[Bug 2137698] Re: ubuntu_blktrace_smoke_test fails on questing with rust coreutils

This bug is awaiting verification that the linux-
nvidia-6.17/6.17.0-1008.8 kernel in -proposed solves the problem. Please
test the kernel and update this bug with the results. If the problem is
solved, change the tag 'verification-needed-noble-linux-nvidia-6.17' to
'verification-done-noble-linux-nvidia-6.17'. If the problem still
exists, change the tag 'verification-needed-noble-linux-nvidia-6.17' to
'verification-failed-noble-linux-nvidia-6.17'.


If verification is not done by 5 working days from today, this fix will
be dropped from the source code, and this bug will be closed.


See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how
to enable and use -proposed. Thank you!


** Tags added: kernel-spammed-noble-linux-nvidia-6.17-v2 verification-needed-noble-linux-nvidia-6.17

--
You received this bug notification because you are subscribed to linux
in Ubuntu.
Matching subscriptions: Bgg, Bmail, Nb
https://bugs.launchpad.net/bugs/2137698

Title:
ubuntu_blktrace_smoke_test fails on questing with rust coreutils

Status in linux package in Ubuntu:
Invalid
Status in linux source package in Questing:
Fix Committed

Bug description:
ubuntu_blktrace_smoke_test fails on questing with rust coreutils
enabled. Executing with GNU coreutils makes the test pass.

2043 ID: com.canonical.kernel::ubuntu_blktrace_smoke_test/blktrace-smoke-test
2044 Category: com.canonical.plainbox::uncategorised
2045 ... 8< -------------------------------------------------------------------------
2046 PASSED (CONFIG_BLK_DEV_IO_TRACE=y in /boot/config-6.17.0-11-generic)
2047
2048 Using block device /dev/loop0 for path /tmp/cwd-hmzznri9.43f7bf5d91458d09dc595f27a2105298dbd7be4196a6f7444786bd783e1e92a1/mnt
2049
2050 Test regime:
2051 dd performing 65536 1K block writes
2052 looking for at least 1024 blktrace events
2053
2054 Sat Dec 20 19:52:16 UTC 2025: blktrace starting
2055 Sat Dec 20 19:52:16 UTC 2025: dd starting
2056 Sat Dec 20 19:52:16 UTC 2025: dd stopped
2057 Sat Dec 20 19:52:16 UTC 2025: waiting for 10 seconds
2058 Sat Dec 20 19:52:26 UTC 2025: blktrace being terminated
2059 Sat Dec 20 19:52:26 UTC 2025: blktrace terminated
2060 Sat Dec 20 19:52:26 UTC 2025: blktrace data parsed
2061
2062 FAILED (expecting at least 1024 block traces events from the dd process, got 0)
2063 FAILED (expecting at least 1024 block read traces events, got 0)
2064 FAILED (expecting at least 1024 block write traces events, got 0)

Regressing the issue, this is the breaking commit.

ommit 6c840c16c7cc0d3464ebf4a6d4c4f03763a45ff0
Author: Jan Kara <jack@suse.cz>
Date: Mon Sep 1 13:27:40 2025 +0200

    ext4: fail unaligned direct IO write with EINVAL

    BugLink: https://bugs.launchpad.net/bugs/2131259

    commit 963845748fe67125006859229487b45485564db7 upstream.

    Commit bc264fea0f6f ("iomap: support incremental iomap_iter advances")
    changed the error handling logic in iomap_iter(). Previously any error
    from iomap_dio_bio_iter() got propagated to userspace, after this commit
    if ->iomap_end returns error, it gets propagated to userspace instead of
    an error from iomap_dio_bio_iter(). This results in unaligned writes to
    ext4 to silently fallback to buffered IO instead of erroring out.

    Now returning ENOTBLK for DIO writes from ext4_iomap_end() seems
    unnecessary these days. It is enough to return ENOTBLK from
    ext4_iomap_begin() when we don't support DIO write for that particular
    file offset (due to hole).

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2137698/+subscriptions