Pi 5 kernel8.img for 6.6.62 causes audio players to freeze with a looping sound

[theflyingape]

Describe the bug

config.txt

[pi5]
kernel=kernel8.img

Steps to reproduce the behaviour

ffplay sound.wav race condition with studdering sound until killed.
If I comment out kernel8.img, ffplay runs fine, but some other software will segfault without kernel8.

Device (s)

Raspberry Pi 5

System

uname -r
6.6.62+rpt-rpi-v8

If I rollback to 6.6.51, it kernel8.img and sound events run fine.

Logs

No response

Additional context

No response

[popcornmix]

What device are you outputting audio to? (e.g. hdmi?)
Is this RPiOS bookworm 64-bit? Lite or desktop image?

[theflyingape]

Yes, latest Bookworm release and output to HDMI-0 -- both from LightDM with Wayland+labwc and from console mode.

[theflyingape]

I can confirm rolling back the linux-image-6.6.62 --> linux-image-6.6.51 with the kernel8.img enabled works as expected.

[pelwell]

I'm not a regular ffpmeg user, but it's working for me to HDMI0. There is no audible difference between kernel8.img and kernel_2712.img.

1. What sample format and rate is sound.wav? If you've used several formats, does the behaviour differ between them?
2. Does speaker-test -c 2 stutter?
3. How about aplay sound.wav?

[theflyingape]

Just tried sudo apt install linux-image-6.6.62+rpt-rpi-2712 linux-image-6.6.62+rpt-rpi-v8 on Pi 5, rebooted with and without kernel8.img, and sound stutters regardless of format (wav, mp3) which includes playing video (flv, mp4) and the studder stalls video playback. I also tested with and without wireplumber and pipewire services.

speaker-test does stutter and I cannot Ctrl-C to stop -- must wait 30-seconds.
aplay the same -- but interesting that ffplay, mplayer, mpv all work on Pi 4 (not Pi 5 as reported) but aplay fails.

Only on Pi4, I must stop wireplumber (not pipewire) else mplayer immediately reports "Audio device got stuck!"
On Pi 5, stopping all and sending it to ALSA device reports "Audio device got stuck!" repeatedly.

[pelwell]

If you have a spare SD card, try a clean install of the latest 64-bit OS - it works with ffplay, speaker-test, aplay and mplayer (which I did have to apt install). Yes, speaker-test doesn't like Ctrl-C, and ffplay doesn't exit and the end of the file, but those are common problems.

[theflyingape]

Happy to oblige -- used 2024-11-19-raspios-bookworm-arm64-full.img.xz
Works perfectly off the installation with the auto updates it applies.
However, after modifying config.txt:

# For more options and information see
# http://rptl.io/configtxt
# Some settings may impact device functionality. See link above for details

# Uncomment some or all of these to enable the optional hardware interfaces
#dtparam=i2c_arm=on
#dtparam=i2s=on
#dtparam=spi=on

# Enable audio (loads snd_bcm2835)
dtparam=audio=on

# Additional overlays and parameters are documented
# /boot/firmware/overlays/README

# Automatically load overlays for detected cameras
camera_auto_detect=1

# Automatically load overlays for detected DSI displays
display_auto_detect=1

# Automatically load initramfs files, if found
auto_initramfs=1

# Enable DRM VC4 V3D driver
dtoverlay=vc4-kms-v3d
max_framebuffers=2

# Don't have the firmware create an initial video= setting in cmdline.txt.
# Use the kernel's default instead.
disable_fw_kms_setup=1

# Run in 64-bit mode
arm_64bit=1

# Disable compensation for displays with overscan
disable_overscan=1

# Run as fast as firmware / board allows
arm_boost=1

[cm4]
# Enable host mode on the 2711 built-in XHCI USB controller.
# This line should be removed if the legacy DWC2 controller is required
# (e.g. for USB device mode) or if USB support is not required.
otg_mode=1

[cm5]
dtoverlay=dwc2,dr_mode=host

[all]

[pi5]
arm_freq=2760
dtparam=nvme
dtparam=pciex1_gen=3
gpu_freq=920
#over_voltage=3
# account for Mupen64Plus-Next limitation
kernel=kernel8.img

ffplay essentially gets stuck and speaker-test studders horribly @ ~60 seconds per period instead of constant pink noise for ~11

[popcornmix]

Can you confirm the issue occurs without the overclock?

[theflyingape]

No problem, I had posted a "clean" config.txt from the start but certainly worth a try off the "clean" install.
So booted off the clean SD card install without any NVME and without any cpu/gpu changes, sound behaves badly as before with the kernel8.img on Pi 5. I comment out config.txt kernel=kernel8.img, it works perfectly fine.

I do see a quick flash on the console from power up with the kernel of a cma failure with allocate multiple times that does not occur from the regular kernel startup. I could not find it in any boot log, but was able to capture it with my phone:

[popcornmix]

That's interesting. Does changing:

dtoverlay=vc4-kms-v3d

to

dtoverlay=vc4-kms-v3d,cma-320

help?

Note: you can see those messages by running dmesg. Posting the output of dmesg here is probably useful.

[theflyingape]

Well that was a magic bullet -- I updated linux images on both the SD card clean install and then my NVME boot (with overclocking) and sound works for all my use-cases using the kernel8.img now.
How did we get here? :)
P.S., I had checked dmesgand there were no like cma entries in it.

[popcornmix]

You should see the cma messages in dmesg if you hit the issue (i.e. without the cma-320 parameter).

The question is, what needs so much cma on your system?
I wouldn't expect it from ffplay (with an audio file).
The nvme drive is something to consider.

Can you show output of grep Cma /proc/meminfo?
I get:

$ grep Cma /proc/meminfo 
CmaTotal:          65536 kB
CmaFree:           48992 kB

[popcornmix]

Post url from raspinfo | pastebinit.

[theflyingape]

CmaTotal:         327680 kB
CmaFree:          245392 kB

... nets only: 82288 kB
Url: raspinfo

[theflyingape]

I checked the Pi 4, it shows 524288 allocated using stock settings off the stock kernel.
On Pi 5, removed cma-320 using kernel8.img and it shows 65536 allocated with 0 free.
This keeps coming back to me as a regression in this kernel build, no?
Here's the dmesg | grep -i cma output:

[    0.000000] Reserved memory: created CMA memory pool at 0x0000000001c00000, size 64 MiB
[    0.000000] OF: reserved mem: initialized node linux,cma, compatible id shared-dma-pool
[    0.000000] OF: reserved mem: 0x0000000001c00000..0x0000000005bfffff (65536 KiB) map reusable linux,cma
[    0.000000] Memory: 8043824K/8380416K available (13504K kernel code, 2206K rwdata, 4292K rodata, 4864K init, 1086K bss, 271056K reserved, 65536K cma-reserved)
[    0.663863] cma: cma_alloc: linux,cma: alloc failed, req-size: 4 pages, ret: -12
[    0.671307] cma: cma_alloc: linux,cma: alloc failed, req-size: 16 pages, ret: -12
[    0.678843] cma: cma_alloc: linux,cma: alloc failed, req-size: 4 pages, ret: -12
[    0.686280] cma: cma_alloc: linux,cma: alloc failed, req-size: 16 pages, ret: -12
[    0.693814] cma: cma_alloc: linux,cma: alloc failed, req-size: 4 pages, ret: -12
[    0.701251] cma: cma_alloc: linux,cma: alloc failed, req-size: 16 pages, ret: -12
[    0.708776] cma: cma_alloc: linux,cma: alloc failed, req-size: 4 pages, ret: -12
[    0.716208] cma: cma_alloc: linux,cma: alloc failed, req-size: 16 pages, ret: -12
[    1.460346] cma: cma_alloc: linux,cma: alloc failed, req-size: 514 pages, ret: -12
[    1.468420] cma: cma_alloc: linux,cma: alloc failed, req-size: 514 pages, ret: -12