it seems no problem with CPU-detection but there is a nullpointer dereference near mt2701_afe_pcm_driver_init
but this is new to me (only comes up in 4.9 and not from any patch)
do you use now 4.9-config from lede-repo?
i hoped that this kernel works better (but was only created for comparision and not tested), because there are not modifications by me…only the official patches from garys lede-repo
it seems that we need to wait for vendors help ;(
had added config (based on 4.9 from lede)+buildscript to 4.9-branch
can compile, but get the dtc warnings
DTC arch/arm/boot/dts/mt7623-evb.dtb
Warning (reg_format): "reg" property in /mt7530@0 has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
DTC arch/arm/boot/dts/mt7623-eMMC.dtb
Warning (reg_format): "reg" property in /switch has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
DTC arch/arm/boot/dts/mt7623-NAND.dtb
DTC arch/arm/boot/dts/mt7623-NAND-ePHY.dtb
DTC arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dtb
Warning (reg_format): "reg" property in /mt7530@0 has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
Warning (reg_format): "reg" property in /switch has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)same issue, glad you can reproduce it. Can you boot on kernel 4.9?
currently it stops after
Starting kernel ...
uart2 is enabled in dts and no other so that this is not the problem…
added pinmux for uart2, but still hang after loading kernel
after adding serial-drivers i get boot-output 
[ 2.884457] VFS: Cannot open root device "mmcblk0p2" or unknown-block(179,2)9
[ 2.892415] Please append a correct "root=" boot option; here are the availa:
[ 2.900766] b300 7634944 mmcblk0 [ 2.904585] driver: mmcblk
[ 2.907366] b318 512 mmcblk0rpmb [ 2.911513] (driver?)
[ 2.913861] b310 4096 mmcblk0boot1 [ 2.918121] (driver?)
[ 2.920471] b308 4096 mmcblk0boot0 [ 2.924717] (driver?)
[ 2.927066] b320 30375936 mmcblk1 [ 2.930869] driver: mmcblk
[ 2.933647] b321 262144 mmcblk1p1 000ed6cd-01[ 2.938764]
[ 2.940251] b322 7270400 mmcblk1p2 000ed6cd-02[ 2.945357]
[ 2.946844] Kernel panic - not syncing: VFS: Unable to mount root fs on unkn)
so i have to swap mmc0/1
hanging now on
/sbin/init: /lib/arm-linux-gnueabihf/libc.so.6: version `LIBPAM_1.0' not found )
/sbin/init: /lib/arm-linux-gnueabihf/libc.so.6: version `LIBKMOD_5' not found ()kernel 4.9 compiled and booting
had some trouble because “make mt7623n_evb_fwu_defconfig” did not import arch/arm/configs/mt7623n_evb_fwu_defconfig…now making manual copy and this works…
Linux bpi-r2 4.9.44-bpi-r2+ #153 SMP Mon Mar 19 19:05:03 CET 2018 armv7l
[18:54] root@bpi-r2:~# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group defau1
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP g0
link/ether 12:5a:b1:db:d9:c5 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qle0
link/ether 4a:b5:c9:5a:a5:7a brd ff:ff:ff:ff:ff:ff
i miss the lan/wan-Ports…maybe some additional driver is missing (had used my config from 4.14)…
added hnat and got compile error…fixed…now the section mismatch
WARNING: drivers/net/ethernet/mediatek/mtk_hnat/mtkhnat.o(.text+0x628): Section mismatch in reference from the function hnat_probe() to the function .init.text:hnat_init_debugfs()
The function hnat_probe() references
the function __init hnat_init_debugfs().
This is often because hnat_probe lacks a __init
annotation or the annotation of hnat_init_debugfs is wrong.Great so you are able to compile kernel 4.9 and boot! We should have all lede kernel feature. For port-separation, you are missing the DSA driver to see multiple lan0-3 nic. However is it need? All lan nic you should act as hardware swicth with gmac eth0, or i am miss-understanding?
Can you confirm, you have both hwnat, hwqos, vlan, enable?
As mention we need a kernel will all feature advertise in the website
mhm…added dsa-driver, but still no lan/wan (only eth0/eth1)
but hnat is loadable
[19:11] root@bpi-r2:~# modprobe mtkhnat
[19:11] root@bpi-r2:~# lsmod
Module Size Used by
mtkhnat 14617 0
NET_DSA_MT7530 seems to be needed…
The module for the dsa is call mt7530 in kernel 4.14.
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP g0
link/ether 22:ef:ec:35:57:71 brd ff:ff:ff:ff:ff:ff
inet6 fe80::20ef:ecff:fe35:5771/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qle0
link/ether 9e:55:02:d6:1f:dc brd ff:ff:ff:ff:ff:ff
4: wan@eth1: <BROADCAST,MULTICAST,M-DOWN> mtu 1500 qdisc noop state DOWN group 0
link/ether 9e:55:02:d6:1f:dc brd ff:ff:ff:ff:ff:ff
5: lan0@eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state 0
link/ether 08:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.11/24 brd 192.168.0.255 scope global lan0
valid_lft forever preferred_lft forever
6: lan1@eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state 0
link/ether 08:00:00:00:00:01 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.10/24 brd 192.168.1.255 scope global lan1
valid_lft forever preferred_lft forever
7: lan2@eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group defaul0
link/ether 22:ef:ec:35:57:71 brd ff:ff:ff:ff:ff:ff
8: lan3@eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group defaul0
link/ether 22:ef:ec:35:57:71 brd ff:ff:ff:ff:ff:ffthat is great! you made it! DSA is fine on 4.9 lede. I got a merge conflict while update your repo, have to re-download the all kernel. Can you test the vlan?
currently i have no time to make further tests…
I run some test on kernel 4.4.120 using your repo.
For clarification, without DSA driver you don’t see any of the LAN/WAN interface, however the system does act as a physical on eth0, so need to bridge. Or i am missing something?
I got the 2 gmac
[ 3.766296] mtk_eth_soc 1b100000.ethernet: generated random MAC address da:ce:d2:f9:b4:78
[ 3.774770] mtk_eth_soc 1b100000.ethernet: connected mac 0 to PHY at fixed-0:00 [uid=00000000, driver=Generic PHY]
[ 3.785654] mtk_eth_soc 1b100000.ethernet eth0: mediatek frame engine at 0xe32c0000, irq 71
[ 3.793998] mtk_eth_soc 1b100000.ethernet: generated random MAC address 3e:38:12:1d:2d:14
[ 3.802430] mtk_eth_soc 1b100000.ethernet: connected mac 1 to PHY at fixed-0:01 [uid=00000000, driver=Generic PHY]
[ 3.813229] mtk_eth_soc 1b100000.ethernet eth1: mediatek frame engine at 0xe32c0000, irq 71
also I found something not present in other kerne 49 and 4.14, MTK AHCI SATA support, AHCI_MTK However my SATA drive are detected but no device…
[ 1.730721] ata1: SATA max UDMA/133 abar m512@0x60000000 port 0x60000100 irq 63
[ 1.737996] ata2: SATA max UDMA/133 abar m512@0x60000000 port 0x60000180 irq 63
...
[ 2.310571] ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 2.311326] ata1.00: ATA-10: INTEL SSDSC2KW480H6, LSF036C, max UDMA/133
[ 2.311331] ata1.00: 937703088 sectors, multi 16: LBA48 NCQ (depth 31/32), AA
[ 2.311859] ata1.00: configured for UDMA/133
[ 2.326174] scsi 0:0:0:0: Direct-Access ATA INTEL SSDSC2KW48 036C PQ: 0 ANSI: 5
...
[ 2.870576] ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
[ 2.877458] ata2.00: ATA-10: INTEL SSDSC2KW480H6, LSF036C, max UDMA/133
[ 2.884135] ata2.00: 937703088 sectors, multi 16: LBA48 NCQ (depth 31/32), AA
[ 2.891747] ata2.00: configured for UDMA/133
[ 2.896486] scsi 1:0:0:0: Direct-Access ATA INTEL SSDSC2KW48 036C PQ: 0 ANSI: 5
/proc/interrupts
# cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
16: 14046 88889 102829 8222 GIC 29 Edge arch_timer
17: 0 0 0 0 GIC 30 Edge arch_timer
20: 0 0 0 0 MT_SYSIRQ 112 Level mtk_timer
23: 0 0 0 0 MT_SYSIRQ 88 Edge mtk-wdt
24: 0 0 0 0 MT_SYSIRQ 78 Level 1100a000.spi
25: 0 0 0 0 MT_SYSIRQ 115 Level mt-pmic-pwrap
26: 4 0 0 0 MT_SYSIRQ 182 Level hdmi hpd
27: 0 0 0 0 MT_SYSIRQ 106 Level 10205000.mmsys_iommu
31: 7323 0 0 0 MT_SYSIRQ 53 Level serial
35: 0 0 0 0 MT_SYSIRQ 44 Level i2c-mt65xx
36: 29 0 0 0 MT_SYSIRQ 45 Level i2c-mt65xx
37: 0 0 0 0 MT_SYSIRQ 46 Level i2c-mt65xx
42: 0 0 0 0 MT_SYSIRQ 32 Level musb-hdrc.1.auto
43: 0 0 0 0 MT_SYSIRQ 104 Level afe-isr
44: 0 0 0 0 MT_SYSIRQ 132 Level asys-isr
45: 52 0 0 0 MT_SYSIRQ 40 Level 11240000.mmc
46: 7483 0 0 0 MT_SYSIRQ 39 Level 11230000.mmc
47: 0 0 0 0 MT_SYSIRQ 170 Level Mali_GP
48: 0 0 0 0 MT_SYSIRQ 171 Level Mali_GP_MMU
49: 0 0 0 0 MT_SYSIRQ 172 Level Mali_PP0
50: 0 0 0 0 MT_SYSIRQ 173 Level Mali_PP0_MMU
51: 0 0 0 0 MT_SYSIRQ 174 Level Mali_PP1
52: 0 0 0 0 MT_SYSIRQ 175 Level Mali_PP1_MMU
53: 0 0 0 0 MT_SYSIRQ 176 Level Mali_PP2
54: 0 0 0 0 MT_SYSIRQ 177 Level Mali_PP2_MMU
55: 0 0 0 0 MT_SYSIRQ 180 Level Mali_PP_Broadcast
56: 0 0 0 0 MT_SYSIRQ 153 Level 14007000.ovl
57: 0 0 0 0 MT_SYSIRQ 152 Level 14008000.rdma
62: 0 0 0 0 MT_SYSIRQ 164 Level 14012000.rdma
63: 18 0 0 0 MT_SYSIRQ 194 Level 0000:01:00.0
69: 0 0 0 0 MT_SYSIRQ 196 Level xhci-hcd:usb1
70: 0 0 0 0 MT_SYSIRQ 197 Level xhci-hcd:usb3
72: 2216 0 0 0 MT_SYSIRQ 199 Level 1b100000.ethernet
73: 4385 0 0 0 MT_SYSIRQ 198 Level 1b100000.ethernet
74: 0 0 0 0 MT_SYSIRQ 82 Level mtk-aes
75: 0 0 0 0 MT_SYSIRQ 83 Level mtk-aes
76: 0 0 0 0 MT_SYSIRQ 84 Level mtk-sha
77: 0 0 0 0 MT_SYSIRQ 91 Level mtk-sha
79: 0 0 0 0 MT_SYSIRQ 87 Level mtk-ir
97: 10 0 0 0 mt-eint 17 Edge gsw
106: 0 0 0 0 mt-eint 26 Level USB_IDDIG
226: 0 0 0 0 mt-eint 146 Edge 11240000.mmc cd
230: 0 0 0 0 mt-eint 150 Level mt6397-pmic
IPI0: 0 1 1 1 CPU wakeup interrupts
IPI1: 0 0 0 0 Timer broadcast interrupts
IPI2: 4417 4267 1491 1578 Rescheduling interrupts
IPI3: 5 6 5 5 Function call interrupts
IPI4: 3 0 0 2 Single function call interrupts
IPI5: 0 0 0 0 CPU stop interrupts
IPI6: 7 1 0 0 IRQ work interrupts
IPI7: 0 0 0 0 completion interrupts
Err: 0
Re-compile kernel 4.9 form your repository, and i still can’t boot, kernel panic. Seem like my config is wrong.
[ 1.315091] Unable to handle kernel NULL pointer dereference at virtual address 00000000
[ 1.323174] pgd = c0004000
[ 1.325880] [00000000] *pgd=00000000
[ 1.329445] Internal error: Oops: 5 [#1] SMP ARM
[ 1.334033] Modules linked in:
[ 1.337076] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.9.44-00022-g1cff8763e07a #1
[ 1.344688] Hardware name: Mediatek Cortex-A7 (Device Tree)
[ 1.350224] task: de8e8000 task.stack: de8e2000
[ 1.354733] PC is at clk_hw_round_rate+0x28/0x54
[ 1.359322] LR is at divider_round_rate+0x1b4/0x550
[ 1.364170] pc : [<c0412da4>] lr : [<c0418c1c>] psr: 60000013
[ 1.364170] sp : de8e3a68 ip : de8e3a98 fp : de8e3a94
[ 1.375578] r10: 00000000 r9 : de83c380 r8 : 00000000
[ 1.380769] r7 : 00000100 r6 : 02ee0000 r5 : 02ee0000 r4 : 00000000
[ 1.387253] r3 : de883d80 r2 : de8e3a74 r1 : de8e3a70 r0 : 00000000
[ 1.393739] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
[ 1.400827] Control: 10c5387d Table: 8000406a DAC: 00000051
[ 1.406535] Process swapper/0 (pid: 1, stack limit = 0xde8e2218)
[ 1.412502] Stack: (0xde8e3a68 to 0xde8e4000)
hi xbgmsharp
We will support Kernel 4.14 in next release, both 2GMAC and HDMI will supported. but no plan to support sound(I2S) yet.
Thanks Gary
Do you use my buil.sh to import config?
make xxx_defconfig does not work with 4.9…i have no clue why…if i cp the defconfig to .config it works
Thanks for the info, when is the next release? I don’t care much about the Kernel version, however i am looking to have all feature working properly. It can be 4.14, .4.9, 4.4 but need support for:
This time i try your config on kernel4.9, same compiling warning but it does boot
CC drivers/net/ethernet/mediatek/mtk_eth_soc.o
drivers/net/ethernet/mediatek/mtk_eth_soc.c: In function 'mtk_start_xmit':
drivers/net/ethernet/mediatek/mtk_eth_soc.c:677:9: warning: 'fport' may be used uninitialized in this function [-Wmaybe-uninitialized]
fport |= 0x4 << TX_DMA_FPORT_SHIFT;
drivers/net/ethernet/mediatek/mtk_eth_soc.c:668:26: note: 'fport' was declared here
u32 txd3 = 0, txd4 = 0, fport;
^~~~~
LD drivers/net/ethernet/mediatek/built-in.o
CC [M] drivers/net/ethernet/mediatek/mtk_hnat/hnat_nf_hook.o
LD [M] drivers/net/ethernet/mediatek/mtk_hnat/mtkhnat.o
WARNING: drivers/net/ethernet/mediatek/mtk_hnat/mtkhnat.o(.text+0x604): Section mismatch in reference from the function hnat_probe() to the function .init.text:hnat_init_debugfs()
The function hnat_probe() references
the function __init hnat_init_debugfs().
This is often because hnat_probe lacks a __init
annotation or the annotation of hnat_init_debugfs is wrong.
LD drivers/net/ethernet/built-in.o
I have a the 2Gmac, DSA, HWnat,
[ 1.143585] mtk_soc_eth 1b100000.ethernet: generated random MAC address 32:89:df:c5:ff:5b
[ 1.151752] of_get_named_gpiod_flags: can't parse 'link-gpios' property of node '/ethernet@1b100000/mac@0/fixed-link[0]'
[ 1.163244] mtk_soc_eth 1b100000.ethernet: connected mac 0 to PHY at fixed-0:00 [uid=00000000, driver=Generic PHY]
[ 1.174112] mtk_soc_eth 1b100000.ethernet eth0: mediatek frame engine at 0xe10a0000, irq 211
[ 1.182559] mtk_soc_eth 1b100000.ethernet: generated random MAC address 62:29:88:68:89:ef
[ 1.190717] of_get_named_gpiod_flags: can't parse 'link-gpios' property of node '/ethernet@1b100000/mac@1/fixed-link[0]'
[ 1.201860] mtk_soc_eth 1b100000.ethernet: connected mac 1 to PHY at fixed-0:01 [uid=00000000, driver=Generic PHY]
[ 1.212648] mtk_soc_eth 1b100000.ethernet eth1: mediatek frame engine at 0xe10a0000, irq 211
...
[ 2.401186] Generic PHY dsa-0.0:00: attached PHY driver [Generic PHY] (mii_bus:phy_addr=dsa-0.0:00, irq=-1)
[ 2.411564] Generic PHY dsa-0.0:01: attached PHY driver [Generic PHY] (mii_bus:phy_addr=dsa-0.0:01, irq=-1)
[ 2.421922] Generic PHY dsa-0.0:02: attached PHY driver [Generic PHY] (mii_bus:phy_addr=dsa-0.0:02, irq=-1)
[ 2.432279] Generic PHY dsa-0.0:03: attached PHY driver [Generic PHY] (mii_bus:phy_addr=dsa-0.0:03, irq=-1)
[ 2.442626] Generic PHY dsa-0.0:04: attached PHY driver [Generic PHY] (mii_bus:phy_addr=dsa-0.0:04, irq=-1)
The udev rule don’t assing a MAC by gmac as both as detected as KERNELS=="1b100000.ethernet"
I wonder how should the MAC be assign, same for lan0-3 and eth0 then a second MAC for eth1 and wan?
Should i bridge all lan0-3 or only eth0?
Also as i will vtag the wan traffic, should i use eth1 or WAN nic?
How to get the best network performance?
you only work with lanx/wan…don’t bridge/tag ethx
Manage to have kernel 4.9 with hwnat, second gmac, port-separation and vlan. There is some compilation warning but it does work. Will give feedback on performance.
OBJCOPY arch/arm/boot/zImage
Kernel: arch/arm/boot/zImage is ready
DTC arch/arm/boot/dts/mt2701-evb.dtb
DTC arch/arm/boot/dts/mt6580-evbp1.dtb
DTC arch/arm/boot/dts/mt6589-aquaris5.dtb
DTC arch/arm/boot/dts/mt6592-evb.dtb
DTC arch/arm/boot/dts/mt7623-evb.dtb
Warning (reg_format): "reg" property in /mt7530@0 has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
DTC arch/arm/boot/dts/mt7623-eMMC.dtb
Warning (reg_format): "reg" property in /switch has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
DTC arch/arm/boot/dts/mt7623-NAND.dtb
DTC arch/arm/boot/dts/mt7623-NAND-ePHY.dtb
DTC arch/arm/boot/dts/mt7623n-bananapi-bpi-r2.dtb
Warning (reg_format): "reg" property in /mt7530@0 has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
Warning (reg_format): "reg" property in /switch has invalid length (4 bytes) (#address-cells == 2, #size-cells == 2)
DTC arch/arm/boot/dts/mt8127-moose.dtb
DTC arch/arm/boot/dts/mt8135-evbp1.dtb
drivers/misc/mediatek/connectivity/wlan/Makefile:6: include gen2
drivers/misc/mediatek/connectivity/wlan/gen2/Makefile:220: y
Building modules, stage 2.
MODPOST 110 modules
WARNING: drivers/net/ethernet/mediatek/mtk_hnat/mtkhnat.o(.text+0x604): Section mismatch in reference from the function hnat_probe() to the function .init.text:hnat_init_debugfs()
The function hnat_probe() references
the function __init hnat_init_debugfs().
This is often because hnat_probe lacks a __init
annotation or the annotation of hnat_init_debugfs is wrong.