Spdk/module/bdev/raid/bdev_raid_rpc.c

384 lines
10 KiB
C
Raw Normal View History

/* SPDX-License-Identifier: BSD-3-Clause
* Copyright (C) 2018 Intel Corporation.
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* All rights reserved.
*/
#include "spdk/rpc.h"
#include "spdk/bdev.h"
#include "bdev_raid.h"
#include "spdk/util.h"
#include "spdk/string.h"
#include "spdk/log.h"
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
#include "spdk/env.h"
#define RPC_MAX_BASE_BDEVS 255
/*
* Input structure for bdev_raid_get_bdevs RPC
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
struct rpc_bdev_raid_get_bdevs {
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* category - all or online or configuring or offline */
char *category;
};
/*
* brief:
* free_rpc_bdev_raid_get_bdevs function frees RPC bdev_raid_get_bdevs related parameters
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params:
* req - pointer to RPC request
* returns:
* none
*/
static void
free_rpc_bdev_raid_get_bdevs(struct rpc_bdev_raid_get_bdevs *req)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
free(req->category);
}
/*
* Decoder object for RPC get_raids
*/
static const struct spdk_json_object_decoder rpc_bdev_raid_get_bdevs_decoders[] = {
{"category", offsetof(struct rpc_bdev_raid_get_bdevs, category), spdk_json_decode_string},
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
};
/*
* brief:
* rpc_bdev_raid_get_bdevs function is the RPC for rpc_bdev_raid_get_bdevs. This is used to list
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* all the raid bdev names based on the input category requested. Category should be
* one of "all", "online", "configuring" or "offline". "all" means all the raids
* whether they are online or configuring or offline. "online" is the raid bdev which
* is registered with bdev layer. "configuring" is the raid bdev which does not have
* full configuration discovered yet. "offline" is the raid bdev which is not
* registered with bdev as of now and it has encountered any error or user has
* requested to offline the raid.
* params:
* request - pointer to json rpc request
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params - pointer to request parameters
* returns:
* none
*/
static void
rpc_bdev_raid_get_bdevs(struct spdk_jsonrpc_request *request,
const struct spdk_json_val *params)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
struct rpc_bdev_raid_get_bdevs req = {};
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
struct spdk_json_write_ctx *w;
struct raid_bdev *raid_bdev;
enum raid_bdev_state state;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
if (spdk_json_decode_object(params, rpc_bdev_raid_get_bdevs_decoders,
SPDK_COUNTOF(rpc_bdev_raid_get_bdevs_decoders),
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
&req)) {
spdk_jsonrpc_send_error_response(request, SPDK_JSONRPC_ERROR_PARSE_ERROR,
"spdk_json_decode_object failed");
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
state = raid_bdev_str_to_state(req.category);
if (state == RAID_BDEV_STATE_MAX && strcmp(req.category, "all") != 0) {
spdk_jsonrpc_send_error_response(request, -EINVAL, spdk_strerror(EINVAL));
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
w = spdk_jsonrpc_begin_result(request);
spdk_json_write_array_begin(w);
/* Get raid bdev list based on the category requested */
TAILQ_FOREACH(raid_bdev, &g_raid_bdev_list, global_link) {
if (raid_bdev->state == state || state == RAID_BDEV_STATE_MAX) {
spdk_json_write_object_begin(w);
spdk_json_write_named_string(w, "name", raid_bdev->bdev.name);
raid_bdev_write_info_json(raid_bdev, w);
spdk_json_write_object_end(w);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
}
spdk_json_write_array_end(w);
spdk_jsonrpc_end_result(request, w);
cleanup:
free_rpc_bdev_raid_get_bdevs(&req);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
SPDK_RPC_REGISTER("bdev_raid_get_bdevs", rpc_bdev_raid_get_bdevs, SPDK_RPC_RUNTIME)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/*
* Base bdevs in RPC bdev_raid_create
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
struct rpc_bdev_raid_create_base_bdevs {
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* Number of base bdevs */
size_t num_base_bdevs;
/* List of base bdevs names */
char *base_bdevs[RPC_MAX_BASE_BDEVS];
};
/*
* Input structure for RPC rpc_bdev_raid_create
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
struct rpc_bdev_raid_create {
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* Raid bdev name */
char *name;
/* RAID strip size in KB */
uint32_t strip_size_kb;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* RAID raid level */
enum raid_level level;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* Base bdevs information */
struct rpc_bdev_raid_create_base_bdevs base_bdevs;
/* UUID for this raid bdev */
char *uuid;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
};
/*
* brief:
* free_rpc_bdev_raid_create function is to free RPC bdev_raid_create related parameters
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params:
* req - pointer to RPC request
* returns:
* none
*/
static void
free_rpc_bdev_raid_create(struct rpc_bdev_raid_create *req)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
size_t i;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
free(req->name);
free(req->uuid);
for (i = 0; i < req->base_bdevs.num_base_bdevs; i++) {
free(req->base_bdevs.base_bdevs[i]);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
}
/*
* Decoder function for RPC bdev_raid_create to decode raid level
*/
static int
decode_raid_level(const struct spdk_json_val *val, void *out)
{
int ret;
char *str = NULL;
enum raid_level level;
ret = spdk_json_decode_string(val, &str);
if (ret == 0 && str != NULL) {
level = raid_bdev_str_to_level(str);
if (level == INVALID_RAID_LEVEL) {
ret = -EINVAL;
} else {
*(enum raid_level *)out = level;
}
}
free(str);
return ret;
}
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/*
* Decoder function for RPC bdev_raid_create to decode base bdevs list
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
static int
decode_base_bdevs(const struct spdk_json_val *val, void *out)
{
struct rpc_bdev_raid_create_base_bdevs *base_bdevs = out;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
return spdk_json_decode_array(val, spdk_json_decode_string, base_bdevs->base_bdevs,
RPC_MAX_BASE_BDEVS, &base_bdevs->num_base_bdevs, sizeof(char *));
}
/*
* Decoder object for RPC bdev_raid_create
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
static const struct spdk_json_object_decoder rpc_bdev_raid_create_decoders[] = {
{"name", offsetof(struct rpc_bdev_raid_create, name), spdk_json_decode_string},
{"strip_size_kb", offsetof(struct rpc_bdev_raid_create, strip_size_kb), spdk_json_decode_uint32, true},
{"raid_level", offsetof(struct rpc_bdev_raid_create, level), decode_raid_level},
{"base_bdevs", offsetof(struct rpc_bdev_raid_create, base_bdevs), decode_base_bdevs},
{"uuid", offsetof(struct rpc_bdev_raid_create, uuid), spdk_json_decode_string, true},
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
};
/*
* brief:
* rpc_bdev_raid_create function is the RPC for creating RAID bdevs. It takes
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* input as raid bdev name, raid level, strip size in KB and list of base bdev names.
* params:
* request - pointer to json rpc request
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params - pointer to request parameters
* returns:
* none
*/
static void
rpc_bdev_raid_create(struct spdk_jsonrpc_request *request,
const struct spdk_json_val *params)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
struct rpc_bdev_raid_create req = {};
struct raid_bdev *raid_bdev;
int rc;
size_t i;
struct spdk_uuid *uuid = NULL;
struct spdk_uuid decoded_uuid;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
if (spdk_json_decode_object(params, rpc_bdev_raid_create_decoders,
SPDK_COUNTOF(rpc_bdev_raid_create_decoders),
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
&req)) {
spdk_jsonrpc_send_error_response(request, SPDK_JSONRPC_ERROR_PARSE_ERROR,
"spdk_json_decode_object failed");
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
if (req.uuid) {
if (spdk_uuid_parse(&decoded_uuid, req.uuid)) {
spdk_jsonrpc_send_error_response(request, SPDK_JSONRPC_ERROR_INVALID_PARAMS,
"Failed to parse bdev UUID");
goto cleanup;
}
uuid = &decoded_uuid;
}
rc = raid_bdev_create(req.name, req.strip_size_kb, req.base_bdevs.num_base_bdevs,
req.level, &raid_bdev, uuid);
if (rc != 0) {
spdk_jsonrpc_send_error_response_fmt(request, rc,
"Failed to create RAID bdev %s: %s",
req.name, spdk_strerror(-rc));
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
for (i = 0; i < req.base_bdevs.num_base_bdevs; i++) {
const char *base_bdev_name = req.base_bdevs.base_bdevs[i];
rc = raid_bdev_add_base_device(raid_bdev, base_bdev_name, i);
if (rc == -ENODEV) {
SPDK_DEBUGLOG(bdev_raid, "base bdev %s doesn't exist now\n", base_bdev_name);
} else if (rc != 0) {
raid_bdev_delete(raid_bdev, NULL, NULL);
spdk_jsonrpc_send_error_response_fmt(request, rc,
"Failed to add base bdev %s to RAID bdev %s: %s",
base_bdev_name, req.name,
spdk_strerror(-rc));
goto cleanup;
}
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
spdk_jsonrpc_send_bool_response(request, true);
cleanup:
free_rpc_bdev_raid_create(&req);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
SPDK_RPC_REGISTER("bdev_raid_create", rpc_bdev_raid_create, SPDK_RPC_RUNTIME)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/*
* Input structure for RPC deleting a raid bdev
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
struct rpc_bdev_raid_delete {
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/* raid bdev name */
char *name;
};
/*
* brief:
* free_rpc_bdev_raid_delete function is used to free RPC bdev_raid_delete related parameters
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params:
* req - pointer to RPC request
* params:
* none
*/
static void
free_rpc_bdev_raid_delete(struct rpc_bdev_raid_delete *req)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
free(req->name);
}
/*
* Decoder object for RPC raid_bdev_delete
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
*/
static const struct spdk_json_object_decoder rpc_bdev_raid_delete_decoders[] = {
{"name", offsetof(struct rpc_bdev_raid_delete, name), spdk_json_decode_string},
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
};
struct rpc_bdev_raid_delete_ctx {
struct rpc_bdev_raid_delete req;
struct spdk_jsonrpc_request *request;
};
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
/*
* brief:
* params:
* cb_arg - pointer to the callback context.
* rc - return code of the deletion of the raid bdev.
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* returns:
* none
*/
static void
bdev_raid_delete_done(void *cb_arg, int rc)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
struct rpc_bdev_raid_delete_ctx *ctx = cb_arg;
struct spdk_jsonrpc_request *request = ctx->request;
if (rc != 0) {
SPDK_ERRLOG("Failed to delete raid bdev %s (%d): %s\n",
ctx->req.name, rc, spdk_strerror(-rc));
spdk_jsonrpc_send_error_response(request, SPDK_JSONRPC_ERROR_INTERNAL_ERROR,
spdk_strerror(-rc));
goto exit;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
spdk_jsonrpc_send_bool_response(request, true);
exit:
free_rpc_bdev_raid_delete(&ctx->req);
free(ctx);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
/*
* brief:
* rpc_bdev_raid_delete function is the RPC for deleting a raid bdev. It takes raid
* name as input and delete that raid bdev including freeing the base bdev
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* resources.
* params:
* request - pointer to json rpc request
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
* params - pointer to request parameters
* returns:
* none
*/
static void
rpc_bdev_raid_delete(struct spdk_jsonrpc_request *request,
const struct spdk_json_val *params)
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
{
struct rpc_bdev_raid_delete_ctx *ctx;
struct raid_bdev *raid_bdev;
ctx = calloc(1, sizeof(*ctx));
if (!ctx) {
spdk_jsonrpc_send_error_response(request, -ENOMEM, spdk_strerror(ENOMEM));
return;
}
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
if (spdk_json_decode_object(params, rpc_bdev_raid_delete_decoders,
SPDK_COUNTOF(rpc_bdev_raid_delete_decoders),
&ctx->req)) {
spdk_jsonrpc_send_error_response(request, SPDK_JSONRPC_ERROR_PARSE_ERROR,
"spdk_json_decode_object failed");
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
raid_bdev = raid_bdev_find_by_name(ctx->req.name);
if (raid_bdev == NULL) {
spdk_jsonrpc_send_error_response_fmt(request, -ENODEV,
"raid bdev %s not found",
ctx->req.name);
goto cleanup;
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
ctx->request = request;
raid_bdev_delete(raid_bdev, bdev_raid_delete_done, ctx);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
return;
cleanup:
free_rpc_bdev_raid_delete(&ctx->req);
free(ctx);
bdev: add raid bdev module Raid module: ============ - SPDK raid bdev module is a new bdev module which is responsible for striping various NVMe devices and expose the raid bdev to bdev layer which would enhance the performance and capacity. - It can support theoretically 256 base devices (currently it is being tested max upto 8 base devices) - Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is supported. Most of the current testing is focused on 64KB strip size. - New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs" are introduced to configure raid bdev dynamically in a running SPDK system. - Currently raid bdev configuration parameters are persisted in the current SPDK configuration file for across reboot support. DDF will be introduced later. High level testing done: ======================= - Raid bdev is created with 8 base NVMe devices via configuration file and is exposed to initiator via existing methods. Initiator is able to see a single NVMe namespace with capacity equal to sum of the minimum capacities of 8 devices. Initiator was able to run raw read/write workload, file system workload etc (tested with XFS file system workload). - Multiple raid bdevs are also created and exposed to initiator and tested with file system and other workloads for read/write IO. - LVS / LVOL are created over raid bdev and exposed to initiator. Testing was done for raw read/write workloads and XFS file system workloads. - RPC testing is done where on the running SPDK system raid bdevs are created out of NVMe base devices. These raid bdevs (and LVOLs over raid bdevs) are then exposed to initiator and IO workload was tested for raw read/write and XFS file system workload. - RPC testing is done for delete raid bdevs where all raid bdevs are deleted in running SPDK system. - RPC testing is done for get raid bdevs where existing list of raid bdev names is printed (it can be all raid bdevs or only online or only configuring or only offline). - RPC testing is done where raid bdevs and underlying NVMe devices relationship was returned in JSON RPC commands Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8 Signed-off-by: Kunal Sablok <kunal.sablok@intel.com> Reviewed-on: https://review.gerrithub.io/410484 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2018-05-08 11:30:29 +00:00
}
SPDK_RPC_REGISTER("bdev_raid_delete", rpc_bdev_raid_delete, SPDK_RPC_RUNTIME)