commit 89878f60f4 x86: lift kernel minimum CPU requirement to Pentium MMX
caused kconfig havoc. Fix this and make sure PAE is enabled even on legacy
CPUs as the minimum required CPU has been Pentium MMX for a while now and
hence PAE is supported even on the x86_legacy target.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Anything older than that isn't supported since commit f4f8f4a180,
hence also switch to Pentium MMX when building the kernel.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
commit 4b4f739373 switched on HIGHMEM4G which implicitely disabled
PAE and hence also NX and other useful and security-relevant features.
Re-enable PAE by switching to HIGHMEM64G.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
On the more sophisticated (i.e. deeper FIFO) serial controllers,
flow-control might be needed to avoid dropping output.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
This driver supports CPU-specific idle features on recent Intel
processors. It does not conflict with the ACPI idle driver and
that driver will continue to be used for unsupported and non-Intel
processors.
Signed-off-by: Michael Marley <michael@michaelmarley.com>
The x86_64 build already has the k10temp driver for AMD processors
built in, so this patch adds the coretemp driver for the same
functionality on Intel processors.
Signed-off-by: Michael Marley <michael@michaelmarley.com>
It's unused since commit 7427007193 ("x86: remove the olpc subtarget,
it has been unmaintained for a long time").
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
CONFIG_FB_EFI and CONFIG_FB_SIMPLE are needed to display console text on
EFI framebuffer.
CONFIG_FB_EFI is needed when the kernel is directly launched via EFI
shell or EFI startup.nsh script.
CONFIG_FB_SIMPLE is needed when launching the kernel via grub2 efi. In
this case, grub2 has prepared a gfxterm framebuffer and the kernel just
need to use the already prepared grub's gfxterm framebuffer to display
console text.
Signed-off-by: Alif M. Ahmad <alive4ever@live.com>
This makes the Geode images actually useful again. The Geos profile
should include the relevant hardware for that board, and the Default
profile adds the via-rhine adapter which seems to have been present in
the net5501 and alix targets killed in commit 9e0759ea26 ("x86: merge
all geode based subtargets into one").
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
This adds the default LED and network settings for the PC Engines APU2
when running under the x86 target.
[dwmw2: Change Ethernet port setup]
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
Based on a patch from Chris Blake <chrisrblake93@gmail.com>, except let's
do it by using the LED configuration instead of hard-coding it for each
board type. And try using /bin/board_detect to do the default behaviour,
on the first boot where the config hasn't yet been generated.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
This change moves the files in 657418d to the root of the x86 target.
This is done in preperation for adding more devices under other
subtargets.
CC: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
When we merged all the Geode boards into one generic target, the default
network and LED configuration was lost. Put it back.
Fixes: 9e0759ea26 ("x86: merge all geode based subtargets into one")
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Bump kernel to 4.4.44. Compile-tested on ar71xx, ramips/mt7621 and x86/64.
.44 has been run-tested on the 17.01 branch here on ar71xx and mt7621.
Signed-off-by: Stijn Segers <francesco.borromini@inventati.org>
The following changes enables GPIO sysfs as well as the LEDS_GPIO option
within the kernel. This is required to enable LEDs over a GPIO
interface.
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
Even when the disk uses 4k blocks, the partition table still uses units
of 512 byte sectors. Always use ibs=512 for the offsets
Signed-off-by: Felix Fietkau <nbd@nbd.name>
According to some reports, -march=pentium-mmx is a better choice for
older Geode CPUs than -march=geode anyway.
Bump the minimum architecture of the legacy target from i486 to
pentium-mmx. Anything older is not worth supporting anyway.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
This change adds the sp5100_tco driver as a kernel module for the x86
target. Specifically, this can be used by the PCEngines APU2/APU3. The
reason for having this as a kernel module is to allow users to
load/unload it on demand, as the I2C interface on the APU2/APU3 will not
work while this module is loaded. More info can be found on GitHub at
https://github.com/riptidewave93/LEDE-APU2/pull/5#issuecomment-255667736
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
This adds the following patches to the x86 target:
sp5100_tco: Add AMD Mullins platform support
sp5100_tco: Add AMD Carrizo platform support
sp5100_tco: fix the device check for SB800 and later chipsets
watchdog: sp5100_tco: properly check for new register layouts
With these added, the sp5100_tco driver can then be used on newer AMD
platforms, such as the PCEngines APU2/APU3 boards.
Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
It could cause crashes with some forms of virtualization, and it is
unlikely to work properly with most systems.
It's safer to just disable it.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Is it used by VMware Fusion by default. This allows images to boot
without further config changes in VMware.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
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>
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>
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>
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>
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>
Change grub's root device to xen/xvda,msdos1 for the x86_xen_domu
target so that it will boot without further changes.
Signed-off-by: Wilfried Klaebe <w+lede-project@chaos.in-kiel.de>
[Jo-Philipp Wich: fixed and rebased patch from FS#264, added subject]
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
MMC support has been added to x86-64 a while ago, there is no reason not
to support it in x86-generic as well.
Signed-off-by: Andreas Ziegler <github@andreas-ziegler.de>
[Matthias Schiffer: add commit message]
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
PATA support has been removed from x86-generic without any note in LEDE
r538. Not including them makes the generated images incompatible with older
(and some newer) hardware without any significant gain.
Add it back, and also add the same drivers (as far as available) to x86-64.
Signed-off-by: Andreas Ziegler <github@andreas-ziegler.de>
[Matthias Schiffer: add back x86-generic, update commit message]
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
This will allow dynamically adding/removing at least virtio-net pci
devices which are quite the norm in cloud environment with QEMU/KVM
netdev_add bridge,id=wan2,br=br-wan,helper=/home/yousong/.usr/libexec/qemu-bridge-helper
device_add virtio-net-pci,id=devwan2,netdev=wan2,mac=11:22:33:22:11:00
The config was formed by selecting target x86/64 first, then select
CONFIG_HOTPLUG_PCI and CONFIG_HOTPLUG_PCI_ACPI with
make kernel_menuconfig CONFIG_TARGET=subtarget
The following text tries to explain how the current config was formed
1. CONFIG_PCI_LABEL and CONFIG_ATA_PIIX were removed because they were
already enabled in x86 platform config
2. CONFIG_ATA_SFF was removed because it was enabled in generic config
3. CONFIG_NLS was removed because it will be selected by CONFIG_PCI_LABEL
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
No functional change, it's just to have the same style everywhere. This
way I don't need to use any regex magic to extract all subtargets for
compile tests.
Signed-off-by: Mathias Kresin <dev@kresin.me>
CPU frequency scaling enables the operating system to scale the CPU
frequency up or down in order to save power. CPU frequencies can be
scaled automatically depending on the system load, in response to ACPI
events, or manually by userspace programs.
Signed-off-by: Lucian Cristian <lucian.cristian@gmail.com>
* build for pentium4 instead of i486
* enable PAE
* enable EFI support
* enable KVM guest and host support
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Use the DMI data available in sysfs to extract manufacturer and model info
and write it to /tmp/sysinfo/.
The data will be picked up by board_detect and can be used by e.g. LuCI to
display a more appropriate model description.
On an APU board the files will contain the following values:
# cat /tmp/sysinfo/model
PC Engines APU
# cat /tmp/sysinfo/board_name
pc-engines-apu
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Enable support for DMI decoding in the kernel so that we can access
manufacturer and model information via sysfs.
Also remove redundant per-subtarget DMI overrides and preset a few
previously unset symbols popping up due to the now enabled DMI support.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Generalize the partition discovery in sysupgrade in order to fix sysupgrade
and config backup/recovery on MMC block devices which use a different naming
scheme compared to mtdblock or sd* devices.
The change also adds the find applet to the ramdisk utilities so that upgrade
code can rely on it.
The commit is based on the initial submission by Russell Senior at
http://patchwork.ozlabs.org/patch/625440/ .
Signed-off-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
The kmod-acpi-button functionality is included in the kernel where needed
and the corresponding kmod package has been deleted long ago, therfor drop
the remaining references now.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Virtual disk devices in a Xen virtual machine (DomU) can be /dev/xvda,
/dev/xvdb and so on with partitions like /dev/xdva1. Devices named like
this where not considered before. This resulted in a non working
sysupgrade, because the boot partition could not be found.
Signed-off-by: Alexander Dahl <alex@netz39.de>
Suggested-by: Matthias Schiffer <mschiffer@universe-factory.net>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 49190
The buildbots complained about these config options being missing for arm64:
CONFIG_DUMMY_CONSOLE_COLUMNS=80
CONFIG_DUMMY_CONSOLE_ROWS=25
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48929
With this patch sysupgrade will write directly to the partitions
instead of to the main disk. The UUID is copied from the image
to the MBR as well. This prevents the mbr from being completely
overwritten and losing the partition table. The -p option has
been added to maintain the original behavior and overwite the
entire disk with the new image. Tests have been added to ensure
that the image partitions match up with the active partitions.
Signed-off-by: Rob Mosher <nyt-openwrt@countercultured.net>
SVN-Revision: 48682
This enables booting from devices that use an ATI PATA controller for
the boot device, such as the embedded CF cards in Fujitsu-Siemens Futro
thin-clients.
Signed-off-by: Felix Kaechele <heffer@fedoraproject.org>
SVN-Revision: 47304
Since r40909 openwrt can automatically shutdown when you press the
power button or try to shutdown a kvm virtual machine.
When booting a recent CC-rc2 or trunk image, it doesn't work,
as two kernel modules are missing.
Add these two kernel modules to the default packages, so users
don't need to install them manually.
Signed-off-by: Stefan Hellermann <stefan@the2masters.de>
SVN-Revision: 46366
To improve reproducibility, prevent the inclusion of timestamps
in the gzip header.
Signed-off-by: Reiner Herrmann <reiner@reiner-h.de>
SVN-Revision: 46361
Add features ext4 targz to target x86/xen_domu in order to
generate images in defconfig.
This fixes#18074.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
SVN-Revision: 46233
This changes the x86 image generation to match x86_64, using the PARTUUID for
the rootfs instead of explicitly configuring the device.
It unbreaks KVM with VirtIO, which uses /dev/vda2 instead of /dev/sda2.
Tested in QEMU/KVM with VirtIO, VirtualBox and VMware.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 44966
Build all geode platform drivers into the kernel
Remove kmod-* packages for these drivers
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44809
copied target/linux/x86/config-3.14 and target/linux/x86/patches-3.14 to
3.18 equivalents and then tweaked until it built.
Tested on alix2, soekris net4826 and soekris net4521.
Still having trouble with net4826 booting from Grub, alix2 and net4521
are fine.
Signed-off-by: Russell Senior <russell@personaltelco.net>
SVN-Revision: 44758
Disable ARCH_NEEDS_CPU_IDLE_COUPLED by-default in generic config, since
only one platfrom (omap) needs them.
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44614
HAMRADIO enabled in all generic configs, but no one platform directly
use related drivers/protocols. This symbol is only used for kmod-ax25
package modules. Furthermore, half of platforms explicitly disables
this symbol, what silently disables build of modules for kmod-ax25
package.
So disable HAMRADIO by-default in generic config, add it to kmod-ax25
package and remove it from platform specific configs.
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44613
CONFIG_MTD_SPLIT_SUPPORT symbol default value is 'y' and many platform
specific configs explicitly enables it, while no one platform disables
this symbol. So place it in generic config and remove from platform
specific configs.
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44612
modules build during the kernel compile phase are ignored anyway,
all modules should be built using KernelPackage in
package/kernel/linux/modules/*
selecting the appropriate config symbols there rather than in
target/linux/*/config-*
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 43842
replace all occurences of LINUX_VERSION with the cleaner
approach. future kernel upgrades must mostly touch only
one file. the only platform left is netlogic, because it
uses a intermediate kernel 3.14.16
Signed-off-by: Bastian Bittorf <bittorf@bluebottle.com>
SVN-Revision: 43047
This allows IPv6 to set up without IPv4 being up thus
IPv6-only or IPv6+DS-Lite working with the default config.
Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 42848
CONFIG_BLK_DEV does not activate any functionality, but many code, but
makes it possible to activate other options. Deactivating this breaks
the build of kmod-zram.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 42528
205-fix-headers_install.patch is obseleted by upstream commit 3246a0352e3d58380b9386570f1db1faf7edf8a8
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 41351
ARGC is a 'C-ism', but not known/valid in shell-syntax - insert the correct
var $# (=number of args) here. under normal conditions this had no impact,
but we should at least correct it. the error was observable like this:
root@box:~ [ -e "/etc/functions.sh" ] && . /etc/functions.sh
root@box:~ [ -e "/lib/functions.sh" ] && . /lib/functions.sh
root@box:~ . /lib/upgrade/platform.sh
root@box:~ . /lib/upgrade/common.sh
root@box:~ platform_check_image /tmp/myfirmware.bin
ash: bad number
root@box:~ echo $?
0
Signed-off-by: Bastian Bittorf <bittorf@bluebottle.com>
SVN-Revision: 40915
According to Christophe, the kernel boots on the Soekris net5501
board.
Reported-by: Christophe Prevotaux <cprevotaux@nltinc.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 40468
Place the previous selection (3.3.8) into the only subtarget that did
not override it to 3.10
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 40454
Function move_config should be called after or during preinit_mount_root
hook in do_mount_root function. At this state sysupgrade.tgz is not in its
place during that time when do_mount_root is called. Function move_config
is called later so the sysupgrade.tgz stays in root directory to the second
restart when it is unpacked properly.
This patch adds move_config function to preinit_mount_root hook instead
of preinit_main and changes the filename of script to be called before
80_mount_root is called. It will prepare the sysupgrade archive for do_mount_root
within preinit_mount_root hook.
This patch solves ticket #15042 and #14088.
Signed-off-by: Jiri Slachta <slachta@cesnet.cz>
SVN-Revision: 39996
The 3.6 kernel isn't used by any of the x86 subtargets, so remove it.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 39403
Based on 'kernel: update x86 generic to 3.10.13' patch
from http://patchwork.openwrt.org/patch/4129/
Signed-off-by: Ulrich Weber <uw@xyne.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38500
Based on 'kernel: update x86 generic to 3.10.13' patch
from http://patchwork.openwrt.org/patch/4129/
Signed-off-by: Ulrich Weber <uw@xyne.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38499
TARGET_x86_generic_Soekris48xx is not defined,
CONFIG_TARGET_x86_generic_Soekris48xx is
This changes makes it possible to override the network config for soekris net48xx
and net45xx targets.
Signed-off-by: Bruno Randolf <br1@einfach.org>
SVN-Revision: 38450
Without this patch, the patch
target/linux/generic/patches-3.10/770-bgmac-backport.patch
would not apply.
Signed-off-by: Tobias Steinicke <tsteinicke@net.t-labs.tu-berlin.de>
SVN-Revision: 38438
Taken from the 'alix2: update to LINUX_VERSION 3.10.9'
patch: http://patchwork.openwrt.org/patch/3971/.
Russell says in the original patch:
"Initial whack at updating alix2 to a current kernel. It builds
and runs without problems so far. I copied target/linux/config-3.8,
target/linux/x86/alix2/config-3.8, and target/linux/patches-3.8 to
their 3.10 equivalents and tweaked until it built. The patches
required no changes."
Signed-off-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 37869
Taken from the 'alix2: update to LINUX_VERSION 3.10.9'
patch: http://patchwork.openwrt.org/patch/3971/.
Russell says in the original patch:
"Initial whack at updating alix2 to a current kernel. It builds
and runs without problems so far. I copied target/linux/config-3.8,
target/linux/x86/alix2/config-3.8, and target/linux/patches-3.8 to
their 3.10 equivalents and tweaked until it built. The patches
required no changes."
The current patch contains additional configuration tweaks.
Signed-off-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 37868
Add kernel configuration and patches for Linux 3.10.
Taken from the 'alix2: update to LINUX_VERSION 3.10.9'
patch: http://patchwork.openwrt.org/patch/3971/.
Russell says in the original patch:
"Initial whack at updating alix2 to a current kernel. It builds
and runs without problems so far. I copied target/linux/config-3.8,
target/linux/x86/alix2/config-3.8, and target/linux/patches-3.8 to
their 3.10 equivalents and tweaked until it built. The patches
required no changes."
The current patch contains additional configuration tweaks,
and the kernel patches were refreshed.
Signed-off-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 37867
The USB controller is sitting on the PCI bus and
without the package the EHCI controller is not
usable.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 37835
With
.config:CONFIG_X86_GRUB_SERIAL=""
which (AFAICT) is the way to tell GRUB not to use a serial console, in
target/linux/x86/image/Makefile:ifneq ($(CONFIG_X86_GRUB_SERIAL),)
$(CONFIG_X86_GRUB_SERIAL) expands to `""' (a literal double double-quote),
making the condition unconditionally false.
This patch fixes the situation by passing CONFIG_X86_GRUB_SERIAL through
qstrip before testing.
Signed-off-by: Tamas TEVESZ <ice@extreme.hu>
SVN-Revision: 37749
Job control is turned off because it cannot obtain controlling terminal on
/dev/console, so it is impossible to do any job control operation behind
real x86 device (keyboard+monitor or virtualized).
This patch switches /dev/console to tty devices for ash on x86 generic target.
Signed-off-by: Jiri Slachta <slachta@cesnet.cz>
SVN-Revision: 37693
In order to support both normal images and initramfs, ensure that each
target sets KERNELNAME properly so that the generic kernel building code
can copy the corresponding files over $(KDIR) with the appropriate
extension. Update the various paths to the kernel and wrapper images
from $(LINUX_DIR)/arch/$(ARCH)/boot/$(foo) to $(KDIR)/$(foo).
Signed-off-by: Florian Fainelli <florian@openwrt.org>
SVN-Revision: 37049
The behaviour of calling 'mount' differed depending on whether it called
the busybox-mount, the mount of util-linux, the mount defined in
/lib/functions.sh and /lib/functions/boot.sh
/etc/preinit even included /lib/functions.sh and /lib/functions/boot.sh,
both re-defining 'mount'.
SVN-Revision: 34792
Sysupgrade broke on x86 when grub2 superceded grub. This patch
corrects tha magic_word to be expected in grub2 combined images.
Signed-off-by: Russell Senior <russell@personaltelco.net>
SVN-Revision: 33890
The following patch corrects a hard reset that occurs on alix and
soekris net5501 hardware. Specifically, this is an issue with the via
rhine NIC driver. Under periods of extreme load, the via rhine driver
can cause a reset of the entire system. When this happens no output is
seen on the console, the device simply reboots. This was reported in the
following ticket:
Consistent crash on Soekris NET5501
https://dev.openwrt.org/ticket/11882
I was able to reproduce the problem locally with numerous alix 2d13
platforms. Under heavy 64 byte frame size load from a Smartbits traffic
generator, the alix hardware will hard reset in under 5 minutes. The
same hardware, under the same load, on FreeBSD 9.0 does not reset.
The patch below backports most of the via rhine changes from kernel 3.5.
With OpenWrt trunk being on kernel 3.3.8, it seemed prudent to bring
these changes in to avoid this serious issue.
I've tested this heavily in my test environment and was able to keep the
device stable under load for extended periods of time without any resets.
Also, a nice side effect of this change is that it significantly raises
the performance the platform. Without this patch the alix 2d13 can move
approximately 29,000 packets per second at 64 byte frame sizes. After
this patch the alix can move 52,000 packets per second at 64 byte frame
sizes.
I put this patch under the x86 patches as those are the only platforms
with via rhine hardware that I'm aware of. If it needs to go some place
else please let me know.
Thanks,
Adam
Signed-off-by: Adam Gensler <openwrt@kristenandadam.net>
SVN-Revision: 33072
This patch enables few extra kernel options for the kvm_guest subtarget:
- Tickless kernel to avoid timer ticks in idle guests, reduces CPU usage
- Enable paravirtualization steal time support
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
SVN-Revision: 32436
The alix2 target has the VGA console for grub enabled by default. This
prevents the grub menu from appearing on bootup. It is disabled by
default for the other embedded x86 platforms. This patch corrects that
behavior for the alix2 target.
Signed-off-by: Adam Gensler <openwrt@kristenandadam.net>
SVN-Revision: 31270
The Geos2 includes 2 ADSL+ interfaces, and as such it needs to have the TEQL scheduler for bonding.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
SVN-Revision: 31094
List of patches that Jo-Philipp groveled out of git.
Redux: defuzzed.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
SVN-Revision: 30366
