cptofs hangs with 100% CPU usage for >64G volumes

[dfordivam]

The issue was observed while copying to an ext4 FS :
NixOS/nixpkgs#59867

The actual usage is in this file https://github.com/NixOS/nixpkgs/blob/1dac88dbd3d48c5706fd85d58608d8ff168424e1/nixos/lib/make-disk-image.nix#L188

[RaitoBezarius]

I have to add that this issue was observed under QEMU to build an aarch64 image of 2033573888 bytes, i.e. < 2G.

Some log:

Creating an EXT4 image of 2033573888 bytes (numInodes=79580, numDataBlocks=337318)
mke2fs 1.45.4 (23-Sep-2019)
ext2fs_check_if_mount: Can't check if filesystem is mounted due to missing mtab file while determining whether /nix/store/lgnm1lsvw4md4gkdwilabay66iq0mzig-ext4-fs.img is mounted.
Unsupported ioctl: cmd=0xffffffff80200204
Discarding device blocks: done         
Creating filesystem with 496478 4k blocks and 124160 inodes
Filesystem UUID: 44444444-4444-4444-8888-888888888888
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912

Allocating group tables: done 
Writing inode tables: done 
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done 

(original ref: NixOS/nixpkgs#76780)

[RaitoBezarius]

If someone is super interested, here's a full strace output right when cptofs is executed under qemu-aarch64-static:

https://gist.github.com/RaitoBezarius/83c565401b995df33acd31381eec345c

[flokli]

@RaitoBezarius isn't this another bug? The initial report is about 100% CPU usages for volumes >64G, on x86 (I guess?).

@evanjs, @dfordivam can you confirm?

[RaitoBezarius]

@flokli You're definitely right.

[con-f-use]

I'm still/again encountering this issue with NixOS 25.11 even with Images smaller than 64G.

[jsoo1]

I had this issue and looked at cptofs.c:

linux/tools/lkl/cptofs.c

Line 667 in f6d6f7d

ret = lkl_start_kernel("mem=100M");

Patching this to 1000M seemed to help...

[con-f-use]

Thank you jsoo1! Amazingly, this has indeed helped.

[jsoo1]

If I may suggest a fix: can we add a command line argument for the kernel command line options?

[ddiss]

If I may suggest a fix: can we add a command line argument for the kernel command line options?

How about a --mb <memory_in_MB> option, similar to what we have for lklfuse?

[jsoo1]

If I may suggest a fix: can we add a command line argument for the kernel command line options?

How about a --mb <memory_in_MB> option, similar to what we have for lklfuse?

Sure, whatever works. Though --kernel-parameters <anything> is a lot more flexible. What do you prefer?