ivampiresp/Spdk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

misc: Fix spelling mistakes

Browse Source 
Found with misspell-fixer.

Signed-off-by: Michal Berger <michal.berger@intel.com>
Change-Id: If062df0189d92e4fb2da3f055fb981909780dc04
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15207
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This commit is contained in:
Michal Berger 2022-11-02 16:09:36 +01:00 committed by Tomasz Zawadzki
parent acd2cc94e7
commit 3f912cf0e9
62 changed files with 90 additions and 90 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
CHANGELOG.md
View File

@ -2075,7 +2075,7 @@ A new `spdk_bdev_open_ext` function has been added and `spdk_bdev_open` function
The new open function introduces requirement to provide callback function that will be called by The new open function introduces requirement to provide callback function that will be called by
asynchronous event such as bdev removal. `spdk_bdev_open_ext` function takes bdev name as asynchronous event such as bdev removal. `spdk_bdev_open_ext` function takes bdev name as
an argument instead of bdev structure to avoid a race condition that can happen when the bdev an argument instead of bdev structure to avoid a race condition that can happen when the bdev
is being removed between a call to get its structure based on a name and actually openning it. is being removed between a call to get its structure based on a name and actually opening it.
New 'resize' event has been added to notify about change of block count property of block device. New 'resize' event has been added to notify about change of block count property of block device.
Event is delivered only if block device was opened with `spdk_bdev_open_ext` function. Event is delivered only if block device was opened with `spdk_bdev_open_ext` function.
@ -3001,7 +3001,7 @@ Net framework initialization and finish is now done asynchronously.
Added `spdk_rpc_is_method_allowed` function for checking whether method is permitted in a given state. Added `spdk_rpc_is_method_allowed` function for checking whether method is permitted in a given state.
Added `spdk_rpc_get_state` to check current state of RPC server. Added `spdk_rpc_get_state` to check current state of RPC server.
RPC `wait_subsystem_init` has been added to allow clients to block untill all subsystems are initialized. RPC `wait_subsystem_init` has been added to allow clients to block until all subsystems are initialized.
### json rpc ### json rpc
@ -3299,7 +3299,7 @@ to link only the minimal set of components required.
### git pre-commit and pre-push hooks ### git pre-commit and pre-push hooks
The pre-commit hook will run `scripts/check_format.sh` and verify there are no The pre-commit hook will run `scripts/check_format.sh` and verify there are no
formating errors before allowing `git commit` to run. The pre-push hook runs formatting errors before allowing `git commit` to run. The pre-push hook runs
`make CONFIG_WERROR=y` with and without `CONFIG_DEBUG=y` using both the gcc and `make CONFIG_WERROR=y` with and without `CONFIG_DEBUG=y` using both the gcc and
clang compiler before allowing `git push` to run. Following each DEBUG build clang compiler before allowing `git push` to run. Following each DEBUG build
`test/unit/unittest.sh` is run and verified. Results are recorded in the `test/unit/unittest.sh` is run and verified. Results are recorded in the

2
CONFIG
View File

@ -146,7 +146,7 @@ CONFIG_CRYPTO=n
# Build spdk shared libraries in addition to the static ones. # Build spdk shared libraries in addition to the static ones.
CONFIG_SHARED=n CONFIG_SHARED=n
# Build with VTune suport. # Build with VTune support.
CONFIG_VTUNE=n CONFIG_VTUNE=n
CONFIG_VTUNE_DIR= CONFIG_VTUNE_DIR=

4
doc/accel_fw.md
View File

@ -91,7 +91,7 @@ Design software to deal with the inability to get a channel.
There are several dependencies to leverage the Linux idxd driver for driving DSA devices. There are several dependencies to leverage the Linux idxd driver for driving DSA devices.
1 Linux kernel support: You need to have a Linux kernel with the `idxd` driver 1 Linux kernel support: You need to have a Linux kernel with the `idxd` driver
loaded. Futher, add the following command line options to the kernel boot loaded. Further, add the following command line options to the kernel boot
commands: commands:
```bash ```bash
@ -170,5 +170,5 @@ The following RPCs would accomplish the copy override:
} }
``` ```
To detemine the name of available modules and their supported operations use the To determine the name of available modules and their supported operations use the
RPC `accel_get_module_info`. RPC `accel_get_module_info`.

2
doc/ftl.md
View File

@ -145,7 +145,7 @@ well as on the base devices. The following types of metadata are persisted:
After power failure, FTL needs to rebuild the whole L2P using the address maps (`P2L`) stored within each band/chunk. After power failure, FTL needs to rebuild the whole L2P using the address maps (`P2L`) stored within each band/chunk.
This needs to done, because while individual L2P pages may have been paged out and persisted to the cache device, This needs to done, because while individual L2P pages may have been paged out and persisted to the cache device,
there's no way to tell which, if any, pages were dirty before the power failure occured. The P2L consists of not only there's no way to tell which, if any, pages were dirty before the power failure occurred. The P2L consists of not only
the mapping itself, but also a sequence id (`seq_id`), which describes the relative age of a given logical block the mapping itself, but also a sequence id (`seq_id`), which describes the relative age of a given logical block
(multiple writes to the same logical block would produce the same amount of P2L entries, only the last one having the current data). (multiple writes to the same logical block would produce the same amount of P2L entries, only the last one having the current data).

2
doc/gdb_macros.md
View File

@ -175,7 +175,7 @@ iterate the spdk lists and build iterable representations of the list objects.
This will result in errors if these are not available which is very possible if This will result in errors if these are not available which is very possible if
gdb is used for reasons other than debugging spdk core dumps. gdb is used for reasons other than debugging spdk core dumps.
In the example bellow, I attempted to load the macros when the globals are not In the example below, I attempted to load the macros when the globals are not
available causing gdb to fail loading the gdb_macros: available causing gdb to fail loading the gdb_macros:
~~~{.sh} ~~~{.sh}

12
doc/jsonrpc.md
View File

@ -4343,7 +4343,7 @@ This method is available only if SPDK was build with Ceph RBD support.
Name | Optional | Type | Description Name | Optional | Type | Description
----------------------- | -------- | ----------- | ----------- ----------------------- | -------- | ----------- | -----------
name | Required | string | Registerd Rados cluster object name name | Required | string | Registered Rados cluster object name
user_id | Optional | string | Ceph ID (i.e. admin, not client.admin) user_id | Optional | string | Ceph ID (i.e. admin, not client.admin)
config_param | Optional | string map | Explicit librados configuration config_param | Optional | string map | Explicit librados configuration
config_file | Optional | string | File path of libraodos configuration file config_file | Optional | string | File path of libraodos configuration file
@ -7374,7 +7374,7 @@ io_unit_size | Optional | number | I/O unit size (bytes)
max_aq_depth | Optional | number | Max number of admin cmds per AQ max_aq_depth | Optional | number | Max number of admin cmds per AQ
num_shared_buffers | Optional | number | The number of pooled data buffers available to the transport num_shared_buffers | Optional | number | The number of pooled data buffers available to the transport
buf_cache_size | Optional | number | The number of shared buffers to reserve for each poll group buf_cache_size | Optional | number | The number of shared buffers to reserve for each poll group
num_cqe | Optional | number | The number of CQ entires. Only used when no_srq=true (RDMA only) num_cqe | Optional | number | The number of CQ entries. Only used when no_srq=true (RDMA only)
max_srq_depth | Optional | number | The number of elements in a per-thread shared receive queue (RDMA only) max_srq_depth | Optional | number | The number of elements in a per-thread shared receive queue (RDMA only)
no_srq | Optional | boolean | Disable shared receive queue even for devices that support it. (RDMA only) no_srq | Optional | boolean | Disable shared receive queue even for devices that support it. (RDMA only)
c2h_success | Optional | boolean | Disable C2H success optimization (TCP only) c2h_success | Optional | boolean | Disable C2H success optimization (TCP only)
@ -8782,7 +8782,7 @@ ctrlr | string | Controller name
cpumask | string | @ref cpu_mask of this controller cpumask | string | @ref cpu_mask of this controller
delay_base_us | number | Base (minimum) coalescing time in microseconds (0 if disabled) delay_base_us | number | Base (minimum) coalescing time in microseconds (0 if disabled)
iops_threshold | number | Coalescing activation level iops_threshold | number | Coalescing activation level
backend_specific | object | Backend specific informations backend_specific | object | Backend specific information
### Vhost block {#rpc_vhost_get_controllers_blk} ### Vhost block {#rpc_vhost_get_controllers_blk}
@ -10171,7 +10171,7 @@ Example response:
### notify_get_notifications {#notify_get_notifications} ### notify_get_notifications {#notify_get_notifications}
Request notifications. Returns array of notifications that happend since the specified id (or first that is available). Request notifications. Returns array of notifications that happened since the specified id (or first that is available).
Notice: Notifications are kept in circular buffer with limited size. Older notifications might be inaccessible Notice: Notifications are kept in circular buffer with limited size. Older notifications might be inaccessible
due to being overwritten by new ones. due to being overwritten by new ones.
@ -10679,7 +10679,7 @@ Example response:
Send NVMe command directly to NVMe controller or namespace. Parameters and responses encoded by base64 urlsafe need further processing. Send NVMe command directly to NVMe controller or namespace. Parameters and responses encoded by base64 urlsafe need further processing.
Notice: bdev_nvme_send_cmd requires user to guarentee the correctness of NVMe command itself, and also optional parameters. Notice: bdev_nvme_send_cmd requires user to guarantee the correctness of NVMe command itself, and also optional parameters.
Illegal command contents or mismatching buffer size may result in unpredictable behavior. Illegal command contents or mismatching buffer size may result in unpredictable behavior.
#### Parameters #### Parameters
@ -11042,7 +11042,7 @@ uuid | Optional | string | UUID of new bdev
oclass | Optional | string | DAOS object class (default SX) oclass | Optional | string | DAOS object class (default SX)
To find more about various object classes please visit [DAOS documentation](https://github.com/daos-stack/daos/blob/master/src/object/README.md). To find more about various object classes please visit [DAOS documentation](https://github.com/daos-stack/daos/blob/master/src/object/README.md).
Please note, that DAOS bdev module uses the same CLI flag notation as `dmg` and `daos` commmands, Please note, that DAOS bdev module uses the same CLI flag notation as `dmg` and `daos` commands,
for instance, `SX` or `EC_4P2G2` rather than in DAOS header file `OC_SX` or `OC_EC_4P2G2`. for instance, `SX` or `EC_4P2G2` rather than in DAOS header file `OC_SX` or `OC_EC_4P2G2`.
#### Result #### Result

2
doc/vagrant.md
View File

@ -116,7 +116,7 @@ Bringing machine 'default' up with 'virtualbox' provider...
Use vagrant "destroy" followed by "rm -rf ubuntu18" to destroy all trace of vm. Use vagrant "destroy" followed by "rm -rf ubuntu18" to destroy all trace of vm.
~~~ ~~~
Check the enviroment. Check the environment.
~~~{.sh} ~~~{.sh}
user@dev-system:~/spdk/scripts/vagrant$ cd ubuntu18 user@dev-system:~/spdk/scripts/vagrant$ cd ubuntu18

2
docker/README.md
View File

@ -75,7 +75,7 @@ docker-compose run traffic-generator-nvme
docker-compose run traffic-generator-virtio docker-compose run traffic-generator-virtio
~~~ ~~~
Enviroment variables to containers can be passed as shown in Environment variables to containers can be passed as shown in
[docs](https://docs.docker.com/compose/environment-variables/). [docs](https://docs.docker.com/compose/environment-variables/).
For example extra arguments to fio can be passed as so: For example extra arguments to fio can be passed as so:

2
examples/sock/hello_world/hello_sock.c
View File

@ -74,7 +74,7 @@ hello_sock_usage(void)
printf(" -T tls_ver TLS version, e.g., -T 12 or -T 13. If omitted, auto-negotiation will take place\n"); printf(" -T tls_ver TLS version, e.g., -T 12 or -T 13. If omitted, auto-negotiation will take place\n");
printf(" -k disable KTLS for the given sock implementation (default)\n"); printf(" -k disable KTLS for the given sock implementation (default)\n");
printf(" -K enable KTLS for the given sock implementation\n"); printf(" -K enable KTLS for the given sock implementation\n");
printf(" -V print out additional informations\n"); printf(" -V print out additional information\n");
printf(" -z disable zero copy send for the given sock implementation\n"); printf(" -z disable zero copy send for the given sock implementation\n");
printf(" -Z enable zero copy send for the given sock implementation\n"); printf(" -Z enable zero copy send for the given sock implementation\n");
} }

2
include/spdk/accel.h
View File

@ -261,7 +261,7 @@ int spdk_accel_submit_decompress(struct spdk_io_channel *ch, struct iovec *dst_i
spdk_accel_completion_cb cb_fn, void *cb_arg); spdk_accel_completion_cb cb_fn, void *cb_arg);
/** /**
* Return the name of the module assigned to a specfic opcode. * Return the name of the module assigned to a specific opcode.
* *
* \param opcode Accel Framework Opcode enum value. Valid codes can be retrieved using * \param opcode Accel Framework Opcode enum value. Valid codes can be retrieved using
* `accel_get_opc_assignments` or `spdk_accel_get_opc_name`. * `accel_get_opc_assignments` or `spdk_accel_get_opc_name`.

4
include/spdk/bdev.h
View File

@ -392,7 +392,7 @@ typedef int (*spdk_for_each_bdev_fn)(void *ctx, struct spdk_bdev *bdev);
* \param ctx Context passed to the callback function. * \param ctx Context passed to the callback function.
* \param fn Callback function for each block device. * \param fn Callback function for each block device.
* *
* \return 0 if operation is sucessful, or suitable errno value one of the * \return 0 if operation is successful, or suitable errno value one of the
* callback returned otherwise. * callback returned otherwise.
*/ */
int spdk_for_each_bdev(void *ctx, spdk_for_each_bdev_fn fn); int spdk_for_each_bdev(void *ctx, spdk_for_each_bdev_fn fn);
@ -405,7 +405,7 @@ int spdk_for_each_bdev(void *ctx, spdk_for_each_bdev_fn fn);
* callback function for each unclaimed bdev internally. * callback function for each unclaimed bdev internally.
* *
* \param ctx Context passed to the callback function. * \param ctx Context passed to the callback function.
* \param fn Callback funciton for each block device without virtual block devices on top. * \param fn Callback function for each block device without virtual block devices on top.
* *
* \return 0 if operation is successful, or suitable errno value one of the * \return 0 if operation is successful, or suitable errno value one of the
* callback returned otherwise. * callback returned otherwise.

2
include/spdk/blob.h
View File

@ -515,7 +515,7 @@ void spdk_blob_opts_init(struct spdk_blob_opts *opts, size_t opts_size);
* \param bs blobstore. * \param bs blobstore.
* \param opts The structure which contains the option values for the new blob. * \param opts The structure which contains the option values for the new blob.
* \param cb_fn Called when the operation is complete. * \param cb_fn Called when the operation is complete.
* \param cb_arg Argument passed to funcion cb_fn. * \param cb_arg Argument passed to function cb_fn.
*/ */
void spdk_bs_create_blob_ext(struct spdk_blob_store *bs, const struct spdk_blob_opts *opts, void spdk_bs_create_blob_ext(struct spdk_blob_store *bs, const struct spdk_blob_opts *opts,
spdk_blob_op_with_id_complete cb_fn, void *cb_arg); spdk_blob_op_with_id_complete cb_fn, void *cb_arg);

4
include/spdk/dif.h
View File

@ -294,7 +294,7 @@ int spdk_dix_inject_error(struct iovec *iovs, int iovcnt, struct iovec *md_iov,
* This function removes the necessity of data copy in the SPDK application * This function removes the necessity of data copy in the SPDK application
* during DIF insertion and strip. * during DIF insertion and strip.
* *
* When the extended LBA payload is splitted into multiple data segments, * When the extended LBA payload is split into multiple data segments,
* start of each data segment is passed through the DIF context. data_offset * start of each data segment is passed through the DIF context. data_offset
* and data_len is within a data segment. * and data_len is within a data segment.
* *
@ -321,7 +321,7 @@ int spdk_dif_set_md_interleave_iovs(struct iovec *iovs, int iovcnt,
/** /**
* Generate and insert DIF into metadata space for newly read data block. * Generate and insert DIF into metadata space for newly read data block.
* *
* When the extended LBA payload is splitted into multiple data segments, * When the extended LBA payload is split into multiple data segments,
* start of each data segment is passed through the DIF context. data_offset * start of each data segment is passed through the DIF context. data_offset
* and data_len is within a data segment. * and data_len is within a data segment.
* *

2
include/spdk_internal/virtio.h
View File

@ -250,7 +250,7 @@ int virtio_dev_reset(struct virtio_dev *vdev, uint64_t req_features);
* \param max_queues number of queues to allocate. The max number of * \param max_queues number of queues to allocate. The max number of
* usable I/O queues is also limited by the host device. `vdev` will be * usable I/O queues is also limited by the host device. `vdev` will be
* started successfully even if the host supports less queues than requested. * started successfully even if the host supports less queues than requested.
* \param fixed_queue_num number of queues preceeding the first * \param fixed_queue_num number of queues preceding the first
* request queue. For Virtio-SCSI this is equal to 2, as there are * request queue. For Virtio-SCSI this is equal to 2, as there are
* additional event and control queues. * additional event and control queues.
*/ */

2
lib/accel/accel.c
View File

@ -593,7 +593,7 @@ spdk_accel_initialize(void)
/* Create our priority global map of opcodes to modules, we populate starting /* Create our priority global map of opcodes to modules, we populate starting
* with the software module (guaranteed to be first on the list) and then * with the software module (guaranteed to be first on the list) and then
* updating opcodes with HW modules that have been initilaized. * updating opcodes with HW modules that have been initilaized.
* NOTE: all opcodes must be suported by software in the event that no HW * NOTE: all opcodes must be supported by software in the event that no HW
* modules are initilaized to support the operation. * modules are initilaized to support the operation.
*/ */
TAILQ_FOREACH(accel_module, &spdk_accel_module_list, tailq) { TAILQ_FOREACH(accel_module, &spdk_accel_module_list, tailq) {

2
lib/bdev/bdev.c
View File

@ -6039,7 +6039,7 @@ _bdev_abort(struct spdk_bdev_io *parent_io)
/* matched_ios is returned and will be kept by the caller. /* matched_ios is returned and will be kept by the caller.
* *
* This funcion will be used for two cases, 1) the same cb_arg is used for * This function will be used for two cases, 1) the same cb_arg is used for
* multiple I/Os, 2) a single large I/O is split into smaller ones. * multiple I/Os, 2) a single large I/O is split into smaller ones.
* Incrementing split_outstanding directly here may confuse readers especially * Incrementing split_outstanding directly here may confuse readers especially
* for the 1st case. * for the 1st case.

2
lib/blob/blobstore.c
View File

@ -3864,7 +3864,7 @@ bs_load_used_clusters_cpl(spdk_bs_sequence_t *seq, void *cb_arg, int bserrno)
struct spdk_blob_md_page) * 8)); struct spdk_blob_md_page) * 8));
/* /*
* The length of the mask must be equal to or larger than the total number of clusters. It may be * The length of the mask must be equal to or larger than the total number of clusters. It may be
* larger than the total nubmer of clusters due to a failure spdk_bs_grow. * larger than the total number of clusters due to a failure spdk_bs_grow.
*/ */
assert(ctx->mask->length >= ctx->bs->total_clusters); assert(ctx->mask->length >= ctx->bs->total_clusters);
if (ctx->mask->length > ctx->bs->total_clusters) { if (ctx->mask->length > ctx->bs->total_clusters) {

2
lib/event/reactor.c
View File

@ -566,7 +566,7 @@ event_queue_run_batch(void *arg)
} }
/* Execute the events. There are still some remaining events /* Execute the events. There are still some remaining events
* that must occur on an SPDK thread. To accomodate those, try to * that must occur on an SPDK thread. To accommodate those, try to
* run them on the first thread in the list, if it exists. */ * run them on the first thread in the list, if it exists. */
lw_thread = TAILQ_FIRST(&reactor->threads); lw_thread = TAILQ_FIRST(&reactor->threads);
if (lw_thread) { if (lw_thread) {

2
lib/event/scheduler_static.c
View File

@ -15,7 +15,7 @@
static int static int
init_static(void) init_static(void)
{ {
/* There is no scheduling perfomed by static scheduler, /* There is no scheduling performed by static scheduler,
* do not set the scheduling period. */ * do not set the scheduling period. */
spdk_scheduler_set_period(0); spdk_scheduler_set_period(0);
return 0; return 0;

6
lib/ftl/ftl_core.c
View File

@ -468,7 +468,7 @@ spdk_ftl_unmap(struct spdk_ftl_dev *dev, struct ftl_io *io, struct spdk_io_chann
uint64_t lba, uint64_t lba_cnt, spdk_ftl_fn cb_fn, void *cb_arg) uint64_t lba, uint64_t lba_cnt, spdk_ftl_fn cb_fn, void *cb_arg)
{ {
int rc; int rc;
uint64_t aligment = dev->layout.l2p.lbas_in_page; uint64_t alignment = dev->layout.l2p.lbas_in_page;
if (lba_cnt == 0) { if (lba_cnt == 0) {
return -EINVAL; return -EINVAL;
@ -486,7 +486,7 @@ spdk_ftl_unmap(struct spdk_ftl_dev *dev, struct ftl_io *io, struct spdk_io_chann
return -EBUSY; return -EBUSY;
} }
if (lba % aligment || lba_cnt % aligment) { if (lba % alignment || lba_cnt % alignment) {
if (!io) { if (!io) {
/* This is management/RPC path, its parameters must be aligned to 1MiB. */ /* This is management/RPC path, its parameters must be aligned to 1MiB. */
return -EINVAL; return -EINVAL;
@ -665,7 +665,7 @@ ftl_process_io_queue(struct spdk_ftl_dev *dev)
/* /*
* Unmap operation requires generating a sequence id for itself, which it gets based on the open chunk * Unmap operation requires generating a sequence id for itself, which it gets based on the open chunk
* in nv cache. If there are no open chunks (because we're in the middle of state transistion or compaction * in nv cache. If there are no open chunks (because we're in the middle of state transition or compaction
* lagged behind), then we need to wait for the nv cache to resolve the situation - it's fine to just put the * lagged behind), then we need to wait for the nv cache to resolve the situation - it's fine to just put the
* unmap and try again later. * unmap and try again later.
*/ */

2
lib/ftl/ftl_internal.h
View File

@ -69,7 +69,7 @@ struct ftl_p2l_map_entry {
* Mapping of physical (actual location on disk) to logical (user's POV) addresses. Used in two main scenarios: * Mapping of physical (actual location on disk) to logical (user's POV) addresses. Used in two main scenarios:
* - during relocation FTL needs to pin L2P pages (this allows to check which pages to pin) and move still valid blocks * - during relocation FTL needs to pin L2P pages (this allows to check which pages to pin) and move still valid blocks
* (valid map allows for preliminary elimination of invalid physical blocks, but user data could invalidate a location * (valid map allows for preliminary elimination of invalid physical blocks, but user data could invalidate a location
* during read/write operation, so actual comparision against L2P needs to be done) * during read/write operation, so actual comparison against L2P needs to be done)
* - After dirty shutdown the state of the L2P is unknown and needs to be rebuilt - it is done by applying all P2L, taking * - After dirty shutdown the state of the L2P is unknown and needs to be rebuilt - it is done by applying all P2L, taking
* into account ordering of user writes * into account ordering of user writes
*/ */

2
lib/ftl/ftl_l2p_cache.c
View File

@ -121,7 +121,7 @@ struct ftl_l2p_cache {
struct ftl_mempool *page_sets_pool; struct ftl_mempool *page_sets_pool;
TAILQ_HEAD(, ftl_l2p_page_set) deferred_page_set_list; /* for deferred page sets */ TAILQ_HEAD(, ftl_l2p_page_set) deferred_page_set_list; /* for deferred page sets */
/* Process unmap in backgorund */ /* Process unmap in background */
struct { struct {
#define FTL_L2P_MAX_LAZY_UNMAP_QD 1 #define FTL_L2P_MAX_LAZY_UNMAP_QD 1
/* Unmap queue depth */ /* Unmap queue depth */

2
lib/ftl/ftl_nv_cache.h
View File

@ -38,7 +38,7 @@
* zero if the number of free chunks is at the threshold, negative if below and positive if above. * zero if the number of free chunks is at the threshold, negative if below and positive if above.
*/ */
/* Interval in miliseconds between write throttle updates. */ /* Interval in milliseconds between write throttle updates. */
#define FTL_NV_CACHE_THROTTLE_INTERVAL_MS 20 #define FTL_NV_CACHE_THROTTLE_INTERVAL_MS 20
/* Throttle modifier proportional gain */ /* Throttle modifier proportional gain */
#define FTL_NV_CACHE_THROTTLE_MODIFIER_KP 20 #define FTL_NV_CACHE_THROTTLE_MODIFIER_KP 20

2
lib/ftl/mngt/ftl_mngt.h
View File

@ -30,7 +30,7 @@ typedef void (*ftl_mngt_fn)(struct spdk_ftl_dev *dev, struct ftl_mngt_process *m
typedef void (*ftl_mngt_completion)(struct spdk_ftl_dev *dev, void *ctx, int status); typedef void (*ftl_mngt_completion)(struct spdk_ftl_dev *dev, void *ctx, int status);
/** /**
* The FTL management step descriptior * The FTL management step descriptor
*/ */
struct ftl_mngt_step_desc { struct ftl_mngt_step_desc {
/** /**

2
lib/ftl/mngt/ftl_mngt_md.c
View File

@ -334,7 +334,7 @@ ftl_mngt_persist_md(struct spdk_ftl_dev *dev, struct ftl_mngt_process *mngt)
} }
/* /*
* Fast clean shutdown path - skips the persistance of most metadata regions and * Fast clean shutdown path - skips the persistence of most metadata regions and
* relies on their shared memory state instead. * relies on their shared memory state instead.
*/ */
static const struct ftl_mngt_process_desc desc_fast_persist = { static const struct ftl_mngt_process_desc desc_fast_persist = {

2
lib/ftl/mngt/ftl_mngt_recovery.c
View File

@ -742,7 +742,7 @@ ftl_mngt_complete_unmap(struct spdk_ftl_dev *dev, struct ftl_mngt_process *mngt)
assert(seq_id <= dev->sb->seq_id); assert(seq_id <= dev->sb->seq_id);
FTL_NOTICELOG(dev, "Uncomplete unmap detected lba: %"PRIu64" num_blocks: %"PRIu64"\n", FTL_NOTICELOG(dev, "Incomplete unmap detected lba: %"PRIu64" num_blocks: %"PRIu64"\n",
start_lba, num_blocks); start_lba, num_blocks);
ftl_set_unmap_map(dev, start_lba, num_blocks, seq_id); ftl_set_unmap_map(dev, start_lba, num_blocks, seq_id);

2
lib/ftl/utils/ftl_md.h
View File

@ -331,7 +331,7 @@ void ftl_md_read_entry(struct ftl_md *md, uint64_t start_entry, void *buffer, vo
void ftl_md_clear(struct ftl_md *md, int pattern, union ftl_md_vss *vss_pattern); void ftl_md_clear(struct ftl_md *md, int pattern, union ftl_md_vss *vss_pattern);
/** /**
* @brief Gets the number of blocks that are transfered in a single IO operation * @brief Gets the number of blocks that are transferred in a single IO operation
* *
* @param dev The FTL device * @param dev The FTL device
* *

6
lib/init/json_config.c
View File

@ -335,7 +335,7 @@ app_json_config_load_subsystem_config_entry(void *_ctx)
SPDK_DEBUG_APP_CFG("Subsystem '%.*s': configuration done.\n", ctx->subsystem_name->len, SPDK_DEBUG_APP_CFG("Subsystem '%.*s': configuration done.\n", ctx->subsystem_name->len,
(char *)ctx->subsystem_name->start); (char *)ctx->subsystem_name->start);
ctx->subsystems_it = spdk_json_next(ctx->subsystems_it); ctx->subsystems_it = spdk_json_next(ctx->subsystems_it);
/* Invoke later to avoid recurrency */ /* Invoke later to avoid recurrence */
spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem, ctx); spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem, ctx);
return; return;
} }
@ -356,7 +356,7 @@ app_json_config_load_subsystem_config_entry(void *_ctx)
cur_state_mask = spdk_rpc_get_state(); cur_state_mask = spdk_rpc_get_state();
if ((state_mask & cur_state_mask) != cur_state_mask) { if ((state_mask & cur_state_mask) != cur_state_mask) {
SPDK_DEBUG_APP_CFG("Method '%s' not allowed -> skipping\n", cfg.method); SPDK_DEBUG_APP_CFG("Method '%s' not allowed -> skipping\n", cfg.method);
/* Invoke later to avoid recurrency */ /* Invoke later to avoid recurrence */
ctx->config_it = spdk_json_next(ctx->config_it); ctx->config_it = spdk_json_next(ctx->config_it);
spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem_config_entry, ctx); spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem_config_entry, ctx);
goto out; goto out;
@ -365,7 +365,7 @@ app_json_config_load_subsystem_config_entry(void *_ctx)
/* Some methods are allowed to be run in both STARTUP and RUNTIME states. /* Some methods are allowed to be run in both STARTUP and RUNTIME states.
* We should not call such methods twice, so ignore the second attempt in RUNTIME state */ * We should not call such methods twice, so ignore the second attempt in RUNTIME state */
SPDK_DEBUG_APP_CFG("Method '%s' has already been run in STARTUP state\n", cfg.method); SPDK_DEBUG_APP_CFG("Method '%s' has already been run in STARTUP state\n", cfg.method);
/* Invoke later to avoid recurrency */ /* Invoke later to avoid recurrence */
ctx->config_it = spdk_json_next(ctx->config_it); ctx->config_it = spdk_json_next(ctx->config_it);
spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem_config_entry, ctx); spdk_thread_send_msg(ctx->thread, app_json_config_load_subsystem_config_entry, ctx);
goto out; goto out;

2
lib/iscsi/iscsi.c
View File

@ -3031,7 +3031,7 @@ iscsi_transfer_in(struct spdk_iscsi_conn *conn, struct spdk_iscsi_task *task)
sequence_end = spdk_min(((i + 1) * conn->sess->MaxBurstLength), sequence_end = spdk_min(((i + 1) * conn->sess->MaxBurstLength),
transfer_len); transfer_len);
/* send data splitted by segment_len */ /* send data split by segment_len */
for (; offset < sequence_end; offset += segment_len) { for (; offset < sequence_end; offset += segment_len) {
len = spdk_min(segment_len, (sequence_end - offset)); len = spdk_min(segment_len, (sequence_end - offset));

2
lib/nvme/nvme_opal.c
View File

@ -792,7 +792,7 @@ opal_discovery0_end(struct spdk_opal_dev *dev, void *payload, uint32_t payload_s
supported = true; supported = true;
break; break;
default: default:
SPDK_INFOLOG(opal, "Unknow feature code: %d\n", feat_code); SPDK_INFOLOG(opal, "Unknown feature code: %d\n", feat_code);
} }
cpos += feat_hdr->length + sizeof(*feat_hdr); cpos += feat_hdr->length + sizeof(*feat_hdr);
} }

2
lib/nvme/nvme_rdma.c
View File

@ -2003,7 +2003,7 @@ nvme_rdma_stale_conn_retry(struct nvme_rdma_qpair *rqpair)
rqpair->stale_conn_retry_count++; rqpair->stale_conn_retry_count++;
SPDK_NOTICELOG("%d times, retry stale connnection to qpair (cntlid:%u, qid:%u).\n", SPDK_NOTICELOG("%d times, retry stale connection to qpair (cntlid:%u, qid:%u).\n",
rqpair->stale_conn_retry_count, qpair->ctrlr->cntlid, qpair->id); rqpair->stale_conn_retry_count, qpair->ctrlr->cntlid, qpair->id);
_nvme_rdma_ctrlr_disconnect_qpair(qpair->ctrlr, qpair, nvme_rdma_stale_conn_disconnected); _nvme_rdma_ctrlr_disconnect_qpair(qpair->ctrlr, qpair, nvme_rdma_stale_conn_disconnected);

4
lib/nvme/nvme_tcp.c
View File

@ -449,7 +449,7 @@ pdu_data_crc32_compute(struct nvme_tcp_pdu *pdu)
/* Data Digest */ /* Data Digest */
if (pdu->data_len > 0 && g_nvme_tcp_ddgst[pdu->hdr.common.pdu_type] && if (pdu->data_len > 0 && g_nvme_tcp_ddgst[pdu->hdr.common.pdu_type] &&
tqpair->flags.host_ddgst_enable) { tqpair->flags.host_ddgst_enable) {
/* Only suport this limited case for the first step */ /* Only support this limited case for the first step */
if ((nvme_qpair_get_state(&tqpair->qpair) >= NVME_QPAIR_CONNECTED) && if ((nvme_qpair_get_state(&tqpair->qpair) >= NVME_QPAIR_CONNECTED) &&
(tgroup != NULL && tgroup->group.group->accel_fn_table.submit_accel_crc32c) && (tgroup != NULL && tgroup->group.group->accel_fn_table.submit_accel_crc32c) &&
spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0))) { spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0))) {
@ -1121,7 +1121,7 @@ nvme_tcp_pdu_payload_handle(struct nvme_tcp_qpair *tqpair,
/* But if the data digest is enabled, tcp_req cannot be NULL */ /* But if the data digest is enabled, tcp_req cannot be NULL */
assert(tcp_req != NULL); assert(tcp_req != NULL);
tgroup = nvme_tcp_poll_group(tqpair->qpair.poll_group); tgroup = nvme_tcp_poll_group(tqpair->qpair.poll_group);
/* Only suport this limitated case that the request has only one c2h pdu */ /* Only support this limitated case that the request has only one c2h pdu */
if ((nvme_qpair_get_state(&tqpair->qpair) >= NVME_QPAIR_CONNECTED) && if ((nvme_qpair_get_state(&tqpair->qpair) >= NVME_QPAIR_CONNECTED) &&
(tgroup != NULL && tgroup->group.group->accel_fn_table.submit_accel_crc32c) && (tgroup != NULL && tgroup->group.group->accel_fn_table.submit_accel_crc32c) &&
spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0) spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0)

2
lib/nvme/nvme_transport.c
View File

@ -41,7 +41,7 @@ nvme_get_next_transport(const struct spdk_nvme_transport *transport)
* Unfortunately, due to NVMe PCIe multiprocess support, we cannot store the * Unfortunately, due to NVMe PCIe multiprocess support, we cannot store the
* transport object in either the controller struct or the admin qpair. THis means * transport object in either the controller struct or the admin qpair. THis means
* that a lot of admin related transport calls will have to call nvme_get_transport * that a lot of admin related transport calls will have to call nvme_get_transport
* in order to knwo which functions to call. * in order to know which functions to call.
* In the I/O path, we have the ability to store the transport struct in the I/O * In the I/O path, we have the ability to store the transport struct in the I/O
* qpairs to avoid taking a performance hit. * qpairs to avoid taking a performance hit.
*/ */

2
lib/nvmf/fc.c
View File

@ -3051,7 +3051,7 @@ nvmf_fc_adm_evnt_hw_port_offline(void *arg)
free(arg); free(arg);
/* Wait untill all the hwqps are removed from poll groups. */ /* Wait until all the hwqps are removed from poll groups. */
return; return;
} else { } else {
SPDK_ERRLOG("Unable to find the SPDK FC port %d\n", args->port_handle); SPDK_ERRLOG("Unable to find the SPDK FC port %d\n", args->port_handle);

4
lib/nvmf/tcp.c
View File

@ -78,7 +78,7 @@ enum spdk_nvmf_tcp_req_state {
/* The request is currently executing at the block device */ /* The request is currently executing at the block device */
TCP_REQUEST_STATE_EXECUTING = 8, TCP_REQUEST_STATE_EXECUTING = 8,
/* The request is waiting for zcopy buffers to be commited */ /* The request is waiting for zcopy buffers to be committed */
TCP_REQUEST_STATE_AWAITING_ZCOPY_COMMIT = 9, TCP_REQUEST_STATE_AWAITING_ZCOPY_COMMIT = 9,
/* The request finished executing at the block device */ /* The request finished executing at the block device */
@ -999,7 +999,7 @@ pdu_data_crc32_compute(struct nvme_tcp_pdu *pdu)
/* Data Digest */ /* Data Digest */
if (pdu->data_len > 0 && g_nvme_tcp_ddgst[pdu->hdr.common.pdu_type] && tqpair->host_ddgst_enable) { if (pdu->data_len > 0 && g_nvme_tcp_ddgst[pdu->hdr.common.pdu_type] && tqpair->host_ddgst_enable) {
/* Only suport this limitated case for the first step */ /* Only support this limitated case for the first step */
if (spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0) if (spdk_likely(!pdu->dif_ctx && (pdu->data_len % SPDK_NVME_TCP_DIGEST_ALIGNMENT == 0)
&& tqpair->group)) { && tqpair->group)) {
rc = spdk_accel_submit_crc32cv(tqpair->group->accel_channel, &pdu->data_digest_crc32, pdu->data_iov, rc = spdk_accel_submit_crc32cv(tqpair->group->accel_channel, &pdu->data_digest_crc32, pdu->data_iov,

2
lib/scsi/scsi_bdev.c
View File

@ -40,7 +40,7 @@ bdev_scsi_set_naa_ieee_extended(const char *name, uint8_t *buf)
/* /*
* see spc3r23 7.6.3.6.2, * see spc3r23 7.6.3.6.2,
* NAA IEEE Extended identifer format * NAA IEEE Extended identifier format
*/ */
id_a = local_value & 0x0000000fff000000ull; id_a = local_value & 0x0000000fff000000ull;
id_a = id_a << 24; id_a = id_a << 24;

6
lib/vhost/rte_vhost_user.c
View File

@ -113,7 +113,7 @@ vhost_log_req_desc(struct spdk_vhost_session *vsession, struct spdk_vhost_virtqu
do { do {
if (vhost_vring_desc_is_wr(desc)) { if (vhost_vring_desc_is_wr(desc)) {
/* To be honest, only pages realy touched should be logged, but /* To be honest, only pages really touched should be logged, but
* doing so would require tracking those changes in each backed. * doing so would require tracking those changes in each backed.
* Also backend most likely will touch all/most of those pages so * Also backend most likely will touch all/most of those pages so
* for lets assume we touched all pages passed to as writeable buffers. */ * for lets assume we touched all pages passed to as writeable buffers. */
@ -1456,7 +1456,7 @@ extern_vhost_pre_msg_handler(int vid, void *_msg)
vsession = vhost_session_find_by_vid(vid); vsession = vhost_session_find_by_vid(vid);
if (vsession == NULL) { if (vsession == NULL) {
SPDK_ERRLOG("Received a message to unitialized session (vid %d).\n", vid); SPDK_ERRLOG("Received a message to uninitialized session (vid %d).\n", vid);
assert(false); assert(false);
return RTE_VHOST_MSG_RESULT_ERR; return RTE_VHOST_MSG_RESULT_ERR;
} }
@ -1512,7 +1512,7 @@ extern_vhost_post_msg_handler(int vid, void *_msg)
vsession = vhost_session_find_by_vid(vid); vsession = vhost_session_find_by_vid(vid);
if (vsession == NULL) { if (vsession == NULL) {
SPDK_ERRLOG("Received a message to unitialized session (vid %d).\n", vid); SPDK_ERRLOG("Received a message to uninitialized session (vid %d).\n", vid);
assert(false); assert(false);
return RTE_VHOST_MSG_RESULT_ERR; return RTE_VHOST_MSG_RESULT_ERR;
} }

2
lib/vhost/vhost_blk.c
View File

@ -1627,7 +1627,7 @@ spdk_vhost_blk_construct(const char *name, const char *cpumask, const char *dev_
* be started/stopped many times, related to the queues num, as the * be started/stopped many times, related to the queues num, as the
* exact number of queues used for this device is not known at the time. * exact number of queues used for this device is not known at the time.
* The target has to stop and start the device once got a valid IO queue. * The target has to stop and start the device once got a valid IO queue.
* When stoping and starting the vhost device, the backend bdev io device * When stopping and starting the vhost device, the backend bdev io device
* will be deleted and created repeatedly. * will be deleted and created repeatedly.
* Hold a bdev reference so that in the struct spdk_vhost_blk_dev, so that * Hold a bdev reference so that in the struct spdk_vhost_blk_dev, so that
* the io device will not be deleted. * the io device will not be deleted.

2
mk/spdk.common.mk
View File

@ -74,7 +74,7 @@ ifeq ($(TARGET_MACHINE),x86_64)
ifneq (,$(shell $(CC) --target-help | grep -qe -mavx512f && echo 1)) ifneq (,$(shell $(CC) --target-help | grep -qe -mavx512f && echo 1))
# Don't use AVX-512 instructions in SPDK code - it breaks Valgrind for # Don't use AVX-512 instructions in SPDK code - it breaks Valgrind for
# some cases where compiler decides to hyper-optimize a relatively # some cases where compiler decides to hyper-optimize a relatively
# simple operation (like int-to-float converstion) using AVX-512 # simple operation (like int-to-float conversion) using AVX-512
COMMON_CFLAGS += -mno-avx512f COMMON_CFLAGS += -mno-avx512f
endif endif
endif endif

4
module/bdev/crypto/vbdev_crypto.c
View File

@ -117,7 +117,7 @@ uint8_t g_number_of_claimed_volumes = 0;
#define AESNI_MB_NUM_QP 64 #define AESNI_MB_NUM_QP 64
/* Common for suported devices. */ /* Common for supported devices. */
#define DEFAULT_NUM_XFORMS 2 #define DEFAULT_NUM_XFORMS 2
#define IV_OFFSET (sizeof(struct rte_crypto_op) + \ #define IV_OFFSET (sizeof(struct rte_crypto_op) + \
sizeof(struct rte_crypto_sym_op) + \ sizeof(struct rte_crypto_sym_op) + \
@ -241,7 +241,7 @@ create_vbdev_dev(uint8_t index, uint16_t num_lcores)
} }
/* Before going any further, make sure we have enough resources for this /* Before going any further, make sure we have enough resources for this
* device type to function. We need a unique queue pair per core accross each * device type to function. We need a unique queue pair per core across each
* device type to remain lockless.... * device type to remain lockless....
*/ */
if ((rte_cryptodev_device_count_by_driver(cdrv_id) * if ((rte_cryptodev_device_count_by_driver(cdrv_id) *

8
module/bdev/ftl/bdev_ftl.c
View File

@ -616,7 +616,7 @@ bdev_ftl_finish(void)
} }
static void static void
bdev_ftl_create_defered_cb(const struct ftl_bdev_info *info, void *ctx, int status) bdev_ftl_create_deferred_cb(const struct ftl_bdev_info *info, void *ctx, int status)
{ {
struct ftl_deferred_init *opts = ctx; struct ftl_deferred_init *opts = ctx;
@ -636,8 +636,8 @@ bdev_ftl_examine(struct spdk_bdev *bdev)
int rc; int rc;
LIST_FOREACH(opts, &g_deferred_init, entry) { LIST_FOREACH(opts, &g_deferred_init, entry) {
/* spdk_bdev_module_examine_done will be called by bdev_ftl_create_defered_cb */ /* spdk_bdev_module_examine_done will be called by bdev_ftl_create_deferred_cb */
rc = bdev_ftl_create_bdev(&opts->conf, bdev_ftl_create_defered_cb, opts); rc = bdev_ftl_create_bdev(&opts->conf, bdev_ftl_create_deferred_cb, opts);
if (rc == -ENODEV) { if (rc == -ENODEV) {
continue; continue;
} }
@ -645,7 +645,7 @@ bdev_ftl_examine(struct spdk_bdev *bdev)
LIST_REMOVE(opts, entry); LIST_REMOVE(opts, entry);
if (rc) { if (rc) {
bdev_ftl_create_defered_cb(NULL, opts, rc); bdev_ftl_create_deferred_cb(NULL, opts, rc);
} }
return; return;
} }

2
module/bdev/iscsi/bdev_iscsi.c
View File

@ -177,7 +177,7 @@ bdev_iscsi_finish(void)
/* clear out pending connection requests here. We cannot /* clear out pending connection requests here. We cannot
* simply set the state to a non SCSI_STATUS_GOOD state as * simply set the state to a non SCSI_STATUS_GOOD state as
* the connection poller wont run anymore * the connection poller won't run anymore
*/ */
TAILQ_FOREACH_SAFE(req, &g_iscsi_conn_req, link, tmp) { TAILQ_FOREACH_SAFE(req, &g_iscsi_conn_req, link, tmp) {
_bdev_iscsi_conn_req_free(req); _bdev_iscsi_conn_req_free(req);

2
python/spdk/sma/qmp.py
View File

@ -198,7 +198,7 @@ class QMPClient():
:return command exec response or optionally execute result event :return command exec response or optionally execute result event
:raise QMPRequestError: on response from QMP server being of error type :raise QMPRequestError: on response from QMP server being of error type
:raise QMPSocketError: on timeout or socket errors :raise QMPSocketError: on timeout or socket errors
:raise QMPError: on id missmatch and JSONdecoder errors :raise QMPError: on id mismatch and JSONdecoder errors
''' '''
cmd_id = self._get_next_exec_id() cmd_id = self._get_next_exec_id()
msg = {'execute': cmd, 'id': cmd_id} msg = {'execute': cmd, 'id': cmd_id}

2
scripts/perf/vhost/run_vhost_test.sh
View File

@ -33,7 +33,7 @@ fio_conf() {
fi fi
if [[ -e $fio_extra_conf ]]; then if [[ -e $fio_extra_conf ]]; then
# Overriden through cmdline|env # Overridden through cmdline|env
cat "$fio_extra_conf" cat "$fio_extra_conf"
elif [[ ! -t $fio_extra_conf ]]; then elif [[ ! -t $fio_extra_conf ]]; then
# Attached to stdin # Attached to stdin

2
scripts/rpc.py
View File

@ -2260,7 +2260,7 @@ Format: 'user:u1 secret:s1 muser:mu1 msecret:ms1,user:u2 secret:s2 muser:mu2 mse
p.add_argument('-b', '--buf-cache-size', help='The number of shared buffers to reserve for each poll group', type=int) p.add_argument('-b', '--buf-cache-size', help='The number of shared buffers to reserve for each poll group', type=int)
p.add_argument('-z', '--zcopy', action='store_true', help='''Use zero-copy operations if the p.add_argument('-z', '--zcopy', action='store_true', help='''Use zero-copy operations if the
underlying bdev supports them''') underlying bdev supports them''')
p.add_argument('-d', '--num-cqe', help="""The number of CQ entires. Only used when no_srq=true. p.add_argument('-d', '--num-cqe', help="""The number of CQ entries. Only used when no_srq=true.
Relevant only for RDMA transport""", type=int) Relevant only for RDMA transport""", type=int)
p.add_argument('-s', '--max-srq-depth', help='Max number of outstanding I/O per SRQ. Relevant only for RDMA transport', type=int) p.add_argument('-s', '--max-srq-depth', help='Max number of outstanding I/O per SRQ. Relevant only for RDMA transport', type=int)
p.add_argument('-r', '--no-srq', action='store_true', help='Disable per-thread shared receive queue. Relevant only for RDMA transport') p.add_argument('-r', '--no-srq', action='store_true', help='Disable per-thread shared receive queue. Relevant only for RDMA transport')

2
test/bdev/blockdev.sh
View File

@ -550,7 +550,7 @@ function stat_test_suite() {
trap - SIGINT SIGTERM EXIT trap - SIGINT SIGTERM EXIT
} }
# Inital bdev creation and configuration # Initial bdev creation and configuration
#----------------------------------------------------- #-----------------------------------------------------
QOS_DEV_1="Malloc_0" QOS_DEV_1="Malloc_0"
QOS_DEV_2="Null_1" QOS_DEV_2="Null_1"

2
test/common/autotest_common.sh
View File

@ -1006,7 +1006,7 @@ function daos_cleanup() {
function _start_stub() { function _start_stub() {
# Disable ASLR for multi-process testing. SPDK does support using DPDK multi-process, # Disable ASLR for multi-process testing. SPDK does support using DPDK multi-process,
# but ASLR can still be unreliable in some cases. # but ASLR can still be unreliable in some cases.
# We will reenable it again after multi-process testing is complete in kill_stub(). # We will re-enable it again after multi-process testing is complete in kill_stub().
# Save current setting so it can be restored upon calling kill_stub(). # Save current setting so it can be restored upon calling kill_stub().
_randomize_va_space=$(< /proc/sys/kernel/randomize_va_space) _randomize_va_space=$(< /proc/sys/kernel/randomize_va_space)
echo 0 > /proc/sys/kernel/randomize_va_space echo 0 > /proc/sys/kernel/randomize_va_space

2
test/json_config/config_filter.py
View File

@ -19,7 +19,7 @@ def sort_json_object(o):
sorted_o[key] = sort_json_object(o[key]) sorted_o[key] = sort_json_object(o[key])
return sorted_o return sorted_o
if isinstance(o, list): if isinstance(o, list):
""" Keep list in the same orded but sort each item """ """ Keep list in the same order but sort each item """
return [sort_json_object(item) for item in o] return [sort_json_object(item) for item in o]
else: else:
return o return o

2
test/nvme/cuse/nvme_ns_manage_cuse.sh
View File

@ -71,7 +71,7 @@ function clean_up() {
# This assumes every NVMe controller contains single namespace, # This assumes every NVMe controller contains single namespace,
# encompassing Total NVM Capacity and formatted as 512 block size. # encompassing Total NVM Capacity and formatted as 512 block size.
# 512 block size is needed for test/vhost/vhost_boot.sh to # 512 block size is needed for test/vhost/vhost_boot.sh to
# succesfully run. # successfully run.
tnvmcap=$($NVME_CMD id-ctrl ${nvme_dev} | grep tnvmcap | cut -d: -f2) tnvmcap=$($NVME_CMD id-ctrl ${nvme_dev} | grep tnvmcap | cut -d: -f2)
blksize=512 blksize=512

2
test/nvme/cuse/spdk_smartctl_cuse.sh
View File

@ -62,7 +62,7 @@ if [ "$CUSE_SMART_ERRLOG" != "$KERNEL_SMART_ERRLOG" ]; then
exit 1 exit 1
fi fi
# Data integrity was checked before, now make sure other commads didn't fail # Data integrity was checked before, now make sure other commands didn't fail
${SMARTCTL_CMD} -i /dev/spdk/nvme0n1 ${SMARTCTL_CMD} -i /dev/spdk/nvme0n1
${SMARTCTL_CMD} -c /dev/spdk/nvme0 ${SMARTCTL_CMD} -c /dev/spdk/nvme0
${SMARTCTL_CMD} -A /dev/spdk/nvme0 ${SMARTCTL_CMD} -A /dev/spdk/nvme0

2
test/nvme/perf/common.sh
View File

@ -255,7 +255,7 @@ function create_fio_config() {
total_disks_per_core=$disks_per_core total_disks_per_core=$disks_per_core
# Check how many "stray" disks are unassigned to CPU cores # Check how many "stray" disks are unassigned to CPU cores
# Assign one disk to current CPU core and substract it from the total of # Assign one disk to current CPU core and subtract it from the total of
# unassigned disks # unassigned disks
if [[ "$disks_per_core_mod" -gt "0" ]]; then if [[ "$disks_per_core_mod" -gt "0" ]]; then
total_disks_per_core=$((disks_per_core + 1)) total_disks_per_core=$((disks_per_core + 1))

2
test/setup/devices.sh
View File

@ -153,7 +153,7 @@ dm_mount() {
dm_mount=$SPDK_TEST_STORAGE/dm_mount dm_mount=$SPDK_TEST_STORAGE/dm_mount
dm_dummy_test_file=$dm_mount/test_dm dm_dummy_test_file=$dm_mount/test_dm
# Each partition is 1G in size, join their halfs # Each partition is 1G in size, join their halves
dmsetup create "$dm_name" <<- DM_TABLE dmsetup create "$dm_name" <<- DM_TABLE
0 1048576 linear /dev/$pv0 0 0 1048576 linear /dev/$pv0 0
1048576 1048576 linear /dev/$pv1 0 1048576 1048576 linear /dev/$pv1 0

4
test/sma/nvmf_tcp.sh
View File

@ -121,7 +121,7 @@ delete_device "$devid1"
NOT rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode1 NOT rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode1
[[ $(rpc_cmd nvmf_get_subsystems | jq -r '. | length') -eq 1 ]] [[ $(rpc_cmd nvmf_get_subsystems | jq -r '. | length') -eq 1 ]]
# Finally check that removing a non-existing device is also sucessful # Finally check that removing a non-existing device is also successful
delete_device "$devid0" delete_device "$devid0"
delete_device "$devid1" delete_device "$devid1"
@ -147,7 +147,7 @@ detach_volume "$devid0" "$uuid"
[[ $(rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode0 | jq -r '.[0].namespaces | length') -eq 0 ]] [[ $(rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode0 | jq -r '.[0].namespaces | length') -eq 0 ]]
[[ $(rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode1 | jq -r '.[0].namespaces | length') -eq 0 ]] [[ $(rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:cnode1 | jq -r '.[0].namespaces | length') -eq 0 ]]
# Detach it again and verify it suceeds # Detach it again and verify it succeeds
detach_volume "$devid0" "$uuid" detach_volume "$devid0" "$uuid"
cleanup cleanup

2
test/sma/vfiouser_qemu.sh
View File

@ -204,7 +204,7 @@ NOT rpc_cmd nvmf_get_subsystems nqn.2016-06.io.spdk:vfiouser-1
[[ $(rpc_cmd nvmf_get_subsystems | jq -r '. | length') -eq 1 ]] [[ $(rpc_cmd nvmf_get_subsystems | jq -r '. | length') -eq 1 ]]
[[ $(vm_count_nvme ${vm_no}) -eq 0 ]] [[ $(vm_count_nvme ${vm_no}) -eq 0 ]]
# Finally check that removing a non-existing device is also sucessful # Finally check that removing a non-existing device is also successful
delete_device "$device0" delete_device "$device0"
delete_device "$device1" delete_device "$device1"

2
test/sma/vhost_blk.sh
View File

@ -131,7 +131,7 @@ delete_device "$devid1"
NOT rpc_cmd vhost_get_controllers -n sma-1 NOT rpc_cmd vhost_get_controllers -n sma-1
[[ $(rpc_cmd vhost_get_controllers | jq -r '. | length') -eq 0 ]] [[ $(rpc_cmd vhost_get_controllers | jq -r '. | length') -eq 0 ]]
# Finally check that removing a non-existing device is also sucessful # Finally check that removing a non-existing device is also successful
delete_device "$devid0" delete_device "$devid0"
delete_device "$devid1" delete_device "$devid1"

8
test/unit/lib/bdev/bdev.c/bdev_ut.c
View File

@ -2582,7 +2582,7 @@ bdev_io_mix_split_test(void)
/* IO crossing the IO boundary requires split. /* IO crossing the IO boundary requires split.
* The 1st child IO segment size exceeds the max_segment_size, * The 1st child IO segment size exceeds the max_segment_size,
* So 1st child IO will be splitted to multiple segment entry. * So 1st child IO will be split to multiple segment entry.
* Then it split to 2 child IOs because of the max_num_segments. * Then it split to 2 child IOs because of the max_num_segments.
* Total 3 child IOs. * Total 3 child IOs.
*/ */
@ -2658,7 +2658,7 @@ bdev_io_mix_split_test(void)
/* IO crossing the IO boundary requires split. /* IO crossing the IO boundary requires split.
* The 1st child IO segment size exceeds the max_segment_size and after * The 1st child IO segment size exceeds the max_segment_size and after
* splitting segment_size, the num_segments exceeds max_num_segments. * splitting segment_size, the num_segments exceeds max_num_segments.
* So 1st child IO will be splitted to 2 child IOs. * So 1st child IO will be split to 2 child IOs.
* Total 3 child IOs. * Total 3 child IOs.
*/ */
@ -2708,7 +2708,7 @@ bdev_io_mix_split_test(void)
/* IO crossing the IO boundary requires split. /* IO crossing the IO boundary requires split.
* 80 block length can split 5 child IOs base on offset and IO boundary. * 80 block length can split 5 child IOs base on offset and IO boundary.
* Each iov entry needs to be splitted to 2 entries because of max_segment_size * Each iov entry needs to be split to 2 entries because of max_segment_size
* Total 5 child IOs. * Total 5 child IOs.
*/ */
@ -2750,7 +2750,7 @@ bdev_io_mix_split_test(void)
} }
TAILQ_INSERT_TAIL(&g_bdev_ut_channel->expected_io, expected_io, link); TAILQ_INSERT_TAIL(&g_bdev_ut_channel->expected_io, expected_io, link);
/* 5th child IO and because of the child iov entry it should be splitted /* 5th child IO and because of the child iov entry it should be split
* in next round. * in next round.
*/ */
expected_io = ut_alloc_expected_io(SPDK_BDEV_IO_TYPE_WRITE, 64, 16, 8); expected_io = ut_alloc_expected_io(SPDK_BDEV_IO_TYPE_WRITE, 64, 16, 8);

2
test/unit/lib/bdev/nvme/bdev_nvme.c/bdev_nvme_ut.c
View File

@ -5788,7 +5788,7 @@ test_set_preferred_path(void)
CU_ASSERT(io_path->nvme_ns->ctrlr->ctrlr == ctrlr2); CU_ASSERT(io_path->nvme_ns->ctrlr->ctrlr == ctrlr2);
/* If io_path to ctrlr3 is set to the preferred path and then a new I/O channel is /* If io_path to ctrlr3 is set to the preferred path and then a new I/O channel is
* aquired, find_io_path() should return io_path to ctrlr3. * acquired, find_io_path() should return io_path to ctrlr3.
*/ */
spdk_put_io_channel(ch); spdk_put_io_channel(ch);

6
test/unit/lib/blob/blob.c/blob_ut.c
View File

@ -1585,7 +1585,7 @@ _blob_io_read_no_split(struct spdk_blob *blob, struct spdk_io_channel *channel,
uint8_t *buf; uint8_t *buf;
uint64_t page_size = spdk_bs_get_page_size(blob->bs); uint64_t page_size = spdk_bs_get_page_size(blob->bs);
/* To be sure that operation is NOT splitted, read one page at the time */ /* To be sure that operation is NOT split, read one page at the time */
buf = payload; buf = payload;
for (i = 0; i < length; i++) { for (i = 0; i < length; i++) {
spdk_blob_io_read(blob, channel, buf, i + offset, 1, blob_op_complete, NULL); spdk_blob_io_read(blob, channel, buf, i + offset, 1, blob_op_complete, NULL);
@ -1609,7 +1609,7 @@ _blob_io_write_no_split(struct spdk_blob *blob, struct spdk_io_channel *channel,
uint8_t *buf; uint8_t *buf;
uint64_t page_size = spdk_bs_get_page_size(blob->bs); uint64_t page_size = spdk_bs_get_page_size(blob->bs);
/* To be sure that operation is NOT splitted, write one page at the time */ /* To be sure that operation is NOT split, write one page at the time */
buf = payload; buf = payload;
for (i = 0; i < length; i++) { for (i = 0; i < length; i++) {
spdk_blob_io_write(blob, channel, buf, i + offset, 1, blob_op_complete, NULL); spdk_blob_io_write(blob, channel, buf, i + offset, 1, blob_op_complete, NULL);
@ -2561,7 +2561,7 @@ bs_load_after_failed_grow(void)
snprintf(opts.bstype.bstype, sizeof(opts.bstype.bstype), "TESTTYPE"); snprintf(opts.bstype.bstype, sizeof(opts.bstype.bstype), "TESTTYPE");
/* /*
* The bdev_size is 64M, cluster_sz is 1M, so there are 64 clusters. The * The bdev_size is 64M, cluster_sz is 1M, so there are 64 clusters. The
* blobstore will create 64 md pages by defualt. We set num_md_pages to 128, * blobstore will create 64 md pages by default. We set num_md_pages to 128,
* thus the blobstore could grow to the double size. * thus the blobstore could grow to the double size.
*/ */
opts.num_md_pages = 128; opts.num_md_pages = 128;

2
test/unit/lib/nvmf/tcp.c/tcp_ut.c
View File

@ -1240,7 +1240,7 @@ test_nvmf_tcp_pdu_ch_handle(void)
(unsigned)sizeof(struct spdk_nvme_tcp_term_req_hdr)); (unsigned)sizeof(struct spdk_nvme_tcp_term_req_hdr));
CU_ASSERT(tqpair.mgmt_pdu->hdr.term_req.fei[0] == 3); CU_ASSERT(tqpair.mgmt_pdu->hdr.term_req.fei[0] == 3);
/* Test case: All parameters is conformed to the functon. Expect: PASS */ /* Test case: All parameters is conformed to the function. Expect: PASS */
tqpair.recv_state = NVME_TCP_PDU_RECV_STATE_AWAIT_PDU_CH; tqpair.recv_state = NVME_TCP_PDU_RECV_STATE_AWAIT_PDU_CH;
tqpair.pdu_in_progress->hdr.common.pdu_type = SPDK_NVME_TCP_PDU_TYPE_IC_REQ; tqpair.pdu_in_progress->hdr.common.pdu_type = SPDK_NVME_TCP_PDU_TYPE_IC_REQ;
tqpair.state = NVME_TCP_QPAIR_STATE_INVALID; tqpair.state = NVME_TCP_QPAIR_STATE_INVALID;

2
test/unit/lib/sock/sock.c/sock_ut.c
View File

@ -1150,7 +1150,7 @@ override_impl_opts(void)
opts.impl_opts = &impl_opts; opts.impl_opts = &impl_opts;
opts.impl_opts_size = sizeof(impl_opts); opts.impl_opts_size = sizeof(impl_opts);
/* Use send_buf_size to verify that impl_opts get overriden */ /* Use send_buf_size to verify that impl_opts get overridden */
send_buf_size = impl_opts.send_buf_size; send_buf_size = impl_opts.send_buf_size;
impl_opts.send_buf_size = send_buf_size + 1; impl_opts.send_buf_size = send_buf_size + 1;

2
test/unit/lib/util/cpuset.c/cpuset_ut.c
View File

@ -121,7 +121,7 @@ test_cpuset_parse(void)
rc = spdk_cpuset_parse(NULL, "[1]"); rc = spdk_cpuset_parse(NULL, "[1]");
CU_ASSERT(rc < 0); CU_ASSERT(rc < 0);
/* Wrong formated core lists */ /* Wrong formatted core lists */
rc = spdk_cpuset_parse(core_mask, ""); rc = spdk_cpuset_parse(core_mask, "");
CU_ASSERT(rc < 0); CU_ASSERT(rc < 0);

2
test/unit/lib/vhost/vhost.c/vhost_ut.c
View File

@ -68,7 +68,7 @@ DEFINE_STUB(rte_vhost_get_vring_base_from_inflight, int,
DEFINE_STUB(rte_vhost_extern_callback_register, int, DEFINE_STUB(rte_vhost_extern_callback_register, int,
(int vid, struct rte_vhost_user_extern_ops const *const ops, void *ctx), 0); (int vid, struct rte_vhost_user_extern_ops const *const ops, void *ctx), 0);
/* rte_vhost_user.c shutdowns vhost_user sessions in a separte pthread */ /* rte_vhost_user.c shutdowns vhost_user sessions in a separate pthread */
DECLARE_WRAPPER(pthread_create, int, (pthread_t *thread, const pthread_attr_t *attr, DECLARE_WRAPPER(pthread_create, int, (pthread_t *thread, const pthread_attr_t *attr,
void *(*start_routine)(void *), void *arg)); void *(*start_routine)(void *), void *arg));
int int