Conversation
| media_id: u32, | ||
| lba: Lba, | ||
| token: Option<NonNull<BlockIO2Token>>, | ||
| len: usize, |
There was a problem hiding this comment.
ping @Virv12 FYI. Unfortunately, I missed this in the review.
There was a problem hiding this comment.
I don't like this change, using a reference introduces a lifetime, implying we are going to use the buffer only until the lifetime is alive, which is wrong.
There has been a similar discussion on DiskIO2.
There was a problem hiding this comment.
Thanks a lot for pointing that out to me, I wasn't aware of this anymore!
There was a problem hiding this comment.
Thanks for catching that, I had forgotten about that too :)
phip1611
changed the title
|
@Virv12 you implemented the blocks i/o 2 protocol in #1841 - I tried to write a very basic integration test for it. The hardest part was to understand how the async interface of the |
phip1611
force-pushed
the
block-io
branch
2 times, most recently
from
10a0ec8 to
250d209
Compare
- improve the ergonomics of the Blocks I/O 2 protocol - figure out how to use it at all and put that into a very basic test
3 participants
add basic integration test for BlockIo and BlockIo2 protocols
Steps to Undraft
Checklist