nvme: add --block-size
option to IO interface to bypass identify command
#2920
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Sometimes it is desired to only issue the underlying I/O command through submit_io (compare, read, write). Currently, at least one identify namespace is used to discover the logical block size.
This change adds a new
--block-size=
option to the submit_io interface. When provided, the--block-size=
is used as the logical block size in bytes, and the--metadata-size=
is also used as the metadata size in bytes. Storage and Reference tag checks are bypassed.(I am less certain about the
--metadata-size=
acting per-block, and how best to support or document it. One idea would be to take the metadata-size as a the total size across all blocks, and settingms = cfg.metadata_size / (cfg.block_count + 1)
. I'm open to ideas or changes here)