Commit Graph

14024 Commits

Author SHA1 Message Date
Felix Fietkau
d0ba3bb1e2 kernel: finally remove layer7 filter support
it has been non-functional for years and caused numerous memleaks and
crashes for people that tried to enable it.
it has no maintained upstream source, and it does not look like it's
going to be fixed any time soon

Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45423
2015-04-13 22:23:14 +00:00
Felix Fietkau
87f854059a kernel: fix more bgmac dma init/cleanup issues
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45422
2015-04-13 20:08:30 +00:00
Imre Kaloz
232c61742e mvebu: copy the initramfs files to the bin dir
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45421
2015-04-13 20:06:47 +00:00
Felix Fietkau
1bf8e555bb kernel: sync bgmac changes with latest upstream submission
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45420
2015-04-13 17:32:33 +00:00
Felix Fietkau
24e9f52106 bcm53xx: add support for the Buffalo WXR-1900DHP
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45419
2015-04-13 17:32:25 +00:00
Felix Fietkau
1e48a48915 bcm53xx: fix GPIO pins > 16
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45418
2015-04-13 17:32:18 +00:00
Felix Fietkau
6bc743ef6f kernel: add bgmac fixes for bcm53xx (fix ethernet bringup on a bcm4709 based device)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45417
2015-04-13 17:32:12 +00:00
Felix Fietkau
1f70f452cc bcm53xx: add missing l2 cache controller AUXCTL bit
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45416
2015-04-13 17:32:05 +00:00
Imre Kaloz
8c684f42e5 mvebu: add support for the in-CPU RTC on the Armada 38x
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45415
2015-04-13 13:06:34 +00:00
Imre Kaloz
c9e6c7c836 mvebu: add comments to image generation
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45414
2015-04-13 12:15:40 +00:00
Imre Kaloz
e3bd4f0969 mvebu: stop legacy image generation
If you can't find the firmware for you board, send proper patches.

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45413
2015-04-13 12:08:42 +00:00
Imre Kaloz
96c793f529 mvebu: generate a single firmware for the 385-db-ap
This has been done without having a board, but should work.

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45412
2015-04-13 12:06:56 +00:00
Imre Kaloz
4da8f96754 mvebu: reflect Armada 38x support in the name
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45411
2015-04-13 11:30:39 +00:00
Felix Fietkau
d24388da44 bcm53xx: add a comment about the gpio base patch
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45408
2015-04-12 22:10:44 +00:00
Felix Fietkau
afafbc0d74 kernel: bgmac: add more DMA related fixes
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45407
2015-04-12 22:10:40 +00:00
Imre Kaloz
09722d0bdc ppc40x: upgrade to 3.18
left "broken" as I'm not sure if my only board is to blame.. testers welcomed

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45406
2015-04-12 20:48:13 +00:00
Imre Kaloz
e55532b7af orion: upgrade to 3.18 (compile tested only)
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45405
2015-04-12 20:45:55 +00:00
Rafał Miłecki
77c75e6fa3 bcm53xx: try to handle Netgear R8000 VLANs config
It seems to have few ports connected to CPU (only for CPU sending data?)
as part of "SMP dual core 3 GMAC setup" feature.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45403
2015-04-12 20:11:25 +00:00
Rafał Miłecki
06ac2f5c74 b53: improve overriding CPU port state on BCM5301X
On BCM5301X there are two different cases to handle: CPU port 8 vs. any
other one. Support for CPU port 8 was already partially implemented but
it lacked setting some extra bit for 2G speed. It also will need to be
extended to implement "SMP dual core 3 GMAC setup". That's the reason
for handling it in separated code block.
This patch also adds overriding CPU port state for port other than 8. It
requires using recently defined GMII_PORT registers.
It was tested for regressions on BCM53011 revs 2 & 3. It was also
confirmed to fix switch on some internal Broadcom board.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Acked-by: Jonas Gorski <jogo@openwrt.org>

SVN-Revision: 45402
2015-04-12 20:00:42 +00:00
Imre Kaloz
ceb7321e13 mvebu: refresh patches
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45401
2015-04-12 17:30:41 +00:00
Imre Kaloz
b0ad24f0d5 mvebu: migrate mamba support to the upstreamed code
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45400
2015-04-12 17:29:50 +00:00
Felix Fietkau
2dbe131a62 bcm53xx: fix bcma gpio chip base
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45399
2015-04-12 17:22:04 +00:00
Felix Fietkau
808fb7a0bd kernel: add missing memory barrier to bgmac
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45398
2015-04-12 16:39:24 +00:00
Imre Kaloz
6f01238398 mvebu: handle other SKUs as well
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45396
2015-04-12 15:04:14 +00:00
Imre Kaloz
d1bedfbdf5 mvebu: include the ahci driver in the default profile
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45395
2015-04-12 15:03:25 +00:00
Imre Kaloz
6cfe1539da generic: make sure we only try to mangle bootargs we do have
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45393
2015-04-12 14:46:19 +00:00
Rafał Miłecki
95ef64a779 bcm53xx: allow sysupgrade firmwares without UBI containing rootfs
In most cases it allows reverting back to the vendor firmware (as they
usually don't use UBI). If users wants to do that we can't do anything
anyway. Erease counters will be just lost. The only thing we do is warn:
"Flashing firmware without UBI for rootfs. All erase counters will be
lost."
It still requires forcing sysupgrade.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45392
2015-04-12 12:04:13 +00:00
Rafał Miłecki
24d40980b0 bcm53xx: add platform_pre_upgrade to prepare for kernel+UBI sysupgrade
We can now detect that provided firmware contains kernel and UBI image
partitions. Flashing it in a sane way (keeping erase counters) still
needs to be implemented.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45391
2015-04-12 11:46:16 +00:00
Rafał Miłecki
a04b71d330 bcm53xx: rework vendor fw handling to don't duplicate upgrade calls
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45389
2015-04-12 10:42:27 +00:00
Felix Fietkau
8904cd1dcc kernel: add bgmac fixes for various issues
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45388
2015-04-12 10:35:21 +00:00
Rafał Miłecki
503ccc502d bcm53xx: fix UBI EOF patch to allow sysupgrade after first installation
With previous version of patch info about need of erasing blocks was
stored once per boot. It was breaking in following scenario:
1) First boot after installation (erasing blocks after 0xdeadc0de)
2) Doing sysupgrade (with ubidetach & ubiformat)
3) Attaching UBI again (it caused all blocks to be erased)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45387
2015-04-12 10:20:59 +00:00
Imre Kaloz
a62fa2d365 mvebu: reformat the damaged syscfg partition on the mamba
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>

SVN-Revision: 45386
2015-04-12 08:35:33 +00:00
Felix Fietkau
1a797043e5 kernel: revert section sorting change from r41951
It currently does not seem to make a difference anymore, except by
increasing compressed kernel image size

Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45385
2015-04-11 21:21:03 +00:00
Luka Perkov
bb828e30b0 imx6: add missing patches from 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45378
2015-04-11 10:46:20 +00:00
Felix Fietkau
e8126cbfbd kernel: remove kernel module checks/dependencies for 3.14
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45377
2015-04-11 10:43:46 +00:00
Luka Perkov
8865505665 generic: drop 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45375
2015-04-11 10:28:14 +00:00
Luka Perkov
b66b0ae632 brcm47xx: drop 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45374
2015-04-11 10:28:11 +00:00
Luka Perkov
fc08d7a3eb adm8668: drop 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45373
2015-04-11 10:28:06 +00:00
Luka Perkov
b976ac8b86 mcs814x: drop 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45372
2015-04-11 10:28:03 +00:00
Luka Perkov
d21b92af8e imx6: drop 3.14
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45371
2015-04-11 10:28:00 +00:00
Luka Perkov
912a6371f8 imx6: switch to 3.18
Signed-off-by: Luka Perkov <luka@openwrt.org>

SVN-Revision: 45370
2015-04-11 10:27:49 +00:00
Zoltan Herpai
4fd21b203e mxs: reinstate rootfs image creation and dtb copy
Signed-off-by: Harald Geyer <harald@ccbib.org>
Acked-by: Michael Heimpold <mhei@heimpold.de>

SVN-Revision: 45369
2015-04-10 21:36:33 +00:00
Zoltan Herpai
db4ce3d16c mxs: Include the right kernel image in the rootfs
Signed-off-by: Harald Geyer <harald@ccbib.org>
Acked-by: Michael Heimpold <mhei@heimpold.de>

SVN-Revision: 45368
2015-04-10 21:34:21 +00:00
John Crispin
6d0ad03fac kernel: remove references to kmod-of-i2c
these break the IB since the module is part of i2c-core now

Signed-off-by: John Crispin <blogic@openwrt.org>

SVN-Revision: 45363
2015-04-10 13:39:33 +00:00
Felix Fietkau
9428c5609b kernel: add jffs2 RENAME_WHITEOUT support, fixes overlayfs renames (#19432)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45362
2015-04-10 11:52:59 +00:00
Felix Fietkau
e1f0a91dfe kernel: remove linux 3.10 support
Signed-off-by: Felix Fietkau <nbd@openwrt.org>

SVN-Revision: 45360
2015-04-10 11:52:47 +00:00
Jonas Gorski
4e826d8303 b53: clean up code to match kernel style better
* properly enclose macro arguments in paranthesis on use
* remove trailing white space
* convert C99 // comments
* add missing blank lines after declaration
* remove braces from single statement blocks
* split lines > 80 chars (except for one)

Signed-off-by: Jonas Gorski <jogo@openwrt.org>

SVN-Revision: 45356
2015-04-10 10:29:04 +00:00
Jonas Gorski
61885f95f0 b53: define registers available and needed on BCM5301X
They are also present on some BCM63xx switches.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45355
2015-04-10 10:28:46 +00:00
Rafał Miłecki
9d6e04a4f1 bcm53xx: build image for ASUS RT-AC56U
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45353
2015-04-10 09:58:00 +00:00
Rafał Miłecki
71fcbffc91 bcm53xx: Luxul XWC-1000 has only 128 MiB of RAM
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>

SVN-Revision: 45352
2015-04-10 09:42:48 +00:00