I can confirm that after switching to Fedora 44, which includes the
upstream alloc_page fix, the issue is completely resolved.
I was pinned to 6.17 to work on an internal NVMe NTFS partition shared
with Windows, until Ubuntu updated to 7.0.0-29 and removed the working
6.17.0-35.35~24.04.1 kernel. Instead of reinstalling and holding 6.17, I
switched distros. The exact same hardware and parallel build workload on
that shared NVMe drive now completes without any hangs or kernel panics.
This appears consistent with the upstream fix identified in comment #5:
70d3855594cf6e8791970714b65cac3202d6160e ("ntfs3: Allocate iomap
inline_data using alloc_page").
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2161222
Title:
Kernel BUG at fs/iomap/buffered-io.c:1061 (iomap_write_end) via
ntfs_file_write_iter on 7.0.0-28-generic — regression, reproducible
under parallel build load on ntfs3
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-hwe-7.0/+bug/2161222/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
