Posts by offbeat

offbeat · Allwinner H6 HEVC? offbeat February 20, 2022 at 5:03 PM

Quick test of iommu commit:

1) Kodi starts with a few drm_gem errors, but keeps running

Code

[   17.546100] ------------[ cut here ]------------
[   17.546117] WARNING: CPU: 3 PID: 738 at drivers/gpu/drm/drm_gem.c:1054 drm_gem_mmap_obj+0x158/0x170
[   17.546140] Modules linked in: 8021q garp stp mrp llc rfkill dwmac_sun8i hantro_vpu(C) sunxi_cir crct10dif_ce v4l2_vp9 v4l2_h264 sun50i_di panfrost gpu_sched videobuf2_vmalloc fuse
[   17.546181] CPU: 3 PID: 738 Comm: kodi.bin Tainted: G         C        5.16.5 #1
[   17.546186] Hardware name: Beelink GS1 (DT)
[   17.546190] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   17.546196] pc : drm_gem_mmap_obj+0x158/0x170
[   17.546200] lr : drm_gem_mmap_obj+0x78/0x170
[   17.546204] sp : ffff800009d03bb0
[   17.546207] x29: ffff800009d03bb0 x28: ffff000002250a80 x27: 0000000000000008
[   17.546214] x26: 00000000ea902000 x25: ffff000006313730 x24: ffff0000061cdf18
[   17.546222] x23: ffff000003516a00 x22: ffff000006dc6800 x21: 0000000000000000
[   17.546229] x20: ffff000006313730 x19: ffff0000061db000 x18: 0000000000000000
[   17.546237] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000002
[   17.546244] x14: 0000000000069f42 x13: 0000000000000001 x12: 0000000000000000
[   17.546251] x11: 00000000ea902000 x10: 00000000eb0eb000 x9 : 00000000000007e9
[   17.546259] x8 : 0000000000000001 x7 : 0000000000000012 x6 : ffff800076950000
[   17.546266] x5 : ffff800009d03990 x4 : ffff8000095274c0 x3 : 0000000000000000
[   17.546273] x2 : 0000000000000000 x1 : ffff000003443500 x0 : 00000000100000fb
[   17.546281] Call trace:
[   17.546284]  drm_gem_mmap_obj+0x158/0x170
[   17.546289]  drm_gem_mmap+0x12c/0x200
[   17.546293]  mmap_region+0x388/0x574
[   17.546300]  do_mmap+0x2c8/0x49c
[   17.546304]  vm_mmap_pgoff+0xd8/0x100
[   17.546312]  ksys_mmap_pgoff+0x194/0x1d4
[   17.546316]  __arm64_compat_sys_aarch32_mmap2+0x1c/0x24
[   17.546324]  invoke_syscall+0x44/0x110
[   17.546329]  el0_svc_common.constprop.0+0x40/0xe0
[   17.546334]  do_el0_svc_compat+0x18/0x30
[   17.546339]  el0_svc_compat+0x14/0x4c
[   17.546348]  el0t_32_sync_handler+0x88/0x11c
[   17.546354]  el0t_32_sync+0x15c/0x160
[   17.546360] ---[ end trace daa0f7c2d1a9e063 ]---
[   19.070650] ------------[ cut here ]------------

Display More

2) Some HEVC videos give iommu faults, requiring reboot. Uploaded F1.x265.2160p50.ts sample to GDrive at the link above.

Code

[  111.890421] sun50i-iommu 30f0000.iommu: Page fault for 0x000000000000e000 (master 1, dir rd)
[  111.890445] IOMMU: Level 1 page error
[  113.890831] cedrus 1c0e000.video-codec: frame processing timed out!

3) VP9 10bit videos are failing with iommu fault as well

Code

[  363.631608] sun50i-iommu 30f0000.iommu: Page fault for 0x00000000f17ff000 (master 5, dir rd)
[  363.631651] IOMMU: Level 2 page error
[  365.661445] hantro_watchdog:119: frame processing timed out!
[  367.677482] hantro_watchdog:119: frame processing timed out!
[  369.697446] hantro_watchdog:119: frame processing timed out!
[  371.713483] hantro_watchdog:119: frame processing timed out!
[  373.725456] hantro_watchdog:119: frame processing timed out!
[  375.741441] hantro_watchdog:119: frame processing timed out!
[  375.743783] sun50i-iommu 30f0000.iommu: TLB Flush timed out!
[  375.746577] sun50i-iommu 30f0000.iommu: TLB Flush timed out!
[  375.749433] sun50i-iommu 30f0000.iommu: TLB Flush timed out!
[  375.752228] sun50i-iommu 30f0000.iommu: TLB Flush timed out!
[  375.754967] sun50i-iommu 30f0000.iommu: TLB Flush timed out!
[  375.757727] sun50i-iommu 30f0000.iommu: TLB Flush timed out!

Display More

offbeat · Allwinner H6 HEVC? offbeat February 19, 2022 at 11:25 PM

Quote from jernej

if you have video sample which causes out of CMA memory issue, I would appreciate if you can provide it to me.

Samples – Google Drive

Two HEVC 4K@60 samples, one seems to be 8-bit, another is 10-bit. Both captured from NASA UHD TV channel, free-to-air on Hotbird 13E satellite.

H6 can't play any of these with default CMA settings, partially decodes first frame, green screen rest, kernel log full of cma errors. No more errors after booting with cma=1024M, but playback isn't smooth. Also, for some reason Kodi switches from 60p to 30p output after a second of playback.

I'm outputing to 1080p60 TV, so there is some downscaling involved as well.

Code

[   51.564148] cma: cma_alloc: reserved: alloc failed, req-size: 3038 pages, ret: -12
[   51.564222] ------------[ cut here ]------------
[   51.564226] WARNING: CPU: 3 PID: 856 at mm/page_alloc.c:5346 __alloc_pages+0x624/0xc3c
[   51.564254] Modules linked in: 8021q garp stp mrp llc rfkill hantro_vpu(C) v4l2_vp9 dwmac_sun8i sunxi_cir v4l2_h264 crct10dif_ce panfrost gpu_sched sun50i_di videobuf2_vmalloc fuse
[   51.564312] CPU: 3 PID: 856 Comm: VideoPlayer Tainted: G         C        5.16.5 #1
[   51.564322] Hardware name: Beelink GS1 (DT)
[   51.564329] pstate: 20000005 (nzCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   51.564337] pc : __alloc_pages+0x624/0xc3c
[   51.564342] lr : __dma_direct_alloc_pages+0x15c/0x1a4
[   51.564362] sp : ffff800009bbb7d0
[   51.564364] x29: ffff800009bbb7d0 x28: 000000000000000c x27: 0000000000000001
[   51.564372] x26: ffff00000c254090 x25: ffff00000c254028 x24: ffff00000190ec10
[   51.564380] x23: ffff8000080be248 x22: 0000000000bde000 x21: 0000000000000cc1
[   51.564388] x20: 0000000000bde000 x19: ffff00000190ec10 x18: 0000000000000020
[   51.564396] x17: 6572202c73656761 x16: 702038333033203a x15: 657a69732d716572
[   51.564404] x14: 202c64656c696166 x13: ffff8000094109f8 x12: 00000000000003e7
[   51.564412] x11: 000000000000014d x10: ffff8000094689f8 x9 : 00000000fffff000
[   51.564420] x8 : 0000000000000000 x7 : ffff8000094689f8 x6 : 0000000000000000
[   51.564428] x5 : 0000000000000001 x4 : 000000000000003f x3 : 0000000000000000
[   51.564436] x2 : ffff00000719b500 x1 : 0000000000000000 x0 : 0000000000000cc1
[   51.564444] Call trace:
[   51.564449]  __alloc_pages+0x624/0xc3c
[   51.564456]  __dma_direct_alloc_pages+0x15c/0x1a4
[   51.564462]  dma_direct_alloc+0x6c/0x310
[   51.564469]  dma_alloc_attrs+0x78/0xe0
[   51.564475]  vb2_dc_alloc+0x98/0x16c
[   51.564493]  __vb2_queue_alloc+0x190/0x450
[   51.564498]  vb2_core_create_bufs+0x110/0x354
[   51.564503]  vb2_create_bufs+0x15c/0x1f0
[   51.564507]  v4l2_m2m_ioctl_create_bufs+0x44/0x50
[   51.564521]  v4l_create_bufs+0x98/0xdc
[   51.564532]  __video_do_ioctl+0x174/0x3d0
[   51.564537]  video_usercopy+0x220/0x6ec
[   51.564543]  video_ioctl2+0x14/0x20
[   51.564548]  v4l2_ioctl+0x3c/0x60
[   51.564553]  v4l2_compat_ioctl32+0x88/0xb0
[   51.564559]  __arm64_compat_sys_ioctl+0x140/0x16c
[   51.564569]  invoke_syscall+0x44/0x110
[   51.564581]  el0_svc_common.constprop.0+0x40/0xe0
[   51.564587]  do_el0_svc_compat+0x18/0x30
[   51.564593]  el0_svc_compat+0x14/0x4c
[   51.564608]  el0t_32_sync_handler+0x88/0x11c
[   51.564615]  el0t_32_sync+0x15c/0x160
[   51.564625] ---[ end trace 925325d3669239df ]---
[   51.564649] cedrus 1c0e000.video-codec: dma alloc of size 12443648 failed
[   51.573792] cma: cma_alloc: reserved: alloc failed, req-size: 3038 pages, ret: -12
[   51.573845] cedrus 1c0e000.video-codec: dma alloc of size 12443648 failed
[   51.576929] cma: cma_alloc: reserved: alloc failed, req-size: 1024 pages, ret: -12
[   51.577346] cma: cma_alloc: reserved: alloc failed, req-size: 3038 pages, ret: -12

Display More

Quote from jernej

Anyway, if you want to see current progress (some H264 videos and HW deinterlacing doesn't work), you can check https://github.com/jernejsk/LibreELEC.tv/commits/iommu

Cherry picked iommu commit, xtest.patch doesn't apply, probably something already in master.

offbeat

Quote from HiassofT

If you want to get this solved please try the suggested addon change.

I did a full clean and rebuild of latest master with with Python 3.9 and DEBUG=yes.
Made Kodi crash relatively quickly, by scrolling around, trace attached.

And yes, disabling script.embuary.info seems to help, can't crash Kodi by navigating in video lists anymore.

Then I've reenabled embuary and got this error in kodi.log

Python

2022-02-19 23:09:24.638 T:2101    error <general>: EXCEPTION Thrown (PythonToCppException) : -->Python callback/script returned the following error<--
                                                    - NOTE: IGNORING THIS CAN LEAD TO MEMORY LEAKS!
                                                   Error Type: <class 'AttributeError'>
                                                   Error Contents: 'datetime.timedelta' object has no attribute 'timestamp'
                                                   Traceback (most recent call last):
                                                     File "/storage/.kodi/addons/script.embuary.info/service.py", line 8, in <module>
                                                       from resources.lib.helper import *
                                                     File "/storage/.kodi/addons/script.embuary.info/resources/lib/helper.py", line 15, in <module>
                                                       import arrow
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/__init__.py", line 2, in <module>
                                                       from .api import get, now, utcnow
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/api.py", line 12, in <module>
                                                       from arrow.arrow import TZ_EXPR, Arrow
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/arrow.py", line 34, in <module>
                                                       from arrow import formatter, locales, parser, util
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/parser.py", line 26, in <module>
                                                       from arrow.util import next_weekday, normalize_timestamp
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/util.py", line 6, in <module>
                                                       from arrow.constants import (
                                                     File "/storage/.kodi/addons/script.module.arrow/lib/arrow/constants.py", line 16, in <module>
                                                       _MAX_TIMESTAMP = datetime.max.timestamp()
                                                   AttributeError: 'datetime.timedelta' object has no attribute 'timestamp'
                                                   -->End of Python script error report<--

Display More

offbeat

Problem isn't RPi specific btw, on my Allwinner H6 Kodi was crashing every few seconds while scrolling through the list of videos, libpython3.9 in crashlogs.

Doing my own builds, so reverting Python 3.9 back to 3.8 solved this issue, at least for now.

Python3: update to 3.9.10 · LibreELEC/LibreELEC.tv@1f01979

github.com

offbeat · IR remote and longpress offbeat February 17, 2022 at 6:00 PM

In most cases LibreELEC uses eventlircd as a shim to pass IR button presses from kernel to Kodi. This approach is stable and reliable, but has one rather major drawback: it’s impossible to use longpress events, which in fact halves the amount of actions you can bind to your IR remote buttons.

Alternative approach is to disable eventlircd altogether and use IR remote as a Linux keyboard device, so this is a way I went.

LibreELEC documentation says “Kodi ... doesn't cope well with Linux input events”, I’d say this is the understatement of the century, to put it mildly. Kodi keyboard handling system is archaic, weaving layer upon layer of abstractions, throwing away more and more of your keypresses, the deeper it goes.

To my understanding, the way IR keypress travels towards Kodi in is:

1) Linux kernel decodes IR signal and translates keypress to evdev scancode, based on IR remote .toml definition, loaded by ir-keytable.

2) Xkbcommon library assumes we got a standard pc105 keyboard and translates evdev code to something resembling xfree86 keycode, according to /usr/share/X11/xkb/keycodes/evdev

3) Xkbcommon library futher translates xfree86 keycode to xkb key name, mapping non-printable keys according to /usr/share/X11/xkb/symbols/inet

4) Kodi accepts a very limited subset of xkb key names, filtering them and mapping to Kodi internal keysyms, according to code in xbmc/platform/linux/input/LibInputKeyboard.cpp

5) Kodi goes through another layer of translations, from Kodi internal keysyms to Kodi internal vkeys and Kodi keynames, relevant code in xbmc/input/XBMC_keytable.cpp

6) Kodi once again translates, keynames are mapped to actions, based on default rules in /usr/share/kodi/system/keymaps/keyboard.xml and user customizations in ~/.kodi/userdata/keymaps/keyboard.xml

Every layer of this onion has some random historical subset or superset of non-printable keycodes hardcoded in, throwing away everything else, sometimes in completely illogical way. There is no way for user to bind directly to kernel evdev scancode, is all buried and hidden away. I’m afraid to look at the Windows or Android keyboard abstractions, probably not much saner.

But in the end I managed to dig through this spaghetti and made a .toml keymap for my IR remote, scraping together some 50 evdev keys Kodi actually lets through the cracks. Longpress events work for me now, yey. Hope my keymap can help someone with the same problem.

Code

[[protocols]]
name = "myoldremote"
protocol = "nec"
variant = "necx"
[protocols.scancodes]
# IR      EVDEV name           EVDEV  xkbcommon name                Kodi internal keysym       Kodi name             Notes
#-------------------------------------------------------------------------------------------------------------------------------------------
0xbf10 = "KEY_MAIL"          # 155    XKB_KEY_XF86Mail              XBMCK_LAUNCH_MAIL          launch_mail           actual button is EJECT
0xbf46 = "KEY_MUTE"          # 113    XKB_KEY_XF86AudioMute         XBMCK_VOLUME_MUTE          volume_mute
0xbf45 = "KEY_FAVORITES"     # 364    XKB_KEY_XF86Favorites         XBMCK_FAVORITES            favorites             actual button is MODE
0xbf43 = "KEY_POWER"         # 116    XKB_KEY_XF86PowerOff          XBMCK_POWER                power
0xbf40 = "KEY_F1"            # 59     XKB_KEY_F1                    XBMCK_F1                   f1                    actual button is RED
0xbf1f = "KEY_F2"            # 60     XKB_KEY_F2                    XBMCK_F2                   f2                    actual button is GREEN
0xbf00 = "KEY_F3"            # 61     XKB_KEY_F3                    XBMCK_F3                   f3                    actual button is YELLOW
0xbf41 = "KEY_F4"            # 62     XKB_KEY_F4                    XBMCK_F4                   f4                    actual button is BLUE
0xbf0b = "KEY_1"             # 2      XKB_KEY_1                     XBMCK_1                    one
0xbf0c = "KEY_2"             # 3      XKB_KEY_2                     XBMCK_2                    two
0xbf0d = "KEY_3"             # 4      XKB_KEY_3                     XBMCK_3                    three
0xbf0e = "KEY_4"             # 5      XKB_KEY_4                     XBMCK_4                    four
0xbf0f = "KEY_5"             # 6      XKB_KEY_5                     XBMCK_5                    five
0xbf01 = "KEY_6"             # 7      XKB_KEY_6                     XBMCK_6                    six
0xbf11 = "KEY_7"             # 8      XKB_KEY_7                     XBMCK_7                    seven
0xbf12 = "KEY_8"             # 9      XKB_KEY_8                     XBMCK_8                    eight
0xbf13 = "KEY_9"             # 10     XKB_KEY_9                     XBMCK_9                    nine
0xbf05 = "KEY_DELETE"        # 111    XKB_KEY_Delete                XBMCK_DELETE               delete
0xbf0a = "KEY_0"             # 11     XKB_KEY_0                     XBMCK_0                    zero
0xbf42 = "KEY_CAPSLOCK"      # 58     XKB_KEY_Caps_Lock             XBMCK_CAPSLOCK             capslock              actual button is SELECT/CAP/NUM
0xbf52 = "KEY_VOLUMEUP"      # 115    XKB_KEY_XF86AudioRaiseVolume  XBMCK_VOLUME_UP            volume_up
0xbf53 = "KEY_VOLUMEDOWN"    # 114    XKB_KEY_XF86AudioLowerVolume  XBMCK_VOLUME_DOWN          volume_down
0xbf06 = "KEY_SEARCH"        # 217    XKB_KEY_XF86Search            XBMCK_BROWSER_SEARCH       browser_search
0xbf02 = "KEY_MEDIA"         # 226    XKB_KEY_XF86AudioMedia        XBMCK_LAUNCH_MEDIA_CENTER  launch_media_select   actual button is ZOOM
0xbf4e = "KEY_F12"           # 88     XKB_KEY_F12                   XBMCK_F12                  f12                   actual button is SETUP                                        actual button is SETUP
0xbf4b = "KEY_PAGEUP"        # 104    XKB_KEY_Page_Up               XBMCK_PAGEUP               pageup                actual button is CHANNELUP
0xbf4c = "KEY_PAGEDOWN"      # 109    XKB_KEY_Page_Down             XBMCK_PAGEDOWN             pagedown              actual button is CHANNELUP
0xbf50 = "KEY_F10"           # 68     XKB_KEY_F10                   XBMCK_F10                  f10                   actual button is INFO
0xbf07 = "KEY_F11"           # 87     XKB_KEY_F11                   XBMCK_F11                  f11                   actual button is POPUP MENU
0xbf15 = "KEY_UP"            # 103    XKB_KEY_Up                    XBMCK_UP                   up
0xbf16 = "KEY_DOWN"          # 108    XKB_KEY_Down                  XBMCK_DOWN                 down
0xbf17 = "KEY_LEFT"          # 105    XKB_KEY_Left                  XBMCK_LEFT                 left
0xbf18 = "KEY_RIGHT"         # 106    XKB_KEY_Right                 XBMCK_RIGHT                right
0xbf14 = "KEY_ENTER"         # 28     XKB_KEY_Return                XBMCK_RETURN               return                actual button is OK
0xbf04 = "KEY_BACK"          # 158    XKB_KEY_XF86Back              XBMCK_BROWSER_BACK         browser_back          actual button is RETURN
0xbf51 = "KEY_HOMEPAGE"      # 172    XKB_KEY_XF86HomePage          XBMCK_HOMEPAGE             homepage              actual button is TOP MENU
0xbf48 = "KEY_PLAY"          # 207    XKB_KEY_XF86AudioPlay         XBMCK_PLAY                 play_pause
0xbf1e = "KEY_PAUSECD"       # 201    XKB_KEY_XF86AudioPause        XBMCK_MEDIA_PLAY_PAUSE     play_pause
0xbf49 = "KEY_PREVIOUSSONG"  # 165    XKB_KEY_XF86AudioPrev         XBMCK_MEDIA_PREV_TRACK     prev_track
0xbf1d = "KEY_NEXTSONG"      # 163    XKB_KEY_XF86AudioNext         XBMCK_MEDIA_NEXT_TRACK     next_track
0xbf19 = "KEY_STOPCD"        # 166    XKB_KEY_XF86AudioStop         XBMCK_MEDIA_STOP           stop
0xbf1a = "KEY_PROG1"         # 148    XKB_KEY_XF86Launch1           XBMCK_LAUNCH_APP2          launch_app2_pc_icon   actual button is SLOW
0xbf1c = "KEY_REWIND"        # 168    XKB_KEY_XF86AudioRewind       XBMCK_MEDIA_REWIND         rewind
0xbf1b = "KEY_FASTFORWARD"   # 208    XKB_KEY_XF86AudioForward      XBMCK_MEDIA_FASTFORWARD    fastforward
0xbf54 = "KEY_F5"            # 63     XKB_KEY_F5                    XBMCK_F5                   f5                    actual button is SUBTITLE
0xbf4d = "KEY_F6"            # 64     XKB_KEY_F6                    XBMCK_F6                   f6                    actual button is ANGLE
0xbf44 = "KEY_F7"            # 65     XKB_KEY_F7                    XBMCK_F7                   f7                    actual button is AUDIO
0xbf4f = "KEY_F8"            # 66     XKB_KEY_F8                    XBMCK_F8                   f8                    actual button is REPEAT
0xbf47 = "KEY_F9"            # 67     XKB_KEY_F9                    XBMCK_F9                   f9                    actual button is SHUFFLE
0xbf03 = "KEY_WWW"           # 150    XKB_KEY_XF86WWW               XBMCK_LAUNCH_FILE_BROWSER  launch_file_browser

Display More

offbeat

Beelink GS1 Ethernet sometimes fails with a "transmit queue 0 timed out" error, resulting total loss of connection.

Code

[19438.239679] skoll kernel: ------------[ cut here ]------------
[19438.239698] skoll kernel: NETDEV WATCHDOG: eth0 (dwmac-sun8i): transmit queue 0 timed out
[19438.239759] skoll kernel: WARNING: CPU: 0 PID: 13693 at net/sched/sch_generic.c:477 dev_watchdog+0x2f8/0x300
[19438.239779] skoll kernel: Modules linked in: mt7601u mac80211 cfg80211 libarc4 8021q garp stp mrp llc rfkill hantro_vpu(C) crct10dif_ce v4l2_vp9 sunxi_cir v4l2_h264 dwmac_sun8i videobuf2_vmalloc panfrost gpu_sched sun50i_di fuse
[19438.239832] skoll kernel: CPU: 0 PID: 13693 Comm: LanguageInvoker Tainted: G         C        5.16.5 #1
[19438.239839] skoll kernel: Hardware name: Beelink GS1 (DT)
[19438.239842] skoll kernel: pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[19438.239848] skoll kernel: pc : dev_watchdog+0x2f8/0x300
[19438.239853] skoll kernel: lr : dev_watchdog+0x2f8/0x300
[19438.239857] skoll kernel: sp : ffff800009c1bc50
[19438.239859] skoll kernel: x29: ffff800009c1bc50 x28: 0000000000000007 x27: 00000000ffffffff
[19438.239867] skoll kernel: x26: ffff000002df0480 x25: 0000000000000000 x24: ffff0000018f1a00
[19438.239875] skoll kernel: x23: ffff0000018f1940 x22: ffff000002df03dc x21: ffff8000093f7000
[19438.239883] skoll kernel: x20: ffff000002df0000 x19: 0000000000000000 x18: 0000000000000020
[19438.239890] skoll kernel: x17: ffff80007691a000 x16: ffff800008004000 x15: ffffffffffffffff
[19438.239898] skoll kernel: x14: ffff800009553d93 x13: ffff8000094109f8 x12: 0000000000000774
[19438.239905] skoll kernel: x11: 000000000000027c x10: ffff8000094689f8 x9 : 00000000fffff000
[19438.239913] skoll kernel: x8 : ffff8000094109f8 x7 : ffff8000094689f8 x6 : 0000000000000000
[19438.239920] skoll kernel: x5 : ffff00007fb74878 x4 : 0000000000000000 x3 : 0000000000000027
[19438.239928] skoll kernel: x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff00000197c240
[19438.239936] skoll kernel: Call trace:
[19438.239939] skoll kernel:  dev_watchdog+0x2f8/0x300
[19438.239944] skoll kernel:  call_timer_fn.constprop.0+0x20/0x7c
[19438.239953] skoll kernel:  __run_timers.part.0+0x1f0/0x224
[19438.239958] skoll kernel:  run_timer_softirq+0x38/0x70
[19438.239964] skoll kernel:  _stext+0x124/0x29c
[19438.239970] skoll kernel:  __irq_exit_rcu+0xdc/0xfc
[19438.239976] skoll kernel:  irq_exit_rcu+0xc/0x14
[19438.239979] skoll kernel:  el0_interrupt+0x38/0xa4
[19438.239987] skoll kernel:  __el0_irq_handler_common+0x14/0x1c
[19438.239993] skoll kernel:  el0t_32_irq_handler+0xc/0x14
[19438.239999] skoll kernel:  el0t_32_irq+0x15c/0x160
[19438.240004] skoll kernel: ---[ end trace 59ca441fbd9f9b65 ]---
[19438.240066] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: Reset adapter.
[19438.243692] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: PHY [stmmac-0:01] driver [RTL8211E Gigabit Ethernet] (irq=POLL)
[19439.084927] skoll connmand[422]: eth0 {TX} 13052139 packets 2969467971 bytes
[19438.256584] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: Register MEM_TYPE_PAGE_POOL RxQ-0
[19438.257256] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: No Safety Features support found
[19438.257269] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: No MAC Management Counters available
[19438.257275] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: PTP not supported by HW
[19438.257290] skoll kernel: dwmac-sun8i 5020000.ethernet eth0: configuring for phy/rgmii-id link mode

Display More

The only way to get it back alive without rebooting is to rmmod/modprobe dwmac_sun8i

Rmmod gives two additional errors "bus-emac already disabled" and "bus-emac already unprepared"

Code

[19940.714877] skoll kernel: ------------[ cut here ]------------
[19940.714892] skoll kernel: bus-emac already disabled
[19940.714948] skoll kernel: WARNING: CPU: 0 PID: 13972 at drivers/clk/clk.c:952 clk_core_disable+0x98/0xa0
[19940.714976] skoll kernel: Modules linked in: mt7601u mac80211 cfg80211 libarc4 8021q garp stp mrp llc rfkill hantro_vpu(C) crct10dif_ce v4l2_vp9 sunxi_cir v4l2_h264 dwmac_sun8i(-) videobuf2_vmalloc panfrost gpu_sched sun50i_di fuse
[19940.715026] skoll kernel: CPU: 0 PID: 13972 Comm: rmmod Tainted: G        WC        5.16.5 #1
[19940.715032] skoll kernel: Hardware name: Beelink GS1 (DT)
[19940.715036] skoll kernel: pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[19940.715042] skoll kernel: pc : clk_core_disable+0x98/0xa0
[19940.715046] skoll kernel: lr : clk_core_disable+0x98/0xa0
[19940.715051] skoll kernel: sp : ffff8000096b3c10
[19940.715053] skoll kernel: x29: ffff8000096b3c10 x28: ffff000001a7c240 x27: 0000000000000000
[19940.715060] skoll kernel: x26: 0000000000000000 x25: 0000000000000000 x24: 0000000000000000
[19940.715066] skoll kernel: x23: ffff000001a04c90 x22: ffff800000b99028 x21: 0000000000000000
[19940.715073] skoll kernel: x20: ffff000002a16c00 x19: ffff000002a16c00 x18: 0000000000000028
[19940.715080] skoll kernel: x17: ffff80007691a000 x16: ffff00000347e000 x15: fffffffffffeb330
[19940.715087] skoll kernel: x14: ffff8000094109a8 x13: 0a64656c62617369 x12: 642079646165726c
[19940.715093] skoll kernel: x11: 656820747563205b x10: fffffffffffe0000 x9 : ffff8000096b3c10
[19940.715100] skoll kernel: x8 : 00000000ffffffff x7 : 656c626173696420 x6 : 0000000000000000
[19940.715107] skoll kernel: x5 : ffff00007fb74878 x4 : 0000000000000000 x3 : 0000000000000027
[19940.715113] skoll kernel: x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff000001a7c240
[19940.715120] skoll kernel: Call trace:
[19940.715123] skoll kernel:  clk_core_disable+0x98/0xa0
[19940.715128] skoll kernel:  clk_disable+0x2c/0x44
[19940.715135] skoll kernel:  stmmac_pltfr_remove+0x48/0x90
[19940.715143] skoll kernel:  sun8i_dwmac_remove+0x30/0xc0 [dwmac_sun8i]
[19940.715155] skoll kernel:  platform_remove+0x24/0x54
[19940.715161] skoll kernel:  __device_release_driver+0x180/0x230
[19940.715167] skoll kernel:  driver_detach+0xbc/0xfc
[19940.715171] skoll kernel:  bus_remove_driver+0x54/0xb0
[19940.715178] skoll kernel:  driver_unregister+0x2c/0x54
[19940.715182] skoll kernel:  platform_driver_unregister+0x10/0x20
[19940.715187] skoll kernel:  sun8i_dwmac_driver_exit+0x14/0x7c0 [dwmac_sun8i]
[19940.715193] skoll kernel:  __arm64_sys_delete_module+0x178/0x1f0
[19940.715201] skoll kernel:  invoke_syscall+0x44/0x110
[19940.715209] skoll kernel:  el0_svc_common.constprop.0+0x40/0xe0
[19940.715213] skoll kernel:  do_el0_svc_compat+0x18/0x30
[19940.715218] skoll kernel:  el0_svc_compat+0x14/0x4c
[19940.715226] skoll kernel:  el0t_32_sync_handler+0x88/0x11c
[19940.715232] skoll kernel:  el0t_32_sync+0x15c/0x160
[19940.715238] skoll kernel: ---[ end trace 59ca441fbd9f9b66 ]---
[19940.717425] skoll kernel: ------------[ cut here ]------------
[19940.717441] skoll kernel: bus-emac already unprepared
[19940.717510] skoll kernel: WARNING: CPU: 0 PID: 13972 at drivers/clk/clk.c:810 clk_core_unprepare+0xd8/0x100
[19940.717542] skoll kernel: Modules linked in: mt7601u mac80211 cfg80211 libarc4 8021q garp stp mrp llc rfkill hantro_vpu(C) crct10dif_ce v4l2_vp9 sunxi_cir v4l2_h264 dwmac_sun8i(-) videobuf2_vmalloc panfrost gpu_sched sun50i_di fuse
[19940.717593] skoll kernel: CPU: 0 PID: 13972 Comm: rmmod Tainted: G        WC        5.16.5 #1
[19940.717599] skoll kernel: Hardware name: Beelink GS1 (DT)
[19940.717603] skoll kernel: pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[19940.717609] skoll kernel: pc : clk_core_unprepare+0xd8/0x100
[19940.717614] skoll kernel: lr : clk_core_unprepare+0xd8/0x100
[19940.717619] skoll kernel: sp : ffff8000096b3c10
[19940.717621] skoll kernel: x29: ffff8000096b3c10 x28: ffff000001a7c240 x27: 0000000000000000
[19940.717628] skoll kernel: x26: 0000000000000000 x25: 0000000000000000 x24: 0000000000000000
[19940.717635] skoll kernel: x23: ffff000001a04c90 x22: ffff800000b99028 x21: 0000000000000000
[19940.717641] skoll kernel: x20: ffff000005569700 x19: ffff000002a16c00 x18: 0000000000000028
[19940.717649] skoll kernel: x17: ffff80007691a000 x16: ffff000002f8e1c0 x15: fffffffffffebc98
[19940.717655] skoll kernel: x14: ffff8000094109a8 x13: 0a64657261706572 x12: 706e752079646165
[19940.717663] skoll kernel: x11: 656820747563205b x10: fffffffffffe0000 x9 : ffff8000096b3c10
[19940.717670] skoll kernel: x8 : 00000000ffffffff x7 : 61706572706e7520 x6 : 0000000000000000
[19940.717677] skoll kernel: x5 : ffff00007fb74878 x4 : 0000000000000000 x3 : 0000000000000027
[19940.717683] skoll kernel: x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff000001a7c240
[19940.717690] skoll kernel: Call trace:
[19940.717693] skoll kernel:  clk_core_unprepare+0xd8/0x100
[19940.717699] skoll kernel:  clk_unprepare+0x28/0x40
[19940.717704] skoll kernel:  stmmac_pltfr_remove+0x50/0x90
[19940.717712] skoll kernel:  sun8i_dwmac_remove+0x30/0xc0 [dwmac_sun8i]
[19940.717726] skoll kernel:  platform_remove+0x24/0x54
[19940.717734] skoll kernel:  __device_release_driver+0x180/0x230
[19940.717739] skoll kernel:  driver_detach+0xbc/0xfc
[19940.717743] skoll kernel:  bus_remove_driver+0x54/0xb0
[19940.717751] skoll kernel:  driver_unregister+0x2c/0x54
[19940.717755] skoll kernel:  platform_driver_unregister+0x10/0x20
[19940.717760] skoll kernel:  sun8i_dwmac_driver_exit+0x14/0x7c0 [dwmac_sun8i]
[19940.717766] skoll kernel:  __arm64_sys_delete_module+0x178/0x1f0
[19940.717775] skoll kernel:  invoke_syscall+0x44/0x110
[19940.717783] skoll kernel:  el0_svc_common.constprop.0+0x40/0xe0
[19940.717788] skoll kernel:  do_el0_svc_compat+0x18/0x30
[19940.717792] skoll kernel:  el0_svc_compat+0x14/0x4c
[19940.717802] skoll kernel:  el0t_32_sync_handler+0x88/0x11c
[19940.717809] skoll kernel:  el0t_32_sync+0x15c/0x160
[19940.717815] skoll kernel: ---[ end trace 59ca441fbd9f9b67 ]---

Display More

offbeat · Allwinner H6 HEVC? offbeat February 11, 2022 at 11:07 PM

I got H6 Beelink GS1, using it daily for a last few weeks with 1080p TV, so can give some feedback.

It’s certainly usable, smooth hardware playback 1080p60 8bit for all three relevant codecs H.264/HEVC/VP9. No problems seeking videos.

10bit HEVC plays just fine, 10bit VP9 isn’t really working yet (decodes to garbled pixels). Can’t vouch for colors, most likely TV only gets 8bit for now.

4K60 is probably only realistic with HEVC, both H.264 and VP9 hardware decoders seem to only handle 4K30. Had to increase kernel CMA memory reservation while trying 4K60 10bit HEVC, defaults are too low.

Now, some some bugs and annoyances:

Hardware deinterlaced videos tend to stuck after resuming from pause, need to seek to get them going again.

HDMI output can randomly drop out for a second, both during playback and in menu. Probably some output timings issues, since sometimes (rare though) TV gets confused and refuses to display certain Hz videos, requiring on/off cycle, which never happened outputing from my other STBs.

H6 still needs algorithmdirtyregions=0 set in Kodi, otherwise any video will stutter.

HDMI CEC seems to be very stable and usable now. But I had to disable it, because Panfrost GPU drivers ability to sleep is still semi-broken, every second or third attempt to suspend fails, so TV gets woken up by CEC, etc.

Crust co-processor firmware rocks as well, this box consumes pretty much zero watts in sleep. If Panfrost will let you to suspend it, that is

H6 box can also be configured to support pretty much any IR remote, including wakeup button if you bother to rebuild Crust. I certainly did, reused big comfortable remote from old mediaplayer.

Theoretically, if display output stack gets some upgrades, H6 can become as capable video player as RPI4: 4K60 with 10bits and HDR should be real.

offbeat · Hardware accelerated video in 2022 offbeat January 31, 2022 at 9:17 PM

Quote from powerarmour

but you've got to have a developer who has the time to put in hard work

Developers on payroll of Realtek and other vendors put their time in either way. Its management who wasn't interested in going extra mile mainlining code, adjusting own deadlines, etc.

I'm happy that some companies are starting to see the benefits of mainlining things, others are getting pushed there by Google because of Android sertification.

For us, end users, it will hopefully result less fragmentation, less e-waste. And maybe one or two less video decoding standard to care about, if we stay on topic.

offbeat · Hardware accelerated video in 2022 offbeat January 31, 2022 at 8:27 PM

Quote from isamudaison

Maybe a bit off-topic and maybe too philosophical, but do you see the vendor-specific ARM-based stuff as the 'future' of Kodi?

Got two rather ancient old media devices running in my living room.

First is media player, MIPS Sigma SOC. Fantastic hardware, plays any 8bit H.264 with dynamic Hz changing, every older codec, not a single visible framedrop and glitch. But welcome back to year 2007 and say hello to kernel 2.6.22. If you want to play something off network share, open up SMB1.

Another one is DVB-S2 SAT receiver, MIPS Realtek SOC. Once again, vendor blobs, kernel 3.13, you get what you get. Want to plug additional DVB-T USB stick supported in mainline? Tough luck.

Both devices got hardware beefy enough to run something like LibreElec (3D acceleration, hardware video decoding, enough RAM), yet both are rotting away, feature and security wise, waiting to become e-waste.

Who wants such 'future'?

offbeat

Wake on LAN isn't supported on Allwinner boxes, right?

Ethtool says "nope", Sunxi wiki says no WoL for GMAC + external PHY.
Crust firmware seems to support wakeup on button/RTC/IR/CEC only.

There are some old IRC logs about h3fakeoff supporting WoL, but RAM has to remain powered.
https://irclog.whitequark.org/linux-sunxi/2018-01-15

offbeat

Quote from jernej

offbeat everlanes I just found the issue and potential solution. It seems that width and height need to be multiple of 32 when converting from original NV12 tiled format to standard NV12. Please test following update

Confirming, latest master build decodes 1552px wide VP9 video just fine.
Thank you for your work!

offbeat

Quote from jernej

It's entirely possible that 144 Hz output needs some things to be done in more precise ways

According to calculator, 1440p@144 pushes the very limits of HDMI 2.0 bandwidth, assuming H6 outputs full RGB.
At least simple OpenGLES things like kmscube seem to render and output at 144FPS just fine, so it's only video plane output that has some glitch.

Data Rate Required: 14.08 Gbit/s

Video Format:

2560 × 1440 (16∶9 ratio) at 144 Hz

8 bpc (24 bit/px) RGB color

Uncompressed

CVT-R2 timing format

HDMI 2.0 14.40 Gbit/s

offbeat

Quote from jernej

Wait, are you testing this on LE?

Yes, on LE, latest master with some packages like mpv added to build.

offbeat

Quote from jernej

ok, so issue is either with rendering code in Kodi or with DRM driver.

Quick test in console:

mpv --drm-mode=2560x1440@120 --- GOOD
mpv --drm-mode=2560x1440@144 --- BAD

So its DRM driver, guess it can be reported to mesa bugtracker?

offbeat

Quote from jernej

That might be DRM rendering issue. Can you try with switching DRMPRIME rendering from "direct to plane" to EGL? If that doesn't help, I don't know... I would need such monitor to test directly.

Yes, prime decoder=ON + prime render=EGL helps, video is OK.

offbeat

Quote from jernej

You expect too much from A53 cores. H6 predates AV1, so no HW decoding support.

Yea, but it looks like H6 can handle 720p AV1, dav1d software decoder seems to be pretty optimized.

Here is a different issue I mentioned in previous comment, broken hardware decoding outputing to 144Hz, video is plain h264.

Switching to 120Hz or 60Hz fixes it, turning off DRM decoder solves the problem as well.

offbeat · Hardware accelerated video in 2022 offbeat January 27, 2022 at 6:31 PM

Thank you, I hope V4L2 path related changes will get mainlined soon!

offbeat

Downloaded some random popular youtube playlist, 1080p res, each video in h264, vp9 and av1.

Latest master build for Beelink GS1, tested in both kodi and mpv.

AV1 software decoding get H6 to its knees, all 4 cores loaded pretty much to 100%.

H264 hardware decoding, no problems detected.

VP9 hardware decoding, 49 of 50 videos played just fine, one produces garbled picture. Also, mpv sometimes hangs on VP9 videos, not a hard lock though, can Ctrl+C and play again.

This VP9 clip fails to decode correctly

External Content www.youtube.com

Content embedded from external sources will not be displayed without your consent.

Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

Ah yes, H6 hardware decoding path also garbles everything when outputing to 1440p@144Hz my monitor HDMI defaults at.

Downgrading video output to 1080p@60Hz solved the issue (except that one VP9 clip)

Posts by offbeat

Allwinner H6 HEVC?

Allwinner H6 HEVC?

LibreELEC-RPi4.arm-11.0-nightly-20220213-fcba5db is restarting every 20 seconds

LibreELEC-RPi4.arm-11.0-nightly-20220213-fcba5db is restarting every 20 seconds

IR remote and longpress

Nightly images for A64, H3, H5, H6 and R40 boards

Allwinner H6 HEVC?

Hardware accelerated video in 2022

Hardware accelerated video in 2022

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Nightly images for A64, H3, H5, H6 and R40 boards

Hardware accelerated video in 2022

Nightly images for A64, H3, H5, H6 and R40 boards