Received-SPF: pass (google.com: domain of linux-kernel+bounces-65192-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: Nilay Shroff <nilay@linux.ibm.com>
To: john.g.garry@oracle.com
Cc: alan.adamson@oracle.com, axboe@kernel.dk, brauner@kernel.org,
        bvanassche@acm.org, dchinner@redhat.com, djwong@kernel.org, hch@lst.de,
        jack@suse.cz, jbongio@google.com, jejb@linux.ibm.com,
        kbusch@kernel.org, linux-block@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-nvme@lists.infradead.org, linux-scsi@vger.kernel.org,
        linux-xfs@vger.kernel.org, martin.petersen@oracle.com,
        ming.lei@redhat.com, ojaswin@linux.ibm.com, sagi@grimberg.me,
        tytso@mit.edu, viro@zeniv.linux.org.uk
Subject: Re: [PATCH v3 14/15] nvme: Support atomic writes
Date: Wed, 14 Feb 2024 17:57:19 +0530
Message-ID: <20240214122719.184946-1-nilay@linux.ibm.com>
In-Reply-To: <20240124113841.31824-15-john.g.garry@oracle.com>
References: <20240124113841.31824-15-john.g.garry@oracle.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable

>Support reading atomic write registers to fill in request_queue=0D
>properties.=0D
=0D
>Use following method to calculate limits:=0D
>atomic_write_max_bytes =3D flp2(NAWUPF ?: AWUPF)=0D
>atomic_write_unit_min =3D logical_block_size=0D
>atomic_write_unit_max =3D flp2(NAWUPF ?: AWUPF)=0D
>atomic_write_boundary =3D NABSPF=0D
=0D
In case the device doesn't support namespace atomic boundary size (i.e. NAB=
SPF =0D
is zero) then while merging atomic block-IO we should allow merge.=0D
 =0D
For example, while front/back merging the atomic block IO, we check whether=
 =0D
boundary is defined or not. In case if boundary is not-defined (i.e. it's z=
ero) =0D
then we simply reject merging ateempt (as implemented in =0D
rq_straddles_atomic_write_boundary()).  =0D
=0D
I am quoting this from NVMe spec (Command Set Specification, revision 1.0a,=
 =0D
Section 2.1.4.3) : "To ensure backwards compatibility, the values reported =
for =0D
AWUN, AWUPF, and ACWU shall be set such that they  are  supported  even  if=
  a  =0D
write  crosses  an  atomic  boundary.  If  a  controller  does  not  guaran=
tee =0D
atomicity across atomic boundaries, the controller shall set AWUN, AWUPF, a=
nd =0D
ACWU to 0h (1 LBA)." =0D
=0D
Thanks,=0D
--Nilay=0D
=0D