When an I/O with children is being freed, also free its child I/O
requests that were allocated via spdk_bdev_get_child_io().
Change-Id: I2d44aed845c1035ae8f8cb07c5992da855f1dc99
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This callback was only used for freeing buffers, but the buffers are now
managed by the bdev core, so none of the free_request callbacks actually
do anything.
Change-Id: Icfe2e6169e829159dda5e3d75a27d8f040de07c6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Add unmap support to the ramdisk block device for testing purposes.
Change-Id: Ibeb5530b2b5a31603d09d2d1de07760f32dea0f8
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The bdev layer can be used independently of iSCSI, so fix the
misleading names.
Change-Id: I3fd5b113403acdd7578ce93234dde0fd4f148e96
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Check that the number of blocks/ranges in the command fits within the
length specified by the SGL.
Change-Id: I21aded797dc1f1e752fe0bc9cec27310a4fb106a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The Dataset Management command allows several operations to be specified
at once; the virtual controller only supports deallocate for now, but it
should just ignore the other bits in order to be spec compliant: "If the
Dataset Management command is supported, all combinations of attributes
[...] may be set".
The spec also explicitly states that it is acceptable for controllers to
choose to take no action based on information provided, so not
implementing the other attributes is fine.
Change-Id: Ia989dc1faa9c852660bf1299ea18fa8e7bdf4053
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Also add a diagnostic message if the requested log page ID is not
supported.
Change-Id: I7551b5905d5ebc29356839f0f9153dc86f237106
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Rather than comparing the bdev name against "NVMe", use the new I/O type
supported API to query whether the unmap operation is supported.
Change-Id: I62c7a1ea5529366ff2ae4723b62f24ea78aa8193
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Bdev modules need a separate interface than public
consumers of the blockdevs.
Change-Id: I581ee493570c114f7e96b31a425bc077a791c71e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This compilation unit depends on bdev.h definitions, but
was only getting them due to #include ordering elsewhere.
Change-Id: I4fcbdb2582a40836bcabc3539cc558614fbfacfd
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Some block devices do not support the unmap operation, and we may add
other optional I/O types in the future. Add a method to check which I/O
types a specific block device supports.
Change-Id: I6e6414bf6b6482ea0224022d8326b252bd363c7f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Switch from the non-portable <sys/endian.h> functions (htobeXX/beXXtoh)
to the SPDK endian conversion functions.
Change-Id: Id49b87f2e536c68f0d5d567e78e1990c0a37ef14
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Intel DC P3*** NVMe devices specify a desired stripe size, which was
used for splitting I/O. Not all devices, however, specify a desired
stripe size (such as the Intel DC D3*** line), and for only these
devices there was a logic mistake that overwrote the maximum I/O
size with a 2MB default. This patch corrects that error.
Change-Id: I94b72a3a3dd1dfa18bd638daf7e01a592eb6ed17
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Move the NQN validation into the subsytem creation function, and fix the
allowed size to match the spec.
The spec is not clear about the allowed NQN size; for now, interpret it
as 223 bytes, including the null terminator (222 bytes of actual NQN
plus one terminator byte).
Change-Id: If9743ab2fe009d9d852e8b03317d9b38d8af18dc
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
DPDK 16.07 introduced a new PCI ID field for matching by class code
instead of vendor/device ID. Use it to match all NVMe devices instead of
explicitly listing vendor and device ID pairs.
Change-Id: Ib2a5cc6833bf2b793d37d77caab97207f365df8f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
SUBNQN is a UTF-8 null terminated string according to the NVMe base
spec, so pad it with zeroes using strncpy().
Change-Id: I486161b26d91f3ea1fd17428e220b9f20a874732
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
These are specified as "ASCII string", which means they should be
left-aligned and padded with spaces, according to the NVMe base
specification.
Change-Id: I25babe0ca417c2e16137b0bfc41fc7834277114e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will be useful outside of the SCSI code, so put it in the common
string utility file.
Also reorder the parameters so they match the order used in strncpy().
Change-Id: I9e25a59b64e4bedf04e5a96de463b1d8aa0ddac3
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Clean up the poller and only then free the associated subsystem's
memory. This prepares for future dynamic subsystem creation/deletion.
Change-Id: I9e56cbf8822814930fdbb662095c51b6ad40fbc4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Currently the NVMf target listens for new connections on any address.
Instead, listen only on the addresses specified by the user.
Change-Id: Idb6d37c422e442fc70a8673bd3fcfb9c27b57828
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
assert is part of the C standard library and is available
on any platform we'd consider porting to. Don't put a
wrapper around it.
Change-Id: I0acfdd6a8a269d6c37df38fb7ddf4f1227630223
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
pthreads are widely supported and are available on any
platform we currently foresee porting to. Use that API
instead of attempting to abstract it away to simplify
the code.
Change-Id: I822f9c10910020719e94cce6fca4e1600a2d9f2a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
pthreads are widely supported and are available on any
platform we currently foresee porting to. Use that API
instead of attempting to abstract it away to simplify
the code.
Change-Id: I28123d427ea8da07c6329b0233f0702f2d85c2a0
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Comments are not allowed in the JSON RFC, but some JSON libraries accept
JavaScript-style comments.
Add a flag that enables non-spec-compliant comment parsing.
Change-Id: I9dfb66bb46ecff1a22d8af5a9c50620686a4707c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use the event framework's new delay parameter to allow
for idle cores to sleep for up to 1ms at a time.
Change-Id: I665f38e590c07338418892afe0e75b0b2c79706e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
It is no longer needed, since the nvmf_tgt app handles initialization
and shutdown.
Change-Id: I051afe2b4fcbd09b32998386c63f591a0ab343c2
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The user can now specify a maximum delay, in microseconds, that
defines the maximum amount of time a reactor will sleep for
between polling for new events. By default, the time is 0
which means the reactor will never sleep.
Change-Id: I94cddb69c832524878cad97b66673daa4bd5c721
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This will be used in future patches outside the library.
Change-Id: I1fcf5709944a884e161e5a6a9eaec033a995a812
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The NVMe over Fabrics target library now exposes a simple function call
that polls the acceptor once, and the application handles registration
of the poller.
Also rename the transport function pointers related to the acceptor so
they better reflect their purpose.
Change-Id: I5fa0d516586bf17e73afeb88ff3c2d5b0d46794d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will become more important when other transports are added.
For now, it is also useful to be able to start nvmf_tgt on systems
without RDMA hardware.
Change-Id: I6b9002cc7711f928c4e6b73adcd9b677349ebdd6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
spdk_shutdown_nvmf_subsystems() was removing the subsystem from the
list, but nvmf_delete_subsystem() also wants to remove it, so drop the
extra removal.
Also rewrite the shutdown loop as a TAILQ_FOREACH_SAFE() to make the
static analyzer happy (and make it more obvious that the loop will
terminate).
Change-Id: Iccadafa77d9cd3e26be21c0f11e62cfc1ef0197c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Verify that the record format is the one we support (only 0 is defined
by the spec for now).
Change-Id: Iddf038b381e540134abf572e0545c97a0ef71d5f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The spec requires that NQNs are null terminated and maximum of 223 bytes
long, despite the Connect command fields being larger (256 bytes), so
add checks for both subsystem NQN and host NQN before using them as null
terminated strings.
Change-Id: I343d9e44a09ab4d0f6654feba460b31e976c4e56
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Since we bind the NVMe device to UIO driver to protect against native
NVMe driver, but for Admin queue, there are still INTx interrupts
exist, as all the completion for Admin queue will be processed in
user space, so we don't need INTx anymore.
Change-Id: Ife5b3e410ae95690ed0f3f9a2f2dfaf55a7797b5
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Users can specify the core for each subsystem and the acceptor listen routine
to run on different cores for performance consideration.
Change-Id: I4bd1a96f39194c870863b4b778e6ea7cf8fc1a2d
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
This is causing issues during shutdown because the poller removal is not
synchronized with the rest of the cleanup path.
This reverts commit 7dfc5e922d.
Change-Id: If95c4b72c5d120f18bdc3db6d7d532ad1aada642
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The lcore_id field in the get_iscsi_connections RPC was removed in
commit 5d8c94536a7d1d4c1f0ee3349188bf0e7e8c9e74; add a field to
spdk_iscsi_conn to track the lcore so this can be re-added.
Change-Id: I6c9574829466b168880728f4620401987fc7dd3c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This should enhance performance, since the hardware admin queue poll
function takes a mutex and should not be in the performance path.
Change-Id: I7e4acde0337aaf7079811612cba5348acf0a467d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This leaves more flexibility for future changes to the poller
representation without requiring API changes (after this one).
It also prevents the user from accidentally using poller fields in a
non-thread-safe way, since they can't be accessed directly anymore.
Change-Id: I7677d5b93668665d29ae39c5e0ba74333ad3f878
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Replace other critical rte_zmalloc() sites that actually depend on the
memory being zeroed.
Change-Id: If6856ad44a4c50869811d3ce9411c993ce88018d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Linux block layer driver will use the maximum transfer length field to
split IOs larger than this value. We should set the field according to
iSCSI target limitation.
Change-Id: I03ee35bb96f0949418bb976a6c8013f88622a324
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Allow the tables to be in the read-only data section.
Change-Id: I58199a86d4d44dbad7baed397b2e148c45b3a3de
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
rte_zmalloc() is broken and does not actually return zeroed memory on at
least DPDK 16.07 on FreeBSD, so do it ourselves.
Change-Id: If8da93ead0b3911c8bca24aa27ed90dc00b8a9a4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
For VPD page 0xB1 and 0xB2, the scsi target did not return correct
value to the initiator, so return the length with correct value.
Change-Id: Ic17d804ca00d490fd6a2f833db5c9b73ce8dc160
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
This value was incremented and decremented, but it was never used
otherwise.
Change-Id: I6e83a504cf2ef4043363ca04b77556c612068658
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
In files that don't otherwise use DPDK, switch to the standard C library
assert().
Change-Id: I79756908ecf9a2e141b036321e42309db30b5e0f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The NVMe submission queue head wraparound point can be determined in the
generic NVMe over Fabrics layer; it should not be using the RDMA
connection queue depth.
Change-Id: I9da8f09e4f057f8fdc1ff4c6cc5f48cea7123e11
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Report the maximum admin queue size correctly.
Change-Id: I52cad654bf59806e0abb8d869c22973647056617
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use the max_queue_depth parameter rather than rdma_conn->max_queue_depth
so that we can start to eliminate rdma_conn->max_queue_depth.
Change-Id: I1670c634e6d12aa004fb5a10338b7624850fbc4a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
There were two unchecked allocations in the nvmf library. Check
for allocation failures.
Change-Id: Ic6b3104d825dba1ee6bd1748fa99e132702f300c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This fixes a static analysis warning for unsigned/signed
mismatch.
Change-Id: I49bd8d6d195f13b402e14a85503a5de6114f5b7f
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This is the size of a logical block in bytes; 4 GB is more than plenty.
Also allows cleaning up casts to uint32_t in the SCSI translation layer.
Change-Id: I3ec2e2f41fd378f1a83f31aac25c46ef780f63e9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The large buffer pool allocation was using the per-connection queue
depth, whereas the RDMA memory region registration was using the global
RDMA max queue depth. These sizes need to match, so use the global RDMA
max queue depth for both calls.
Change-Id: Iae161b719e09e19ca3e81df6593b68a4a2e86614
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is a step towards enabling sharing SPDK NVMe
device access from multiple processes using DPDK's
multi-process framework.
Change-Id: I57d5eec158b42addc1036bd2583596471a467a95
Signed-off-by: GangCao <gang.cao@intel.com>
Similar to our NVMf target, this is an iSCSI target that
can interoperate with the Linux and Windows standard iSCSI
initiators.
Change-Id: I6961c5ef99f7b161c396330ed5b543ea29b0ca7b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This is a useful abstraction when you want to plug in
a userspace networking layer instead of using the kernel.
Change-Id: I7039d2987e6abad9dcd1987fa105282b1598e2f5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The public header file was missing some required definitions.
Change-Id: Ic4f8028367b1e21ea00c02660ca36be28da54e37
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Use the new timer-based poller functionality to replace rte_timer.
Change-Id: Ic40653306cc73b40139fe18e06bab29b35721a43
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Allow pollers to be scheduled to be run periodically every N
microseconds instead of every iteration of the reactor loop.
Change-Id: Iaea3e98965d81044e6dc5ce5f406bcb7a455289e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Just getting a reference to a bdev should not claim it.
Change-Id: I21e07160662490ec95b52fa31ea1d2ae93a21f09
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Combine the necessary functionality with the main bdev file.
Change-Id: I96d796bc87ac2a8688cdf1fd3c16d2a7c8aef730
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The rte_ring used for pollers is already single-producer and
single-consumer, so it is not providing any thread safety guarantees.
ALl modifications to the active_pollers ring are done from the core that
is running the reactor (via events). This means the rte_ring can be
replaced with a simpler intrusive linked list.
This simplifies the removal of pollers in the middle of the list and
avoids extra allocations for the ring.
Change-Id: Ica149b7a1668a8af1e6ca8f741c48f2217f6f9bf
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
We reported virtualized NVMe devices through NVMe over Fabric specification,
with 1.2.1 NVMe version. For direct mode, the NVMe device maybe has lower
version, such as 1.0, the identify namespace list can not support in those
devices, so we need to add helper function here to simulate such commands
from initiator.
Change-Id: I226f4f34bf61017f538d2dd80332f1d054a501f1
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Allow higher queue depths by allowing many more send/recv
operations than read/write.
Change-Id: I66c424a6463e5e09be6d5463667241ce9271404b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The target can only provide updates to sq_head inside
of completions. Therefore, we must update sq_head prior
to sending the completion or we'll incorrectly get into
queue full scenarios.
Change-Id: If2925d39570bbc247801219f352e690d33132a2d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This allows the target to poll for internal completions
at higher priority.
Change-Id: I895c33a594a7d7c0545aa3a8405a296be3c106fb
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This ensures that the data buffers are not in use
when we go to send the completion.
Change-Id: I30467b3e3964001150f81b21e5b695dcd0974b0c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This is useful for holding session-wide buffer pools.
Change-Id: I7024da24b210a2205bf1e159d5935e0093b81120
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
For small SGLs, even if they are keyed and not inline, use the
buffer we allocated for inline data.
Change-Id: I5051c43aabacb20a4247b2feaf2af801dba5f5a9
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Read/Write depth is much lower than Send/Recv depth.
Calculate them separately to prepare for supporting
a larger number of receives than read/writes.
Currently, the target still only exposes a queue depth
equal to the read/write depth.
Change-Id: I08a7434d4ace8d696ae7e1eee241047004de7cc5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
These don't actually work quite yet, but pipe the
configuration file data through to where it will
be needed.
Change-Id: I95512d718d45b936fa85c03c0b80689ce3c866bc
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
For each connection, allocate a single buffer each
of requests, inline data buffers, commands, and
completions.
Change-Id: Ie235a3c0c37a3242831311fa595c8135813ae49e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This can be used to release requests that don't
require a completion to be sent.
Change-Id: I8fb932ea8569bf3c45342d9fa4e270af5510c60c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
PORT IDs indicate hardware failure domains according
to the NVMf specification, which means they should
indicate which transport addresses are on the same
NIC. Unfortunately, that doesn't really make sense for
IP-based fabrics because IP addresses can move. The
safest way to present this is to show all IP addresses
as part of different subsystem ports.
Change-Id: I056a50c69be70b4fbf1f896e684ce65bd792241e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The NVMe over Fabrics 1.0 spec corresponds to the NVMe base spec version
1.2.1, so we should pretend to be at least that new.
Change-Id: I36fc44c780de01d6c666e87b803cd47dba0e74c5
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
These belong in nvme_spec.h anyway and are not used.
Change-Id: I889dfebee523dc5ae503fd0370bb800f1d17fb5d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is a leftover from a previous controller numbering scheme that is
no longer used.
Change-Id: I3058802f0324b0e38708111634ee993c6e884087
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Move the ctrlr and io_qpair out of spdk_nvmf_subsystem, package them
as a new data structure. Union the direct and virtual mode namespaces.
Change-Id: I839aee3372c6c57aa03a0be76f8aaeb5045ecdaf
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
CAP.CQR indicates whether contiguous queues are required; this is
meaningless in NVMe over Fabrics, since queue creation is handled
implicitly for each connection, but the spec requires it to be set to 1.
Change-Id: I6b05954eefa6928beecd7a640bbbdbd835c6b69a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use the size of the applicable structs directly.
Change-Id: I4a65de548d409c9962b11a75d3fde2bfe434a3ec
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
nvmf_create_subsystem() already copies the name, so the strdup() in the
caller is unnecessary.
Change-Id: I225f0f077fee30051b197a4b1d7276b113ec6b01
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
It isn't actually necessary to drain the cq before
destroying it.
Change-Id: I6f77ae578176a14b5de935274a14cfd165229ec5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This logically belongs inside the session handling code, not
in the transport-specific layer.
Change-Id: I93b2271f38dbfc742162c98c40acb153c7e9022a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Track and print out the currently outstanding I/O in debug
mode with rdma tracing enabled.
Change-Id: I0a1f0cd6e22dbf21e18ca0ec7d0c2c6d194509e3
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Instead of reimplementing handling for checking the
completion queue, nvmf_rdma_accept can now call
the general purpose poller.
Change-Id: Id2c899d1e500a8cb8491e51cc101a1bf0e167764
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
AER breaks our current model of requests/completion pairs.
Temporarily handle it by immediately re-posting the
capsule while we work on a real solution.
Change-Id: Ie7a4d88030b6fff5a11c4697eec0f024f9737f27
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Inline this code into the places that called it. These two
spots will be combined into a single path in a later patch.
Change-Id: Ice2f009ad56b783dc28ebbf1abbb877ce6000293
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This is an RDMA-specific operation, so hide it inside
the transport-specific layer.
Change-Id: Iaa097e8dde78d820547b3a39e9717c992581340b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
These can be done at the same time now that the queue depth
is known ahead of time.
Change-Id: I7ecef30ebb4311e0a1c88f37461d34534f8600bf
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Calculate queue depth into a local variable without
touching the rdma_conn.
Change-Id: Ie804ed39ddecbf59015a4e4f7aa127f1381d9080
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
