The hardware queue scheduling is apparently configured with fixed
priorities, which creates a nasty fairness issue where traffic from one
CPU can starve traffic from all other CPUs.
Work around this issue by forcing all tx packets to go through one CPU,
until this issue is fixed properly.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Commit af79fdbe4a changed the code to use tx_nbits for dual SPI tx
transfers, however the SPI stack only allows this when the device mode
includes the relevant bit as well
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This code was marked as incompatible to Linux 4.4 well over a year ago
and nobody cared, and now it's breaking builds.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This target has consistently failed builds for a long time and it does
not look like it has any users left, nor any maintainers.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Rework the bundle-libraries.sh implementation to use a more robust approach
for executing host binaries through the shipped ELF loader and libraries.
The previous approach relied on symlinks pointing to a wrapper script which
caused various issues, especially with multicall binaries as the original
argv[0] name was not preserved through the ld.so invocation. Another down-
side was the fact that the actual binaries got moved into another directory
which caused executables to fail looking up resources with paths relative
to the executable location.
The new library wrapper implements the following improvements:
- Instead of symlinks pointing to a common wrapper, each ELF executable
is now replaced by a unqiue shell script which retains the original
program name getting called
- Instead of letting ld.so invoke the ELF executable directly, launch
the final ELF binary through a helper program which fixes up the argv[0]
argument for the target program
- Support sharing a common location for the bundled libraries instead of
having one copy in each directory containing wrapped binaries
Finally modify the SDK build to wrap the staging_dir and toolchain binaries
which allows to use the SDK on systems with a different glibc version.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Without this patch swconfig will only allow setting up a total of 16
VLANs, with VLAN ID range of 0-15.
Tested on ubnt-erx.
Signed-off-by: Antonis Kanouras <antonis@metadosis.eu>
Fixing pvs1 bin voltage as found in GPL tarball and Qualcomm SDK.
Also adjusting smb208_s2a/b minimum voltage to reflect lowest value in pvs table.
Current ipq8065 devices in LEDE choose (based on eFuse data) pvs bin n4, that has
correct values in DT, but in future a new device may choose n1 or n6.
Without this change n1 is semi-correct and n6 is not available, because OPP driver
disregards bins with voltage values that regulator cannot supply.
Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
This adds USB initialization fixes for Danube, Amazon SE and xrx300 and
should fix the clock on at least Danube which hopefully closes FS#351.
The xrx200 usb driver now uses more memory for the dwc2 fifos, this was
increased in the chip compared to ar9.
This is based in part on the vendor documentation and the vendor code
base as a reference.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The DMA controller used only local locks, which will not lock the
section against other CPUs, fix this by using spin locks.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Now that we generate images with metadata attached, require it in order
to prevent accidentally flashing the wrong image to a board.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Add support for the integrated AC97 sound device on motherboards
with AMD CS5535/CS5536 chipsets.
Tested on Wyse Winterm S30.
Signed-off-by: Cezary Jackiewicz <cezary@eko.one.pl>
The boot loader of many boards provides a kernel boot argument with a
broken console parameter to the kernel. This will be taken before the
stdout-path defined in device tree and the serial will not work on this
board. Multiple boards are affected by this problem, so revert this
patch for all boards. Later we can remove the boot arguments from the
device tree again, after the individual board was checked to work.
This partly reverts this commit:
commit 4995c64857
Author: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: Sun Dec 11 21:55:00 2016 +0100
lantiq: specify console using stdout-path instead of cmdline argument
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
To achieve this, device tree compatible string was used as boardname and
the value of it will be checked against supported_devices list.
It should be noted that we do not distinguish between
sun5i-a13-olimex-som and sun5i-a13-olinuxino as they share the same dts
file.
The other thing is that we need to gunzip the generated firmware to do
fwtool check.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Enalbe builtin support for FAT filesystem as we need to mount boot
partition to store sysupgrade.tgz there
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Two things that need to be noted
- There is no partition type id allocated for squashfs yet. In the
case of sunxi, any non-zero value should work and we keep it 83 (the
value for ext4)
- Remaining spare space within the rootfs partition, not the entire
sdcard space will be formated as either f2fs or ext4 and mounted as
overlay to serve the role of rootfs_data.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
The new Device/xxx were transformed automatically from old profiles.
Most device names are now taken from basename of the corresponding
kernel device tree file. Device/sun5i-a13-olimex-som is an exception
because it is not explicitly supported in the kernel yet and shares the
same dts file with Device/sun5i-a13-olinuxino
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
To squash error messages at boot time
mv: can't rename '/mnt/sysupgrade.tgz': No such file or directory
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
The kernel will detect if the host supports this, so we can just enable
it in the kernel config.
Tested on an APU2 with AES-NI support and a KVM VM on a Xeon E5520 host
without AES-NI support.
Throughput over an IPsec tunnel between these 2 hosts increased from
~63Mbps to ~140Mbps. Ciphers: AES_GCM_16_256/PRF_HMAC_SHA2_512/ECP_521.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
WL500GPv1 profile included ath5k which made it usable only for people
who decided to replace default BCM4318 card with Atheros one. We can't
have profile for every possible configuration. If someone adjusts hw in
such a way he can always install a proper package.
WRTSL54GS profile got extra packages for a specific USB usage. Our
standard profile provides basic USB and we should stick to this. We
can't make everyone happy by including packages for all common USB use
cases and all common filesystems.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
The powerpc boot wrapper Makefile is not parallel build safe, causing fixdep
to fail reading dependency files of the addnote, hack-coff and mktree
utilities when concurrently building different image targets.
A typical failure looks like:
Building modules, stage 2.
HOSTCC arch/powerpc/boot/addnote
HOSTCC arch/powerpc/boot/hack-coff
DTC arch/powerpc/boot/taishan.dtb
HOSTCC arch/powerpc/boot/addnote
HOSTCC arch/powerpc/boot/hack-coff
MODPOST 800 modules
fixdep: error opening depfile: arch/powerpc/boot/.hack-coff.d: No such file or directory
scripts/Makefile.host:91: recipe for target 'arch/powerpc/boot/hack-coff' failed
make[5]: *** [arch/powerpc/boot/hack-coff] Error 2
make[5]: *** Waiting for unfinished jobs....
fixdep: error opening depfile: arch/powerpc/boot/.addnote.d: No such file or directory
scripts/Makefile.host:91: recipe for target 'arch/powerpc/boot/addnote' failed
make[5]: *** [arch/powerpc/boot/addnote] Error 2
rm arch/powerpc/boot/taishan.dtb
arch/powerpc/Makefile:263: recipe for target 'cuImage.taishan' failed
make[4]: *** [cuImage.taishan] Error 2
make[4]: *** Waiting for unfinished jobs....
Add a GNU make specific .NOTPARALLEL pseudo rule to enforce sequential building
of the addnote, hack-coff and mktree executables.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This allows more feature complete images. Of course it affect the size,
e.g. enabling b43 bumped rootfs from 1569618 to 2029122 for me.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
This chip has write protection enabled on power-up, so this flag is
necessary to support write operations.
Signed-off-by: Victor Shyba <victor1984@riseup.net>
This flag was added to 4.9 with upstream commit
76a4707de5e18dc32d9cb4e990686140c5664a15.
Signed-off-by: Victor Shyba <victor1984@riseup.net>
[refresh and adjust platform patches, fix commit message]
Signed-off-by: Mathias Kresin <dev@kresin.me>
LS1088A is an ARMv8 implementation combining eight ARM A53 processor
cores. The LS1088ARDB is an evaluatoin platform that supports the
LS1088A family SoCs.
Features summary:
- Eight 64-bit ARM v8 Cortex-A53 CPUs
- Data path acceleration architecture 2.0 (DPAA2)
- Ethernet interfaces
- QUADSPI flash, 3 PCIe, 2 USB, 1 SD, 2 DUARTs etc
Signed-off-by: Yutang Jiang <yutang.jiang@nxp.com>
In case the soft reset in dwc2_core_reset() timeouts, the
hsotg->core_params are freed albeit it is owned by the core. This
results into a kernel panic as shown in FS#351.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use the the dt-bindings macros and add the reset button.
Set the correct polarity for the LEDs and drop the default state.
Remove all trigger for the LEDs. According to the manual the LEDs are
only used to show the operation state, where blue means normal
operation.
Use the MAC-Addresses stored in EEPROM for the ethernet and the
wireless interface.
Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
[use leds only for boot status indication, add proper commit message]
Signed-off-by: Mathias Kresin <dev@kresin.me>
This commit introduced a syntax error in ox820-akitio.dts which is
fixed now:
commit 5cde94d9ab
Author: Daniel Golle <daniel@makrotopia.org>
Date: Sat Sep 24 01:14:53 2016 +0200
oxnas: backport upstream NAND driver
This caused the folowing error message in the build bot:
Error: arch/arm/boot/dts/ox820-akitio.dts:146.3-147.1 syntax error
FATAL ERROR: Unable to parse input tree
scripts/Makefile.lib:293: recipe for target 'arch/arm/boot/dts/ox820-akitio.dtb' failed
make[5]: *** [arch/arm/boot/dts/ox820-akitio.dtb] Error 1
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
We have profile for this device thanks to DEVICE_PACKAGES now.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Cc: Russell Senior <russell@personaltelco.net>
We don't use this driver since commit 741715331a ("bcm53xx: switch to
m25p80 and drop bcm53xxspiflash").
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
There was a typo in Makefile that prevented using these profiles.
Fixes: a75ce960ac ("ramips: use different board names for variants")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
depending packages have been moved to kernel-config
- kmod-xen-kbddev in 9fde361
- kmod-xen-fs, kmod-xen-evtchn, kmod-xen-netdev in 018807d
this will also fix imagebuilder
Signed-off-by: Sven Roederer <devel-sven@geroedel.de>
crypto4xx_probe() is in the __init section and referenced by code form
other sections, which causes a build error. Backport a patch from
mainline kernel to fix this.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The following commit changed the build templates name but forgot to
update the TARGET_DEVICES variable properly.
commit f9226158be (lantiq: rename EASY98000 to EASY98000NOR)
Signed-off-by: John Crispin <john@phrozen.org>
The generic lzma-loader uses vmlinux.lzma as an input to compress so we
need to provide a lzma compressed version of the initramfs kernel.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
The power LED can not be used for indicating the xDSL line state and
the in system (boot) state at the same time in LEDE. As soon as the
xDSL Line goes down, the power LED is switched off.
During xDSL handshake the power led would blink with the same or a
similar pattern as during boot.
Both might confuse users.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Due to a hardware bug of Atheros 8030 phys, the driver need to reset
the phys on link state change.
Use the correct compatible string for the at903x phys connected to
switch port 0 and 1.
Fix the pinmux of the gpio lines connected to the reset pin of the phys
and define the reset-pins to let the driver do the fixups.
Fixes FS#343
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use different names for flash size related board variants, to make sure
that only images for the actual flash size are considered as valid by
the image validation code.
Remove the flash size suffix from the string returned by
ramips_board_detect() to ensure that existing scripts relying on the
former used boardname are still working.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The name from the Device define will be used in the metadata. Due to
typos/different spelling, this name might not match the one exported in
/lib/ramips.sh.
Fix all name mismatches which were found by checking if the name used
in the metadata exists in /lib/ramips.sh.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Use a distinct board name even if the board is near to identical to
the WeVO W2914NS v2.
To make sure that a 11AC NAS image can not be installed on a
WeVO W2914NS v2, both board need to use different names.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Remove dt: it is implicitly enabled based on the kernel config
Remove tar.gz and jffs2_nand: Legacy build code has been removed,
NAND devices are only supported with UBI now.
Remove ubifs: deprecated, use squashfs images instead
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The priv->vlan_id member is of size AR8X16_MAX_VLANS, not AR8X16_MAX_PORTS,
so check for the proper maximum value in order to avoid capping valid VLAN IDs
to 7 (AR8X16_MAX_PORTS - 1).
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Enabled Hyper-V network interface card driver, display adapter driver,
storage driver, keyboard driver, mouse driver and Hyper-V utility and
EFI boot support in the kernel for subtarget x86/64.
Convert the img file to vhd by Ubuntu qemu-img, rather than by the buildroot's
built-in qemu-img.
Tested on Windows Server 2008 r2 and 2012 r2 Gen1 and Gen2 VMs.
Signed-off-by: Tedaz <tedaz99999@hotmail.com>
Add full support for TP-Link RE450.
The wireless SoC is similiar to Archer C7: QCA9558 + QCA9880 (pci).
The ethernet interface is AR8035, but the mdio is connected to the gpio and
the chipset builtin mdio bus isn't used, which is unique (and weird), so the
kernel gpio mdio module is used.
The two ethernet leds are connected to the GPIO, so they are both configured,
one to indicate link status and the other to indicate data transfer.
The generation of the image was added to the image Makefile.
The return value of tplink-safeloader is not ignored anymore (to fail on error)
The result factory image is flashable from the stock web ui.
Signed-off-by: Tal Keren <kooolk@gmail.com>
[rd@radekdostal.com: ar71xx: RE450: do not build RE450 for nand]
Signed-off-by: Radek Dostál <rd@radekdostal.com>
[hyniu@o2.pl: ar71xx: RE450: Renaming LED accordance with the standard]
Signed-off-by: Henryk Heisig <hyniu@o2.pl>
[jo@mein.io: squash fixes from Radek and Henryk into original commit]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The firmware image that is used in TP-Link RE450 (and some more devices from
the RE series) is tplink-safeloader.
In the kernel partition, the kernel is compressed in a regular tp-link
firmware that is just used for booting. Since it is only used for compressing
and booting, only four fields are filled in the header:
Vendor, version, kernel load address and kernel entry point.
mktplinkfw-kernel is a simpler version of mktpolinkfw that generate such
images. It also specifies the hardware id (as it is in the product info
section), so when doing a sysupgrade - the existing code will check for
hardware compatibility.
Signed-off-by: Tal Keren <kooolk@gmail.com>
[rd@radekdostal.com: add build target to .../image/tp-link.mk]
Signed-off-by: Radek Dostál <rd@radekdostal.com>
mktplinkfw ignores -j in combined mode, the EOF marker is added by
tplink-safeloader.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
In the latest version of grub-mkimage, the prefix option is mandatory.
Not supplying it fails with:
```
Prefix not specified (use the -p option).
```
In grub-2.02-beta2 a DEFAULT_DIRECTORY was defined
in `include/grub/osdep/hostfile_unix.h` as:
```
#if defined (__NetBSD__)
/* NetBSD uses /boot for its boot block. */
# define DEFAULT_DIRECTORY "/"GRUB_DIR_NAME
#else
# define DEFAULT_DIRECTORY "/"GRUB_BOOT_DIR_NAME"/"GRUB_DIR_NAME
#endif
```
Where:
* GRUB_BOOT_DIR_NAME == boot
* GRUB_DIR_NAME == grub
This was used if the -p option was omitted.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
This patch is part of a series adding support for 0x9200 and 0x9300. The
prior was merged into the upstream kernel while the latter was not due
to lack of testers. Drop the patch as it is untested and most likely
unused.
Signed-off-by: John Crispin <john@phrozen.org>
When running sysupgrade for the first time over the second partition
(that still had stock firmware) the rootfs wasn't flashed as there
wasn't enough space.
This happend because stock also uses UBI, but the volume name wasn't
recognised and wasn't deleted before flashing.
Signed-off-by: Adrian Panella <ianchi74@outlook.com>
The si3210 is a SLIC device providing a complete analog telephone
interface and therefore frequently used in soho router.
The si3210 have a native spi interface to be controlled by the CPU
but currently there is no dedicated driver in lede.
Adding a registration for this device in spidev allow to control the
device in user space.
This way of patching is also in line with the rationale of the spidev
driver, see: http://marc.info/?t=148145791900001&r=1&w=2
The si3210 has been also added in the DWR-512 DT to properly describe
the HW.
Signed-off-by: Giuseppe Lippolis <giu.lippolis@gmail.com>
While researching for the armvirt target, I looked at the
existing arm platforms. It turns out that the mediatek target
with its sole MT7623N/A chip is sold as a "highly integrated
multimedia network router system-on-chip". To that end, it
lists support for the "NEON multimedia processing engine with
SIMDv2 / VFPv4 ISA support".
<http://topics.mediatek.com/en/products/connectivity/wifi/home-network/wifi-ap/mt7623na/>
So this patch enables the CPU_SUBTYPE to use this information.
This should have the nice side effect that LEDE's phase2 builders
no longer need to built a separate "cortex-a7" target, so this
should free up some resources.
Cc: John Crispin <john@phrozen.org>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Bump & refresh patches for all 4.4 targets.
Compile & run tested: ar71xx Archer C7 v2
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
Fix the mtd partition layout and enable both radios in the dts
Tested-by Andrius Štikonas <andrius@stikonas.eu>
Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
