Package architecture aarch64_generic [1] can be used just with three
devices. One is NanoPI R2S and then there are two development boards
from NXP. Let's change armvirt/64 to Cortex A53 (aarch64_cortex-a53)
[2]. It has wider support by multiple devices like NanoPI Neo Plus2/Core2,
ESPRESSObin, Pine64, and Raspberry Pi 2&3.
While looking at ARMvirt/32 it has set CPU_TYPE and CPU_SUBTYPE to be
arm_cortex-a15_neon-vfpv4 [3]. It has support to devices like
Linksys EA8500 v1, Linksys EA7500 v1, Netgear D7800, Netgear R7500 and so on.
Tested with:
qemu-system-aarch64 -m 1024 -smp 2 -cpu cortex-a57 -M virt -nographic \
-kernel openwrt-armvirt-64-Image-initramfs
Successfully compiled and booted.
Here goes the output:
root@OpenWrt:/# uname -a
Linux OpenWrt 5.4.82 #0 SMP Sun Dec 13 12:52:10 2020 aarch64 GNU/Linux
root@OpenWrt:/# cat /etc/openwrt_release
DISTRIB_ID='OpenWrt'
DISTRIB_RELEASE='SNAPSHOT'
DISTRIB_REVISION='r15207-96fca0f807'
DISTRIB_TARGET='armvirt/64'
DISTRIB_ARCH='aarch64_cortex-a53'
DISTRIB_DESCRIPTION='OpenWrt SNAPSHOT r15207-96fca0f807'
DISTRIB_TAINTS='no-all'
Also, change BOARDNAME to be the same as it is in armvirt/32.
[1] https://openwrt.org/docs/techref/instructionset/aarch64_generic
[2] https://openwrt.org/docs/techref/instructionset/aarch64_cortex-a53
[3] https://openwrt.org/docs/techref/instructionset/arm_cortex-a15_neon-vfpv4
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
(cherry picked from commit 222dc0ad91)
Tegra 2 processors have only 16 double-precision registers. The change
introduced by 8dcc108760 ("toolchain: ARM: Fix toolchain compilation
for gcc 8.x") switched accidentally the toolchain for tegra target to cpu
type with 32 double-precision registers. This stems from gcc defaults
which assume "vfpv3-d32" if only "vfpv3" as mfpu is specified. That
change resulted in unusable image, in which kernel will kill userspace as
soon as it causing "Illegal instruction".
Ref: https://forum.openwrt.org/t/gcc-was-broken-on-mvebu-armada-370-device-after-commit-on-2019-03-25/43272
Fixes: 8dcc108760 ("toolchain: ARM: Fix toolchain compilation for
gcc 8.x")
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit 43d1d88510)
Armada 370 processors have only 16 double-precision registers. The
change introduced by 8dcc108760 ("toolchain: ARM: Fix toolchain
compilation for gcc 8.x") switched accidentally the toolchain for mvebu
cortexa9 subtarget to cpu type with 32 double-precision registers. This
stems from gcc defaults which assume "vfpv3-d32" if only "vfpv3" as mfpu
is specified. That change resulted in unusable image, in which kernel
will kill userspace as soon as it causing "Illegal instruction".
Ref: https://forum.openwrt.org/t/gcc-was-broken-on-mvebu-armada-370-device-after-commit-on-2019-03-25/43272
Fixes: 8dcc108760 ("toolchain: ARM: Fix toolchain compilation for
gcc 8.x")
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
(cherry picked from commit 2d61f8821c)
The root file system is getting too big for this device and this breaks
the ramips/rt305x build.
Do not build images for this board by default to fix this problem.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit f46a3c7b14)
Devices with 4M flash are not built be default for 20.xx anymore.
Building them with buildbot settings does not work anymore anyway.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
(cherry picked from commit ad22f6a8aa)
Removed patches:
- bcm27xx/patches-4.19/950-0475-usb-xhci-Disable-the-XHCI-5-second-timeout.patch
Should be fixed by upstream commit 1c713660c51f3bc64fc1c709ba2bfb358551f3da
("xhci: make xhci_handshake timeout for xhci_reset() adjustable").
Refreshed all patches.
Signed-off-by: Tianling Shen <cnsztl@immortalwrt.org>
The MediaTek's Crypto Engine module is only available for mt7623, in
which case it is built into the kernel.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
Previously, grub2 was hardcoded to always look on "hd0" for the
kernel.
This works well when the system only had a single disk.
But if there was a second disk/stick present, it may have look
on the wrong drive because of enumeration races.
This patch utilizes grub2 search function to look for a filesystem
with the label "kernel". This works thanks to existing setup in
scripts/gen_image_generic.sh. Which sets the "kernel" label on
both the fat and ext4 filesystem variants.
Signed-off-by: Jax Jiang <jax.jiang.007@gmail.com>
Suggested-by: Alberto Bursi <bobafetthotmail@gmail.com> (MX100 WA)
(word wrapped, slightly rewritten commit message, removed MX100 WA)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 1050e66c8f)
1) Add BACKUP_FILE and use it when copying an archive to be restored
after sysupgrade (on the next preinit).
2) Use CONF_TAR for copying backup prepared by the /sbin/sysupgrade
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
According to the vendor [1] these HATs share the same DT overlay:
hifiberry-dacplus. The PCM512x-compatible control unit is attached to
I2C, so the additional snd-soc-pcm512x-i2c kernel module is required.
Also explicitly note the Amp2 support to reduce confusion for those
users.
[1] <https://www.hifiberry.com/docs/software/configuring-linux-3-18-x/>
Signed-off-by: Torsten Duwe <duwe@lst.de>
(added bcm27xx tag, changed commit message)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(cherry picked from commit 7ea9936f7f)
Added support to generate dynamic-sized VHDX images for Hyper-V.
Compile-tested on x86 and run-tested on Windows 10 21H2 (Hyper-V).
Signed-off-by: Oldřich Jedlička <oldium.pro@gmail.com>
Extensive testing on the board showed that ethernet does
not work when forced to 10Mbps.
Trial-and-error revealed that the correct PLL value
should be altered to 0x00001313 (iso 0x00001616)
The change is done for this specific board only as I do not have
other boards using this specific SoC.
The board now works correctly in 1000, 100 and 10 Mode
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
(cherry picked from commit 5e8b9624f1791d193e8d5e007ce4c873454f014a)
Both targets miss a subtarget causing an image naming style which is
different from other all othe targets, even tho it already uses
`x/generic/` as subfolder as if the subtarget would exist.
This commit adds the Generic subtarget resulting in consistent naming.
Fixes: c9e45c2 ("uboot-fritz4040: update PKG_MIRROR_HASH")
Fixes: bb43571 ("uboot-fritz4040: update to latest HEAD")
Signed-off-by: Paul Spooren <mail@aparcar.org>
(cherry picked from commit 853e4dd306)
