trace: do not define fn in SPDK_TRACE_REGISTER_FN

Newer lcov try to detect issues with function definitions, and throw
warnings if function definitions have overlapping lines. Since
SPDK_TRACE_REGIS

trace: do not define fn in SPDK_TRACE_REGISTER_FN

Newer lcov try to detect issues with function definitions, and throw
warnings if function definitions have overlapping lines. Since
SPDK_TRACE_REGISTER_FN defines both the constructor function and
the trace function, it treats this as an overlap.

Note this is the only one of our REGISTER functions in SPDK that
effectively defines two functions in the same macro.

So change the macro so that the user actually defines the
trace function instead.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Ib9d0d1af7674c5ff9b65c12c1e4a5b05e6e110f5
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/24836
Reviewed-by: Michal Berger <michal.berger@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot

show more ...

scsi pr: only registrants are consider holders when reservation in mode
7/8

When reservation is in 'all registrant' mode (7/8), we should
consider all registrants as reservation holders. but host wh

scsi pr: only registrants are consider holders when reservation in mode
7/8

When reservation is in 'all registrant' mode (7/8), we should
consider all registrants as reservation holders. but host who is not
a registered cannot be treated as a reservation holder,
other wise it might get read/write access when it should not.

Change-Id: Ifa4bd677a49d62e8459d6da22535b5652b18fb50
Signed-off-by: Oded Mashbach <oded.mashbach@silk.us>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23615
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>

show more ...

scsi pr: passig all registrant reservation to next registrant inline

When a registrant releases an "all registrant" reservation (mode 7/8),
we should pass the reservation holder to the first registr

scsi pr: passig all registrant reservation to next registrant inline

When a registrant releases an "all registrant" reservation (mode 7/8),
we should pass the reservation holder to the first registrant in the
lun's
registrant list that is not the same registrant that is releasing the
reservation.
other wise we might face the following bugs:
1. when A is the first registrant and B is the second:
Once A will release his mode 7/8 reservation - it will stay under A
and not move to B.
2. when A is the only registrant that exist, Once A will try to release
the reservation of mode 7/8 - it will not work.

Change-Id: Ib4f1e82283725638c5d2ea57b6ec7c19fff543aa
Signed-off-by: Oded Mashbach <oded.mashbach@silk.us>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/23614
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

trace: rename owner to owner_type

We will be adding the real concept of an "owner" in upcoming patches. So change
the existing "owner" to "owner_type" which more accurate reflects its meaning.
For e

trace: rename owner to owner_type

We will be adding the real concept of an "owner" in upcoming patches. So change
the existing "owner" to "owner_type" which more accurate reflects its meaning.
For example, an "owner_type" can be OWNER_BDEV, but an actual "owner" would
be some specific bdev identified by its name.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Ieb25eb5a574a88b3812291f2345796e5bb66a54f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22645
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

lib/scsi: fixed potential expression overflow.

Fixed potential issue found by Coverity.
The function _bytes_to_blocks() takes num_bytes parameter,
which is 64-bit long (uint64_t). In bdev_scsi_write

lib/scsi: fixed potential expression overflow.

Fixed potential issue found by Coverity.
The function _bytes_to_blocks() takes num_bytes parameter,
which is 64-bit long (uint64_t). In bdev_scsi_write_same()
function, the parameter was calculated as
task->length * xfer_len
Both of the values are 32-bit long, so the expression is
evaluated using 32-bit arithmetic potentially overflowing
the result (truncating to 32-bit). To avoid this, we need
to cast at least one parameter to 64-bit value.

Change-Id: Idd8a52d088a0c971158515a0a542a44595633321
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/22219
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>

show more ...

scsi: simplify split path

Currently the splitting code for unmap and write_same is a bit complicated,
since it needs to handle case where we get a 0-length unmap descriptor and
spdk_bdev_unmap_block

scsi: simplify split path

Currently the splitting code for unmap and write_same is a bit complicated,
since it needs to handle case where we get a 0-length unmap descriptor and
spdk_bdev_unmap_blocks() didn't handle 0-length calls.

But spdk_bdev_unmap_blocks() has been updated now to handle 0-length calls,
so we can eliminate this extra complexity.

This also alleviates a small regression identified when this splitting code
path was updated.

Fixes issue #3249.
Fixes b735c429 ("scsi: Improve child IO split mechanism").

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I93cdac1c449a17fcc22e9909beb0feafe2b08b54
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21773
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Ben Walker <ben@nvidia.com>

show more ...

scsi: Re-order struct spdk_scsi_lun

Move *removed* and *resizing* to the front of struct spdk_scsi_lun,
there are 2 benefits:
- reduce memory
- the two fields work in hot path, they are tested on ea

scsi: Re-order struct spdk_scsi_lun

Move *removed* and *resizing* to the front of struct spdk_scsi_lun,
there are 2 benefits:
- reduce memory
- the two fields work in hot path, they are tested on each task
execution

Also move *ref* to the neighbor of *pr_generation*. Finally, we reduce
16 bytes(1080 -> 1064).

Change-Id: Ia3705a0e765fc5183fce836eb8f6684a79cd1353
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21923
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Mellanox Build Bot

show more ...

scsi: Move registrant fields to the end of struct spdk_scsi_lun

Registrant related fields use the biggest part of struct
spdk_scsi_lun, and it's usually *not* hot code path. Move them to
the end of

scsi: Move registrant fields to the end of struct spdk_scsi_lun

Registrant related fields use the biggest part of struct
spdk_scsi_lun, and it's usually *not* hot code path. Move them to
the end of structure.

Also swap 'uint32_t pr_generation', is's possible to reduce memory
in the next step.

Suggested-by: Jim Harris <jim.harris@samsung.com>
Change-Id: Ie544efe01ca42162c056dc84deec5e5c338a4eb9
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20709
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot

show more ...

scsi: Improve child IO split mechanism

Improve the child IO split mechanism with the points:
- simplify error handing for splitting loop.
- ctx->fn() does not have to touch any counter. then
ctx->

scsi: Improve child IO split mechanism

Improve the child IO split mechanism with the points:
- simplify error handing for splitting loop.
- ctx->fn() does not have to touch any counter. then
ctx->remaining_count is removed from _bdev_scsi_unmap and
_bdev_scsi_write_same.

Change-Id: I543fcbb6eede662ec248e68c2d377d3b3a488961
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20559
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Mellanox Build Bot

show more ...

so_ver: increase all major versions after SPDK 24.01

To allow SO_MINOR updates on LTS for the whole year it is supported,
the major version for all components needs to be increased.
This is to preve

so_ver: increase all major versions after SPDK 24.01

To allow SO_MINOR updates on LTS for the whole year it is supported,
the major version for all components needs to be increased.
This is to prevent scenario where two versions exists with matching
versions, but conflicting ABI.
Ex. Next SPDK release adds an API call increasing the minor version,
then LTS needs just a subset of those additions.

Increasing major so version after LTS, allows the future releases
to update versions as needed. Yet allowing LTS to increase minor
version separately.

This patch:
- increases SO_VER by 1 for all components
- resets SO_MINOR to 0 for all components

Due to patch below being introduced after v24.01 code freeze,
bdev lib version should not be increased:
61623c5ca9 bdev: add spdk_bdev_io padding in place of io_submit_ch

Short reference to how the versions were changed:
MAX=$(git grep "SO_VER := " | cut -d" " -f 3 | sort -ubnr | head -1)
for((i=$MAX;i>0;i-=1)); do find . -name "Makefile" -exec \
sed -i -e "s/SO_VER := $i\$/SO_VER := $(($i+1))/g" {} +; done
find . -name "Makefile" -exec \
sed -i -e "s/SO_MINOR := .*/SO_MINOR := 0/g" {} +

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I0166d6be6c17bbbf985d48b5dfcb36f1d4af1b48
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/21582
Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

lib/scsi: fixed potential out-of-bonds read

This patch fixes issue detected by Coverity scan.
The spdk_scsi_task_names table is accessed using
'task->function' as an index. Earlier check allows the

lib/scsi: fixed potential out-of-bonds read

This patch fixes issue detected by Coverity scan.
The spdk_scsi_task_names table is accessed using
'task->function' as an index. Earlier check allows the
'task->function' to be equal to
SPDK_COUNTOF(spdk_scsi_task_names), while
the table can only be accessed using indexes from
0..(SPDK_COUNTOF()-1) range.
The 'if' statement needs to be modified to reject index
equal to SPDK_COUNTOF(spdk_scsi_task_names)

Change-Id: I569ea530cee3721a085ddd8c7f90dd7761d644be
Signed-off-by: Marcin Spiewak <marcin.spiewak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20932
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>
Community-CI: Mellanox Build Bot

show more ...

scsi: round task->iov.iov_base allocation up to nearest 4 bytes

Some SCSI commands such as INQUIRY may generate an odd number
of bytes in its response, not evenly divisible by 4.

But iSCSI must ali

scsi: round task->iov.iov_base allocation up to nearest 4 bytes

Some SCSI commands such as INQUIRY may generate an odd number
of bytes in its response, not evenly divisible by 4.

But iSCSI must align all transfer-in PDUs to 4-byte multiple.
After enabling DPDK ASAN, we discover that the SCSI task
response buffer was allocating exactly the size of the INQUIRY
reponse, but iSCSI rounds this up, and ends up doing a sendmsg()
off the end of the buffer for those trailing bytes.

We could try to fix this in iSCSI layer, but it makes the code
more complex - it would need to allocate some separate buffer for
those extra bytes, or allocate a new buffer and copy the bytes
over.

So instead fix this down in the SCSI layer when the buffer is
allocated, still report the iov_len as the smaller amount, but
allocate an actual buffer that is rounded up to the nearest 4.

Fixes issue #3203.

Signed-off-by: Jim Harris <jim.harris@samsung.com>
Change-Id: I702fe1b543c95388fa1dfa8c7cafb2538917479f
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20853
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

scsi: Add notice log on bdev reset

Once we hit the following logs:
nvme_ctrlr.c:1639:nvme_ctrlr_disconnect: *NOTICE*: [0000:b2:00.0] resetting controller
bdev_nvme.c:2027:_bdev_nvme_reset_ctrlr_comp

scsi: Add notice log on bdev reset

Once we hit the following logs:
nvme_ctrlr.c:1639:nvme_ctrlr_disconnect: *NOTICE*: [0000:b2:00.0] resetting controller
bdev_nvme.c:2027:_bdev_nvme_reset_ctrlr_complete: *NOTICE*: Resetting controller successful.

We also need a notice log to find out why the bdev resets.

Change-Id: I5cf9e3fe8bfa950ab6e33e2c038d83fa8ba9d52d
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20652
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

scsi: Add SPDK_SCSI_TASK_FUNC_TARGET_RESET

SCSI/iSCSI triggers bdev_scsi_reset() in two reasons:
- mgmt command from initiator side
- target side cleanup

For example, once we hit the following logs

scsi: Add SPDK_SCSI_TASK_FUNC_TARGET_RESET

SCSI/iSCSI triggers bdev_scsi_reset() in two reasons:
- mgmt command from initiator side
- target side cleanup

For example, once we hit the following logs:
nvme_ctrlr.c:1639:nvme_ctrlr_disconnect: *NOTICE*: [0000:b2:00.0] resetting controller
bdev_nvme.c:2027:_bdev_nvme_reset_ctrlr_complete: *NOTICE*: Resetting controller successful.

We need to know why the bdev/nvme resets, rather than a single
SPDK_SCSI_TASK_FUNC_LUN_RESET reason, separate it into two to
distinguish the root cause of the lower layer bdev reset.

Change-Id: I087add31c31e46780c6e7c60cc8af14290e84636
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20361
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

vhost-scsi: fix incorrect parameter in compare_and_write IO

We should not pass task->iov to spdk_bdev_comparev_and_writev_blocks
because it is NULL.

Change-Id: Iac12b5b24e040e6414c7ef9bd2d4e3dc5ec3

vhost-scsi: fix incorrect parameter in compare_and_write IO

We should not pass task->iov to spdk_bdev_comparev_and_writev_blocks
because it is NULL.

Change-Id: Iac12b5b24e040e6414c7ef9bd2d4e3dc5ec31894
Signed-off-by: Haoqian He <haoqian.he@smartx.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20602
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

scsi: Drop 'name' field in spdk_bdev_scsi_split_ctx

Child IO split mechanism[1] uses 'name' field to dump the necessary
message on failure, now we can use spdk_scsi_sbc_opcode_string
instead.

[1]SP

scsi: Drop 'name' field in spdk_bdev_scsi_split_ctx

Child IO split mechanism[1] uses 'name' field to dump the necessary
message on failure, now we can use spdk_scsi_sbc_opcode_string
instead.

[1]SPDK Commit: f6613e370d75 ("scsi: Abstract child IO split mechanism")
Change-Id: I6f7cca10f7340d6008022671a990ff8b5d157b62
Suggested-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20465
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot

show more ...

scsi: Add helper function spdk_scsi_sbc_opcode_string

Translate SCSI operation code and service action into string, human
readable string is helpful for debug. Exporting this function as
public, it'

scsi: Add helper function spdk_scsi_sbc_opcode_string

Translate SCSI operation code and service action into string, human
readable string is helpful for debug. Exporting this function as
public, it's possible to be used outside(ex, virtio scsi).

Change-Id: I74231c63bc73fcd6acfee385021e0bcc460cbf7f
Suggested-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20464
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.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: Changpeng Liu <changpeng.liu@intel.com>

show more ...

scsi: Check descriptor count for UNMAP

There is no need to call split IO(bdev_scsi_split) on two case:
- invalid descriptor count
- zero descriptor count(newly added)

Suggested-by: Shuhei Matsumoto

scsi: Check descriptor count for UNMAP

There is no need to call split IO(bdev_scsi_split) on two case:
- invalid descriptor count
- zero descriptor count(newly added)

Suggested-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I80a7bae711fedc957319d5d44c55dac8c59f5ef5
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20513
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

vhost-scsi: support SBC-3 compare_and_write IO

Support compare_and_write from SBC-3 for 1 block since
spdk_bdev_comparev_and_writev_blocks function supports
compare_and_write of only 1 block current

vhost-scsi: support SBC-3 compare_and_write IO

Support compare_and_write from SBC-3 for 1 block since
spdk_bdev_comparev_and_writev_blocks function supports
compare_and_write of only 1 block currently.

Change-Id: I1069fa6c9070a5f16b8867f62740a2dce8a282da
Signed-off-by: Haoqian He <haoqian.he@smartx.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19873
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

scsi: Fix MAXIMUM WRITE SAME LENGTH field

Originally, the 'MAXIMUM WRITE SAME LENGTH' feild has dependence on
Bdev UNMAP, once the Bdev has no support on UNMAP(Ex, AIO), the
initiator side gets 0.

scsi: Fix MAXIMUM WRITE SAME LENGTH field

Originally, the 'MAXIMUM WRITE SAME LENGTH' feild has dependence on
Bdev UNMAP, once the Bdev has no support on UNMAP(Ex, AIO), the
initiator side gets 0.

According to spec:
A MAXIMUM WRITE SAME LENGTH field set to a non-zero value indicates
the maximum value that the device server accepts in the NUMBER OF
LOGICAL BLOCKS field for a WRITE SAME command. A MAXIMUM WRITE SAME
LENGTH field set to zero indicates that the device server does not
report a limit on the number of logical blocks that may be requested
for a single WRITE SAME command.

Unlimited maximum blocks on a single request is not expected for SPDK,
fix 'MAXIMUM WRITE SAME LENGTH' field.

Change-Id: I2005156daa198afe0738236f32bfd61eeb3b8356
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20401
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>

show more ...

scsi: Implement WRITE SAME

Implement SPDK_SBC_WRITE_SAME_10 and SPDK_SBC_WRITE_SAME_16 commands.

Change-Id: Ie1a82cea8e8afd8c8e8e490eadb358c9c825b865
Signed-off-by: zhenwei pi <pizhenwei@bytedance.

scsi: Implement WRITE SAME

Implement SPDK_SBC_WRITE_SAME_10 and SPDK_SBC_WRITE_SAME_16 commands.

Change-Id: Ie1a82cea8e8afd8c8e8e490eadb358c9c825b865
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20400
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>

show more ...

scsi: Abstract child IO split mechanism

Child IO split mechanism[1] for UNMAP allows a single SCSI command
to split into several SPDK Bdev IOs.
Separate the common child IO split mechanism from the

scsi: Abstract child IO split mechanism

Child IO split mechanism[1] for UNMAP allows a single SCSI command
to split into several SPDK Bdev IOs.
Separate the common child IO split mechanism from the UNMAP codes,
it's possible to implement SCSI WRITE SAME easily in the next step.

[1] SPDK commit: 73bc324b0df('scsi: Refine split processing for unmap')

Change-Id: I3248f51d2cdc1380e20ddaccc074369bf2b48ec8
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/20399
Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Community-CI: Mellanox Build Bot
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>

show more ...

scsi: Refine split processing for unmap

If any child I/O failed, we should set the status of the SCSI task
to failed and stop further split processing.

However, ctx->count meant number of outstandi

scsi: Refine split processing for unmap

If any child I/O failed, we should set the status of the SCSI task
to failed and stop further split processing.

However, ctx->count meant number of outstanding children I/Os. We
could not stop further split processing. We updated the status of the
SCSI task only if child I/O submission failed.

Fix these in this patch. Verify the changes by refining unit tests
except for error cases.

This idea was derived from the splitting process of the generic bdev
layer.

Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: Ifc74c09c349567b0311fe2a7435a5452d4c68251
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19437
Reviewed-by: Jim Harris <jim.harris@samsung.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot

show more ...

scsi: use notice log for unsupported VPD page

Very new Linux kernel sends VPD 0xb9 (concurrent positioning ranges)
even SPDK doesn't report this page is supported, we can see this
error log in iSCSI

scsi: use notice log for unsupported VPD page

Very new Linux kernel sends VPD 0xb9 (concurrent positioning ranges)
even SPDK doesn't report this page is supported, we can see this
error log in iSCSI and vhost-scsi target, here use a notice log
instead.

Change-Id: I26fd2afaee2152ffbfea6ef4e0f254538360b626
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19714
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...

scsi: Factor out internal of bdev_scsi_unmap() into a helper function

Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I789d7dbf6585e7aeb49b888557f940148c6378f3
Reviewed-on: https:

scsi: Factor out internal of bdev_scsi_unmap() into a helper function

Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
Change-Id: I789d7dbf6585e7aeb49b888557f940148c6378f3
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/19436
Reviewed-by: Jim Harris <jim.harris@gmail.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>

show more ...