Thread (3 messages) 3 messages, 2 authors, 2017-03-13

RE: blk_init_queue vs blk_alloc_queue

From: Umesh Patel <hidden>
Date: 2017-03-13 14:23:17

Currently we are using below API to initialize request queues for our devic=
e which is not calling blk_queue_split. But there is one more function blk_=
init_queues which is internally calling blk_queue_split.=20
So I was thinking to replace blk_alloc_queue with blk_init_queue but since =
I was not aware of impact I shoot this mail.
I wanted to use blk_queue_split to split BIO that block layer is sending si=
nce it it too large.


Thx


dev->osdev.queue =3D blk_alloc_queue(GFP_KERNEL);

-----Original Message-----
From: Christoph Hellwig [mailto:hch@infradead.org]=20
Sent: Monday, March 13, 2017 7:06 PM
To: Umesh Patel
Cc: linux-block@vger.kernel.org
Subject: Re: blk_init_queue vs blk_alloc_queue

On Mon, Mar 13, 2017 at 09:14:12AM +0000, Umesh Patel wrote:
Hello,
=20
I Would like to know difference between blk_init_queue and blk_alloc_queu=
e API.
Can anyone explain when to use what ?
The short answer is that you should neither.  The long answer is that you s=
hould send a link to your driver source if you want anyone to help you.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help