When PRACT is set, if metadata size is 8 bytes, PI is stripped
(read) or inserted (write). Hence block size must not include
metadata size for extended LBA payload. This patch fixes the issue
by reducing metadata size from block size for this case.
On the other hand, When PRACT is set, if metadata size is larger
than 8 bytes, PI is passed (read) or replaced (write). So block
size is not necessary to change for this case.
Signed-off-by: James Bergsten <jamesx.bergsten@intel.com>
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I930c8a07519a4742c44240801b068fac2c4802a7
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/465708
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Add SPDK dependency installation instructions for ArchLinux OS.
When possible dependencies are installed directly via pacman
package manager. If not available - use AUR and makepkg.
One exception is pmem dependency which is installed using
Github sources, as it seems that AUR repo is not frequently
update.
Additionally for pmem installation ldconfig needs to be
updated manually, otherwise it will break fio installation
(if done with vm_setup.sh) which detects pmdk headers, but
cannot find libraries on runtime.
Change-Id: Ib9045ba55b10c2b671e01b48dda8ec439899bc06
Signed-off-by: Kay Bouché <kay-b@live.de>
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/433913
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add Arch Linux option to our Vagrant scripts.
We will probably need this for nightly tests against
latest versions of software we depend on.
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Change-Id: I2dd1d69cbb286625343fd5e70ace0d8da1218825
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466832
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Seth Howell <seth.howell@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
We want to use Vagrant scripts in our CI.
It will allow us to remove statically configured
virtual machines and spawn new instances dynamically.
This patch allows us to use custom version of emulator
instead of preinstalled on host.
Signed-off-by: Maciej Wawryk <maciejx.wawryk@intel.com>
Change-Id: Iff79cd28d4d5b21fcf327db455ef060182f0f4ef
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/465925
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Test scripts are no longer failing and can be
re-enabled in nightly testing.
Change-Id: I07e682466f98a63c65b1d17cca47a4136809e8c9
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468203
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Removing "shift" command because it was not needed and
caused vhost run command to be not properly constructed.
Needed to fix if contition for --no-pci param while at it.
There was no $no_pci variable defined after optargs were
removed.
Change-Id: I72c615fdb2a5fd507efedae100b9b79c5cfd8935
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468202
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This line called vhost_kill with 2 arguments:
$1 - string "0"
$2 - string "vhost_name"
Problem was that vhost_kill only accepts 1 argument, which
is "vhost_name". This was not noticed because in seems in
most of our tests we just use the name "0".
This would fail to kill a vhost instance in case it would
be named anything else than "0"
Change-Id: I635c5b92673abb7350d8bd6c636f6b794651f7ac
Signed-off-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468201
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
If PRACT is enabled, DIF context was not initialized. However it was
expected that PRACT is passed through DIF flags of the DIF context.
Hence PRACT was not set in NVMe command even if user set PRACT.
This patch fixes the issue by passing fio_qpair->io_flags instead
of dif_ctx->dif_flags.
Signed-off-by: James Bergsten <jamesx.bergsten@intel.com>
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ibcb74fc8f74f863d8b53d53484fdea66f4b5db8e
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468016
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: James Bergsten <jrb@thebergstens.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
DIF context has to be initialized both for read and write I/O. However,
it had been initialized only for write I/O unintentionally after
refining error processing.
This patch fixes the issue.
Signed-off-by: James Bergsten <jamesx.bergsten@intel.com>
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I363da40ddba186e52fd0dfce37cfb0dea325040d
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/468015
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: James Bergsten <jrb@thebergstens.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
IB Architecture Specification vol.1 rel.13. in ch.10.3.1 "QUEUE PAIR
AND EE CONTEXT STATES" suggests the following destroy procedure for
QPs associated with SRQ:
- Put the QP in the Error State;
- wait for the Affiliated Asynchronous Last WQE Reached Event;
- either:
* drain the CQ by invoking the Poll CQ verb and either wait for CQ
to be empty or the number of Poll CQ operations has exceeded CQ
capacity size; or
* post another WR that completes on the same CQ and wait for this WR
to return as a WC;
- and then invoke a Destroy QP or Reset QP.
Without the drain step it is possible that LAST_WQE_REACHED event is
received and QP is destroyed before the last receive WR completion is
polled from the CQ.
In SPDK there is no risk of resource leakage in this case. So, instead
of draining we can destroy QP and then just ignore receive completions
without QP and post receive WRs back to SRQ.
Fixes#903
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: Ice6d3d5afc205c489f768e3b51c6cda8809bee9a
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/465747
Reviewed-by: Seth Howell <seth.howell@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
In prep for replacing some of the internal r/w calls with function
pointers based on RAID level, just call spdk_bdev_io_get_buf()
directly in the submit path for reads. This:
* will reduce the number of places where unique calls to the
upcoming function pointer will be
* bring it in line with how the majority of other bdev modules look
* actually increase UT coverage by about 10% as we're now calling
spdk_bdev_io_get_buf() and it's callback.
Change-Id: I7e6da0dab80687988ba52f57b0d9e2dbf20676dc
Signed-off-by: paul luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467538
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Function multiversioning conflicts with LTO when applied
to a function defined in a header file included from
multiple compilation units.
Change-Id: I65bed3903a717b7e982ab185c314d2118ae0e795
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/465972
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
The unit test mock macro function doesn't mesh well with the scan-build
static analysis tool. So we need to reset cc to the user provided value
or default before running the unit tests.
Change-Id: Ib636f6e396b09f9bf1cb8f930e6befa56ce99848
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466816
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
A sub-team over a year ago agreed that there was not enough value
in checking test plans in to the repo as they require maintenance
and their primary purpose is not to document how the module is tested
but to facilitate discussion during test development. It was agreed
that we would use the review system to iterate on test plans but once
the actual tests were developed that the plan would not get checked in.
This patch just removes those that were likely in the repo before that
discussion.
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I75dcdd8b4754b7ecb4a21079b251c707557a3280
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467394
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This is a preparation to use zcopy APIs for write I/O.
When we use zcopy APIs for write I/O, submit_task() will be called
in the callback to spdk_bdev_zcopy_start().
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I15bfce5b39ddd65459274fad7628f2c8fbb22327
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467899
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This is a preparation to use zcopy APIs for write I/O.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I685d0e1e5860c7ec72d4680c48ee7c0b5e9071bd
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467898
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This is a preparation to use zcopy APIs for write I/O.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I7c544fc08db029e1b5ff32458e53dcbb014a5722
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467897
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Two return code checks are added to bdevperf_complete().
Checking if return code is not false will be enough.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: Ic097e37c2d74aa3860bb06931f320c9f18c45640
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467896
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Support for these options was not introduced until DPDK commit
7f0bb634a1406b132ff15c9cd56a0a9f33e5f11d
Signed-off-by: Benjamin Saunders <bsaunders@google.com>
Change-Id: Id6db73dd48ac01aa1b05eca4c920c5753e8cc6f0
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467703
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>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The default behavior is to set it to 2MB, so this isn't
required anymore.
Change-Id: I62d7605cd4d5bc41347128f32f9a1aa373a15744
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466993
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
The socket now automatically sets the recvbuf size
to 2MB, so this isn't necessary.
Change-Id: Id2196f4038f6835118047233f18c0395fa3f2670
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466992
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Everywhere we use this, we increase the size to 2MB.
Just make that the default behavior.
Change-Id: Ie174419a09df1792a0c7311eddd0c2dcefa267b5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466991
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Invalid flags are not handled. Add a *) case.
Change-Id: Ib3e0d978ed9a131a5ab23a2bc5013bbe0698d1be
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467268
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
The dependencies between vhost and rte_vhost were not added during
earlier changes. This change moves the rte_vhost directory up to the
level of the other libraries and adds the proper dependencies for when
it is linked.
Change-Id: I089de1cd945062b64975a0011887700c0e38bb0f
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467700
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
There is at least one example of an SPDK shared library dependency that
is only linked against on certain conditions, so add a framework for
dealing with those conditions now.
Change-Id: I63ad767994c5f56f2908f70016e700f5bb74a5f4
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467699
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Removing queuing tests as they are really only testing mocked functions,
minimal coverage is gained here and the queueing logic from a UT
perspective is part of bdev, not a bdev module.
This is a prelude to adding RAID 1E which will start with abstracting the
RAID r/w functions and it may not look like it now, but getting rid of
this UT code will make the move to function pointers for r/w based on
RAID level a bit smoother.
Change-Id: I01eef2aa99b2bdd4308ddef9bad9dd1121ef7470
Signed-off-by: paul luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467537
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
If there are enough subsystems, the discovery log page spills
over to a second page. Add a test to confirm that is working.
Change-Id: I1df57b0b9a543736fe4c48f2cfe5628cb963a631
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466820
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Karol Latecki <karol.latecki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
This reverts commit 6129e78d26.
When the initiator sends the discovery log page, if the log page
exceeds the size of its data buffer, it will break it up into
multiple log page commands with appropriate offsets. However,
supporting offsets in log pages is an optional feature in NVMe
and reported by the EDLP bit in the identify data.
This commit changed the discovery process to no longer send an
identify command prior to doing the discovery log page command,
so the values in the identify data are always 0. If the discovery
log page exceeds the size of the data buffer (4k), it will then
fail to send the second log page with an offset because it
believes the controller does not support the feature.
Revert this change to fix it. An identify should always be sent
as part of the discovery process. A test case is included in a
follow up patch the demonstrates the bug.
Reported-by: Zahra Khatami <zahra.k.khatami@oracle.com>
Reported-by: Akshay Shah <akshay.shah@oracle.com>
Change-Id: Iefd512a7521e0fea90541b3eb547671cfa816ea6
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466819
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Most modern shells know how to interpret the string version of commands
when parsing them from a variable, but some shells like the one centos 7
uses misinterpret the command causing the dpdk version check in
configure to fail erroneously. This can be observed in the CentOS logs
of recent vs dpdk master jobs on the CI.
This method looks to be the more conservative way of doing the same
thing and fixes the issue on my dev machine.
Change-Id: Ib51c537ec88c781eb62519e08e4252ae05e554ef
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/467714
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>