The user should not see the bdev_io status directly; the NVMe and SCSI
error code wrappers provide the ability to translate to the desired
format regardless of what kind of error is stored inside the bdev_io.
Replace the spdk_bdev_io_completion_cb status parameter with a bool
simply indiciating whether the I/O completed successfully.
Change-Id: Iad18c2dac4374112c41b7a656154ed3ae1a68569
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362047
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This will make sure the hotplug instance not exit before all the hot insert and
remove events done, the setup.sh for hot insert events will use several seconds
sometimes.
Change-Id: I4ad61937760529547cdcf7701f3c3d9033121e45
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/362131
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This is no longer used anywhere. For the places where we previously
used it, we've since found alternate solutions that do not
require it.
Change-Id: I738a80b95ef50348ce1c14969a3812b0a625b3fd
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362064
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This isn't strictly necessary, but follow the
pattern for sync and I/O devices.
Change-Id: I34b1bd6f9da06008e419f55ee9810022f5fed3d5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361674
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This allows us to no longer use the unique=true flag as well
as assign sync channels different queue depths.
Change-Id: I0b3129a3139bf39d745c2a4dc1d9f402c0a8faa8
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361673
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This allows us to have a different queue depth
on I/O channels than on metadata/sync channels
Change-Id: I30dfc95af550082efffec4b12ed6c4cd4b8c11cf
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361672
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This temporarily increases the size of the sync channel
to 512 (like all of the other channels). This will be
fixed in a later patch.
Change-Id: I679f1cb7c9f7afdad0acf9a3a26f9632f60cc510
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361670
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This allows us to allocate different size channels and
not require the unique flag.
Change-Id: I4b1ffd244b60e9e9ab06f9ab4da8161ab57e1169
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361668
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Piotr Pelpliński <piotr.pelplinski@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
An ECN to the NVMe 1.3 specification has clarified that the NQN may
contain 223 bytes before the null terminator. Make all of our NQN
length checks consistently enforce this behavior.
Change-Id: Iebfd57d11abea64964c7a6ad9d886e40efa243c3
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The channel memory isn't allocated by these
libraries, so they can't free it.
Change-Id: I30909fa4e77bc5a41b45230f04ba5fe75b172dbf
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Separate the maximum metadata operations from the
maximum channel operations.
Change-Id: I1bbd440ab094a2a2e19c9a5b71724ac91ba88e42
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This will need to be configured globally for all channels.
Change-Id: I773252f220373617f8d09d1f24243db8095cf8a4
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Adding group write permission is needed in case of using vfio
driver and using libvirtd for vm provisioning to avoid
permission errors.
Change-Id: I5348e74c65bd8639071c1a2a164f28849dde9678
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
Specify the function as a parameter to spdk_scsi_dev_queue_mgmt_task.
This makes the API clearer by making it explicit that the management
function is required for management tasks.
Change-Id: I92d893aadb6faebea81dd79729894d2c2fddf088
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The type of a task is known due to the function used to submit it:
- spdk_scsi_dev_queue_task() for normal SCSI command tasks
- spdk_scsi_dev_queue_mgmt_task() for management tasks
Change-Id: I183a1f89ab85f3fce1de2491e77d95d4b147fd72
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The SCSI read and write functions use the same LBA range check, so move
it up to spdk_bdev_scsi_readwrite().
spdk_bdev_scsi_read() previously passed task->transfer_len / blen rather
than xfer_len to the LBA range checking function, but this should
actually be using the transfer length field from the CDB (which is the
xfer_len parameter).
Change-Id: I8285abf936a18a0baf7cc25709945637e4e6b87d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Collect coverage info when running UT locally when lcov
is available.
Change-Id: I34a8b60056775a25cd4a3906d59c4a42e7cc845f
Signed-off-by: paul luse <paul.e.luse@intel.com>
Added new rpc commands together with underlying vhost API and tests.
Change-Id: Ib9c6a530d0909193ea5115aaac4920c44f39613c
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
If more than one controller is added to guest the same memory table is
set multiple times making VFIO registration/unregistration failures
Change-Id: Ib55c38e292495e549c070beb0234e73f269e63d5
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
This allows the iscsi_tgt tests to be run more easily
in isolation.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1fa87e8b02f6e6ed62c75f07471d93ce84da30c2
It'll trigger interrupts even if VRING_AVAIL_F_NO_INTERRUPT is present.
Also delegate notifying check to separate function vq_should_notify. It
will be expanded even more once we implement VIRTIO_RING_F_EVENT_IDX
support.
Change-Id: Ic41fd18c9c005c6266b109b0eda4a6dfaef5f06a
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Added new struct field and an rte_vhost_get_negotiated_features() call.
Change-Id: I5937026720b5341c7377c27fee682996b51e836e
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
This will allow a git clone to default to an SPDK-specific
version of the DPDK repository. Users can still override
to use a separate DPDK repository/installation with the
--with-dpdk configure script options.
While here, remove gzip option for the git-archive operations
in autopackage.sh. We need to add a git-archive for the DPDK
submodule if we are using it, and compressing at -9 adds a lot
of unnecessary time. Since we are not archiving these packages,
there is no need to compress them. Also explicitly disable
coverage and ubsan for the autopackage build, since this build
is only to test compilation and is not actually used for any
test execution.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3cf8a2ed984003a175cdece6542636ede8cb2479
The latency tracking is done with ranges of bucket arrays.
The bucket for any given I/O is determined solely by TSC
deltas - any translation to microseconds is only done after
the test is finished and statistics are printed.
Each range has a number of buckets determined by a
NUM_BUCKETS_PER_RANGE value which is currently set to 128.
The buckets in ranges 0 and 1 each map to one specific TSC
delta. The buckets in subsequent ranges each map to twice
as many TSC deltas as buckets in the previous range:
Range 0: 1 TSC each - 128 buckets cover deltas 0 to 127
Range 1: 1 TSC each - 128 buckets cover deltas 128 to 255
Range 2: 2 TSC each - 128 buckets cover deltas 256 to 511
Range 3: 4 TSC each - 128 buckets cover deltas 512 to 1023
Range 4: 8 TSC each - 128 buckets cover deltas 1024 to 2047
Range 5: 16 TSC each - 128 buckets cover deltas 2048 to 4095
etc.
While here, change some variable names and usage
messages to differentiate between the existing latency
tracking via vendor-specific NVMe log pages on Intel
NVMe SSDs, and the newly added latency tracking done
in software.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I299f1c1f6dbfa7ea0e73085f7a685e71fc687a2b
The bdev generation count is an internal implementation detail;
applications should not be reading or writing it.
Change-Id: Ic4455b7b72dc80babbc410420c89f52ef5d77e6d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Move some of the fields in spdk_bdev_io (the layout of which is no
longer public API) to put related fields together and improve struct
packing.
Change-Id: I0ca5d9441b00c65703c35391d125059eaa4bcee3
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Add a helper function to find the iSCSI task given its nested SCSI task
structure, and use it to remove all casting between spdk_scsi_task and
spdk_iscsi_task.
Change-Id: Idc7c6d0a3b4d53041916d25a1bdecedfb56b94f9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch address the issue:
https://github.com/spdk/spdk/issues/151.
For cache_append_no_cache in cache_ut testcase,
there is resource contention for buffer among two
threads in the following two functions.
Thread 0: cache_free_buffers
Thread1: __file_flush_done
When the thread1 execuctes __file_flush_done,
it calls the call back: __sem_post defined in
following statement in spdk_file_sync
_file_sync(file, channel, __sem_post, &channel->sem);
Thus Thread 0 will execute next function
cache_buffers, and it frees the buffer.
Then Thread 1 continues executing the remaining statements
in __file_flush_done with the assert function, and touches
the space already freed.
So it will be safe to move ahead the next buffer check.
Change-Id: Ic007b3481f4e3a17d47eeca5c9c802001949a5ab
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Also change the discovery/nvmf.sh test to use it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I56bce9a84bd46f13b6d4f34da81abf23413f2598
Factor out the common data direction Check Condition logic from read
and write.
Change-Id: Ia2fbd5f69eefd826ebb5ab2a8e99aeb380f1b7ea
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
What was previously called "maxlba" was actually the block count
of the device, which is the maximum LBA plus one.
Change-Id: I5e141a3eb0c79b2aa70006739c5586cb50744d3d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
