NOGUCHI Hiroshi [Fri, 24 Aug 2018 01:36:13 +0000 (10:36 +0900)]
ramips: add support for Netgear R6350
Netgear R6350 is a wireless router, aka Netgear AC1750.
Specification:
- SoC: Mediatek MT7621AT (2 CPU cores, 4 threads)
- RAM: 128MiB (Nanya NT5CC64M16GP-DI)
- ROM: 128MiB NAND Flash (Macronix MX30LF1G18AC-TI)
- Wireless:
for 11b/g/n (upto 300Mbps): MT7603
for 11a/ac (upto 1450Mbps) : MT7615, is not avaliable now
- Ethernet LAN speed: up to 1000Mbps
- Ethernet LAN ports: 4
- Ethernet WAN speed: up to 1000Mbps
- Ethernet WAN ports: 1
- USB ports: 1 (USB 2.0)
- LEDs: 4 (all can be controlled by SoC's GPIO)
- buttons: 2
- serial ports: unknown
Installation through telnet:
- Copy kernel.bin and rootfs.bin to a USB flash disk,
plug to usb port on the router. link: http://192.168.1.1/setup.cgi?todo=debug
(login if required, default: admin password)
- You will see "Debug Enabled!"
- Telnet 192.168.1.1 and login with "root"
- ls /mnt/shares/ to find out path of your USB disk.
'myUdisk' for example.
- cd /mnt/shares/myUdisk
- mtd_write write rootfs.bin Rootfs
- mtd_write write kernel.bin Kernel
- reboot
recovery when bricked:
nmrpflash can be used to recover to the netgear firmware
if a broken image was flashed.
The SC_PART_MAP partition suggests that an on flash partition table
exists. After implementing a partition parser/builder for the sercom
partition format, the definitions don't match the flash layout used by
the stock firmware.
It either means the partition format has not yet been completely
understood or it isn't used by the stock firmware. For now, use fixed
partitions instead.
Signed-off-by: NOGUCHI Hiroshi <drvlabo@gmail.com>
[apply latest ramips changes and document the on flash partition map
issues] Signed-off-by: Mathias Kresin <dev@kresin.me>
Daniel Santos [Fri, 19 Oct 2018 08:59:20 +0000 (03:59 -0500)]
jffs2: Fix use of uninitialized delayed_work, lockdep breakage
I've sent this one upstream. This patch is critical if you want to run
with "prove lock correctness" (lockdep) and you happen to have certain
mtd devices. The misuse of the uninitialized object is undefined
behaviour, but being zeroed it does not appear to have actually broken
anything other than the lockdep engine.
Signed-off-by: Daniel Santos <daniel.santos@pobox.com>
kernel: mv88e6060: disable hardware level MAC learning
Disable Marvell's hardware level MAC learning because it breaks station roaming.
When enabled it drops all frames that arrive from a MAC address
that is on a different port at learning table.
Problem description:
Multiple APs with inter-AP roaming connected to different ports,
when station moves from one AP on one port to another AP on another port,
traffic flow breaks down because the learning table is not updated.
Signed-off-by: Anderson Luiz Alves <alacn1@gmail.com>
[Replaced with upstream version, move to generic] Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
ipq806x: add ath10k calibration data MAC addresses patching
Ben Greear reported in his patch:
|Subject: netgear r7800: Fix mac address of radios.
|
|Reloading the driver causes the phyX to change, and that
|caused the MAC address to change.
This is because all ODM/OEMs except QCA bothered to write
the correct MAC address for the ath10k wifi into the
calibration data.
This patch copies over the MAC patching helper functions from ipq40xx's
target/linux/ipq40xx/base-files/etc/hotplug.d/firmware/11-ath10k-caldata
file and converts all the devices to patch the correct MACs into the
extracted calibration data before it gets sent to the driver, which sets
up the device with the correct MAC address. It also removes the entries
in the 10_fix_wifi_mac file as they have served their purpose for good.
Please note the C2600: There is conflicting information on what
the offset for the second wifi is supposed to be. This patch uses
what was specified in 10_fix_wifi_mac.
According to Ben Greear this method is save to use with the
stock firmware too. As he explained that the stock firmware
messes up rx-bssid mask calculation when the MAC is changed
after the first vif is created.
Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
ipq40xx: integrate ath10kcal_patch_mac into ath10kcal_patch_mac_crc
In the thread: "ipq806x: add ath10k calibration data MAC addresses patching"
Chuanhong Guo <gch981213@gmail.com> noted that:
"All ath10k calibration data have a checksum at 0x2.
ath10kcal_patch_mac works for QCA9880/QCA9882 only because
the ath10k firmware for these two chips doesn't check the
checksum value. (QCA proprietary driver checks this and
refuses to use caldata with incorrect checksum.)"
This patch updates 11-ath10k-caldata of the ipq40xx target accordingly.
Reported-by: Chuanhong Guo <gch981213@gmail.com> Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Hauke Mehrtens [Fri, 2 Nov 2018 23:31:22 +0000 (00:31 +0100)]
kernel: netfilter: chain filters merged into nf_tables.ko
In mainline kernel commit 02c7b25e5f5 ("netfilter: nf_tables: build-in
filter chain type") all chain filters were merged into one file and into
one kernel module to save some memory. The code protected by these
configuration options CONFIG_NF_TABLES_BRIDGE, CONFIG_NF_TABLES_IPV4,
CONFIG_NF_TABLES_ARP, CONFIG_NF_TABLES_IPV6, CONFIG_NF_TABLES_NETDEV and
CONFIG_NF_TABLES_INET was merged into the nft_chain_filter.c file which
is now always compiled into the nf_tables.ko file.
This only happened in kernel 4.19 and OpenWrt has to select these as
modules in older kennel versions. Mark them as build-in in the kernel
4.19 specific kernel configuration file which will then not be
overwritten by the package specific settings which try to make them
modular again.
In kernel commit 0168e8b361 ("netfilter: nat: merge ipv4/ipv6 masquerade
code into main nat module") the CONFIG_NF_NAT_MASQUERADE_IPV4 and
CONFIG_NF_NAT_MASQUERADE_IPV6 kernel configuration option were changed
to bool and the code will not be compiled as a own module any more, but
it will be integrated into nf_nat_ipv4.ko or nf_nat_ipv6.ko to save some
memory.
Activate these options as bool in the generic kernel 4.19 configuration
only, to always build them into the nf_nat_ipv*.ko modules. The kmod
file will still try to select them as module, but the generic
configuration will not be overwritten.
Hauke Mehrtens [Sun, 18 Nov 2018 17:15:56 +0000 (18:15 +0100)]
kernel: netfilter: Add nf_tproxy_ipv{4,6} and nf_socket_ipv{4,6}
The nf_socket.ko module was split in commit 8db4c5be88f ("netfilter:
move socket lookup infrastructure to nf_socket_ipv{4,6}.c") into a
common, n IPv4 and an IPv6 part.
The nf_tproxy.ko module was split in commit 45ca4e0cf27 ("netfilter:
Libify xt_TPROXY") into a common, an IPv4 and an IPv6 part.
Hauke Mehrtens [Sun, 18 Nov 2018 13:40:07 +0000 (14:40 +0100)]
kernel: netfilter: Add nf_defrag_ipv6.ko to NF_CONNTRACK on 4.19
In kernel commit a0ae2562c6c ("netfilter: conntrack: remove l3proto
abstraction") The modules nf_conntrack_ipv4.ko, nf_conntrack_ipv6.ko and
nf_conntrack.ko were squashed together into one module. This module now
depends on nf_defrag_ipv6 when IPv6 support was activated. This is part
of the main netfilter packages, so add nf_defrag_ipv6.ko also to the
default netfilter packages on kernel 4.19 and later.
Hauke Mehrtens [Fri, 2 Nov 2018 23:32:39 +0000 (00:32 +0100)]
kernel: Make CONFIG_NVMEM tristate
CONFIG_OF_NET depends on CONFIG_NVMEM in kernel 4.19. To fix some build
problems in mainline Linux kernel CONFIG_NVMEM was changed from tristate
to bool in commit 2a37ce25d9 ("nvmem: disallow modular CONFIG_NVMEM").
This patch in OpenWrt revert the upstream commit and changes
CONFIG_NVMEM back to tristate we just have to make sure to build this in
for all targets which select CONFIG_OF_NET.
Hauke Mehrtens [Fri, 2 Nov 2018 19:41:37 +0000 (20:41 +0100)]
kernel: Adapt to new location of video-videobuf2 in kernel 4.19
The video-videobuf2 kernel modules were moved to a new folder in kernel
4.19. videobuf2-v4l2.ko is only available since kernel 4.4, blacklist
this kmod completely on kernel 3.18.
Hauke Mehrtens [Sat, 8 Dec 2018 13:47:11 +0000 (14:47 +0100)]
kernel: Use crypto_simd.ko instead of lrw.ko for x86
The x86 optimized cryptographic algorithm kernel modules now mostly use
crypto_simd.ko instead of lrw.ko in kernel 4.19. Add the new module to
the kmod-crypto-misc package.
Hauke Mehrtens [Fri, 2 Nov 2018 13:34:12 +0000 (14:34 +0100)]
kernel: Add kmod-phy-realtek
The r8169 driver uses the phy lib with the realtek phy driver in kernel
4.19 instead of integrating the phy driver into the mac driver.
Add the new phy driver and add this missing dependency.
Hauke Mehrtens [Thu, 1 Nov 2018 17:01:44 +0000 (18:01 +0100)]
kernel: Make the patches apply on top of 4.19
This makes the patches which were just copied in the previous commit
apply on top of kernel 4.19.
The patches in the backports-4.19 folder were checked if they are really
in kernel 4.19 based on the title and only removed if they were found in
the upstream kernel.
The following additional patches form the pending folder went into
upstream Linux 4.19:
pending-4.19/171-usb-dwc2-Fix-inefficient-copy-of-unaligned-buffers.patch
pending-4.19/190-2-5-e1000e-Fix-wrong-comment-related-to-link-detection.patch
pending-4.19/478-mtd-spi-nor-Add-support-for-XM25QH64A-and-XM25QH128A.patch
pending-4.19/479-mtd-spi-nor-add-eon-en25qh32.patch
pending-4.19/950-tty-serial-exar-generalize-rs485-setup.patch
pending-4.19/340-MIPS-mm-remove-mips_dma_mapping_error.patch
Bigger changes were introduced to the m25p80 spi nor driver, as far as I
saw it in the new code, it now has the functionality provided in this
patch:
pending-4.19/450-mtd-m25p80-allow-fallback-from-spi_flash_read-to-reg.patch
Part of this patch went upstream independent of OpenWrt:
hack-4.19/220-gc_sections.patch
This patch was reworked to match the changes done upstream.
The MIPS DMA API changed a lot, this patch was rewritten to match the
new DMA handling:
pending-4.19/341-MIPS-mm-remove-no-op-dma_map_ops-where-possible.patch
I did bigger manual changes to the following patches and I am not 100% sure if they are all correct:
pending-4.19/0931-w1-gpio-fix-problem-with-platfom-data-in-w1-gpio.patch
pending-4.19/411-mtd-partial_eraseblock_write.patch
pending-4.19/600-netfilter_conntrack_flush.patch
pending-4.19/611-netfilter_match_bypass_default_table.patch
pending-4.19/670-ipv6-allow-rejecting-with-source-address-failed-policy.patch
hack-4.19/211-host_tools_portability.patch
hack-4.19/221-module_exports.patch
hack-4.19/321-powerpc_crtsavres_prereq.patch
hack-4.19/902-debloat_proc.patch
This is based on patchset from Marko Ratkaj <marko.ratkaj@sartura.hr>
Hauke Mehrtens [Thu, 1 Nov 2018 16:57:55 +0000 (17:57 +0100)]
kernel: Copy patches from kernel 4.14 to 4.19
This just copies the files from the kernel 4.14 specific folders into
the kernel 4.19 specific folder, no changes are done to the files in
this commit.
Hauke Mehrtens [Wed, 14 Nov 2018 20:37:47 +0000 (21:37 +0100)]
kernel: Move modifications of b53.h into patch
The b53 driver was added as a dsa driver into the mainline Linux kernel,
but we still use the swconfig based driver. The header file b53.h is
used by both drivers, but the swconfig one needs an extra member, add
this one in a patch to not overwrite the version shipped with the
mainline kernel.
Hauke Mehrtens [Wed, 14 Nov 2018 20:22:39 +0000 (21:22 +0100)]
kernel: Replace ledtrig-netdev with upstream backport
The ledtrig-netdev was added to upstream Linux kernel 4.16, replace our
own version with the patch based on the upstream version.
This will remove the ledtrig-netdev support from kernel 3.18, because I
not want to spend time on backporting it to 3.18. This will make it
easier to use the upstream version with kernel 4.19, by just not
applying this patch.
Hans Dedecker [Fri, 14 Dec 2018 15:06:03 +0000 (16:06 +0100)]
glibc: update to latest 2.27 commit (BZ #23927)
f6d0e8c36f Fix rwlock stall with PREFER_WRITER_NONRECURSIVE_NP (bug 23861) 2794474c65 powerpc: Add missing CFI register information (bug #23614) d8ca7a0c71 Fix _dl_profile_fixup data-dependency issue (Bug 23690) f1e211096b inet/tst-if_index-long: New test case for CVE-2018-19591 [BZ #23927] 4e9f34e54f support: Implement <support/descriptors.h> to track file descriptors e75481a7a7 support: Close original descriptors in support_capture_subprocess 0f79dc0be3 support_quote_string: Do not use str parameter name 6b2dd53aa0 support: Implement support_quote_string 50477165b9 malloc: Add another test for tcache double free check. 4b246928bd malloc: tcache double free check
Hauke Mehrtens [Thu, 6 Dec 2018 22:39:52 +0000 (23:39 +0100)]
mac80211: Update to version 4.19.7-1
This updates the backports package used in mac80211 to version 4.19.7-1
which is based on kernel 4.19.7. This integrates all the stable fixes
introduces in this kernel version.
The deleted patches are not needed any more because they are either
included in the upstream Linux kernel 4.19.7 or in backports 4.19.7-1.
Stijn Tintel [Wed, 12 Dec 2018 20:00:44 +0000 (22:00 +0200)]
kmod-dma-buf: fix build with external kernel
In hack/904-debloat_dma_buf.patch, DMA_SHARED_BUFFER is changed from
bool to tristate. As this patch is not applied to external kernel
sources, build fails if kmod-dma-buf is enabled. Fix this by only
including the module file if CONFIG_EXTERNAL_KERNEL_TREE and
CONFIG_KERNEL_GIT_CLONE_URI are not enabled.
Jo-Philipp Wich [Wed, 12 Dec 2018 15:32:53 +0000 (16:32 +0100)]
rpcd: update to latest Git head
3aa81d0 file: access exec timeout via daemon ops structure 7235f34 plugin: store pointer to exec timeout value in the ops structure ccd7c0a treewide: rename exec_timeout to rpc_exec_timeout
Hans Dedecker [Wed, 12 Dec 2018 11:32:46 +0000 (12:32 +0100)]
iproute2: backport upstream patch to fix print_0xhex on 32 bit
The argument to print_0xhex is converted to unsigned long long
so the format string give for normal printout has to be some
variant of %llx. Backport the patch as otherwise, bogus values
will be printed on 32 bit platforms.
Registering the GPIO chip without a parent device completely breaks the
ath9k GPIOs for device tree targets.
As long as boards using the devicetree don't have the gpio-controller
property set for the ath9k node, the unloading of the driver works as
expected.
Register the GPIO chip with the ath9k device as parent only for OF
targets to find a trade-off between the needs of driver developers and
the broken LEDs and buttons seen by users.
Mathias Kresin [Thu, 29 Nov 2018 22:40:12 +0000 (23:40 +0100)]
ramips: use new image build code for D-Link DCS-930 family
Drop the factory images and the firmware tool to create them. They don't
work any more, since the factory image has an uImage header covering the
whole kernel + rootfs. This way the uImage splitter will not be able to
find the rootfs and the kernel will panic later on.
The factory images were most likely added at a time the board had
distinct partitions for kernel and rootfs.
Daniel Golle [Thu, 22 Nov 2018 19:53:09 +0000 (20:53 +0100)]
hostapd: add support for WPS pushbutton station
similar to hostapd, also add a ubus interface for wpa_supplicant
which will allow handling WPS push-button just as it works for hostapd.
In order to have wpa_supplicant running without any network
configuration (so you can use it to retrieve credentials via WPS),
configure wifi-iface in /etc/config/wireless:
This patch updates the uboot-fritz4040 package to the latest version.
The portability and private-libgcc patches, as well as the
upload-to-f4040.sh script have been added to the upstream repository.
Furthermore, the upload-to-f4040 has been updated to take the first
parameter as the file it is supposed to flash, otherwise it defaults
to the previous "uboot-fritz4040.bin". Furthermore the error messages
have been improved and ftp will now dump some "progress information"
to the user's console.
Also included is support for gcc 8+ and a fix for the obnoxous error
that currently breaks the builders:
| fritz/src/lzma2eva.c:23:30: fatal error: zlib.h: No such file or directory
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Fix the imagetag on the HG655b to allow a correct partition detection at boot time.
It turns out that it was defined at the wrong partition. Just move the imagetag to
the linux firmware partition.
The bug is present since the 18.06 release. Without this fix, the board won't boot.
Fixes: a27d59bb4274 ("brcm63xx: switch to new partition layout specification") Signed-off-by: Daniel Gonzalez Cabanelas <dgcbueu@gmail.com> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Backport upstream commits. Most interesting 122392e which changes how
SERVFAIL is handled especially in event of genuine server down/failure
scenarios with multiple servers. a799ca0 also interesting in that
answered received via TCP are now cached, DNSSEC typically using TCP
meant until now answers weren't cached, hence reducing performance.
59e4703 Free config file values on parsing errors. 48d12f1 Remove the NO_FORK compile-time option, and support for uclinux. 122392e Revert 68f6312d4bae30b78daafcd6f51dc441b8685b1e 3a5a84c Fix Makefile lines generating UBUS linker config. 24b8760 Do not rely on dead code elimination, use array instead. Make options bits derived from size and count. Use size of option bits and last supported bit in computation. No new change would be required when new options are added. Just change OPT_LAST constant. 6f7812d Fix spurious AD flags in some DNS replies from local config. cbb5b17 Fix logging in cf5984367bc6a949e3803a576512c5a7bc48ebab cf59843 Don't forward *.bind/*.server queries upstream ee87504 Remove ability to compile without IPv6 support. a220545 Ensure that AD bit is reset on answers from --address=/<domain>/<address>. a799ca0 Impove cache behaviour for TCP connections.
Along with an additional patch to fix compilation without DHCPv6, sent
upstream.
I've been running this for aaaages without obvious issue hence brave
step of opening to wider openwrt community.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
buildbot found an error
option.c: In function 'dhcp_context_free':
option.c:1042:15: error: 'struct dhcp_context' has no member named 'template_interface'
free(ctx->template_interface);
revert for the moment
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Backport upstream commits. Most interesting 122392e which changes how
SERVFAIL is handled especially in event of genuine server down/failure
scenarios with multiple servers. a799ca0 also interesting in that
answered received via TCP are now cached, DNSSEC typically using TCP
meant until now answers weren't cached, hence reducing performance.
59e4703 Free config file values on parsing errors. 48d12f1 Remove the NO_FORK compile-time option, and support for uclinux. 122392e Revert 68f6312d4bae30b78daafcd6f51dc441b8685b1e 3a5a84c Fix Makefile lines generating UBUS linker config. 24b8760 Do not rely on dead code elimination, use array instead. Make options bits derived from size and count. Use size of option bits and last supported bit in computation. No new change would be required when new options are added. Just change OPT_LAST constant. 6f7812d Fix spurious AD flags in some DNS replies from local config. cbb5b17 Fix logging in cf5984367bc6a949e3803a576512c5a7bc48ebab cf59843 Don't forward *.bind/*.server queries upstream ee87504 Remove ability to compile without IPv6 support. a220545 Ensure that AD bit is reset on answers from --address=/<domain>/<address>. a799ca0 Impove cache behaviour for TCP connections.
I've been running this for aaaages without obvious issue hence brave
step of opening to wider openwrt community.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Openwrt no longer uses and has not used since 5acfe55d71 Jun 2016 the
timestamp file (/etc/dnsmasq.time) method of resolving the dnssec/ntp
dnslookup chicken/egg problem, having used signals from ntp since that
change.
Drop the 'dnssec-improve-timestamp-heuristic' patch since it is neither
used nor sent upstream. One less thing to refresh & maintain.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
base-files: add sysupgrade -k to save list of pkgs
When '-k' is used, sysupgrade inserts into backup a new file
/etc/backup/installed_packages.txt which contains pkgname and origin (rom,
overlay, unknown) without touching rootfs.
base-files: add sysupgrade -o to save all overlay files
Add sysupgrade '-o' option in order to include all overlay files in
backup, except for those that are from packages but including files
listed in conffiles, sysupgrade.conf or /lib/upgrade/keep.d.
With '-u' option, it will skip files equals to /rom and conffiles that
were not changed.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
base-files: add sysupgrade -u to skip unchanged files
With '-u', for a file /aaa/bbb/ccc enlisted for backup,
it will only get into backup if /rom/aaa/bbb/ccc does not
exist or /aaa/bbb/ccc is different from /rom/aaa/bbb/ccc.
It also works with '-c', but only effective for files touched
but not modified.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
projects / thirdparty / openwrt.git / log
