Add some options to reduce the number of calls to
get time.
Change-Id: Iceea12b1ad82cba96d1a4bd687af3e9743d5f58a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/393686
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
The test already writes to the entire device twice, so
the trim step does not actually accomplish anything.
Change-Id: Ie761389511196d659358f18f557feb80087628cc
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/393832
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
If it cannot be allocated, we should return error.
Change-Id: I48aa50a8842c35ee112fe7185128b1bc1930176e
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/386369
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Describe the phenoemon for peformance difference
between fio + spdk plugin and spdk nvme perf
on multiple NVMe SSDs while both using single
cpu core.
Change-Id: I80c2cfd51cd1d8ec28a11a1a8653469cfc3335eb
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/379573
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
The reason is that we create duplicated qpair on same subsystem many times,
thus the performance is influenced. And this patch can fix this issue.
Change-Id: I075b028b15e0d34857a89510710642f2d9520235
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-on: https://review.gerrithub.io/375782
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>
With this change, the polling qpairs can be in round-robin manner.
Change-Id: I1926468dc596de2a43f42451525650356f44fbbd
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/375707
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>
This will cause the NVMe driver to use the default qpair size
as opposed to setting it up for the first job in what may end
being a series of jobs in the same fio config file. If we
don't do this and subsequent jobs are specifying a higher queue
depth than earlier jobs, the NVMe qpairs will be too small and
performance results can be way off.
Future fix logged on Trello to correctly size q pairs even when
multiple jobs are specified in a single file but this is a small
optimization.
Change-Id: I73a7b83b10418dd22b663a50ea8e33a6c3d7ecf7
Signed-off-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/371240
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add a new struct spdk_nvme_io_qpair_opts to allow the user to override
controller options on a per-I/O qpair basis.
Existing callers with qprio == 0 can be updated to:
... = spdk_nvme_ctrlr_alloc_io_qpair(ctrlr, NULL, 0);
Callers that need to specify a non-default qprio should be updated to:
struct spdk_nvme_io_qpair_opts opts;
spdk_nvme_ctrlr_get_default_io_qpair_opts(ctrlr, &opts, sizeof(opts));
opts.qprio = SPDK_NVME_QPRIO_...;
... = spdk_nvme_ctrlr_alloc_io_qpair(ctrlr, &opts, sizeof(opts));
Change-Id: I8ac3ea369535cfde759abbe75e1d974b6450a800
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/369676
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
FIO backends are supposed to return a negative value on error.
Add special handling for -ENOMEM (out of requests) case to return
FIO_Q_BUSY, indicating that FIO should resubmit the request later.
This is part of the fix for issue #169, which is related to high queue
depths with large I/Os causing the NVMe library to run out of request
objects.
Change-Id: I4fa4001b078b07c42fcd4d1357434575b2c84023
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/369664
Tested-by: 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>
Documentation update - the plugin just works with 2.21 with no
changes.
Change-Id: If1fb2367a00065bfb5627d0d2af9683e5c909dc1
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/367734
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Checked the code and found that our nvme perf example
added 1 io_queue_depth. Since if you set queue_depth=n,
the real entry is n - 1.
Change-Id: Ie204fc061b98e853e9b810eff21f79565c2ad8d0
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/367931
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Any transport that is not PCIe is a NVMe over Fabrics transport and
requires a subsystem NQN.
Change-Id: I68bfe49086f7f633fea8589ac15069327c4a4aa9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/367290
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>
While here, change the recently added mem_size_mb to
also use the automatic option parsing for integer values
provided by fio.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I889073b3b3b01c7003b21d8ace61e71b672ff2df
Reviewed-on: https://review.gerrithub.io/367284
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
We do not need quotes around the filename.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I31685409eb5d956cdfb2fc30896e9e43312b61a3
Reviewed-on: https://review.gerrithub.io/367280
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
While doing performance testing for the SPDK NVMe driver using fio with
our fio_plugin, I saw the error (transport->ctrlr_create_io_qpair() failed)
when running 18 jobs on my system. The error was happening when trying
to allocate memory for the trackers at line 890 in the lib/nvme/nvme_pcie.c.
Root cause was the fio_plugin.c initializes the environment with only
512 MB of hugepage RAM. I changed opts.mem_size to 1024 and rebuild
the plugin the issue was resolved.
This patch enables setting the mem_size in the fio config file using
parameter named mem_size_mb. E.g. mem_size_mb=1024
Change-Id: I3541b2029a6b36c26f814101313f49c2dd98c9bc
Signed-off-by: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-on: https://review.gerrithub.io/365735
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>
This can be used by threads within SPDK to undo the
affinitization done either by DPDK (for DPDK lcores)
or by inheriting the parent's thread when using
pthread_create().
This will be used by the stub app to unaffinitize
the reactor core to allow the scheduler to flexibly
move it to an idle core.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7e550953cd9dcd7fd9d98bfbb70660994f2eefcd
Reviewed-on: https://review.gerrithub.io/366680
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Remove dpdk_ prefix in spdk_app_opts and spdk_env_opts
Change-Id: I6f231f67072b808e84945d41b1fe31a180beb350
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/365787
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>
fio has a race between reap_threads() and free_ioengine(). free_ioengine()
will call the ioengine's cleanup routine and then dlclose it if it
is dynamically linked (like the spdk fio plugin). free_ioengine() does
not set td->io_ops = NULL though until after dlclose() is complete. If
reap_threads() tries to dereference td->io_ops after our plugin has been
closed but before io_ops was set to NULL, it will segfault.
Solution (until an upstream fio fix is available) is to use LD_PRELOAD
instead.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ide4eb3cb92a636513289107fc211fdf1f98b616f
Reviewed-on: https://review.gerrithub.io/365272
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
- rename spdk_malloc_socket to spdk_dma_malloc_socket
- rename spdk_malloc to spdk_dma_malloc
- rename spdk_zmalloc to spdk_dma_zmalloc
- rename spdk_realloc to spdk_dma_realloc
- rename spdk_free to spdk_dma_free
Change-Id: I52a11b7a4243281f9c56f503e826fd7c4a1fd883
Signed-off-by: John Meneghini <johnm@netapp.com>
Reviewed-on: https://review.gerrithub.io/362604
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Also update the example fio file
Change-Id: I2a9074d524072e95340cc0e87041e87e3bb7d4f8
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362069
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: I9ee1c2ab49691f7e31bbec83a057faa718177c90
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362260
Tested-by: 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: Jim Harris <james.r.harris@intel.com>
These variables are not needed outside of this file.
Change-Id: I577db5140fb4156100e72e6219c8121304a9b137
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch is causing intermittent errors in the automated tests, so
reverting for now until this can be resolved.
This reverts commit 1d9891989c.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I237e31b1b8a1a1e28262851ae0a21cd7345f0f1b
The reaon is that rte_eal_init called in spdk_env_init binds
the master thread on master core, we need to restore the
affinity.
Change-Id: I4eb2d36b74404c1af6d242c4fd5c403bd3d5036c
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
This patch will be prepare to support multi threads.
Change-Id: I2234ae1d564c9cf17b92c85fd0a414adaa639bf2
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Also adds the automation test case for using remote
NVMe devices exported by NVMe-oF target.
Change-Id: I2b839a4eeec33d5b0c30d654e6013ad8c7949e23
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
The SPDK fio plugin only supports jobs as threads, not separate
processes. So log an error and bail if --thread=1 is not
specified.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I318f924e0b7fbb76f11427a447dd7c5fc0ebc221
The previous implementation tried to maintain a ring buffer of completed
I/O retrieved from ->getevents() that had not yet been retrieved by
->event(), but this is not necessary. None of the other FIO engines
implement this sort of ring buffer, and the only code that calls
->getevents() will always retrieve the full number of events via
->event() afterwards.
Instead, just count how many I/Os have been completed within each
->getevents() call in iocq_count, and reset iocq_count on each new
->getevents().
Change-Id: I2776c3a5c89a0d2b24478328bc138ff7a9fb5bdd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
They were very close to the same already, so finish the job.
Change-Id: Ifba9e3b2d11a3e70cbfbe46f57a67552db2757ed
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The probe_info was reduced to just containing a
transport_id, so remove probe_info entirely.
Change-Id: Ica9a22d126cd14e282decd3eea1a0afe0460f099
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Make it easier to use SPDK libraries by putting them all in a single
directory that can be added with -L rather than scattered around the
source tree.
Change-Id: I5c0f5dd6e7058b5f92fa9bc41548190ffc064761
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
spdk_nvme_probe() will now provide a struct spdk_nvme_probe_info to the
probe and attach callbacks in place of the PCI device pointer.
This struct contains the useful information that could be retrieved from
the PCI device during probe.
The goal of this change is to allow expansion of the probe information
in the future when other transports (specifically, NVMe over Fabrics)
are added that do not necessarily use PCI addressing or device IDs.
Change-Id: I59a2a9e874e248ce5fa1d7f4b57c8056962ff3cd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This converts some, but not all, usage of rte_mempool
to spdk_mempool. The remaining rte_mempools use features
we elected not to expose through spdk_mempool such as
constructors, so that will need to be revisited.
Change-Id: I6528809a864ab466b8d19431789bf0f976b648b6
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Use the env library to perform all memory allocations
that previously called DPDK directly.
Change-Id: I6d33e85bde99796e0c85277d6d4880521c34f10d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This allows users to swap their PCI library from
libpciaccess/dpdk to another mechanism using the standard
method for swapping out the env library.
Change-Id: Ib2248f8b43754a540de2ec01897e571f0302b667
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This allows users to swap out SPDK's third party
libraries for an implementation based on their own
framework.
Change-Id: Ia0b7384ce5e31acba5ad0d7002dec9e95b759c52
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
The new env library will wrap all third-party library
calls and be easily swappable with alternate implementations
at build time. For now, it's just the memory library
renamed.
Change-Id: I26a70933289f8137107208ba75f7520fd7a33da0
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
This patch also drops support for automatically unbinding
devices from the kernel - run scripts/setup.sh first.
Our generic pci interface is now hidden behind include/spdk/pci.h
and implemented in lib/util/pci.c. We no longer wrap the calls
in nvme_impl.h or ioat_impl.h. The implementation now only uses
DPDK and the libpciaccess dependency has been removed. If using
a version of DPDK earlier than 16.07, enumerating devices
by class code isn't available and only Intel SSDs will be
discovered. DPDK 16.07 adds enumeration by class code and all
NVMe devices will be correctly discovered.
Change-Id: I0e8bac36b5ca57df604a2b310c47342c67dc9f3c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
While here, also mention that the plugin only supports
a single thread currently.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I265e59a89e77cd8ed26419e811aa2f36afcd37b3
This will allow removal notifications to be propagated to the library
user (e.g. for hotplug).
The callback is currently unused, but this at least prepares the API for
the future hotplug support.
Based on a patch by Dave Jiang <dave.jiang@intel.com>
Change-Id: I20b1c2dbf5e084e0b45a7e51205aba4514ee9a95
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use the knowledge that both the source and destination of
nvme_copy_command() are aligned to emit the aligned variants of the
SSE2/AVX mov instructions.
Change-Id: I0a7e32a3bb10b9a1920cd85691b79fa7172eecb3
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Simplify the build rules so that common libraries are always linked.
Also fix up a couple of -lpciaccess instances that should have been
changed to $(PCIACCESS_LIB).
Change-Id: I4c50fa3aa59cae013d3385e38fbb830794299f6e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
