lib/nvmf: Fix next_cntlid misnomer

Within a subsystem, we track 'next_cntlid' which helps
in choosing controller id when a new controller is added.
This is currently a misnomer, as we actually incre

lib/nvmf: Fix next_cntlid misnomer

Within a subsystem, we track 'next_cntlid' which helps
in choosing controller id when a new controller is added.
This is currently a misnomer, as we actually increment it
before using it. This patch fixes that.

Change-Id: I739bc16055d697a771c526adaa457054a6edca46
Signed-off-by: Abhineet Pandey <abhineet.pandey@nutanix.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25189
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

gscheduler: fix divide-by-zero error

If gscheduler is activated and gets a chance to balance() before all
cores have an spdk_thread assigned, the total_tsc for that core would
be 0, and calculate_bu

gscheduler: fix divide-by-zero error

If gscheduler is activated and gets a chance to balance() before all
cores have an spdk_thread assigned, the total_tsc for that core would
be 0, and calculate_busy_pct() would divide by zero causing a SIGFPE.

Fixes issue #3550.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I5b21bbbd3fb7448edd6d929a36a77b9fbd9d0e43
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25200
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Michal Berger <michal.berger@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Alliswell <hisunzhenliang@outlook.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>

show more ...

test/compress: Add missing io pattern arg to run_bdeveprf()

412fced1b5 forgot to update the last instance that is bound to
nightly tests.

Change-Id: I49215ee921875aba13705022026141e3f67d3d3e
Signed

test/compress: Add missing io pattern arg to run_bdeveprf()

412fced1b5 forgot to update the last instance that is bound to
nightly tests.

Change-Id: I49215ee921875aba13705022026141e3f67d3d3e
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25197
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

event: move struct spdk_lw_thread to internal header

This does not need to be exposed through public API.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I7f2f1bf5d47e9b0ded011a4f3f83

event: move struct spdk_lw_thread to internal header

This does not need to be exposed through public API.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I7f2f1bf5d47e9b0ded011a4f3f832320b106b3ec
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25032
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

module/raid: Assign bdev_io buffers to raid_io

If an application does not allocate data and metadata
buffers for the read operation and relies on the generic
bdev layer to do the allocations, the da

module/raid: Assign bdev_io buffers to raid_io

If an application does not allocate data and metadata
buffers for the read operation and relies on the generic
bdev layer to do the allocations, the data iovecs
and metadata pointer was not copied to the raid_io
and not pushed down the I/O stack.

This fix copies the pointers once the allocations are complete.
If the application did provide the buffers, this flow will
still work since those pointers are already in the bdev_io
and raid_io.

Change-Id: Ib9e212ec97b59c0ed57f2e48a35fdf8fd2a21a52
Signed-off-by: Slawomir Ptak <slawomir.ptak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24961
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

bdevperf: Allocate an md buffer for verify op

For the read operation in the verify operation flow,
if the generic bdev layer cannot allocate the buffers, the data
buffers are allocated by bdevperf.

bdevperf: Allocate an md buffer for verify op

For the read operation in the verify operation flow,
if the generic bdev layer cannot allocate the buffers, the data
buffers are allocated by bdevperf.

This patch adds metadata buffers allocation to be consistent with
the existing data buffers allocation.

Change-Id: Id065c6ec80048e600520c45c097ab63302781721
Signed-off-by: Slawomir Ptak <slawomir.ptak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24960
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot

show more ...

spdk_log: introduce spdk_log_ext API

Current spdk_log implementation allows to provide only log function.
In some cases it might not be enough because user have to take care
of managing resources as

spdk_log: introduce spdk_log_ext API

Current spdk_log implementation allows to provide only log function.
In some cases it might not be enough because user have to take care
of managing resources associated with logger.
Extended API supports providing open and close functions
with set context which will be called by SPDK in corresponded places.

Fixes #3480

Change-Id: Ia1d41fe3b414e2f603c2ae6612b7f9eb8d29bbb7
Signed-off-by: Vasilii Ivanov <iwanovvvasilij@gmail.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24847
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

bdev/compress: unmap support.

Change-Id: Ibdb98f02725615532ebac2e5536d5db55d76c886
Signed-off-by: Yalong Wang <yalong9@staff.sina.com.cn>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/247

bdev/compress: unmap support.

Change-Id: Ibdb98f02725615532ebac2e5536d5db55d76c886
Signed-off-by: Yalong Wang <yalong9@staff.sina.com.cn>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24798
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

nvme: rename spdk_nvme_ctrlr_aer_completion_list

The structure spdk_nvme_ctrlr_aer_completion_list has been renamed
to spdk_nvme_ctrlr_aer_completion, these structs go into a list but
aren't the lis

nvme: rename spdk_nvme_ctrlr_aer_completion_list

The structure spdk_nvme_ctrlr_aer_completion_list has been renamed
to spdk_nvme_ctrlr_aer_completion, these structs go into a list but
aren't the list itself.

Change-Id: I7919b9a37860001349b9379821a8b6f2fef4a16e
Signed-off-by: Damiano Cipriani <damiano.cipriani@suse.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25168
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

blob: use uint64_t for cluster start page index

When creating blobstore on devices with more than 16 TiB space, the
start page index of clusters can execeed UINT32_MAX and result in
overflow in uint

blob: use uint64_t for cluster start page index

When creating blobstore on devices with more than 16 TiB space, the
start page index of clusters can execeed UINT32_MAX and result in
overflow in uint32_t representation.

Use uint64_t instead of uint32_t to avoid the overflow.

Change-Id: I8e2586251b1b31ef7520b870af537c3f5792508e
Signed-off-by: Jinlong Chen <chenjinlong.cjl@alibaba-inc.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24619
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>

show more ...

fsdev/aio: aio_io_poll: correct return value

Change-Id: If2f0640a3d81f6a117d967c1c22aa028a85fd9e2
Signed-off-by: Anton Nayshtut <anayshtut@nvidia.com>
Reviewed-on: https://review.spdk.io/gerrit/c/sp

fsdev/aio: aio_io_poll: correct return value

Change-Id: If2f0640a3d81f6a117d967c1c22aa028a85fd9e2
Signed-off-by: Anton Nayshtut <anayshtut@nvidia.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25156
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Ben Walker <ben@nvidia.com>

show more ...

test/vhost: Attempt to verify vhost status upon termination

This is a bit problematic, since vhost_kill() operates on PID files,
meaning, that it may attempt to kill a process which doesn't belong
t

test/vhost: Attempt to verify vhost status upon termination

This is a bit problematic, since vhost_kill() operates on PID files,
meaning, that it may attempt to kill a process which doesn't belong
to the instance that's calling it. So to make it somewhat sensible,
we need to check first if vhost is our child - only then attempt
to verify it's status.

Change-Id: I9024ab19f62652472ce9113bc3f1cf7326d14454
Signed-off-by: Michal Berger <michal.berger@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25149
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>

show more ...

event: don't print core unlock warnings if cores were never locked

If we fail app init path, we try to unlock any cores that happened
to be locked. But the code path was only checking for -1, and by

event: don't print core unlock warnings if cores were never locked

If we fail app init path, we try to unlock any cores that happened
to be locked. But the code path was only checking for -1, and by
default the core lock array would have 0 values. So only try to
unlock if the core lock fd is not -1 nor 0.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I3a2f7bb8bf11c5bc3d8c59b213a356aff87b30d0
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25059
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

event: help user if env initialization fails as non-root

env_dpdk usually cannot initialize successfully without running as
root. If env init fails and the uid isn't root, print an error message
to

event: help user if env initialization fails as non-root

env_dpdk usually cannot initialize successfully without running as
root. If env init fails and the uid isn't root, print an error message
to help the user if they forgot to run as root.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I15be366ed2c7f67a9f99232ad2dbdf641b58b8d6
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25058
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

event: move function declarations to inside of extern "C" guard

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I7f804752e30b00930a86e430c907afdce083f780
Reviewed-on: https://review.sp

event: move function declarations to inside of extern "C" guard

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I7f804752e30b00930a86e430c907afdce083f780
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25031
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Anisa Su <anisa.su@samsung.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

configure: clarify optional parameter for --with-rdma

The DIR was likely copied from some other option, but for --with-rdma
it is a provider name, not a directory.

Signed-off-by: Jim Harris <jim.ha

configure: clarify optional parameter for --with-rdma

The DIR was likely copied from some other option, but for --with-rdma
it is a provider name, not a directory.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Icd262a823db7bff144b1b3c4ebfc27196902d65f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25126
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

event: shrink size of event message pool

These spdk_event structures are used for passing messages between
reactors. When it was first created, we did not have the concept
of spdk_threads, so this p

event: shrink size of event message pool

These spdk_event structures are used for passing messages between
reactors. When it was first created, we did not have the concept
of spdk_threads, so this pool needed to be very large since all
"per-thread" messaging was done using this pool. But now spdk_threads
are the primary mechanism for per-thread messaging. This event
pool in the reactor code is only for messages sent between cores which
is basically for the scheduler and interrupt mode.

256K pool size was way too big, and ends up consuming over 50MB of
memory. Shrink it to 16K, which is probably still an order of
magnitude more than needed, but still significantly cuts the amount
of memory wasted by this mempool.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Id46aeae7fd6a495d90ab40ed90f2dcb89b6ef885
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24998
Reviewed-by: Pawel Piatek <pawelx.piatek@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

fsdev/aio: add support for null IOs

This patch adds support for null IOs in the AIO fsdev by introducing
a new RPC flag (--enable-skip-rw) during fsdev creation.

When enabled, the AIO fsdev will sk

fsdev/aio: add support for null IOs

This patch adds support for null IOs in the AIO fsdev by introducing
a new RPC flag (--enable-skip-rw) during fsdev creation.

When enabled, the AIO fsdev will skip read/write operations, similar
to how the null bdev operates in the bdev module.

This feature facilitates performance testing across the entire stack
without actual data read/write operations. Implementing this within
the fsdev/aio module, rather than creating a separate fsdev/null module,
avoids unnecessary code duplication and keeps the AIO fsdev focused as a
comprehensive example demonstrating various features, including null IO
capability.

Change-Id: Id7e84711628f23e6faa9128f9e428f759e0032c9
Signed-off-by: Yoray Zack <yorayz@nvidia.com>
Signed-off-by: Vishwanath Venkatesan <vvenkatesan@nvidia.com>
Signed-off-by: Ben Walker <ben@nvidia.com>
Signed-off-by: Anton Nayshtut <anayshtut@nvidia.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24563
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeliu@tencent.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

lib/reduce: Check return code of read superblock

Change-Id: I06e34a5f247bec1868e9c64933d34b0504b64e4a
Signed-off-by: Yankun Li <yankun@staff.sina.com>
Reviewed-on: https://review.spdk.io/gerrit/c/sp

lib/reduce: Check return code of read superblock

Change-Id: I06e34a5f247bec1868e9c64933d34b0504b64e4a
Signed-off-by: Yankun Li <yankun@staff.sina.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25154
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

bdev/nvme: controller failover/multipath doc change

Added information about implemented change of the default controller
mode, now multipath is a new default.
The consistency of configuration of con

bdev/nvme: controller failover/multipath doc change

Added information about implemented change of the default controller
mode, now multipath is a new default.
The consistency of configuration of controllers with the same name is
now verified - all must be configured either for multipath or failover,
it is not possible to configure e.g one for multipath and second for
failover mode.
Added information about spdk_bdev_nvme_create() API function parameter
list change.

Change-Id: I76e95c19a5b97eb5dd74485964d00d6705c0c559
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25030
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

bdev/nvme: removed 'multipath' param from spdk_bdev_nvme_create()

Changed 'spdk_bdev_nvme_create' API function, the 'multipath prameter
was removed as it is redundant to 'multipath' field in
spdk_bd

bdev/nvme: removed 'multipath' param from spdk_bdev_nvme_create()

Changed 'spdk_bdev_nvme_create' API function, the 'multipath prameter
was removed as it is redundant to 'multipath' field in
spdk_bdev_nvme_ctrlr_opts structure passed as a parameter to this
function. If multipathing shall be enabled for nvme bdev, the
'bdev_opts.multipath' field shall be set to 'true'. When
'bdev_opts.multipath' is set to 'false', failover mode is enabled.

Change-Id: I3ba5ac159b4c97a4d8566257b0505b1154a158f8
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25161
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>

show more ...

bdev/nvme: changed default config to multipath

If no '-x' option is specified in bdev_nvme_attach_controller RPC call,
the multipath mode is now assigned as a default.

Change-Id: Ic2f67ed61e2348910

bdev/nvme: changed default config to multipath

If no '-x' option is specified in bdev_nvme_attach_controller RPC call,
the multipath mode is now assigned as a default.

Change-Id: Ic2f67ed61e23489106f02c5529d5fe3858e10943
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25029
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

bdev/nvme: ctrl config consistency check

All controllers created with the same name shall be configured either
for multipath or for failover. Otherwise we have configuration mismatch.
This patch add

bdev/nvme: ctrl config consistency check

All controllers created with the same name shall be configured either
for multipath or for failover. Otherwise we have configuration mismatch.
This patch adds cofiguration consistency check, and all controllers
created with the same name will be forced to have consistent setting,
either '-x multipath' or '-x failover'. No mixing of '-x' options is
allowed.
Deprecation warning has been removed as we are now implementing the
advertised change.

Change-Id: I1bc472f95950e4326ca67bed33a1bf9523683843
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25028
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot

show more ...

bdev/nvme: Change spdk_bdev_reset() to succeed if at least one nvme_ctrlr is reconnected

This patch fixes issue with spdk_bdev_reset() on multipath nvme bdev
with at least one but not all failed nvm

bdev/nvme: Change spdk_bdev_reset() to succeed if at least one nvme_ctrlr is reconnected

This patch fixes issue with spdk_bdev_reset() on multipath nvme bdev
with at least one but not all failed nvme_ctrlrs. For multipath nvme bdev
reset sequentially reconnects all nvme_ctrlrs. Before
this patch, if at least one of nvme_ctrlrs can't be reconnected, the whole
reset request is completed with errors. It doesn't make sense because
bdev may still be operational. This patch changes behaviour to
complete reset request successfully if at least one nvme_ctrlr was
reconnected. Reset will still be failed if all nvme_ctrlrs failed to
reconnect.

Change-Id: I992a9a44f26f45970f9daa120a93be9bb4604091
Signed-off-by: Evgeniy Kochetov <evgeniik@nvidia.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24974
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

test/blob: Add BLOCKLEN macro in blob_ut

blob_ut has a lot of places hardcoded with 4096, mostly used for
data payload operations. These are replaced with BLOCKLEN macro.

It is similar to DEV_BUFFE

test/blob: Add BLOCKLEN macro in blob_ut

blob_ut has a lot of places hardcoded with 4096, mostly used for
data payload operations. These are replaced with BLOCKLEN macro.

It is similar to DEV_BUFFER_BLOCKLEN, but in shortened form.

This is the first patch of the series where metadata support
will be added to blobstore for large IU devices.

Currently SPDK blobstore depends on 4KB atomicity of metadata.
So, Metadata page size will be made variable based on the physical
block size reported by the underlying device. This will enable
blobstore to store metadata in IU-sized chunks, not always 4KB.

Change-Id: I9fb166a32b78b43ecd1db752c24835cee441e665
Signed-off-by: Atul Malakar <a.malakar@samsung.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/25023
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...