No. Installing those packages has nothing to do with the kernel. Those packages has to do with installing firmware only.
Do this to install the firmware and reboot and see if you have wifi. You will need to rename the firmware you put in /usr/lib/firmware/brcm first.
sudo pacman -Syyu Kvim2-firmware linux-firmware
Noted. Will try this much later and revert.
One thing that seems easy to do in Manjaro is testing out new kernel. Not sure whether other linux distros like Ubuntu/Debian is this easy.
Out now and responding via ipad.
Edit: In the case for GT KING PRO, each update is a touch of fate, never know whether kernel panic will strike. If kernel panic strike during writing to the disk the USB disk can be corrupted in worst case require to re-burn image. But so far with linux-vim kernel 5.9.0-1 and 5.9.0-2, it was stable enough to perform updates successfully but with large updates I split the updates using Package Manager into two or three times to avoid kernel panic.
Will test out linux kernel 5.9.6-1.
“Just pushed 5.9.8 to unstable and tried mesa-git-21.0 on it, panfrost is much smooth and no font or color issue but it still seem to get stuck after few mins which means we need 5.10 to get stable panfrost”
On linux kernel 5.9.6-1. Checking whether is stable with Panfrost. Mesa 20.2.1 is already pre-installed does it mean Panfrost should be running?
Hi Darksky,
Just performed “sudo pacman -Syyu Kvim2-firmware linux-firmware”. Error message “target not found: Kvim2-firmware”
[jfl@GTKPro ~]$ sudo pacman -Syyu Kvim2-firmware linux-firmware
[sudo] password for jfl:
:: Synchronizing package databases…
core 225.1 KiB 341 KiB/s 00:01 [######################] 100%
extra 2.4 MiB 811 KiB/s 00:03 [######################] 100%
community 5.5 MiB 496 KiB/s 00:11 [######################] 100%
error: target not found: Kvim2-firmware
warning: linux-firmware-20201023.dae4b4c-1 is up to date – reinstalling
[jfl@GTKPro ~]$
Should be a small k. Not capital K.
In order to get a clearer history for my ongoing testing I have now built a new image (with manjaro-arm-tools) to use for testing on my Ugoos (s922x-j). The files can be found here
image, pkg-list and build-log.
With the new image I had to follow my previous instructions earlier in the thread to get everything working, but now also with two additions: In the latest image two important system services are not enabled by default (sound and Network Manager). These can be enabled with;
sudo systemctl enable NetworkManager.service --now
sudo systemctl enable sound.service --now
(possibly a reboot is needed for the sound to work, I don’t remember). (One might also want to do the same for the sshd.service since it is not running by default.)
Strit,
Thanks. sudo pacman -Syyu kvim2-firmware linux-firmware works after renaming all the 4356 file from /usr/lib/firmware/brcm/ that I had copied there before.
Ethernet and Wifi works with the kvim2-firmware. Just no sound (as I am on linux kernel 5.9.6-1 instead of linux-vim kernel 5.9.0-2). Thanks.
But according to dmesg output there are more “warning” red text compare to my earlier “brcmfmac4356-sdio.bin”.
[ 8.593942] [CRTC:42:meson_crtc] vblank wait timed out
[ 8.594925] WARNING: CPU: 3 PID: 445 at drivers/gpu/drm/drm_atomic_helper.c:1496 drm_atomic_helper_wait_for_vblanks.part.0+0x280/0x2a0 [drm_kms_helper]
[ 8.594931] Modules linked in: snd_soc_hdmi_codec rc_cec dw_hdmi_i2s_audio dw_hdmi_cec btsdio brcmfmac brcmutil cfg80211 hci_uart btqca btrtl btbcm btintel bluetooth meson_dw_hdmi dw_hdmi meson_vdec(C) panfrost cec meson_gxl meson_drm gpu_sched ecdh_generic ecc drm_kms_helper v4l2_mem2mem rfkill videobuf2_dma_contig videobuf2_memops videobuf2_v4l2 snd_soc_meson_axg_frddr snd_soc_meson_axg_sound_card snd_soc_meson_card_utils syscopyarea sysfillrect sysimgblt videobuf2_common meson_ir dwmac_generic snd_soc_meson_g12a_tohdmitx realtek snd_soc_meson_axg_fifo fb_sys_fops meson_canvas snd_soc_meson_codec_glue snd_soc_meson_axg_tdmout rc_core pwm_meson dwmac_meson8b rtc_meson_vrtc stmmac_platform stmmac mdio_xpcs phylink mdio_mux_meson_g12a mdio_mux snd_soc_meson_axg_tdm_interface snd_soc_meson_axg_tdm_formatter display_connector nvmem_meson_efuse drm drm_panel_orientation_quirks crypto_user gpio_keys_polled
[ 8.595100] CPU: 3 PID: 445 Comm: kworker/3:3 Tainted: G C 5.9.6-1-MANJARO-ARM #1
[ 8.595104] Hardware name: Beelink GT-King Pro (DT)
[ 8.595153] Workqueue: events splash_callback_animation
[ 8.595166] pstate: 80000005 (Nzcv daif -PAN -UAO BTYPE=--)
[ 8.595207] pc : drm_atomic_helper_wait_for_vblanks.part.0+0x280/0x2a0 [drm_kms_helper]
[ 8.595234] lr : drm_atomic_helper_wait_for_vblanks.part.0+0x280/0x2a0 [drm_kms_helper]
[ 8.595237] sp : ffff80001076b940
[ 8.595239] x29: ffff80001076b940 x28: 000000000000001b
[ 8.595243] x27: 0000000000000000 x26: 0000000000000001
[ 8.595246] x25: 0000000000000038 x24: 0000000000000000
[ 8.595250] x23: ffff0000cf814000 x22: 0000000000000001
[ 8.595253] x21: ffff0000d692b880 x20: ffff0000d7144400
[ 8.595255] x19: 0000000000000000 x18: 0000000000000020
[ 8.595258] x17: 0000000000000000 x16: 0000000000000000
[ 8.595261] x15: ffff800012dc3000 x14: ffff800012fa98d0
[ 8.595263] x13: 0000000000000000 x12: ffff800012fa8000
[ 8.595266] x11: ffff800012dc3000 x10: ffff800012fa8ec0
[ 8.595268] x9 : 0000000000000000 x8 : ffff800012fb70ec
[ 8.595270] x7 : 0000000000000000 x6 : 0000000000000001
[ 8.595272] x5 : ffff0000d8fc0190 x4 : 0000000000000000
[ 8.595275] x3 : 0000000000000027 x2 : 0000000000000023
[ 8.595277] x1 : 192c61a95afdc400 x0 : 0000000000000000
[ 8.595283] Call trace:
[ 8.595315] drm_atomic_helper_wait_for_vblanks.part.0+0x280/0x2a0 [drm_kms_helper]
[ 8.595342] drm_atomic_helper_commit_tail_rpm+0x78/0x9c [drm_kms_helper]
[ 8.595366] commit_tail+0xa4/0x184 [drm_kms_helper]
[ 8.595392] drm_atomic_helper_commit+0x160/0x410 [drm_kms_helper]
[ 8.596114] drm_atomic_commit+0x4c/0x5c [drm]
[ 8.596181] drm_client_modeset_commit_atomic+0x208/0x260 [drm]
[ 8.596239] drm_client_modeset_commit_locked+0x5c/0xac [drm]
[ 8.596265] drm_fb_helper_pan_display+0xbc/0x1d0 [drm_kms_helper]
[ 8.596292] fb_pan_display+0x9c/0x114
[ 8.596304] dummy_update_start+0x20/0x50
[ 8.596307] fbcon_switch+0x400/0x5c0
[ 8.596325] redraw_screen+0x140/0x280
[ 8.596330] splash_callback_animation+0x80/0x90
[ 8.596357] process_one_work+0x1c4/0x390
[ 8.596360] worker_thread+0x170/0x3f0
[ 8.596368] kthread+0x10c/0x110
[ 8.596392] ret_from_fork+0x10/0x34
[ 8.596398] ---[ end trace 9ce51baafd59390f ]---
[ 10.473883] Bluetooth: hci0: command 0xfc4c tx timeout
[ 12.195218] audit: type=1130 audit(12.330:9): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=lvm2-monitor comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.237060] FAT-fs (sda1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.
[ 12.331516] audit: type=1130 audit(12.460:10): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-tmpfiles-setup comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.358015] audit: type=1127 audit(12.490:11): pid=698 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg=' comm="systemd-update-utmp" exe="/usr/lib/systemd/systemd-update-utmp" hostname=? addr=? terminal=? res=success'
[ 12.369398] audit: type=1130 audit(12.500:12): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-update-utmp comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.401149] audit: type=1130 audit(12.530:13): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-journal-catalog-update comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.526919] audit: type=1130 audit(12.660:14): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=ldconfig comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.551042] audit: type=1130 audit(12.680:15): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-update-done comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.724472] audit: type=1130 audit(1605619310.255:16): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-timesyncd comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.729973] audit: type=1130 audit(1605619310.255:17): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=bluetooth-khadas comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 12.735416] audit: type=1130 audit(1605619310.265:18): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=dbus comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 13.785634] meson8b-dwmac ff3f0000.ethernet eth0: PHY [0.0:00] driver [RTL8211F Gigabit Ethernet] (irq=29)
[ 13.789367] meson8b-dwmac ff3f0000.ethernet eth0: No Safety Features support found
[ 13.789377] meson8b-dwmac ff3f0000.ethernet eth0: PTP not supported by HW
[ 13.789652] meson8b-dwmac ff3f0000.ethernet eth0: configuring for phy/rgmii link mode
[ 17.274013] rc rc0: two consecutive events of type space
[ 17.712720] meson8b-dwmac ff3f0000.ethernet eth0: Link is Up - 1Gbps/Full - flow control rx/tx
[ 17.712768] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 17.920732] kauditd_printk_skb: 33 callbacks suppressed
[ 17.920735] audit: type=1130 audit(1605619315.445:52): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-networkd-wait-online comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 17.998538] audit: type=1130 audit(1605619315.525:53): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=NetworkManager-wait-online comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 18.647936] audit: type=1130 audit(1605619417.134:54): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=nmb comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 19.033953] Bluetooth: hci0: BCM: Patch command fc4c failed (-110)
[ 19.033967] Bluetooth: hci0: BCM: Patch failed (-110)
[ 19.149125] audit: type=1130 audit(1605619417.634:55): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=smb comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 19.899869] audit: type=1131 audit(1605619418.384:56): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=systemd-rfkill comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 21.114153] Bluetooth: hci0: command 0xfc18 tx timeout
[ 24.046837] audit: type=1131 audit(1605619422.534:57): pid=1 uid=0 auid=4294967295 ses=4294967295 subj==unconfined msg='unit=NetworkManager-dispatcher comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 29.274010] Bluetooth: hci0: BCM: failed to write update baudrate (-110)
[ 31.354080] Bluetooth: hci0: command 0x0c03 tx timeout
[ 39.514063] Bluetooth: hci0: BCM: Reset failed (-110)
Darksky,
Ethernet and Wifi works with “sudo pacman -Syyu kvim2-firmware linux-firmware” instruction. Thanks.
Same when I installed Manjaro-ARM-xfce-vim3-20.11.img_xz, NetworkManger.service and sound.service not enabled.
I have tried latest Librelec on my Tv box and Panfrost is enabled in Kodi ¿that means that they have a cheated old kernel? cool !
On the other side first freeze 
and no sound
Yes even I had the same issue when I created the new image and I have found the issue and fixed it.
For those making their own images please install the latest Manjaro-Arm-Tools package and update arm-profiles.
On linux-vim 5.9.8 now. Panfrost up and running without freezing, with mesa-git 21.0 .Video plays much better now.
Problem is that I’ve made changes with a composite TV and now XCFE refuses to lunch and display goes off (or is just panfrost?)Thanks that have kodi installed and can run it instead.
Looks good, BRAVO !!!
PD I need help to recover XCFE 
Editing: it’s not that XCFE refuses to lunch, OS is running but display can not manage composite resolution. How can I change hdmi in composite mode? Display settings only allows to change active display.
Update after 24h on my little experiment starting with a freshly made image:
With the new image (using vim kernel 5.9.0-2), the system is much more stable. I have only had one freeze during the last 24h (listening to a podcast on youtube in Firefox, sound entering into short loop and frozen screen with unresponsive box). During this time I have watched a number of movies and played music and youtube videos.
I have tried to keep the install as vanilla as possible (using only packages/files from Manjaro and as little editing of configutation files (like for mpv) as possible).
With the old image (20.06 upgraded to 20.10) I had frequent freezes, but always during sound playback (several times with youtube videos on Chrome, several times with videos played with mpv and once only playing music with Audacious).
The glitches on the screen are still there, but only when the desktop (xfce) is visible, never during full screen video playback. (The glitches are a only minor issue, I think.)
@spikerguy This looks very promising, perhaps my particular freezes on Ugoos are not due to the kernel, per se, but are somehow sound related (or the interaction btw sound and kernel)? We’ll see, I guess…
I asked myself why linux-vim kernel is unstable when using panfrost while linux kernel is stable. So I thought what is fundamentally different and obviously sound is working with linux-vim while it is not working with linux.
To see if system hanging when doing graphically demanding stuff might be related to sound card/kernel interactions I have prepared a .dtb where I deleted everything that was related to audio/sound.
My box booted and ran completely normal with that .dtb but when I started Supertuxkart and loaded ‘Ravenscroft Manor’ it became unresponsive in the starting sequence. Same behavior as using my normal .dtb file with sound working.
@spikerguy I think we might be able to conclude from that result that instabilities when using panfrost on linux-vim are not caused by interactions of kernel with sound system.
@tripole you might do a similar test to see if your problems on your box are really related to sound.
With the new image the freezes are rare and the only useful/interesting things I can do with the (TV box Ugoos) is really, well, to listen to music or view movies (with sound, using speakers or Bluethooth). I have never experienced a freeze when just surfing the web (which I also do on the box, but not so much) or downloading files, etc. (in particular never during upgrades). Still, I’ll think about it…
Update after 48h of my experiment with freshly minted Manjaro ARM image (for my Ugoos).
Ths system had been stable for over 20h (watching lots of videos, movies, listening to music etc) when this morning I got one freeze, and shortly after rebooting another one. A while after the second reboot I was watching some youtube video when I started getting into one of those short, stuttering sound loops that normally would render the system unresponsive. This time however, I could open a terminal and salvage dmesg. Indeed, there is a kernel Ops there:
[13551.579325] Unable to handle kernel paging request at virtual address ffff80001b93865c
[13551.581843] Mem abort info:
[13551.584600] ESR = 0x96000047
[13551.587643] EC = 0x25: DABT (current EL), IL = 32 bits
[13551.592901] SET = 0, FnV = 0
[13551.595915] EA = 0, S1PTW = 0
[13551.598974] Data abort info:
[13551.601880] ISV = 0, ISS = 0x00000047
[13551.605647] CM = 0, WnR = 1
[13551.608689] swapper pgtable: 4k pages, 48-bit VAs, pgdp=000000000278a000
[13551.615227] [ffff80001b93865c] pgd=00000000dd7ff003, p4d=00000000dd7ff003, pud=00000000dd7fe003, pmd=00000000744c9003, pte=0000000000000000
[13551.627684] Internal error: Oops: 96000047 [#1] PREEMPT SMP
[13551.633141] Modules linked in: uinput rfcomm fuse algif_hash algif_skcipher af_alg bnep hid_logitech_hidpp joydev hid_logitech_dj nls_ascii uas btsdio brcmfmac brcmutil hci_uart meson_vdec© cfg80211 btqca btrtl btbcm v4l2_mem2mem btintel videobuf2_dma_contig videobuf2_memops bluetooth videobuf2_v4l2 videobuf2_common videodev ecdh_generic ir_nec_decoder dw_hdmi_cec rc_khadas ecc rfkill mc meson_rng meson_ir rc_core rtc_meson_vrtc snd_soc_meson_g12a_tohdmitx snd_soc_meson_axg_spdifout mdio_mux_meson_g12a snd_soc_meson_axg_sound_card sch_fq_codel crypto_user ip_tables x_tables
[13551.683910] CPU: 2 PID: 19310 Comm: kworker/2:2 Tainted: G C 5.9.0-2-MANJARO-ARM #1
[13551.692672] Hardware name: Ugoos AM6 (DT)
[13551.696871] Workqueue: events dbs_work_handler
[13551.701128] pstate: 60000085 (nZCv daIf -PAN -UAO BTYPE=–)
[13551.706678] pc : regmap_update_bits_base+0x5c/0xa0
[13551.711360] lr : regmap_update_bits_base+0x6c/0xa0
[13551.716033] sp : ffff80001b933a60
[13551.719329] x29: ffff80001b933a60 x28: ffff0000d5e98200
[13551.724580] x27: ffff0000d5e98a00 x26: ffff0000d5e98f00
[13551.729832] x25: 0000000000000000 x24: 0000000000000000
[13551.735089] x23: 0000000000000000 x22: 0000000000000030
[13551.740354] x21: 00000000040003f0 x20: 0000000000000208
[13551.745613] x19: ffff0000d7ffa400 x18: 0000001dcd64f880
[13551.750871] x17: 0000000000000000 x16: 0000000000000000
[13551.756132] x15: 0000000000000000 x14: 0000000000b71b00
[13551.761397] x13: 00000000016e3600 x12: 0000000165a0bc00
[13551.766651] x11: 0000000000000000 x10: 0000000000000001
[13551.771921] x9 : ffff80001b933978 x8 : 0000000000000006
[13551.777180] x7 : 000000000000001a x6 : 0000000000000000
[13551.782435] x5 : 0000000000000000 x4 : 0000000000000000
[13551.787698] x3 : 0000000000000030 x2 : 0000000000000001
[13551.792959] x1 : 0000000000000000 x0 : ffff0000d7ffa400
[13551.798232] Call trace:
[13551.800758] regmap_update_bits_base+0x5c/0xa0
[13551.805157] meson_clk_cpu_dyndiv_set_rate+0xf4/0x120
[13551.810151] clk_change_rate+0x148/0x2bc
[13551.814013] clk_change_rate+0xf4/0x2bc
[13551.817814] clk_core_set_rate_nolock+0x1bc/0x274
[13551.822460] clk_set_rate+0x38/0x150
[13551.826020] dev_pm_opp_set_rate+0x314/0x770
[13551.830225] set_target+0x44/0x90
[13551.833542] __cpufreq_driver_target+0x1dc/0x644
[13551.838053] od_dbs_update+0xc0/0x1a4
[13551.841677] dbs_work_handler+0x40/0x80
[13551.845489] process_one_work+0x1cc/0x350
[13551.849429] worker_thread+0x13c/0x470
[13551.853165] kthread+0x154/0x160
[13551.856359] ret_from_fork+0x10/0x30
[13551.860013] Code: 2a1603e3 2a1803e5 2a1503e2 2a1403e1 (aa1303e0)
[13551.865921] —[ end trace 1209fb87572b5546 ]—
[13551.870487] note: kworker/2:2[19310] exited with preempt_count 1
To better isolate the problem I have now created (carefully) a Franken-image with boot partition and kernel (5.9.0-rc2) from Khadas, and everything else (i.e. the root partion) from my recent Manjaro image (except, of course, I had to copy the module tree from 5.9.0-rc2 into the Manjaro root partition file hierarchy).
So now I’m running a Manjaro system but with Khadas kernel. So far, so good. To be continued…
Edit: One immediate observation is that the graphics glitches (when the desktop is visible) are still here and therefore they are not due to the kernel, in particular not due to the Manjaro linux-vim kernel. (Perhaps this was already known.)
I have tried Debian/VIM3_Debian-server-buster_Linux-5.9-rc2_arm64_SD-USB_V0.9.3-200907.img.xz on GTKPro before and it was not that stable on GTKPro but it could be different on Ugoos AM6.
@JFL This is a very interesting piece of information. It is consistent with what @spikerguy has said about some incompatibility of Beelink’s version of the bootloader (the ROM part of it, I guess) with the mainline boot process. I have previously used a slightly newer (homebuilt) version of the Khadas image with very good result (completely stable) on my Ugoos. So, even though the differences between the boxes seem small in terms of specs, they are important.
Edit: Clarification about the type of the incompatibility.
@tripole @JFL And @TheMojoManthemo
Thank you for testing different variants of kernel and panfrost and audio packages.
It helps us with more information about the issues which we all seem to face on amlogic devices.
Highly appreciate the time and effort you put in testing and supporting Manjaro arm development.
Cheers.