[PATCH v4 4/5] nvme: wire-up uring-cmd support for io-passthru on char-device.

[Jens Axboe]

On 5/6/22 2:28 AM, Christoph Hellwig wrote:
> On Thu, May 05, 2022 at 11:23:01AM -0600, Jens Axboe wrote:
>> The top three patches here have a proposed solution for the 3 issues
>> that I highlighted:
>>
>> https://git.kernel.dk/cgit/linux-block/log/?h=for-5.19/io_uring-passthrough
>>
>> Totally untested... Kanchan, can you take a look and see what you think?
>> They all need folding obviously, I just tried to do them separately.
>> Should also get tested :-)
> 
> I've also pushed out a tree based on this, which contains my little
> fixups that I'd suggest to be folded.  Totally untested and written
> while jetlagged:
> 
> http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/io_uring-passthrough
> 
> Note that while I tried to keep all my changes in separate patches, the
> main passthrough patch had conflicts during a rebase, which I had to
> fix up, but I tried to touch as little as possible.

Folded most of it, but I left your two meta data related patches as
separate as I they really should be separate. However, they need a
proper commit message and signed-off-by from you. It's these two:

https://git.kernel.dk/cgit/linux-block/commit/?h=for-5.19/io_uring-passthrough&id=b855a4458068722235bdf69688448820c8ddae8e
https://git.kernel.dk/cgit/linux-block/commit/?h=for-5.19/io_uring-passthrough&id=2be698bdd668daeb1aad2ecd516484a62e948547

Can you do those as proper patches?

I did not do your async_size changes, I think you're jetlagged eyes
missed that this isn't a sizeof thing on a flexible array, it's just the
offset of it. Hence for non-sqe128, the the async size is io_uring_sqe -
offsetof where pdu starts, and so forth.

-- 
Jens Axboe