
pfe->ddr_baseaddr was got through phys_to_virt() not ioremap(),
so iounmap() for pfe->ddr_baseaddr should be removed.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
This patch is to add a known issue of make menuconfig
in README file. This is also a common issue for OpenWrt.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
This patch changes the declarations of ndo_get_stats64 handlers
to the previous struct rtnl_link_stats64 * return type instead of
the mainline void return.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Tis patch is to fix up some descriptions in README.
Per-device rootfs had been supported, and a known issue
had been found. These also should be updated accordingly.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Some layerscape devices (ls1012ardb/ls1012afrdm/ls1046ardb/ls1088ardb)
were using ext4 rootfs because there were issues using squashfs.
This patch is to drop using ext4 rootfs and use ubifs rootfs instead
which is more proper for SPI-NOR flash.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
This patch is to remove USB configs for 32-bit device since
they're already in kmod package. Also removed TI platform
specific configs.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Patches changes
- Updated patches-4.9 to NXP LSDK1712 linux-4.9.
- Merged changes of patch 303 into integrated patch 201.
- Split changes of patch 706 into dpaa part and dpaa2
part, and merged these changes into integrated patches
701 and 705.
- Removed patch 819 since ehci-fsl driver could be compiled now.
- Refreshed these patches.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
This device is identical as TP-Link RE450
RE355 is a dual-band AC1200 router, based on Qualcomm/Atheros
QCA9558+QCA9880.
Specification:
720/600/200 MHz (CPU/DDR/AHB)
64/128 MB of RAM (DDR2)
8 MB of FLASH (SPI NOR)
3T3R 2.4 GHz
3T3R 5 GHz
1x 10/100/1000 Mbps Ethernet
7x LED, 3x button
UART header on PCB
Flash instruction:
Web:
Download lede-ar71xx-generic-archer-c60-v2-squashfs-factory.bin
and use OEM System Tools - Firmware Upgrade site.
Signed-off-by: Henryk Heisig <hyniu@o2.pl>
This will introduce a warning on 3.18, but fixes an error when
compiling for 4.9.
Fixes: cf9e0a59aa ("ar7: add kernel 4.9 support")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
When clearfog was renamed to clearfog pro, it broke sysupgrade from
17.04 as the new images now get rejected as incompatible. Fix this by
adding the legacy boardname to the compatible devices.
Fixes: ec4a8c6dee ("mvebu: ClearFog renamed upstream to ClearFog Pro")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Add support for kernel 4.9 based on the more upstream comformant
partition defintions. Increases compressed kernel size by ~95k
compared to 4.4.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Partitions are supposed to be enclosed in an extra partitions node. Allow
the bcm63xxpart parser to cope with that.
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
The lookup needs to be called just "reset" for the common code.
Fixes: 054c0d4e31 ("brcm63xx: backport mdio-bus reset gpio support")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
The arm CPUs uses in the supported Mediatket SoCs have a FPU accordingly
to the datasheet, activate it also. The CPU subtype "neon-vfpv4" is
selected, but the toolcahin generated for this SoC will still be
compiled with soft float and not with the hard float ABI as we haven't
the fpu feature flag set. If this toolchain is reused by other targets
this will even affect other targets.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The BDFs for OpenMesh A42 were upstreamed [1] to the ath10k-firmware
repository and are now part of ath10k-firmware 2018-01-26. The
ipq-wifi-openmesh_a42 package can now be dropped because OpenWrt already
ships the QCA4019 board-2.bin from this version.
[1] https://wireless.wiki.kernel.org/en/users/drivers/ath10k/boardfiles
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
COMFAST CF-E375AC is a ceiling mount AP with PoE support, based on
Qualcomm/Atheros QCA9563 + QCA9886 + QCA8337.
Short specification:
2x 10/100/1000 Mbps Ethernet, with PoE support
128MB of RAM (DDR2)
16 MB of FLASH
3T3R 2.4 GHz, 802.11b/g/n
2T2R 5 GHz, 802.11ac/n/a, wave 2
built-in 5x 3 dBi antennas
output power (max): 500 mW (27 dBm)
1x RGB LED, 1x button
built-in watchdog chipset
Flash instruction:
Original firmware is based on OpenWrt.
Use sysupgrade image directly in vendor GUI.
Signed-off-by: Ding Tengfei <dtf@comfast.cn>
TP-Link Archer C50 v3 is a router with 5-port FE switch and
non-detachable antennas. It's based on MediaTek MT7628N+MT7612E.
Specification:
- MediaTek MT7628N/N (580 Mhz)
- 64 MB of RAM
- 8 MB of FLASH
- 2T2R 2.4 GHz and 2T2R 5 GHz
- 5x 10/100 Mbps Ethernet
- 4x external, non-detachable antennas
- UART (J1) header on PCB (115200 8n1)
- 7x LED (GPIO-controlled*), 2x button, power switch
* WAN LED in this devices is a dual-color, dual-leads type which isn't
(fully) supported by gpio-leds driver. This type of LED requires both
GPIOs state change at the same time to select color or turn it off.
For now, we support/use only the green part of the LED.
Flash instruction:
The only way to flash LEDE image in ArcherC50v3 is to use
tftp recovery mode in U-Boot:
1. Configure PC with static IP 192.168.0.66/24 and tftp server.
2. Rename "openwrt-ramips-mt7628-ArcherC50v3-squashfs-tftp-recovery.bin"
to "tp_recovery.bin" and place it in tftp server directory.
3. Connect PC with one of LAN ports, press the reset button, power up
the router and keep button pressed for around 6-7 seconds, until
device starts downloading the file.
4. Router will download file from server, write it to flash and reboot.
Signed-off-by: Henryk Heisig <hyniu@o2.pl>
The D-Link devices with JBOOT bootloader use their own kernel
image header (stag + sch2 headers).
This driver find jImage header and set rootfs start after kernel file.
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
This commit marks the CPUs switchport explicit as untagged.
Otherwise, an eth0.1 interface is created and the devices
LAN-ports are not working.
Signed-off-by: David Bauer <mail@david-bauer.net>
Use "$(DTS_DIR)", defined at include/image.mk, instead of
"$(LINUX_DIR)/arch/$(LINUX_KARCH)/boot/dts" in order to generalize and
allow a better Device/* device-tree parameterization (i.e. DEVICE_DTS_DIR
and DTS_DIR).
Signed-off-by: Adrià Llaudet <adria.llaudet@gmail.com>
Update Linux kernel version from 4.9 to 4.14 for archs38.
config-4.14 was simply regenerated with "make kernel_menuconfig".
Signed-off-by: Evgeniy Didin <Evgeniy.Didin@synopsys.com>
Cc: Alexey Brodkin <abrodkin@synopsys.com>
Cc: John Crispin <john@phrozen.org>
This config option was renamed in upstream Linux commit 681bec0367
("tracing: Rename update the enum_map file")
Reported-by: Rosen Penev <rosenp@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Convert userspace code to use generic device-tree compatible board
detection method. Users of the existing code will have to use
sysupgrade -F once to switch to the new generic board naming.
Properly setup pinctrl fixing the switch port LEDs.
Fixes commit 9c4fe103cb (ramips: add support for ZBT-WE1226)
Reported-by: Mathias Kresin <dev@kresin.me>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Locally generated packets weren't forwarded to the isolated interfaces in a
bridge. Isolation should only prevent the flooding of incomming packets to
other interfaces in the bridge.
Signed-off-by: Daniel Danzberger <daniel@dd-wrt.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Qualcomm claims this improves the D-cache footprint. Origina commit message below:
From: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Date: Fri, 7 Jun 2013 10:57:28 -0500
Subject: [ag71xx] cluster/align structs for cache perf
Cluster the frequently used, per-packet structures in ag71xx near
to each other, and cacheline-align them. Some other re-ordering
occurred to move "warmer" structures near the per-packet structures.
Signed-off-by: Ben Menchaca <ben.menchaca@qca.qualcomm.com>
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Hauke Mehrtens
Stijn Tintel
John Crispin
Evgeniy Didin
Antony Antony
Mathias Kresin
Zoltan HERPAI
Roman Yeryomin
Jo-Philipp Wich
Dongming Han
Kevin Darbyshire-Bryant
Jonas Gorski
Stefan Lippers-Hollmann
Luis Araneda
Julien Dusser
George Hopkins
Karl Palsson
INAGAKI Hiroshi
Koen Vandeputte
Tim Harvey
Michael Heimpold
Rosen Penev
Yangbo Lu
Henryk Heisig
Matthias Schiffer
Sven Eckelmann
Ding Tengfei
David Bauer
Pawel Dembicki
Arne Zachlod
Adrià Llaudet
Felix Fietkau
Daniel Golle
Daniel Danzberger