The ath10k dt-binding property "qcom,coexist-support" was
explicitly defined as type uint8 since upstream commit
ed09c61eb19d ("dt-bindings: net: Convert ath10k to YAML").
Therefore, this hack patch no longer makes sense.
Link: ed09c61eb1
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/18393
Signed-off-by: Robert Marko <robimarko@gmail.com>
Backport two commits to resolve issues with ath1kk causing it to fail
driver registration on iommuless systems with DRAM outside of 32bit
addressing such as a 4GiB imx8mm:
commit 1bcd20981834 ("wifi: ath11k: Fix DMA buffer allocation
to resolve SWIOTLB issues")
commit eeadc6baf8b3 ("wifi: ath11k: Use dma_alloc_noncoherent
for rx_tid buffer allocation")
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Link: https://github.com/openwrt/openwrt/pull/17751
Signed-off-by: Robert Marko <robimarko@gmail.com>
This new version fixes some memory leak and NULL pointer issues.
Upstreamed patches:
[1] 010-api_update.patch
[2] 201-wifi-ath10k-add-LED-and-GPIO-controlling-support-for.patch
The new 003 patch was introduced to fix the build error on old LTS
kernel. The patch 004 was used to fix the issue of IPQ4019 radio
not being able to start.
Tested on QCA9882 and IPQ4019.
[1] bca8bc0399
[2] 8e1debd824
Signed-off-by: Shiji Yang <yangshiji66@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/18368
Signed-off-by: Robert Marko <robimarko@gmail.com>
Fix compilation error for old stable version caused by
genlmsg_multicast_allns backport fix pushed middle version.
Version 5.15 version 0-169, 6.1 version 0-115, 6.6 version 0-58 have the
old genlmsg_multicast_allns version with flags variable.
Compiling backport project with these version result in a compilation
error. To handle this, introduce a backport function for the affected
kernel version.
Link: https://github.com/openwrt/openwrt/pull/18373
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
MR7500 features IPQ6018 which is CPPE, however, it also utilizes AQR114CB0 as WAN.
Previously, without this patch, aquantia_phy_api_ops was never called.
Therefore, enable `IN_AQUATIA_PHY` so that AQR PHY gets initialized properly on boot.
[ 3.973942] Aquantia AQR114C 90000.mdio-1:08: loading firmware version 'v5.6.5 Cybertan Divo 090221 14:43:44' from 'NVMEM'
[ 16.059953] aquantia_phy_api_ops_init[1485]:INFO:qca probe aquantia phy driver succeeded!
[ 17.930149] Aquantia AQR114C 90000.mdio-1:08: attached PHY driver (mii_bus:phy_addr=90000.mdio-1:08, irq=POLL)
Signed-off-by: Weikai Kong <priv@pppig236.com>
Link: https://github.com/openwrt/openwrt/pull/18185
Signed-off-by: Robert Marko <robimarko@gmail.com>
Sets the boot flag for the i40e network device driver
to load it at a more early stage of the boot process.
With commit 0a47d518df,
I added a boot priority for the mlx4 and mlx5 drivers.
Also, increase those priorities because I think they
are too low since there is currently no "room" for
built-in network device drivers.
That can cause interface order, i.e., name inconsistencies,
when Mellanox ConnectX cards are inserted or removed.
Signed-off-by: Til Kaiser <mail@tk154.de>
Link: https://github.com/openwrt/openwrt/pull/17990
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The CONFIG_NET_SWITCHDEV option is needed by CONFIG_DSA and some other
options. It is boolean, we have to compile it into the kernel it self.
Activate it for all targets in the generic configuration, it is already
activated for most of them. This allows to install DSA drivers as a
module.
On the ramips/mt7620 target the kernel would grown by 4.5kB.
For some small targets which do not support a DSA switch by default the
option is deactivated.
Link: https://github.com/openwrt/openwrt/pull/17668
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
(This was a real problem I encountered with a nanopi R6S device and
an external rtl8152 usb3 network controller - the USB controller would
claim the eth1 name, causing much confusion).
Signed-off-by: Michel Lespinasse <michel@lespinasse.org>
Link: https://github.com/openwrt/openwrt/pull/17638
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
It doesn't depend on either usb-net or usb-net-cdc-ncm. It does, however, depend
on mii. Fix thusly, and make it depend explicitly on usb, not usb-net.
While at it, add a conditional dependency on libphy, for future kernel versions.
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
Now that CONFIG_USB_NET_DRIVERS is unconditionally enabled, remove it from the
usb-net kconfig symbol list.
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
The kmod-mlxsw-spectrum driver activated CONFIG_DCB indirectly already
on all targets which are building this driver. All other DCB capable
driver did not activate their DCB support.
CONFIG_DCB increases the uncompressed kernel size by about 7.8KB.
CONFIG_DCB is only needed some data center Ethernet cards and not used
on normal routers. Activate it only on the x86_64 and the armsr_arm64
target which are used on normal servers or in VMs.
Link: https://github.com/openwrt/openwrt/pull/17672
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
btmtk.ko is used by btusb.ko and btmtkuart.ko, add it into an own
package and make both packages depend on it.
Fixes: 1c42a0be36 ("kernel: modules: bluetooth: separating UART and USB drivers")
Link: https://github.com/openwrt/openwrt/pull/18110
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
There is some new changes since 12.5.r2 to 12.5.5, so refresh
and update patches.
Changes:
modified: package/kernel/qca-nss-dp/Makefile
modified: package/kernel/qca-nss-dp/patches/0006-nss_dp_main-Use-a-phy-handle-property-to-connect-to-.patch
modified: package/kernel/qca-nss-dp/patches/0008-nss-dp-allow-setting-netdev-name-from-DTS.patch
deleted: package/kernel/qca-nss-dp/patches/0011-01-edma_v1-rework-hw_reset-logic-to-permit-rmmod-and-in.patch
deleted: package/kernel/qca-nss-dp/patches/0011-02-nss_dp_switchdev-correctly-unregister-notifier-on-dp.patch
deleted: package/kernel/qca-nss-dp/patches/0011-03-nss_dp_main-swap-dp_exit-function-call.patch
deleted: package/kernel/qca-nss-dp/patches/0011-04-nss_dp_main-call-unregister_netdev-first-in-dp_remov.patch
deleted: package/kernel/qca-nss-dp/patches/0011-05-nss_dp_main-use-phy_detach-instead-of-disconnect-in-.patch
deleted: package/kernel/qca-nss-dp/patches/0011-06-edma_v1-skip-edma_disable_port-in-edma_cleanup-subse.patch
Log:
2024-10-16 |4c9f671| [qca-nss-dp] removed the calling of fal_port_autoneg_status_get
2024-10-16 |fb33119| Merge "[qca-nss-dp] Change Debug Level of TX Complete Errors"
2024-10-15 |e197b9d| [qca-nss-dp] Change Debug Level of TX Complete Errors
2024-10-03 |1b7aeb1| Merge remote-tracking branch origin/AU_LINUX_QSDK_GINGER_OPEN_TARGET_ALL.12.5.5.230.918 into HEAD
2024-10-01 |830ac73| Merge "[qca-nss-dp] Enable rmmod support for qca-nss-dp module."
2024-06-05 |aef242d| [qca-nss-dp] Enable rmmod support for qca-nss-dp module.
2024-09-18 |b1bfeb6| Merge "[qca-nss-dp] Move pr_info to pr_debug"
2024-09-12 |54aee26| Merge "[qca-nss-dp] Fix Loopback ring teardown path"
2024-09-10 |21f6567| Merge "[qca-nss-dp] Extend Tx Complete errors to track individual error type"
2024-09-10 |7fee76f| Merge "[qca-nss-dp] Destination VP xmit support"
2024-09-09 |0d4ecfd| [qca-nss-dp] Fix Loopback ring teardown path
2024-08-29 |438ab8e| [qca-nss-dp] Move pr_info to pr_debug
2024-08-29 |3f45e3d| Merge "[qca-nss-dp] Fix EDMA Recovery API to Prevent Multiple RX NAPI Add Calls"
2024-08-23 |7a792b1| [qca-nss-dp] Fix EDMA Recovery API to Prevent Multiple RX NAPI Add Calls
2024-07-08 |1227746| [qca-nss-dp] Add support for DDR extended buffer
2024-08-20 |d45bceb| [qca-nss-dp] Extend Tx Complete errors to track individual error type
2024-06-17 |a8f7155| [qca-nss-dp] Destination VP xmit support
2024-08-21 |a5d189b| Merge "[qca-nss-dp] Support for the PPE-VP Qdisc feature"
2024-08-16 |da97bc7| Merge "[qca-nss-dp] Restrict compilation of a file in DP module"
2024-06-22 |bf90fe9| [qca-nss-dp] Support for the PPE-VP Qdisc feature
2024-02-07 |6ceeb8f| [qca-nss-dp] Enable EDMA driver to allocate from beyond 4GB space.
2024-03-19 |4a7ff28| [qca-nss-dp] Restrict compilation of a file in DP module
2024-07-09 |73bad34| [qca-nss-dp] Check for DS node state before getting ppe queue
2024-05-03 |22cdbd6| [qca-nss-dp] Add PPE-DS Enqueue vp to queue mapping
2024-07-04 |5d6ef18| Merge "[qca-nss-dp] NAPI Budget change for KPI improvement"
2024-07-02 |cd0b543| Merge "[qca-nss-dp] Read fake_mac header indication from rx_desc"
2024-07-01 |160b988| [qca-nss-dp] NAPI Budget change for KPI improvement
2024-05-22 |54c2fd1| [qca-nss-dp]: Fixing SA warnings as part of qca-nss-dp module.
2024-06-22 |602534b| [qca-nss-dp] Read fake_mac header indication from rx_desc
2023-11-24 |10210e2| [qca-nss-dp] EDMA register changes for ipq54xx
2024-05-22 |1af0d03| Merge "[qca-nss-dp]: Change to support PPE-VP path for MLO Assist"
2024-05-15 |2acd9f3| [qca-nss-dp]: Change to support PPE-VP path for MLO Assist
2023-09-27 |1ca59f2| [qca-nss-dp] DP changes for ipq54xx
2024-04-18 |bc60c5a| Merge "[qca-nss-dp] Add support for the XGMAC latency computation"
2024-04-18 |6a67d6d| Merge "[qca-nss-dp] Fix the packets processing in the Rx NAPI"
Signed-off-by: Kristian Skramstad <kristian+github@83.no>
Link: https://github.com/openwrt/openwrt/pull/17731
Signed-off-by: Robert Marko <robimarko@gmail.com>
This removes -static compile option. The -static option tells GCC to
link this statically with the libc, which we do not want in OpenWrt. We
want to link everything dynamically to the libc. This fixes a compile
problem with glibc.
References: 71bdff91 "ltq-vdsl-mei: Remove static linking"
Signed-off-by: Andre Heider <a.heider@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/18056
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Not all devices need Bluetooth UART and USB drivers to be installed together. Three separate drivers have been created:
- kmod-bluetooth
- kmod-hci-uart
- kmod-btusb
Signed-off-by: Paweł Owoc <frut3k7@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/15118
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Commit 7bc487c12e ("kernel: ltq-vdsl-vr9-mei: fix warning about
field-spanning write") patched the driver to use unsafe_memcpy in
MEI_IoctlCmdMsgWrite.
However, this is not actually necessary. The assignment of the variable
"pDestPtr" can be modified so that the compiler knows about the correct
size. This way, the check in the fortified memcpy works correctly.
While at it, also adjust all places where similar code is used to copy
from a CMV_STD_MESSAGE_T struct.
Also mark all related structs as packed, because the code (and the
driver in general) seems to rely on that anyway.
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250206225444.2521817-2-jan@3e8.eu/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Since the update to kernel 6.1, a warning like this appears in the
kernel log:
[ 49.773953] ------------[ cut here ]------------
[ 49.773998] WARNING: CPU: 3 PID: 2349 at target-arm_cortex-a7+neon-vfpv4_musl_eabi/linux-ipq40xx_generic/dsl_cpe_mei-ugw_8.5.2.10/src/drv_mei_cpe_msg_process.c:3570 MEI_IoctlCmdMsgWrite+0x290/0x2c8 [drv_mei_cpe]
[ 49.777670] memcpy: detected field-spanning write (size 4) of single field "pDestPtr" at target-arm_cortex-a7+neon-vfpv4_musl_eabi/linux-ipq40xx_generic/dsl_cpe_mei-ugw_8.5.2.10/src/drv_mei_cpe_msg_process.c:3570 (size 2)
...
[ 50.087078] ---[ end trace 0000000000000000 ]---
The variable "pDestPtr" points to the field "header.index" in a
CMV_STD_MESSAGE_T struct (header is a CMV_STD_MESSAGE_HEADER_T struct).
The offending code intentionally copies data beyond this field, which is
followed by "header.length" and "payload".
To fix this, change the assignment of "pDestPtr" to use the pointer to
the message plus the offset of the "header.index" field. This way, the
compiler knows about the size and thus the false positive warning
disappears.
While at it, also adjust all places where similar code is used to copy
from a CMV_STD_MESSAGE_T struct.
Also mark all related structs as packed, because the code (and the
driver in general) seems to rely on that anyway.
Fixes: https://github.com/openwrt/openwrt/issues/17142
Signed-off-by: Jan Hoffmann <jan@3e8.eu>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250206225444.2521817-1-jan@3e8.eu/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
The fixes are only for the WRT1900X and WRT1200AC.
It contains:
Deletes the driver's ability to modify the debit table.
Remove skb_get(done_skb) in txdone
Reworking ISR
clean code
Napi replaces tasklet
Add rx_decrypt feature
Signed-off-by: Michael Trinidad <trinidude4@hotmail.com>
Link: https://github.com/openwrt/openwrt/pull/17997
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
The FDB roaming issues were observed on ipq807x and ipq60xx boards.
The fix depends on API exposed only when NSS_DP_PPE_SUPPORT is enabled.
However, this flag applies to above mentioned platforms only and is
causing the logs to be flooded on other QCA platforms, including ipq50xx,
with:
[ 34.893418] nss-dp 39c00000.dp1 lan: cannot get VSI ID for port 1
[ 34.898370] nss-dp 39c00000.dp1 lan: cannot get VSI ID for port 1
[ 34.904598] nss-dp 39c00000.dp1 lan: cannot get VSI ID for port 1
[ 34.910661] nss-dp 39c00000.dp1 lan: cannot get VSI ID for port 1
So let's apply a dependency on the NSS_DP_PPE_SUPPORT flag and contain
the patch code for ipq807x and ipq60xx within conditional directives.
Tested on: Linksys SPNMX56
Signed-off-by: George Moussalem <george.moussalem@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/17966
Signed-off-by: Robert Marko <robimarko@gmail.com>
This downstream patch fixes a bug which could flood the logs with the
following message and would eventually lead to a crash.
ath11k c000000.wifi: failed to send HAL_REO_CMD_UPDATE_RX_QUEUE cmd, tid 0 (-105)
Signed-off-by: George Moussalem <george.moussalem@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/17182
Signed-off-by: Robert Marko <robimarko@gmail.com>
Add QCN6122 platform support.
QCN6122 is a PCIe based solution that is attached to and enumerated
by the WPSS (Wireless Processor SubSystem) Q6 processor.
Though it is a PCIe device, since it is not attached to APSS processor
(Application Processor SubSystem), APSS will be unaware of such a decice
and hence it is registered to the APSS processor as a platform device(AHB).
Because of this hybrid nature, it is called as a hybrid bus device.
As such, QCN6122 is a hybrid bus type device and follows the same codepath
as for WCN6750.
This is a reversed engineered and heavily simplified version of below
downstream patch:
https://git.codelinaro.org/clo/qsdk/oss/system/feeds/wlan-open/-/ \
blob/NHSS.QSDK.12.4.5.r2/mac80211/patches/232-ath11k-qcn6122-support.patch
Co-developed-by: George Moussalem <george.moussalem@outlook.com>
Signed-off-by: Sowmiya Sree Elavalagan <ssreeela@codeaurora.org>
Signed-off-by: George Moussalem <george.moussalem@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/17182
Signed-off-by: Robert Marko <robimarko@gmail.com>
Add support for the Qualcomm IPQ50xx in the QCA NSS dataplane driver.
The QCA implementation uses depracated DMA api calls and a downstream SCM
call, so convert to proper Linux DMA and SCM api calls.
In addition, add fixed-link support to support SGMII which is used to
connect the internal IPQ50xx switch to an external switch (ex. QCA8337)
Co-developed-by: Ziyang Huang <hzyitc@outlook.com>
Signed-off-by: Ziyang Huang <hzyitc@outlook.com>
Signed-off-by: George Moussalem <george.moussalem@outlook.com>
Link: https://github.com/openwrt/openwrt/pull/17182
Signed-off-by: Robert Marko <robimarko@gmail.com>
Uses upstream DSA switch modules (rtl8365mb, rtl8366), similar to
RTL8367C and rtl8366rb swconfig drivers.
The package dependencies exclude targets built without kernel CONFIG_OF.
It also fixes the rtl8366rb LED support.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/17182
Signed-off-by: Robert Marko <robimarko@gmail.com>
The three packages base-files, libc and kernel are special, the former
can't be upgraded in place since it's content are modified on startup,
the latter two are virtual packages only used as constraints for the
package manager.
Historically base-files was "locked" via a special OPKG function, the
latter two were hidden from the package index and thereby never picked
as possible upgrade.
Time moved forward and we now have APK and tools like OWUT. The latter
compares available packages with installed packages and generates user
readable output, requiring versions for libc and kernel, too. At the
same time, APK uses a different looking mechanism, which is set during
installation instead of part of the package metadata.
In short, this patch adds version constraints to the three packages,
allowing them to be part of the package index.
Fixes: #17774Fixes: #17775Fixes: efahl/owut#31
Signed-off-by: Paul Spooren <mail@aparcar.org>
atm_qos struct should be the same both for user and kernel spaces. Via
the __SO_ENCODE() macro it is used to define the SO_ATMQOS socket IOC.
During the VRX518 support introduction, the atm_trafprm sturct nested
into the atm_qos stucture was update with newer fields that are
referenced by the ATM TC layer of the VRX518 TC driver. These new fields
are intended to communicate information for extra traffic classes
supported by the driver. But we are still using vanilla kernel headers
to build the toolchain. Due to the atm.h header incoherency br2684ctl
from linux-atm tools is incapable to configure the ATM bridge netdev:
br2684ctl: Interface "dsl0" created sucessfully
br2684ctl: Communicating over ATM 0.1.2, encapsulation: LLC
br2684ctl: setsockopt SO_ATMQOS 22 <-- EINVAL errno
br2684ctl: Fatal: failed to connect on socket; File descriptor in bad state
There are two options to fix this incoherency. (a) update the header
file in the toolchain to build linux-atm against updated atm_trafprm and
atm_qos structures, or (b) revert atm_trafprm changes.
Since there are no actual users of the extra ATM QoS traffic classes,
just drop these extra traffic classes from vrx518_tc ATM TC layer and
drop the kernel patch updating atm.h.
Besides fixing the compatibility with linux-atm tools, removing the
kernel patch should simplify kernel updates removing unneeded burden of
maintenance.
Run tested with FRITZ!Box 7530 with disabled extra traffic classes and
then removed them entirely before the submission.
CC: John Crispin <john@phrozen.org>
Fixes: cfd42a0098 ("ipq40xx: add Intel/Lantiq ATM hacks")
Suggested-by: Andre Heider <a.heider@gmail.com>
Reported-and-tested-by: nebibigon93@yandex.ru
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250122222654.21833-4-ryazanov.s.a@gmail.com/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
ATM TC layer have some issues which effectively prevent VRX518 from
being used as ADSL modem. Specifically, there one crash during the ATM
layer configuration and wrong PVC ID selection on packet receiving what
breaks RX path. Fix both of the issues. Make subif iface registration
optional to prevent the crash (see more details in the new patch) and
update the hardcoded PVC ID to match the first allocated channel.
Run tested with FRITZ!Box 7530.
Fixes: 474bbe23b7 ("kernel: add Intel/Lantiq VRX518 TC driver")
Reported-and-tested-by: nebibigon93@yandex.ru
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250122222654.21833-3-ryazanov.s.a@gmail.com/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
It looks like VRX518 returns phys addr of data buffer in the 'data_ptr'
field of the RX descriptor and an actual data offset within the buffer
in the 'byte_off' field. In order to map the phys address back to
virtual we need the original phys address of the allocated buffer.
In the same driver applies offset to phys address before the mapping,
what leads to WARN_ON triggering in plat_mem_virt() function with
subsequent kernel panic:
WARNING: CPU: 0 PID: 0 at .../sw_plat.c:764 0xbf306cd0 [vrx518_tc@8af9f5d0+0x25000]
...
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = aff5701e
[00000000] *pgd=00000000
Internal error: Oops: 5 [#1] SMP ARM
Noticed in ATM mode, when chip always returns byte_off = 4.
In order to fix the issue, pass the phys address to plat_mem_virt() as
is and apply byte_off later for proper DMA syncing and on mapped virtual
address when copying RXed data into the skb.
Run tested with FRITZ!Box 7530 on both ADSL and VDSL (thanks Jan) links.
Fixes: 474bbe23b7 ("kernel: add Intel/Lantiq VRX518 TC driver")
Tested-by: Jan Hoffmann <jan@3e8.eu> # VDSL link
Reported-and-tested-by: nebibigon93@yandex.ru # ADSL link
Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
Link: https://patchwork.ozlabs.org/project/openwrt/patch/20250122222654.21833-2-ryazanov.s.a@gmail.com/
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
