When the membase and pci_dev pointer were moved to a new struct in priv,
the actual membase users were left untouched, and they started reading
out arbitrary memory behind the struct instead of registers. This
unfortunately turned the RNG into a constant number generator, depending
on the content of what was at that offset.
To fix this, update geode_rng_data_{read,present}() to also get the
membase via amd_geode_priv, and properly read from the right addresses
again.
Closes#13417.
Reported-by: Timur I. Davletshin <timur.davletshin@gmail.com>
Tested-by: Timur I. Davletshin <timur.davletshin@gmail.com>
Suggested-by: Jo-Philipp Wich <jo@mein.io>
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
(cherry picked from commit 09d13cd8d8)
This compiles the CONFIG_PTP_1588_CLOCK support into the kernel binary
and activates the drivers for KVM and VMware which allow syncing the
host time with the VM when OpenWrt is running in a VM. With this change
the CONFIG_HYPERV_UTILS driver is now build into the kernel, because it
depends on the PTP framework being compiled in. CONFIG_HYPERV_UTILS was
build as a module, but not packages before.
Fixes: #13277
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 54d470ed0e)
This activates PCI Express ASPM control in Linux. Without this option it
is completely controlled by the BIOS, now Linux will take over and apply
some workarounds if needed.
Fixes: #13248
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit ff71035751)
1. Add new symbols to generic config
2. Bump kernel
Changelog: https://lore.kernel.org/stable/2023080818-groin-gradient-a031@gregkh/
All patches automatically rebased.
Signed-off-by: John Audia <therealgraysky@proton.me>
[Refreshed on top of OpenWrt 23.05]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit daed3322d3)
The Amazon ENA network devices are also used on the
AWS Arm (Graviton) instance types, so move it from
the x86-only module file to the top level netdevices.
Signed-off-by: Mathew McBride <matt@traverse.com.au>
(cherry picked from commit 3a7c8fd15e)
Support newer IOMMU_V2 on AMD platforms, useful for DPDK and KVM.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
(cherry picked from commit 1eb02ce325)
The Sophos SG/XG-135 revision 3 has odd numbering of eth ports
where the WAN port (as marked on the case) is:
`eth6` and `eth0`, `eth1`, `eth2`, `eth3`, `eth5`, `eth7`, `eth8` are LAN ports.
Port `eth4` seems to be the SFP port.
Also add the missing LED definition for supported Sophos devices.
Original discussion at:
https://forum.openwrt.org/t/openwrt-on-revision-3-of-sophos-desktop-appliances/152912
Signed-off-by: Stan Grishin <stangri@melmac.ca>
We only use 5.15 kernel. So remove all those unnecessary symbols
referencing 5.10 or 5.15 kernel.
Can be found with:
git grep -E 'LINUX_5_1(0|5)'
Note that we remove the dependency from "sound-soc-chipdip-dac" instead
of removing the complete kernel package. The 5.15 version bump forgot to
delete the "@LINUX_5_10" dependency. The kernel package is still needed
in 5.15 kernel.
Signed-off-by: Nick Hainke <vincent@systemli.org>
All targets are bumped to 5.15. Remove the old 5.10 patches, configs
and files using:
find target/linux -iname '*-5.10' -exec rm -r {} \;
Further, remove the 5.10 include.
Signed-off-by: Nick Hainke <vincent@systemli.org>
Kernel setting CONFIG_IO_URING supports high-performance I/O for file
access and servers, generally for more performant platforms, and adds
~45 KB to kernel sizes. The need for this on less "beefy" devices is
questionable, as is the size cost considering many platforms have kernel
size limits which require tricky repartitioning if outgrown. The size
cost is also large relative to the ~180 KB bump expected between major
OpenWRT kernel releases.
No OpenWrt packages have hard dependencies on this; samba4 and mariadb
can take advantage if available (+KERNEL_IO_URING:liburing) but
otherwise build and work fine.
Since CONFIG_IO_URING is already managed via the KERNEL_IO_URING setting
in Config-kernel.in (default Y), remove it from those target configs
which unconditionally enable it, and update the defaults to enable it
conditionally only on more powerful 64-bit x86 and arm devices. It may
still be manually enabled as needed for high-performance custom builds.
Signed-off-by: Tony Ambardar <itugrok@yahoo.com>
This hack was to bring all existing installations to the newest GRUB
version as fast as possible. Since 19.07.x is EoL we can assume this
task is completed. Now sysupgrade will solely be responsible for
bootloader upgrade.
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Match interface numbers with printed numbers on device enclosure and
assign first port as WAN interface.
Notes
Serial console is available through RJ-45 port with Cisco pinout
baud: 19200, parity: none, flow control: none
The device is setup with UEFI. To enter setup hold DEL or ESC key on
boot. Default UEFI Administrator password is: bcndk1
For users using graphics IC it's advisable to disable display with:
i915.disable_display=1
appending to kernel command line inside bootloader, to save about
0.5-0.6W energy on idle.
For users not using graphics IC, disable it in UEFI, this will save about
1.5W energy on idle.
Pins marked CN19 are ATX power On/Off button.
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
The board is similar to an APU4 except it has an SFP cage for eth0.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Signed-off-by: Petr Štetiar <ynezz@true.cz> [patch refresh]
The CONFIG_DRM_XEN_FRONTEND configuration symbol is also used by the
layerscape target, move it to the generic kernel configuration.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Manually rebased: ath79/patches-5.10/910-unaligned_access_hacks.patch
All other patches automatically rebased.
Signed-off-by: John Audia <therealgraysky@proton.me>
Intel PINCTRL is not enable in the 64bit build, while it is enabled in
the x86/general target, which disables the ability of controlling GPIO
in the 64 bit build.
This commit copies the corresponding part of x86/general config, since
it is already there, so it should be fine to enable the same settings
here.
Signed-off-by: Xiaopo Zhang <xiaopoz@proton.me>
Set 5.15 as default kernel. Testing support was already added on
29.03.2021.
Testing support was added in
9b3bc5d119 ("x86: enable 5.15 as testing kernel").
[0] - edd6021465
Tested-by: Mark Mentovai <mark@mentovai.com>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Tested-by: Thomas Huehn <thomas.huehn@hs-nordhausen.de>
Signed-off-by: Nick Hainke <vincent@systemli.org>
If this module is not set to y, then I get the following compilation
error during geode build.
Package kmod-w83627hf-wdt is missing dependencies for the following libraries:
watchdog.ko
Setting the linux CONFIG_WATCHDOG_CORE to y as in all other targets
fixes this issue.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
If this module is not set to y, then I get the following compilation
error during geode build.
Package kmod-w83627hf-wdt is missing dependencies for the following
libraries:
watchdog.ko
Setting the linux CONFIG_WATCHDOG_CORE to y as in all other targets
fixes this issue.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
This adds some missing IOMMU related options for x86/64 and moves some
of them to generic for all targets.
On x86 IOMMU_DEFAULT_DMA_LAZY is used by default, on all other platforms
IOMMU_DEFAULT_DMA_STRICT is the default. we just follow the default
kernel configuration here.
Fixes: 8fea4a102c ("x86/64: enable IOMMU support")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Enable IOMMU support for Intel and AMD x86 platforms. With this, when the
vfio module is present, physical PCI devices can be passed to VMs, for
example with `qemu-system-x86_64 -device vfio-pci,host=05:00.0 ...`.
IOMMU support increases the kernel size by a small amount (~370KB, from
5239840 B to 5611200 B, a ~7% increase in size).
Signed-off-by: Nicola Corna <nicola@corna.info>
With the commit 5876d6a62f the command under
`/usr/sbin/grub-bios-setup` has been moved to its own package named
`grub-bios-setup`.
The script `81_upgrade_bootloader` under `/lib/preinit` is used by all
x86 targets to update the bootloader. The script is using the command
`grub-bios-setup` for this.
I get the following output at the first boot after the upgrade.
`/etc/preinit: line 9: /usr/sbin/grub-bios-setup: not found`.
To fix this, the DEFAULT_PACKAGES dependency is extended by the entry
`grub2-bios-setup` so that the missing command is installed again.
Signed-off-by: Florian Eckert <fe@dev.tdt.de>
The Meraki MX100 has ten 1000BASE-T and 2 SFP ethernet ports through
3, 4-port PCIe devices. The default enumeration of these network
devices' names does not correspond to their labeling. Fix this by
explicitly naming the devices, mapping against their sysfs path.
Note that these default network names can only be up to 8 characters,
because we can have up to 8 characters of modifiers (e.g. ^br-,
.4096$), and because the maximum network interface name is 16
characters long.
Signed-off-by: Martin Kennedy <hurricos@gmail.com>
[lowercase subject]
Signed-off-by: Paul Spooren <mail@aparcar.org>
All targets expect the malta target already activate the CONFIG_GPIOLIB
option. Move it to generic kernel configuration and also activate it for
malta.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The Lex 3I380NX industrial PC has 4 ethernet controllers on board
which need pmc_plt_clk0 - 3 to function, add it to the critclk_systems
DMI table, so that drivers/clk/x86/clk-pmc-atom.c will mark the clocks
as CLK_CRITICAL and they will not get turned off.
This commit is nearly redundant to 3d0818f5eba8 ("platform/x86:
pmc_atom: Add Lex 3I380D industrial PC to critclk_systems DMI table")
but for all Lex Baytrail devices.
The original vendor firmware is only available using the WaybackMachine:
http://www.lex.com.tw/products/3I380NX.html
Signed-off-by: Michael Schöne <michael.schoene@rhebo.com>
Signed-off-by: Paul Spooren <paul.spooren@rhebo.com>
(Hans broader version for more Lex Baytrail systems, v5.15)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Starting with GCC 12, we have the possibility of mitigating straight-line
speculation vulnerabilities in x86-64 targets. Make it so.
Reported-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
Restore CONFIG_I8K + CONFIG_INTEL_INT0002_VGPIO that got
removed when I refreshed the config. Each x86 target gets
its own CONFIG_CRYPTO_BLAKE2S + LIB settings as only the
x86_64 can use the accelerated x86 version.
Also remove two extra spaces that sneaked into geode's config.
Fixes: 539e60539a ("generic: enable CRYPTO_LIB_BLAKE2S[_X86|_ARM]")
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
This is now built-in, enable so it won't propagate on target configs.
Link: https://lkml.org/lkml/2022/1/3/168
Fixes: 79e7a2552e ("kernel: bump 5.15 to 5.15.44")
Fixes: 0ca9367069 ("kernel: bump 5.10 to 5.10.119")
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
(Link to Kernel's commit taht made it built-in,
CRYPTO_LIB_BLAKE2S[_ARM|_X86] as it's selectable, 5.10 backport)
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
upstream linux have these watchdogs locked behind X86.
These will not build on other architectures. So move them
to target/linux/x86/modules.mk
drivers/watchdog/Kconfig:
|config F71808E_WDT
| tristate "Fintek F718xx, F818xx Super I/O Watchdog"
| depends on X86
|[...]
|config IT87_WDT
| tristate "IT87 Watchdog Timer"
| depends on X86
|[...]
|config ITCO_WDT
| tristate "Intel TCO Timer/Watchdog"
| depends on (X86 || IA64) && PCI
|[...]
|config W83627HF_WDT
| tristate "Watchdog timer for W83627HF/W83627DHG and compatibles"
| depends on X86
|[...]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Grommish reported the dreaded build error that happend with 5.4
since the kernel didn't have the cgpio v2 interface. His reason
for the removed 5.4 was that the octeon target had a memory leak
issue, so he had to backport the removed 5.4 kernel for his tests.
Chen Minqiang chimed in and noted that no matter what (i.e.
@TARGET_x86 in depends) didn't prevent the package from being build
on other targets.
From what I can tell, the reason for this was that +nu801 meant
that kmod-meraki-mx100 pulled in an unconditional dependency as
part of to the kernel build.
|scripts/package-metadata.pl mk tmp/.packageinfo
|
|$(curdir)/kernel/linux/compile += $(curdir)/firmware/linux-firmware/compile \
| $(curdir)/firmware/prism54-firmware/compile \
| $(curdir)/kernel/gpio-button-hotplug/compile \
| >>> $(curdir)/system/gpio-cdev/nu801/compile <<<
change this by making the dependency conditional on the
meraki-mx100 module itself. Note that the nu801 enables/sets
the KCONFIG for the cgpio v2 interface itself, since the
userspace program and not the kernel meraki-mx100 relies on it.
Reference: <https://github.com/openwrt/openwrt/commit/eeb8fd4ce7e9>
Reported-by: Grommish <grommish@gmail.com>
Reported-by: Chen Minqiang <ptpt52@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
QEMU+Libvirt can emulate the ib700wdt watchdogs
which due to its I/O-Port mapping makes it x86
specific.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
