Now that virtual mode namespaces can be chosen arbitrarily, there can be
holes in the ns_list. Make sure all of the virtual mode code paths are
prepared to handle these inactive but valid NSIDs.
Change-Id: I0d70fd9dae37a29a8358e1d38dcc6cac3237fd9a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/365717
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
In this patch, spdk_app_init will be merged into
spdk_app_start.
Due to this change, we need to change event_perf
to use spdk_app_start, since we cannot use spdk_app_init
anymore. So the related changes for event_perf is to
make it work with reactor framework.
Change-Id: Id67edf209fd628cca361a499068c93aeedfe6167
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/364153
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I272c2e56984abbb820758866297ee9e916af625a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/363605
Reviewed-by: John Meneghini <johnm@netapp.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: <ed.rodriguez@netapp.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
MAX_VIRTUAL_NAMESPACE will be removed from the public API, so use an
RPC-specific array size to remove its use in nvmf_tgt.
Also increase the number of namespaces accepted by the RPC call; this
does not actually allow more than MAX_VIRTUAL_NAMESPACE right now, since
the namespaces are still stored in a fixed-size array inside the
subsystem, but at least the RPC side is not limited by that size now.
In the future, we will make the subsystem namespace list dynamically
sized.
Change-Id: I99c77ddeae4c13ddde34e2b945c663dbafdb68d9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363310
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: I61b6579e8698d16e5a8ab74d304af9ea53f9dce4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363307
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
spdk_nvmf_subsystem_add_ns() already checks to see if there is space for
the new namespace.
Change-Id: I9c66311a4f1eb289e87089ab7fa04e8c05b68631
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363305
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Also move the validation of serial number length into the library.
Change-Id: Ibc9384fc1fccb87bd15b75da7f3942570900bd96
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363304
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Change-Id: Icecbc0c8f43ad894e43e6cd1e818978615b56fd6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363302
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
spdk_nvmf_subsystem_exists(nqn) is exactly equivalent to
spdk_nvmf_find_subsystem(nqn) != NULL.
Change-Id: Icafa43166c99d1364b49cef30a5ade4b6625bcd5
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363299
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This patch do the following optimization:
1 when the nvmf target run as secondary process, the nvme probe will
not call the probe callback function, which will skip the step to
set up the ctx->found, so we set up it when we call attach callback.
2 when the nvmf target run as secondary process, every time when we try
to probe one device, it will iterate all the attached devices and call
the attach callback, for nvmf target, it will cause the overwrite of
the nvme controller for the nvmf subsystem, so we need to compare the
attached deivce with the device which we try to attach, if they are not
the same one, just return.
Change-Id: I606ba75b1dab7fd1b1e4e1cfcf63248a506725d1
Signed-off-by: cunyinch <cunyin.chang@intel.com>
Reviewed-on: https://review.gerrithub.io/362807
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This used to be only for RPC, but now it is used in both
the config file and RPC paths.
Change-Id: I138a23f87e9042be60460bc5451470a2443ac3c3
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Added a new function spdk_nvmf_tgt_start() that initializes and
starts the nvmf app. The main function now only parses the command
line args and calls into spdk_nvmf_tgt_start().
Change-Id: I886a4021762a11c3c1f6ebfff5ed72adc409a34e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
spdk_nvmf_tgt_listen() can fail due to invalid user input (e.g. from a
config file), so we can't just assert() that it worked.
Change-Id: Iea2d4ab2c82d79f6eada24e6082456489bf304b8
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
it will free invalid address when we start nvmf using profile,
this patch fix the bug.
Change-Id: I5100eadc2fd129a5b9e3588decac1429d03d78a4
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
SPDK_COUNTOF works like sizeof, except it returns the number of elements
in an array instead of the number of bytes.
Change-Id: I38ff4dd3485ed9b630cc5660ff84851d0031911f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
It is caused by this commit:
4163626c5cd310fa1c7fca2e02ccf41ffaf6d7f4
nvmf_tgt_delete_subsystem registers the poller
with function subsystem_delete_event.
subsystem_delete_event is called asynchronously,
the deletion should happen in this function.
Otherwise, with the current code,
g_subsystems_shutdown = true
TAILQ_EMPTY(&g_subsystems) = true
when subsystem_delete_event is firstly called.
If there are multiple subsystems, the logic is wrong.
Thus other subsystem will never be delete. since
we already execute shutdown_complete().
Also add related test scripts.
Change-Id: I3823563fc9e8611c11a6d798685ff64e2939842e
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
subsystem_delete_event():
Delete TAILQ_REMOVE because nvmf_tgt_shutdown_subsystem_by_nqn has
already called TAILQ_REMOVE.
shutdown_subsystems():
Add TAILQ_REMOVE according to the above change.
nvmf_tgt_delete_subsystems():
Add TAILQ_REMOVE.
By default, all SPDK applications will not share memory.
To share memory, start the applications with the same
shared memory id.
Change-Id: Ib6180369ef0ed12d05983a21d7943e467402b21a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Use the construct_nvmf_subsystem RPC call handler code from the
nvmf.conf parser to factor out some of the duplicated code.
Change-Id: I1b8f9af03f4a774cbb5c06dda7c57286e8968204
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is the same check as the subsystem configuration file parsing uses.
Change-Id: Id301a929e06f51b060256a3038e0c24120ccd308
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The host and port output parameters point into the (non-const) char *ip,
so it makes more sense for them to be non-const as well.
This allows the flexibility to pass non-const char pointers as the
output parameters, which will be used in the nvmf_tgt/conf.c parsing
code.
Change-Id: I1d5b102fc389c06d36432904e4fda944437b659e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
These were repeated a few different places, so pull them into a common
header file.
Change-Id: Id807fa2cfec0de2e0363aeb081510fb801781985
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The phenoemon is that we can not shutdown the nvmf tgt.
The solution is that we need to adjust the shutting down orders of
nvmf tgt subsystem and rdma trasport layer.
Change-Id: Ie39657370b1574960e0ee7cf604cc5872db0bed3
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
These strings are not modified by the functions they are passed to, so
they can be const char *.
Change-Id: I11532f232990a305d706c14aac1b0f8f93b8f576
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
When we do frequent same subsystem add/delete,
we will face the adding issue. For example,
1 Add subsystem A
2 Delete subsystem A
3 Add subsystem A (Fail in this step).
The reason is that we did not correctly free
the listener resources of subsystems, and this patch
can solve this issue.
Change-Id: I6765a306a3f10c9a0f38c95dbba12e2a4073e705
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
The 'next' event pointer was never used in the entire code base (always
NULL).
Change-Id: I75f999d3a2e10512d86edec1a5a46ef263e2635b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use 'struct spdk_event *' directly for consistency with the rest of the
API.
Change-Id: Ib41a9bf47f5b18f4aebf5f4dee055455cb12ef7d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This allows the elimination of the spdk_event_get_arg1() and
spdk_event_get_arg2() macros, which accessed the event structure
directly; this was preventing the event structure definition from being
moved out of the public API header.
Change-Id: I74eced799ad7df61ff0b1390c63fb533e3fae8eb
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The public API user is supposed to retrieve the defaults via the
spdk_app_opts_init() function.
Change-Id: Ie2bd6e809b2d47dbd5d62d396e8715f89f4052d9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The spdk_poller_register() function provides a way to pass an event to
call once the poller is registered, but it is always NULL in the current
code base.
Change-Id: I459bf40ae4d050589577d113b7984f1563aaa9cc
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
