struct raid_bdev has a pointer raid_bdev_config to its configuration,
but raid_bdev is duplicated between raid_bdev and raid_bdev_config
and just config may be enough as a name.
Change-Id: I77f6a69b913cd540c22f05803c486a4caf103f24
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422923
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
In bdev_raid.c and bdev_raid_rpc.c, to hold a pointer to
raid_bdev_config, some have used raid_cfg and other have used
raid_bdev_config.
raid_bdev_config is the name of the struct and using the name of
the struct as a pointer to it may cause any confusion.
Change-Id: Ic87ff04d58ea46c30081aaae8d9cda5a53135be9
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422922
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
base_bdevs_io_channel is good but base_channel may be enough and
fit other bdev modules.
Change-Id: I67a1d224f1ef4ca1fc048b4325333f2552a37150
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422921
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
In bdev_raid.c, to hold a pointer to raid_bdev_io_channel, some
have used ch and other have used raid_bdev_io_channel.
Using raid_ch as that purpose throughout will be consistent and
differentiation to pointers to spdk_io_channel, and will improve
readability.
Change-Id: Ib10729f9fa3ebbfdb7b7cc013386b9b5cc0e93cb
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422920
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
raid_base_bdev_info is sufficiently descriptive as name and so
names of member variables in raid_base_bdev_info can be simplified.
Change-Id: I759a758ec0f3ac21de767ad379a902f1b04cc66d
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422919
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Kunal Sablok <kunal.sablok@intel.com>
A number of modules (RAID, logical volumes) have logical
"stripes" that require splitting an I/O into several
child I/O. For example, on a RAID-0 with 128KB strip size,
an I/O that spans a 128KB boundary will require sending
one I/O for the portion that comes before the boundary to
one member disk, and another I/O for the portion that comes
after the boundary to another member disk. Logical volumes
are similar - data is allocated in clusters, so an I/O that
spans a cluster boundary may need to be split since the
clusters may not be contiguous on disk.
Putting the splitting logic in the common bdev layer ensures
bdev module authors don't have to always do this themselves.
This is especially helpful for cases like splitting an I/O
described by many iovs - we can simplify this a lot by
handling it in the common bdev layer.
Note that currently we will only submit one child I/O
at a time. This could be improved later to submit multiple
child I/O in parallel, but the complexity in the iov splitting
code also increases a lot.
Note: Some Intel NVMe SSDs have a similar characteristic.
We will not use this bdev stripe feature for NVMe though -
we want to primarily use the splitting functionality inside
of the NVMe driver itself to ensure it remains fully
functional. Many SPDK users use the NVMe driver without
the bdev layer.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ife804ecc56f6b2b55345a0d0ae9fda9e68632b3b
Reviewed-on: https://review.gerrithub.io/423024
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Remove mutex to guard discovery_auth_group in
spdk_iscsi_get_authinfo() because update of discovery_auth_group
is atomic even no mutex.
Change-Id: Id9ddcb84c2fc3c820984e52dda7ca73839497e24
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422765
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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 patch series will refactor/improve CHAP authentication and consolidate
it into iscsi_subsystem.c.
To get better reviews, this patch just moves spdk_iscsi_chap_get_authinfo()
from iscsi.c to iscsi_subsystem.c.
Change-Id: I953f5c851bfe67dc02f6f82966132b4216e79228
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/422764
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
An new RPC set_iscsi_target_node_auth enables users to configure CHAP
authentication for the target node dynamically.
New setting is used for new iSCSI sessions. For existing iSCSI sessions,
new setting is not used until user logout and login again.
Try to use descriptive message in the RPC when error occurs.
Change-Id: I0bd40d92262d708c1f7de0effb208078bdf8cc41
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/420974
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
An new RPC set_iscsi_discovery_auth enables users to configure
CHAP authentication for discovery sessions dynamically.
Try to use descriptive message in the RPC when error occurs.
Change-Id: I8883d5e054539cf6db091a5b79d0479e62119811
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/420976
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
Current parameter names of `set_iscsi_options` and `get_iscsi_global_params`
RPC method for CHAP in discovery session may not be clear and long
compared with CHAP in login to target nodes.
This patch changes parameter names of `set_iscsi_options` and
`get_iscsi_global_params` RPC method for CHAP in discovery session
from no_discovery_auth, req_discovery_auth, req_discovery_auth_mutual,
and discovery_auth_group to disable_chap, require_chap, mutual_chap,
and chap_group, respectively.
Old parameters are still usable because decoder of them are not removed.
Change-Id: If3102f1233d57ee3cadfb733a6fc1fca14a0e972
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423050
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
Current variable names for CHAP in discovery session may not be
clear and long compared with CHAP in login to target nodes.
This patch changes variable names for CHAP in discovery session
from no_discovery_auth, req_discovery_auth, req_discovery_auth_mutual,
and discovery_auth_group to disable_chap, require_chap, mutual_chap,
and chap_group, respectively.
By this patch, the term "discovery" is removed but this will not
cause any confusion because the code of the use case of them is
clear and an new RPC set_iscsi_discovery_auth will be added in the
subsequent patches.
Change-Id: Ia57041b54b28a19d5d2d90ea6c6665937c25fefc
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/423049
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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 value is used to calculate the disk utilization of a given bdev.
Change-Id: I4bf101c524b92bdd21573941e17f61db59c5c6b8
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423017
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
In the documentation we indicate that the application specific opt
string cannot have any overlap with the generic opt string. This change
provides a run-time sanity check to ensure that the user has not broken
this rule.
Change-Id: Iaa0d913ad609276b28d5f6baeb4218113e2bb559
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/422914
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Update spdk target config files and codes to only enable
16 trace group at most.
Change-Id: I1bd26ccea05d73cea54cbbcf2fcefa869d621352
Signed-off-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-on: https://review.gerrithub.io/422478
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
qp_context is only available for QP related events.
For other events we should not update ibv state as we try
to access null object data field.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Id8d2fee090d9a40c7e00c866914c2eb164e7587c
Reviewed-on: https://review.gerrithub.io/422941
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When testing blobfs, the allocation of cache buffer may fail
without any error information, users may get segment fault
error after that.
Change-Id: I6dc8ee4b93c6a9109aba193f599197caa4c4c383
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/422943
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This will be needed for using this same descriptor when
splitting an I/O.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Idec759df7ab27f8de567d3c8a4214e25dbe173f5
Reviewed-on: https://review.gerrithub.io/423022
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
In the case of failing to spdk_nvmf_poll_group_add_subsystem()
operation, the subsystem still needs to initialize the related
queue so that later coming request can be properly queued.
Also needs to correctly handle the expected state in this failed
condition so that when destroying the subsystem, it could be
properly handled.
Change-Id: I419f2ac7164c25258c3911952c38b9433fca762b
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/422799
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Instead of waiting until the first listen address is added,
create a protection domain and a memory map for every RDMA
device in the system. This consumes more resources when there
are RDMA devices that aren't used by the target, but it
will simplify some order of operations issues when listen
addresses and poll groups are added and removed at run
time.
Change-Id: Idfe6f8307decbf19e02765dbf67f03c2510a328f
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/422602
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
New function was added in bdev layer to allow
handling spdk_bdev_io buffer exhaustion.
This patch adds that functionality to nvmf bdev.
Change-Id: I5b915abcccec930c3eb684d95bc64482e7887af5
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/418937
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
So we don't need to allocate memory (maybe failed) just for free other
memory.
Change-Id: I2c83f6acc2aa6ed79455bff90f952a2e70b44d59
Signed-off-by: wuzhouhui <wuzhouhui@kingsoft.com>
Reviewed-on: https://review.gerrithub.io/422203
Chandler-Test-Pool: 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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This allows us to avoid creating a shared configuration file in the dpdk
rte directory.
Change-Id: Id57d9111669ee8fd51e82c0b05e8c2c07b6dcd7e
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/421652
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Previously a null buffer on read would get passed on which would
be fine for a simple example but for something more complex where
the module in question might need that buffer at some time after
the read callback, it's possible that an underlying module might
free it too soon. So, to be a good example we now own the buffer
in this module.
Change-Id: I411477e83b1e222c64011688952cfd587c32c16e
Signed-off-by: paul luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/422603
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
Currently, the RXE kernel driver does not support send with invalidate.
There is a change to the kernel making its way downstream that will
enable this feature. At that point, we can conditionally enable
send-with-invalidate based on the kernel version.
Change-Id: I05c7bcbf8ec944be89c10bdf6ccc3229e4586914
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/422579
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The passthru module associates a vbdev and the underlying bdev.
This association lives throughout the lifetime of the module and because
of this, passthru vbdevs are recreated when they shouldn't be.
This patch ensures that the association between a vbdev and a bdev is
broken when the vbdev is removed, so that if the underlying bdev is ever
recreated (after being deleted), the vbdev does not reappear.
Note that this change does not affect hot-remove operations.
Change-Id: I5cb4cf7efcb5e5eeaeef79849e0aaefec4684a4f
Signed-off-by: Wael Halbawi <waelhalbawi@gmail.com>
Reviewed-on: https://review.gerrithub.io/422474
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The lvol->unique_id will never change - so just point the
bdev->name directly to it. That eliminates the need to
free bdev->name when cleaning up the bdev, and removes
an extra strdup.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I6547edb1fd4c248fba696c07326842b073d3716c
Reviewed-on: https://review.gerrithub.io/421408
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Ideally lvols would not point to bdevs (bdev/lvol
should be layered on top of lvol). This is a small
step towards making that happen.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie5dff245a0be0dcfe9a8a0ce3e79677be5faa696
Reviewed-on: https://review.gerrithub.io/421407
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The intents of these arrays was to keep track in the
bdev layer of all base<->virtual bdev relationships -
i.e. which member disk bdevs make up a RAID bdev,
which logical volume bdevs are associated with a
bdev that contains an lvolstore, etc.
Currently none of this is used however. And trying
to keep track in the bdev layer instead of asking
the bdev modules for the relationships has a number
of complications. Early one, we tried to do this
with TAILQs - but that doesn't work since this can't
be done with a single TAILQ_ENTRY in the bdev
structures. So we moved to arrays - that works a bit
better, but then the pointer arrays have to be
realloc'd which isn't ideal.
The biggest problem though with these arrays is that
they held bdev pointers - not bdev descriptor pointers.
It's not really valid to access bdevs without a
descriptor - the descriptors are what make sure active
references are accounted for when a bdev is hotplugged.
Of course the bdev layer knows when a bdev is getting
removed and could go and do the updates to these
arrays separately - but that just seems very convoluted.
So for now just remove these arrays completely. If
there is a future need for the bdev layer to
understand relationships between bdevs, we can add
module APIs so that the generic layer can ask
the modules about the relationships.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I99ef1068240bff1262f64f234260cf2fb44df51d
Reviewed-on: https://review.gerrithub.io/420932
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
The base_bdev and vbdevs arrays are going away in the
common bdev layer in a future patch, so prepare for that
in the split RPCs to use the bdev_part interface instead
to get the names of the split bdevs created from the
base_bdev.
No change in observed functionality.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib31807258b253ded3c9bac4f936ed57aa0df7534
Reviewed-on: https://review.gerrithub.io/420931
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This ensures (for example) that a RAID volume is
unregistered before its member disks.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7a7c16acc351f2d5d4218b64b370e2c77c6e2b5e
Reviewed-on: https://review.gerrithub.io/420812
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
During system shutdown, we want each module to release its
claim and close any open descriptors when one of its bdevs
is unregistered.
This prepares for a future patch which will unregister
bdevs at shutdown in reverse order - i.e the RAID bdev
will be unregistered before its member disks.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib4c2b9d25463e420dbc4ec5d3d7030e3d6e8d981
Reviewed-on: https://review.gerrithub.io/420927
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Fixed various misstatements, mentioned long params, and described
some parameters in more depth.
Change-Id: Id3b37e810c7b9f626b1f7e42ba72af99b8b574ae
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/422537
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
On hot remove, before closing bdev, put its IO channel.
If we don't do this bdev won't be destroyed until controller
is removed.
This fixes GitHub issue #401.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I92600c2f7136f56afd6593f663f2d2cf468a37dc
Reviewed-on: https://review.gerrithub.io/422258
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
The GPT module isn't designed currently to support extended
block sizes like 520 or 4104, and we don't really have a good
way to put GPT down on one of these devices anyways.
So just check the block size when examining a new bdev, and
fail immediately if it's not an even multiple of 512.
Long-term, this should probably be a DEBUG print and not an
ERRLOG, but while we work through the bdev implications
of extended block sizes, keep it as an ERRLOG to facilitate
debugging.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1bce0e3b1f0edee2f4d002edf3aaa7e6c16d9677
Reviewed-on: https://review.gerrithub.io/422321
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Liang Yan <liang.z.yan@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
spdk_nvme_ns_get_sector_size returns the sector
size of the data only. This new function adds
in the metadata size, if any.
While here, modify the bdev/nvme driver to use this
function - this is needed for ongoing extended sector
size work through the bdev layer.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic8070a7f8d29b0b2ac2a2d65e0df5f4736488351
Reviewed-on: https://review.gerrithub.io/422445
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
some vendorse support less send sge then SPDK_NVMF_MAX_SGL_ENTRIES.
Change-Id: I5b550b537b6ff4ae5d7876a3f277f88cf06049e4
Signed-off-by: Leonid Ravich <Leonid.Ravich@dell.com>
Reviewed-on: https://review.gerrithub.io/421012
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
The admin queue pair may get disconnected before
the controller is entirely destroyed and can't
be relied on to obtain the correct thread.
Change-Id: I5e80ef286693d53a161134610dd8354c458f8390
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/422134
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: qun wan <qun.wan@intel.com>
Change-Id: I7716f1748803872cac85e296d60747752ca046f4
Signed-off-by: wuzhouhui <wuzhouhui@kingsoft.com>
Reviewed-on: https://review.gerrithub.io/422273
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
When an SPDK application shuts down, the bdev layer will
automatically unregister all of the bdevs to ensure they
are properly quiesced and cleaned up.
Some modules may want to perform different operations when
a bdev is destructed during normal runtime vs. shutdown.
For example, for lvol, when the last lvol is cleaned up,
it should unload the lvolstore, release and close the bdev
that contains the lvolstore. You never want to do this
during normal runtime though - it is perfectly valid to
have an lvolstore that contains no lvols. RAID and future
bdev modules such as multipath have similar use cases.
So add a new bdev module callback named "fini_start".
If a module specifies a function pointer for this callback,
the bdev layer will call it before it starts the bdev
unregistrations.
This enables some future patches to the bdev layer such
that it will always unregister block devices that are not
claimed (i.e. logical volumes) before block devices that
are claimed (i.e. the bdev containing an lvolstore).
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I6e87f5c2b27f16731ea5def858f26e882a29495a
Reviewed-on: https://review.gerrithub.io/421175
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
If the connection is already in exiting or exited
status, we should not set it again. And this
could prevent the coredump.
Change-Id: Ia506d13b12c3a6cb5619d65e4b3353b149a85947
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/421668
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Now that DPDK can reserve additional memory on the fly,
the initial requested memory only serves as preallocation
and it makes sense to request no preallocated memory.
Right now, passing `--mem-size 0` counterintuitively
tries to reserve all hugepages on the system.
Change-Id: I281900efc7bd0bbb7eed30161ba0b88e16195170
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/422241
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
DPDK 18.08 includes the fixes we proposed
and we're now ready to make the switch.
Change-Id: Ie5abf25480d2449cb2c8e23de10c8090fc7b14fb
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/422240
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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>
We used to support it by default in our DPDK forks,
but starting with DPDK 18.08, a new PCI driver flag
RTE_PCI_DRV_WC_ACTIVATE is required.
We enable now it for NVMe and Virtio, but not for I/OAT,
as our I/OAT driver currently assumes strong memory
ordering, which prefetchable resources do not provide.
Change-Id: I1a13356e28535981153b3d3e52bfe9d66b6172af
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/422239
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Starting with DPDK 18.08, rte_kvargs is a dependency
of rte_eal.
Change-Id: I0cde78f632fc313cec745d41ee519fb8b37de81b
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/422238
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: 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 enables easier testing of variable sector size
block lengths (512 + 8, 4096 + 8, etc.) - otherwise
you need an NVMe namespace specially formatted for
VSS.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ifafe57a2cf63edacb1e0f80a183841acf84fa1e4
Reviewed-on: https://review.gerrithub.io/422320
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reason: If the task is queued in scsi layer to handle, we
should free the task after calling the spdk_iscsi_task_cpl.
Otherwise, if only this task is executed (without subread tasks),
the task will be freed by the loop early, which will cause
the segment fault in spdk_iscsi_task_cpl function.
Change-Id: Ifc42399957b24d976af5fd12f6e33459a3ea86ba
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/421706
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
