When receiving a stream with the large block flag, activate feature

[AustinWise]

Motivation and Context

When sending snapshots back and forth between two datasets, it is currently possible to end up in a state where a receive fails with this error message:

cannot receive incremental stream: incremental send stream requires -L (--large-block), to match previous receive.

Adding the -L flag to the zfs send does not fix the problem.

See issue #18101 for details.

Description

When receiving a stream with the large block feature flag, activate the large_blocks feature in the destination dataset.

I think this is a reasonable way to solve the problem I'm having for a number of reasons:

• This makes the destination dataset more similar to the sending dataset, i.e. the large blocks feature will be active whether or not the dataset currently contains any large blocks.
• If users wish to avoid the large blocks feature being activated in the destination dataset, they can create a send stream without using either -L or --raw.

How Has This Been Tested?

Some new tests have been included in this PR. They fail when run on master currently and pass on this PR. They test that both full sends and incremental sends cause the large blocks feature to be activated in the destination dataset.

I ran these tests on Ubuntu 24.04, with kernel 6.14.0-37.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Performance enhancement (non-breaking change which improves efficiency)
• Code cleanup (non-breaking change which makes code smaller or more readable)
• Quality assurance (non-breaking change which makes the code more robust against bugs)
• Breaking change (fix or feature that would cause existing functionality to change)
• Library ABI change (libzfs, libzfs_core, libnvpair, libuutil and libzfsbootenv)
• Documentation (a change to man pages or other documentation)

Checklist:

• My code follows the OpenZFS code style requirements.
• I have updated the documentation accordingly.
• I have read the contributing document.
• I have added tests to cover my changes.
• I have run the ZFS Test Suite with this change applied.
• All commit messages are properly formatted and contain Signed-off-by.

[Copilot]

Pull request overview

This PR fixes issue #18101 where incremental ZFS send/receive operations can fail with a large_blocks feature mismatch error. The solution automatically activates the large_blocks feature on the destination dataset when receiving a stream that has the large block feature flag set, ensuring feature consistency between source and destination datasets.

Key Changes

• Modified dmu_recv_begin_sync() to activate the large_blocks feature on the destination dataset when receiving streams with the DMU_BACKUP_FEATURE_LARGE_BLOCKS flag
• Added two new test cases to verify the feature activation works correctly for both full and incremental sends

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
module/zfs/dmu_recv.c	Adds logic to activate the large_blocks feature on the destination dataset during receive operations, following the same pattern as the existing longname feature activation
tests/zfs-tests/tests/functional/rsend/send_large_blocks_inital.ksh	New test verifying that full send streams propagate large_blocks feature activation even when no large blocks are present
tests/zfs-tests/tests/functional/rsend/send_large_blocks_incremental.ksh	New test verifying that incremental sends activate the large_blocks feature when the source dataset has it activated

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[amotin]

While the actual change makes sense to me, please take a look on the AI comment about the test name, and also you should probably add the tests into Makefile.am and runfiles. Also I wonder why have you inserted the feature activation at the end? I would do it earlier, considering the flag was introduced earlier that longname.

[Copilot]

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[AustinWise]

While the actual change makes sense to me, please take a look on the AI comment about the test name, and also you should probably add the tests into Makefile.am and runfiles. Also I wonder why have you inserted the feature activation at the end? I would do it earlier, considering the flag was introduced earlier that longname.

Thanks, I addressed the misspelling of the test case and added the new test to the runfile an Makefile.am. I moved the activation of large block feature above the long name feature as requested. I also extended the comment to explain why the feature activation is needed.