GL.iNET Flint 2 (GL-MT6000) discussions

ngl.

make -j$(nproc) V=s

is taking quite some time.

1 Like

yeah even with a i9-9900k it takes a while.

3 Likes

Results with test.zip

It'll take a minute for test2.zip

1 Like

test2.zip gives me errors.

  CC      net/dsa/tag.o
  CC      net/dsa/tag_8021q.o
  CC      net/socket.o
  CC      net/dsa/trace.o
  CC      net/dsa/tag_mtk.o
  CC      net/sysctl_net.o
  AR      net/dsa/built-in.a
  AR      net/built-in.a
make[6]: *** [/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30/Makefile:1913: .] Error 2
make[5]: *** [Makefile:234: __sub-make] Error 2
make[5]: Leaving directory '/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30'
make[4]: *** [Makefile:18: /home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30/.modules] Error 2
make[4]: Leaving directory '/home/user/patch2/openwrt-test2/target/linux/mediatek'
make[3]: *** [Makefile:11: compile] Error 2
make[3]: Leaving directory '/home/user/patch2/openwrt-test2/target/linux'
time: target/linux/compile#1509.52#313.76#949.29
    ERROR: target/linux failed to build.
make[2]: *** [target/Makefile:32: target/linux/compile] Error 1
make[2]: Leaving directory '/home/user/patch2/openwrt-test2'
make[1]: *** [target/Makefile:25: /home/user/patch2/openwrt-test2/staging_dir/target-aarch64_cortex-a53_musl/stamp/.target_compile] Error 2
make[1]: Leaving directory '/home/user/patch2/openwrt-test2'
make: *** [/home/user/patch2/openwrt-test2/include/toplevel.mk:233: world] Error 2
user@ubuntu:~/patch2/openwrt-test2$ df
Filesystem                        1K-blocks     Used Available Use% Mounted on
tmpfs                                401008     1200    399808   1% /run
/dev/mapper/ubuntu--vg-ubuntu--lv  50254368 29426328  18242852  62% /
tmpfs                               2005032        0   2005032   0% /dev/shm
tmpfs                                  5120        0      5120   0% /run/lock
/dev/sda2                           1992552    96548   1774764   6% /boot
tmpfs                                401004       12    400992   1% /run/user/1000

1 Like

make -j1 V=s

1 Like
make -C /home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30 KCFLAGS="-fmacro-prefix-map=/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl=target-aarch64_cortex-a53_musl -fno-caller-saves " HOSTCFLAGS="-O2 -I/home/user/patch2/openwrt-test2/staging_dir/host/include  -Wall -Wmissing-prototypes -Wstrict-prototypes" CROSS_COMPILE="aarch64-openwrt-linux-musl-" ARCH="arm64" KBUILD_HAVE_NLS=no KBUILD_BUILD_USER="" KBUILD_BUILD_HOST="" KBUILD_BUILD_TIMESTAMP="Mon May 13 18:04:36 2024" KBUILD_BUILD_VERSION="0" KBUILD_HOSTLDFLAGS="-L/home/user/patch2/openwrt-test2/staging_dir/host/lib" CONFIG_SHELL="bash" V=''  cmd_syscalls=  CC="aarch64-openwrt-linux-musl-gcc" KERNELRELEASE=6.6.30 Image dtbs modules
make[5]: Entering directory '/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30'
  CC      drivers/net/phy/realtek.o
drivers/net/phy/realtek.c: In function 'rtl8221b_vb_cg_match_phy_device':
drivers/net/phy/realtek.c:812:65: error: 'RTL_8221B_VB_CG' undeclared (first use in this function)
  812 |                         return phydev->c45_ids.device_ids[1] == RTL_8221B_VB_CG;
      |                                                                 ^~~~~~~~~~~~~~~
drivers/net/phy/realtek.c:812:65: note: each undeclared identifier is reported only once for each function it appears in
make[10]: *** [scripts/Makefile.build:243: drivers/net/phy/realtek.o] Error 1
make[9]: *** [scripts/Makefile.build:480: drivers/net/phy] Error 2
make[8]: *** [scripts/Makefile.build:480: drivers/net] Error 2
make[7]: *** [scripts/Makefile.build:480: drivers] Error 2
make[6]: *** [/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30/Makefile:1913: .] Error 2
make[5]: *** [Makefile:234: __sub-make] Error 2
make[5]: Leaving directory '/home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30'
make[4]: *** [Makefile:18: /home/user/patch2/openwrt-test2/build_dir/target-aarch64_cortex-a53_musl/linux-mediatek_filogic/linux-6.6.30/.modules] Error 2
make[4]: Leaving directory '/home/user/patch2/openwrt-test2/target/linux/mediatek'
make[3]: *** [Makefile:11: compile] Error 2
make[3]: Leaving directory '/home/user/patch2/openwrt-test2/target/linux'
time: target/linux/compile#3.02#2.93#6.36
    ERROR: target/linux failed to build.
make[2]: *** [target/Makefile:32: target/linux/compile] Error 1
make[2]: Leaving directory '/home/user/patch2/openwrt-test2'
make[1]: *** [target/Makefile:25: /home/user/patch2/openwrt-test2/staging_dir/target-aarch64_cortex-a53_musl/stamp/.target_compile] Error 2
make[1]: Leaving directory '/home/user/patch2/openwrt-test2'
make: *** [/home/user/patch2/openwrt-test2/include/toplevel.mk:233: world] Error 2
1 Like

@gbx Thank you very much for your help, the test2 compilation is no longer needed. https://github.com/openwrt/openwrt/pull/15453 just got accepted.

2 Likes

In layman's terms, could you explain what this patch does? Is it already included in the latest snapshot?

The tests involved https://github.com/openwrt/openwrt/pull/15453. There was an issue with realtek PHY detection on some routers. The test was to check whether by repairing one we did not damage the other. The fix was approved yesterday and future snapshots will include it.

1 Like

Thank you @mietekn and @daniel for your work!

Seems like this is already in the latest snapshots, which are based on the committ after this with hash eb1b022043.

https://firmware-selector.openwrt.org/?version=SNAPSHOT&target=mediatek%2Ffilogic&id=glinet_gl-mt6000

Edit: although attendedsysupgrade offered me a version based on 4f87a4d just prior to the patch until I added one package and forced a new build.

Seems asu is broken at the moment?

It worked fine after I added a package (tailscaled) that was already a dependency for one I had in the list (tailscale). It gave me a build from the latest commit then. I've seen that trick mentioned by others many times so it's worth a shot.

Edit: just upgraded using Luci Attendedsysupgrade, checked the build version and added in the one package as mentioned previously that triggered the new build version, seems to be running fine. Output to dmesg | grep mtk_soc_eth command from a few posts back almost identical, only two characters different, the b00000 became an e00000 in the first two lines, but I think that is unimportant.

  _______                     ________        __
 |       |.-----.-----.-----.|  |  |  |.----.|  |_
 |   -   ||  _  |  -__|     ||  |  |  ||   _||   _|
 |_______||   __|_____|__|__||________||__|  |____|
          |__| W I R E L E S S   F R E E D O M
 -----------------------------------------------------
 OpenWrt SNAPSHOT, r26305-eb1b022043
 -----------------------------------------------------
root@glint:~# dmesg | grep mtk_soc_eth
[    1.892102] mtk_soc_eth 15100000.ethernet eth0: mediatek frame engine at 0xffffffc081e00000, irq 119
[    1.902030] mtk_soc_eth 15100000.ethernet eth1: mediatek frame engine at 0xffffffc081e00000, irq 119
1 Like

I'd wager a guess that's just a memory address location. If it is, It'd be unlikely you'd get the same address as the software is developed.

Has it been long enough that the patch I tested yesterday is in the daily snap?

1 Like

Yes. See my previous posts. The snapshots are now one commit ahead of the two patches that seem to be related that mention Realtek PHYs, which were not being identified properly by kernel 6.6 that is found in snapshots now. This previously had to be fixed in kernel 6.1, but the fix had to be modified slightly for kernal 6.6

2 Likes

Will all of these fixes make it into v24?

1 Like

OpenWrt stable releases are branched from snapshot. In other words, if it is in snapshot when the next stable release is branched off, it will be in the next stable release.

2 Likes

Recommend that you use a snapshot. Very rarely is there some instability introduced.

3 Likes

This. Had a 30-day uptime on a kernel 6.1 snapshot, now 7-day uptime on a new kernel 6.6 snapshot. Even with tons of added packages it's nearly flawless. This target works so well it's almost boring.

6 Likes

Hi everyone, just received this router and am trying to download a snapshot image. I requested a build with the following packages:

base-files busybox ca-bundle dnsmasq dropbear e2fsprogs f2fsck firewall4 fitblk fstools kmod-crypto-hw-safexcel kmod-gpio-button-hotplug kmod-leds-gpio kmod-mt7915e kmod-mt7986-firmware kmod-nft-offload kmod-phy-aquantia kmod-usb3 libc libgcc libustream-mbedtls logd luci mkf2fs mt7986-wo-firmware mtd netifd nftables odhcp6c odhcpd-ipv6only opkg ppp ppp-mod-pppoe procd procd-seccomp procd-ujail uboot-envtools uci uclient-fetch urandom-seed urngd wpad-basic-mbedtls luci uci-app-attendedsysupgrade auc luci-app-statistics collectd-mod-cpufreq collectd-mod-thermal htop irqbalance luci-app-https-dns-proxy luci-app-banip

and get the error

Collected errors:
 * opkg_install_cmd: Cannot install package uci-app-attendedsysupgrade.
make[2]: *** [Makefile:189: package_install] Error 255
make[1]: *** [Makefile:154: _call_manifest] Error 2
make: *** [Makefile:279: manifest] Error 2

is the uci-app-attendedsysupgrade package not working anymore? Or am I doing something wrong?

You have the wrong name...
luci-app-attendedsysupgrade

1 Like