Re: [PATCH v2] io_uring/net: ensure async prep handlers always initialize ->done_io

[Jens Axboe <axboe@kernel.dk>]

On 3/15/24 5:09 PM, Pavel Begunkov wrote:
> On 3/15/24 22:48, Jens Axboe wrote:
>> If we get a request with IOSQE_ASYNC set, then we first run the prep
>> async handlers. But if we then fail setting it up and want to post
>> a CQE with -EINVAL, we use ->done_io. This was previously guarded with
>> REQ_F_PARTIAL_IO, and the normal setup handlers do set it up before any
>> potential errors, but we need to cover the async setup too.
> 
> You can hit io_req_defer_failed() { opdef->fail(); }
> off of an early submission failure path where def->prep has
> not yet been called, I don't think the patch will fix the
> problem.
> 
> ->fail() handlers are fragile, maybe we should skip them
> if def->prep() wasn't called. Not even compile tested:

Yeah they are a mess honestly. Maybe we're better off just flagging it
like in your below patch, and avoid needing opcode handling for this.
Was going to suggest having a PREP_DONE flag, but it's better to have a
FAIL_EARLY and avoid needing to fiddle with it in the normal path.

-- 
Jens Axboe