[02/26] block: add REQ_HIPRI_ASYNC

Message ID	20181207222016.29387-3-axboe@kernel.dk (mailing list archive)
State	New, archived
Headers	show Return-Path: <linux-block-owner@kernel.org> From: Jens Axboe <axboe@kernel.dk> To: linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-aio@kvack.org Cc: hch@lst.de, jmoyer@redhat.com, clm@fb.com, Jens Axboe <axboe@kernel.dk> Subject: [PATCH 02/26] block: add REQ_HIPRI_ASYNC Date: Fri, 7 Dec 2018 15:19:52 -0700 Message-Id: <20181207222016.29387-3-axboe@kernel.dk> In-Reply-To: <20181207222016.29387-1-axboe@kernel.dk> References: <20181207222016.29387-1-axboe@kernel.dk> Sender: linux-block-owner@vger.kernel.org Precedence: bulk
Series	[01/26] fs: add an iopoll method to struct file_operations \| expand [01/26] fs: add an iopoll method to struct file_operations [02/26] block: add REQ_HIPRI_ASYNC [03/26] block: wire up block device iopoll method [04/26] block: use REQ_HIPRI_ASYNC for non-sync polled IO [05/26] iomap: wire up the iopoll method [06/26] aio: use assigned completion handler [07/26] aio: separate out ring reservation from req allocation [08/26] aio: don't zero entire aio_kiocb aio_get_req() [09/26] aio: only use blk plugs for > 2 depth submissions [10/26] aio: use iocb_put() instead of open coding it [11/26] aio: split out iocb copy from io_submit_one() [12/26] aio: abstract out io_event filler helper [13/26] aio: add io_setup2() system call [14/26] aio: add support for having user mapped iocbs [15/26] aio: support for IO polling [16/26] aio: add submission side request cache [17/26] fs: add fget_many() and fput_many() [18/26] aio: use fget/fput_many() for file references [19/26] aio: split iocb init from allocation [20/26] aio: batch aio_kiocb allocation [21/26] block: add BIO_HOLD_PAGES flag [22/26] block: implement bio helper to add iter bvec pages to bio [23/26] fs: add support for mapping an ITER_BVEC for O_DIRECT [24/26] aio: add support for pre-mapped user IO buffers [25/26] aio: split old ring complete out from aio_complete() [26/26] aio: add support for submission/completion rings

Message ID

20181207222016.29387-3-axboe@kernel.dk (mailing list archive)

State

New, archived

Headers

From: Jens Axboe <axboe@kernel.dk>
To: linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        linux-aio@kvack.org
Cc: hch@lst.de, jmoyer@redhat.com, clm@fb.com,
        Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 02/26] block: add REQ_HIPRI_ASYNC
Date: Fri,  7 Dec 2018 15:19:52 -0700
Message-Id: <20181207222016.29387-3-axboe@kernel.dk>
In-Reply-To: <20181207222016.29387-1-axboe@kernel.dk>
References: <20181207222016.29387-1-axboe@kernel.dk>
Sender: linux-block-owner@vger.kernel.org
Precedence: bulk

Series

[01/26] fs: add an iopoll method to struct file_operations | expand

Commit Message

Jens Axboe Dec. 7, 2018, 10:19 p.m. UTC

For the upcoming async polled IO, we can't sleep allocating requests.
If we do, then we introduce a deadlock where the submitter already
has async polled IO in-flight, but can't wait for them to complete
since polled requests must be active found and reaped.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/blk_types.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 46c005d601ac..921d734d6b5d 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -347,6 +347,7 @@  enum req_flag_bits {
 #define REQ_NOWAIT		(1ULL << __REQ_NOWAIT)
 #define REQ_NOUNMAP		(1ULL << __REQ_NOUNMAP)
 #define REQ_HIPRI		(1ULL << __REQ_HIPRI)
+#define REQ_HIPRI_ASYNC		(REQ_HIPRI | REQ_NOWAIT)
 
 #define REQ_DRV			(1ULL << __REQ_DRV)
 #define REQ_SWAP		(1ULL << __REQ_SWAP)

[02/26] block: add REQ_HIPRI_ASYNC

Commit Message

Patch