From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Niklas Cassel Subject: [PATCH 0/4] Improve fio max open zones handling Date: Wed, 12 May 2021 22:36:32 +0000 Message-ID: <20210512223615.17239-1-Niklas.Cassel@wdc.com> Content-Language: en-US Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 To: "axboe@kernel.dk" Cc: "fio@vger.kernel.org" , Damien Le Moal , Niklas Cassel List-ID: From: Niklas Cassel Improve max open zones handling by introducing a new IO engine operation, which can be implemented by ioengines. Also provide a default implementation for Linux, which parses sysfs for the max open zones value. Update libzbc ioengine to utilize the new callback. Having this callback has the two following advantages: 1) If the user forgets to specify a --max_open_zones value, we will automatically use the maximum the drive supports. (Instead of getting errors because fio tries to open too many zones.) 2) If the user specifies a --max_open_zones value that is too big, we will return an error and fio will exit with a proper error message. (Instead of getting errors because fio tries to open too many zones.) Kind regards, Niklas Niklas Cassel (4): zbd: only put an upper limit on max open zones once oslib/linux-blkzoned: move sysfs reading into its own function ioengines: add get_max_open_zones zoned block device operation engines/libzbc: add support for the get_max_open_zones io op engines/libzbc.c | 21 +++++++++ engines/skeleton_external.c | 13 ++++++ ioengines.h | 4 +- oslib/blkzoned.h | 7 +++ oslib/linux-blkzoned.c | 82 ++++++++++++++++++++++++---------- zbd.c | 88 ++++++++++++++++++++++++++++++++++--- 6 files changed, 186 insertions(+), 29 deletions(-) --=20 2.25.1