l request:
it may be best if I just apply for commit access to the
OpenWrt git repository and import it myself perhaps?
Signed-off-by: Linus Walleij
---
Changes in v3:
- Drop applied patch.
- Move APEX to build from a git URL (Ansuel)
- Add a build rule for lib-chacha20 for armeb
- Split microcod
: Linus Walleij
---
package/firmware/ixp4xx-microcode/Makefile | 77 +++
.../firmware/ixp4xx-microcode/src/IxNpeMicrocode.h | 148 +
package/firmware/ixp4xx-microcode/src/LICENSE.IPL | 27
3 files changed, 252 insertions(+)
diff --git a/package/firmware
The IXP4xx crypto module must be loaded after the rootfs is
up as it depends on loading some NPE microcode from the file
system.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v3:
- Use AutoProbe for loading the module.
---
package/kernel/linux/modules/crypto.mk | 13 +
target/li
To support the IXP42x platforms we need a kernel module
for the Epson R7301 RTC so we can load it as an optional
module.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v3:
- Use AutoProbe, drop preinit hook.
---
package/kernel/linux/modules/other.mk | 15 +++
1 file changed,
To support the IXP42x platforms we need a kernel module
for the X1205 RTC so we can load it as an optional module.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v3:
- Use autoprobe, exclude from preinit.
---
package/kernel/linux/modules/other.mk | 16
1 file changed,
Harte
Signed-off-by: Linus Walleij
---
ChangeLog v2->v3:
- Don't add the apex package on the NSLU2, the dependency
is resolved in the other direction, apex should only build
itself for the targets that need it.
- Alter the path for the apex binaries slightly.
- Split out resurrectio
This brings back USRobotics USR8200 support to the IXP4xx
target.
Signed-off-by: Linus Walleij
---
.../linux/ixp4xx/base-files/etc/board.d/02_network | 5 +
target/linux/ixp4xx/config-6.1 | 1 +
target/linux/ixp4xx/image/Makefile | 13 ++
...p4xx-Handle
On Fri, Oct 13, 2023 at 12:28 AM Andrey Jr. Melnikov
wrote:
> Linus Walleij wrote:
> > This script will make sure to spin down harddrives on IXP4xx
> > NAS devices such as the NSLU2 and siblings after 1 minute
> > of inactivity.
>
> Why not use hotplug events?
Could
e for that.
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
On Fri, Oct 20, 2023 at 1:33 PM Christian Marangi wrote:
> On Fri, Oct 20, 2023 at 01:15:51PM +0200, Linus Walleij wrote:
> > If there is some rough consensus on this, could someone start
> > applying these patches, apart from patch 10/10 where we might
> > want to d
there are several things like that in linux-firmware.
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
On Fri, Oct 20, 2023 at 2:52 PM Christian Marangi wrote:
> On Fri, Oct 20, 2023 at 02:48:33PM +0200, Linus Walleij wrote:
> > On Fri, Oct 20, 2023 at 2:32 PM Christian Marangi
> > wrote:
> >
> > > Yes! Would be ideal then we can just bump the linux-firmware packag
: Linus Walleij
---
package/firmware/ixp4xx-microcode/Makefile | 77 +++
.../firmware/ixp4xx-microcode/src/IxNpeMicrocode.h | 148 +
package/firmware/ixp4xx-microcode/src/LICENSE.IPL | 27
3 files changed, 252 insertions(+)
diff --git a/package/firmware
This is a backport of the patch for byte addressed IO to the
Epson RTC7301 driver. This is used by the IXP4xx-based
USRobotics USR8200.
Signed-off-by: Linus Walleij
---
...6.7-rtc-rtc7301-Support-byte-addressed-IO.patch | 93 ++
1 file changed, 93 insertions(+)
diff --git
This is a backport of the patch to support the Altima AMI101L
PHY which is merged for the v6.7 kernel. This PHY is used in the
IXP4xx-based USRobotics USR8200.
Signed-off-by: Linus Walleij
---
7-net-phy-amd-Support-the-Altima-AMI101L.patch | 82 ++
1 file changed, 82
The IXP4xx crypto module must be loaded after the rootfs is
up as it depends on loading some NPE microcode from the file
system.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v3:
- Use AutoProbe for loading the module.
---
package/kernel/linux/modules/crypto.mk | 13 +
target/li
Harte
Signed-off-by: Linus Walleij
---
ChangeLog v2->v3:
- Don't add the apex package on the NSLU2, the dependency
is resolved in the other direction, apex should only build
itself for the targets that need it.
- Alter the path for the apex binaries slightly.
- Split out resurrectio
l request:
it may be best if I just apply for commit access to the
OpenWrt git repository and import it myself perhaps?
Signed-off-by: Linus Walleij
---
Changes in v4:
- Bring the AMD Phy online at preinit, since this may sit
on a primary network interface.
- Split off some USR8200 patches
This adds a package for the AMD and Altima PHY, found in some
odd devices.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v4:
- Add preinit flag since the network interface may be the only
and primary network interface of the platform.
---
package/kernel/linux/modules/netdevices.mk |
This brings back USRobotics USR8200 support to the IXP4xx
target.
Signed-off-by: Linus Walleij
---
.../linux/ixp4xx/base-files/etc/board.d/02_network | 5 +
target/linux/ixp4xx/config-6.1 | 1 +
target/linux/ixp4xx/image/Makefile | 13 ++
...p4xx-Handle
otplug script for the disk? That I think
> was dropped from the series?
Yes I have to check if I can rework it to a generic hotplug script
and drop the corresponding thing in Gemini, it's useful for any
NAS setup.
Yours,
Linus Walleij
___
o
erial console.
The custom inittab is a bit ugly, but I've seen that this
seems to be what other devices do, other ideas welcome.
Signed-off-by: Linus Walleij
---
Linus Walleij (2):
kernel: usb: Autoprobe g_serial
gemini: Activate serial USB console on the DNS-313
package/kernel/lin
This brings up a serial console on the USB device port of
the DNS-313.
Signed-off-by: Linus Walleij
---
target/linux/gemini/Makefile | 2 +-
target/linux/gemini/base-files/etc/inittab | 4
target/linux/gemini/image/Makefile | 2 +-
3 files changed, 6 insertions(+), 2
t is a bit
catch 22 to have to use a serial console to activate the
serial console.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/usb.mk | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/package/kernel/linux/modules/usb.mk
b/package/kernel/linux/modules/us
On Sun, Nov 19, 2023 at 2:27 PM Chuanhong Guo wrote:
> On Sun, Nov 12, 2023 at 6:04 AM Linus Walleij
> wrote:
> >
> > For devices using USB gadget serial we need to probe the modules
> > usb_f_acm and g_serial in order, then the available UDC (USB device
> > contro
Bump the U-Boot version used for BCM53xx to the 2024.01
version that includes all the needed patches upstream, so we
can get rid of those in the process.
Signed-off-by: Linus Walleij
---
package/boot/uboot-bcm53xx/Makefile| 4 +-
.../0001-nand-brcmnand-add-iproc-support.patch
This patch is needed for ethernet to work on most Gemini
devices so backport it.
Signed-off-by: Linus Walleij
---
Changes in v2:
- Also send the patch to openwrt-devel :/
- Link to v1:
https://lore.kernel.org/r/20240123-gemini-ethernet-fix-v1-1-0bbe0b272...@linaro.org
---
...6.8-net-ethernet
of the bigger "firmware" partition. Which
we use for installation and upgrades.
Signed-off-by: Linus Walleij
---
.../files/drivers/mtd/mtdsplit/mtdsplit_uimage.c | 19 +++
1 file changed, 15 insertions(+), 4 deletions(-)
diff --git a/target/linux/generic/files/d
he end 4 bytes of FCS.
This clearly illustrates that the DSA tag is included in the MTU
which we set up in Linux, but the VLAN tag and ethernet headers and
checksum is not.
Cc: Álvaro Fernández Rojas
Cc: Jonas Gorski
Tested-by: Paul Donald
Signed-off-by: Linus Walleij
---
ChangeLog v6
gt; So I merged it @ 3cf1fe5508ee8a2a2c6e33a829bfb6c81381ccd0
> https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=3cf1fe5508ee8a2a2c6e33a829bfb6c81381ccd0
> https://github.com/openwrt/openwrt/commit/3cf1fe5508ee8a2a2c6e33a829bfb6c81381ccd0
Thanks Álvaro!
Yours,
Linus Walleij
___
se the BCM6328-based Inteno XG6846 won't boot an
> image without a small U-Boot binary first in the partition followed by
> the uImage, then the rootfs.
OK changed it.
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@list
t to
0x3 it works like a charm.
I first thought that maybe it's not as flexible (for example if U-Boot
becomes bigger and pass a 64K boundary) but it doesn't really
affect this system as I control this pretty tightly anyway.
Thanks, I'll dr
, and it depends on e.g. the I2C and SFP
to be loaded as modules first.
We have no examples of DSA switches being packaged as
modules before, all seem to be compiled in, but it
actually works just fine to do this.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/netdevices.mk
e6xxx switch into all
bcm6328 kernels. (Those are footprint-sensitive.)
I updated the initramfs and sysupgrade images on my site especially
for you so you can test it :)
I try to trickle in the required patches a bit at the time.
Yours,
Linus Walleij
_
with a proper RTC
is just surplus.
All Kirkwoods have at least one RTC so add RTC to the features
list for Kirkwood as well.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/other.mk | 17 +
target/linux/kirkwood/Makefile| 2 +-
target/linux/kirkwood/image
These patches have been cooking for some time, let's
get them moving.
The idea is to merge this base so we have base support
for the target and then try to work out remaining issues
such as the LED handling.
To:
Signed-off-by: Linus Walleij
---
Changes in v2:
- Include U-Boot for BMIPS
ured out
the two-level switch hierarchy and settings.
Link: https://forum.openwrt.org/t/help-with-inteno-xg6846/68276/14
Signed-off-by: Linus Walleij
---
.../bcm6328/base-files/etc/board.d/02_network | 3 +
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 312 +++
Since we split the Inteno XG6846 "firmware" partition with the
uImage MTD splitter, we need to compile in support for this
splitting method into the BCM6328.
Signed-off-by: Linus Walleij
---
target/linux/bmips/bcm6328/config-6.1 | 1 +
1 file changed, 1 insertion(+)
diff --git a/ta
instead.
The compiled u-boot.bin becomes around ~300KB and with
LZMA compression it will swiftly fit into 128KB, so
we use two 64KB erase blocks right after the CFE to
store an imagetag:ed U-Boot.
Signed-off-by: Linus Walleij
---
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 3 ++-
target
This is needed to boot the BCM6238-based Inteno XG6846.
Currently this is restricted to the XG6846 board.
Signed-off-by: Linus Walleij
---
.github/labeler.yml | 1 +
package/boot/uboot-bmips/Makefile | 32
2 files changed, 33 insertions(+)
diff
now, please check that I didn't do something
obviously wrong.
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
The Gemini works fine with kernel v6.6.
As per the example for ipq806x, drop support for anything
older than v6.6, there is no point in supporting it,
and the new DTS SoC directory just makes it hard to
maintain.
Signed-off-by: Linus Walleij
---
target/linux/gemini/Makefile
The Gemini works fine with kernel v6.6.
As per the example for ipq806x, drop support for anything
older than v6.6, there is no point in supporting it,
and the new DTS SoC directory just makes it hard to
maintain.
Signed-off-by: Linus Walleij
---
ChangeLog v1->v2:
- Update all patches for qu
igned-off-by: Linus Walleij
---
...-net-ethernet-cortina-Restore-TSO-support.patch | 124 +
...hernet-cortina-Use-TSO-also-on-common-TCP.patch | 95
...ernet-cortina-Rename-adjust-link-callback.patch | 36 ++
...hernet-cortina-Use-negotiated-TX-RX-pause.
igned-off-by: Linus Walleij
---
Changes in v2:
- Added a missing locking fix patch.
- Link to v1:
https://lore.kernel.org/r/20240515-gemini-ethernet-patches-v1-1-9a7c1770e...@linaro.org
---
.../0003-net-ethernet-cortina-Locking-fixes.patch | 73
...-net-ethernet-cortina-Restor
The IXP4xx is well supported upstream and can readily be
supported with kernel v6.6. To simplify things after the
DTS directory was renamed, switch to v6.6 only.
Bring in some outstanding patches.
Tested on the Gateworks GW2348-4.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules
with DHCP as expected for a NAS device.
Signed-off-by: Linus Walleij
---
target/linux/ixp4xx/base-files/etc/board.d/02_network | 1 +
target/linux/ixp4xx/image/Makefile| 13 +
2 files changed, 14 insertions(+)
diff --git a/target/linux/ixp4xx/base-files/etc
further
work this image is not built by default, but made available
for developers who know what they are doing.
The TAR+CRC image generation is a rewritten version of the
earlier support code.
Signed-off-by: Linus Walleij
---
.../linux/ixp4xx/base-files/etc/board.d/02_network | 1 +
target/linux
Several of the IXP4xx machines mount root on external harddrives
so add EXT4 and rootfs-part to the featureset so the right
features are always selected.
Signed-off-by: Linus Walleij
---
target/linux/ixp4xx/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/target
OpenWrt supported the D-Link DSM G600 A in the past. It has
64 MB of RAM and 16 MB of flash so it will run just fine,
and should be quite usable with a rootfs on an external
harddrive.
Signed-off-by: Linus Walleij
---
target/linux/ixp4xx/base-files/etc/board.d/02_network | 1 +
target/linux
/developers/rwhitby
Some devices have low specs but anything with 64MB of memory
and 4MB of flash and a USB port for a USB stick or harddrive
rootfs can boot and run current OpenWrt just fine.
Signed-off-by: Linus Walleij
---
Linus Walleij (4):
ixp4xx: Add ext4 and rootfs-part to features
There are too many RTC drivers in other.mk, they deserve their
own menu and .mk-file, so let's break them out to a separate
entity.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/other.mk | 247
package/kernel/linux/modules/rtc.mk
isabling CPU idle on these altogether stabilize them, then maybe that
is what we need to do?
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
afał Miłecki
Reviewed-by: Linus Walleij
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
On Mon, May 27, 2024 at 2:00 PM Rafał Miłecki wrote:
> From: Rafał Miłecki
>
> OpenWrt One is the first ever OpenWrt product. It's based on MT7981B and
> has entered an early production stage.
>
> Signed-off-by: Rafał Miłecki
Reviewed-by: Linus Walleij
e OpenWrt project.
The legal responsibility and trademark of this product is handled
by OpenWrt, so they are legally the entity producing this board,
I think that is what matters for vendor strings.
Yours,
Linus Walleij
___
openwrt-devel mailing l
On Mon, May 27, 2024 at 2:00 PM Rafał Miłecki wrote:
> From: Rafał Miłecki
>
> OpenWrt project (with the help of MediaTek and Banana Pi) has produced
> its very first own hardware. It needs its own prefix.
>
> Signed-off-by: Rafał Miłecki
Reviewed-by: Linus Walleij
Yo
> Signed-off-by: Rafał Miłecki
> > ---
> > V2: Include .h files
> > Send without 22 people in cc to let is reach openwrt-devel@
Thanks for doing this.
Reviewed-by: Linus Walleij
Yours,
Linus Walleij
___
openwrt-devel mai
if (cpu_idle_force_poll == 1234)
> + arch_cpu_idle();
Idle again.
I would have tried to see what arch_cpu_idle() is doing.
arm_pm_idle() or cpu_do_idle()?
What happens if you just put return in arch_cpu_idle()
so it does nothing?
Yours,
Linus Walleij
_
The recently added D-Link DNS-320L and the Zyxel NSA310S
is missing an RTC module so let's give them the default
Marvell RTC at least.
Signed-off-by: Linus Walleij
---
target/linux/kirkwood/image/Makefile | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/target/
Two fixes requested by pauld.
Would be great to get tested.
Signed-off-by: Linus Walleij
---
Linus Walleij (2):
bmips: intento-xg6846: Add EEPROM node
bmips: inteno-xg6846: Fix pin control for USB LED
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 8 +++-
1 file changed, 7
The USB LED is GPIO16 not GPIO17, as can be seen further
down the device tree.
Signed-off-by: Linus Walleij
---
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/target/linux/bmips/dts/bcm6328-inteno-xg6846.dts
b/target/linux
For some troublesome devices it is necessary to obtain direct access
to the SFP module EEPROM so define it in the device tree.
Suggested-by: Paul Donald
Signed-off-by: Linus Walleij
---
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 6 ++
1 file changed, 6 insertions(+)
diff --git a
This adds the LED definitons for the XG6846 DSA port LEDs.
These are standard properties compatible with the existing
Marvell 88e6xxx DT bindings and fully standardized so this
is fine to add. They will be used by the in-flight Marvell
88e6xxx LEDs support patch.
Signed-off-by: Linus Walleij
For some troublesome devices it is necessary to obtain direct access
to the SFP module EEPROM so define it in the device tree.
Suggested-by: Paul Donald
Signed-off-by: Linus Walleij
---
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 6 ++
1 file changed, 6 insertions(+)
diff --git a
Some versions of the Inteno XG6846 has a USB port mounted.
For these machines the corresponing USB port nodes need to
be enabled.
Suggested-by: Henrik Ginstmark
Signed-off-by: Linus Walleij
---
target/linux/bmips/dts/bcm6328-inteno-xg6846.dts | 12
1 file changed, 12 insertions
Some small fixes to the XG6846 device tree.
A new series instead of the old fixes I sent a bit back:
noltari folded in a LED fix fro the old series and now
there are two new patches.
Signed-off-by: Linus Walleij
---
Linus Walleij (3):
bmips: inteno-xg6846: Activate USB port
bmips
Adding openwrt-devel to this thread to get some traction...
On Thu, Oct 24, 2013 at 12:06 PM, Linus Walleij
wrote:
> On Tue, Oct 1, 2013 at 12:57 PM, Linus Walleij
> wrote:
>
>> This converts the gemini machine to use generic clockevents
>> by rewriting the time
As described by Tobias Waldekranz this patch is obsolete
and the issue is already fixed in the upstream kernels we
are using.
Link: http://lists.openwrt.org/pipermail/openwrt-devel/2024-October/043238.html
Signed-off-by: Linus Walleij
---
...xxx-Request-assisted-learning-on-CPU-port.patch | 27
This solution that is needed for some routers to provide proper
LED activity when controlled directly by the MV88E6xxx switch,
has just been merged in upstream Linux.
Refreshing it affects line numbers in some other mv88e6xxx patches
so update these too.
Signed-off-by: Linus Walleij
offsets in pending patches are augmented as part of
the refresh.
Signed-off-by: Linus Walleij
---
Changes in v2:
- Patch is merged for v6.13 so move the patch over to backports.
- Link to v1:
https://lore.kernel.org/r/20241006-xg6846-marvell-leds-v1-1-ab3ad9329...@linaro.org
---
package/kernel
, so there's no need for
> any assistance.
OK I'll fix!
Thanks Tobias!
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
On Sun, Oct 6, 2024 at 6:23 PM Daniel Golle wrote:
> On Sun, Oct 06, 2024 at 05:32:04PM +0200, Linus Walleij wrote:
> > This solution that is needed for some routers to provide proper
> > LED activity when controlled directly by the MV88E6xxx switch,
> > has just been mer
On Sun, Oct 6, 2024 at 6:23 PM Daniel Golle wrote:
> On Sun, Oct 06, 2024 at 05:32:04PM +0200, Linus Walleij wrote:
> > This solution that is needed for some routers to provide proper
> > LED activity when controlled directly by the MV88E6xxx switch,
> > has just been mer
Upstream stable is slow at picking this up and several systems
are regressing. Add the patch locally in OpenWrt.
Signed-off-by: Linus Walleij
---
...td-spi-nor-winbond-fix-w25q128-regression.patch | 59 ++
1 file changed, 59 insertions(+)
diff --git
a/target/linux/generic
device tree with an MV88E6xxx switch
and an OpenWrt image target.
Signed-off-by: Linus Walleij
---
target/linux/kirkwood/config-6.6 | 4
target/linux/kirkwood/image/Makefile | 8 +---
2 files changed, 5 insertions(+), 7 deletions(-)
diff --git a/target/linux/kirkwood/config-6.6 b
devices with no DSA switch.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/netdevices.mk | 17 +
target/linux/ixp4xx/config-6.6 | 3 ---
target/linux/ixp4xx/image/Makefile | 2 +-
3 files changed, 18 insertions(+), 4 deletions(-)
diff --git a
e
with no DSA switch, as other DSA switches (such as qca8k)
have recently moved toward using a module.
Yours,
Linus Walleij
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
> when today's devices can no longer fit a 'standard' 1GB kernel, and
> 'standard' flash space is like 1TB :D
Heh not quite there, but this worked fairly well.
Someone asked me at some point if this device had to be a
dead paperweight now, but actually it can be used.
as a 100MBit only 2g only WiFi AP. Which
you may or may not think is worth having. But you
can!
Details:
https://dflund.se/~triad/krad/netgear-wg302/
Signed-off-by: Linus Walleij
---
Changes in v2:
- Added more target patches after finding bugs and missing
LEDs in the upstream device tree
The WG302 v1 has 32MB of RAM so it can easily run OpenWrt,
however it lacks much flash: only 8 MB.
By just using the flash for rootfs and booting a kernel over
TFTP it works just fine.
Signed-off-by: Linus Walleij
---
.../linux/ixp4xx/base-files/etc/board.d/02_network | 3 +-
target/linux
This adds the OpenWrt-only LED aliases to the supported IXP4xx
devices.
Signed-off-by: Linus Walleij
---
.../302-ARM-dts-ixp4xx-OpenWrt-LED-aliases.patch | 228 +
1 file changed, 228 insertions(+)
diff --git
a/target/linux/ixp4xx/patches-6.6/302-ARM-dts-ixp4xx-OpenWrt
Some ixp4xx platforms with a proper RedBoot config do not
contain the per-ethernet interface npe_eth0_esa, but rather
a single entry named zcom_npe_esa.
Let's use this if fconfig can't find the primary key.
This is needed on the Netgear WG302 v1.
Signed-off-by: Linus Walleij
---
ta
The WG302 v1 have a separate rootfs partition that we
simply just upgrade with a new rootfs image. The kernel
need to be updated on the TFTP server.
Signed-off-by: Linus Walleij
---
.../ixp4xx/base-files/lib/upgrade/platform.sh | 27 ++
target/linux/ixp4xx/image
bmitting.
>
> PR: https://github.com/openwrt/openwrt/pull/17654
>
> It does not include the vlan/forwarding extra patch, only upstream patches.
After 2 weeks of radio silence I rebased and merged this, it's clearly
what we need and there i
Only the DIR-685 uses a Realtek DSA switch currently so
convert this platform to load it from a module and save
memory on all other devices.
At the same time drop the rt2800 PCI package that this
device actually isn't using, it needs an out-of-tree
wireless driver.
Signed-off-by: Linus Wa
) patch?
We also need to get somewhere with this. I'm not sure I can help
super much, every time I deal with VLAN I feel like a hippo in a porcelain
store, but can you point me to the patches so I can form an
opinion?
Yours,
Linus Walleij
___
The WG302 v1 has 32MB of RAM so it can easily run OpenWrt,
however it lacks much flash: only 8 MB.
By just using the flash for rootfs and booting a kernel over
TFTP it works just fine.
Signed-off-by: Linus Walleij
---
.../linux/ixp4xx/base-files/etc/board.d/02_network | 3 +-
target/linux
as a 100MBit only 2g only WiFi AP. Which
you may or may not think is worth having. But you
can!
Details:
https://dflund.se/~triad/krad/netgear-wg302/
Signed-off-by: Linus Walleij
---
Linus Walleij (3):
ixp4xx: Add back support for Netgear WG302 v1
ixp4xx: Support sysupgrade on WG302
Some ixp4xx platforms with a proper RedBoot config do not
contain the per-ethernet interface npe_eth0_esa, but rather
a single entry named zcom_npe_esa.
Let's use this if fconfig can't find the primary key.
This is needed on the Netgear WG302 v1.
Signed-off-by: Linus Walleij
---
ta
The WG302 v1 have a separate rootfs partition that we
simply just upgrade with a new rootfs image. The kernel
need to be updated on the TFTP server.
Signed-off-by: Linus Walleij
---
.../ixp4xx/base-files/lib/upgrade/platform.sh | 27 ++
target/linux/ixp4xx/image
the
tagger for the driver.
Tested on the BMIPS Inteno XG6846.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/netdevices.mk | 29 +
1 file changed, 9 insertions(+), 20 deletions(-)
diff --git a/package/kernel/linux/modules/netdevices.mk
b/package/kernel
Hi Hiroshi,
On Mon, Jan 13, 2025 at 4:47 AM INAGAKI Hiroshi
wrote:
> > +define KernelPackage/dsa-rtl8366rb/description
>
> Isn't this "KernelPackage/dsa-realtek-smi/description" ?
Indeed, thanks for spotting this! I'll
is is how the kernel deals with them.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/netdevices.mk | 51 ++
1 file changed, 51 insertions(+)
diff --git a/package/kernel/linux/modules/netdevices.mk
b/package/kernel/linux/modules/netdevices.mk
Not all Gemini platforms are using a DSA switch, so
by converting the switches into packages and just selecting
that package for the one device needing it we save a lot
of space (600KB+) on all other devices.
Signed-off-by: Linus Walleij
---
Changes in v2:
- Fix typo in one subpackage
Only the DIR-685 uses a Realtek DSA switch currently so
convert this platform to load it from a module and save
memory on all other devices.
At the same time drop the rt2800 PCI package that this
device actually isn't using, it needs an out-of-tree
wireless driver.
Signed-off-by: Linus Wa
have no dependencies.
Signed-off-by: Linus Walleij
---
Changes in v2:
- Push this out to the actual mailing list so it reaches
patchwork, sorry for the mess :/
---
Linus Walleij (2):
mvebu: Switch to use a module for mv88e6xxx
qoriq: Switch to use a module for mv88e6xxx
target
g the packages I went over both the upstream DTS files
and the OpenWrt-specific DTS files and used grep 'marvell,mv88e6'
to find the devices using these switches.
Signed-off-by: Linus Walleij
---
target/linux/mvebu/config-6.6 | 5 -
target/linux/mvebu/image/cortexa53.mk |
Only the DIR-685 uses a Realtek DSA switch currently so
convert this platform to load it from a module and save
memory on all other devices.
At the same time drop the rt2800 PCI package that this
device actually isn't using, it needs an out-of-tree
wireless driver.
Signed-off-by: Linus Wa
is is how the kernel deals with them.
Signed-off-by: Linus Walleij
---
package/kernel/linux/modules/netdevices.mk | 51 ++
1 file changed, 51 insertions(+)
diff --git a/package/kernel/linux/modules/netdevices.mk
b/package/kernel/linux/modules/netdevices.mk
501 - 600 of 604 matches
Mail list logo
