This is a step towards simplifying the Makefiles for
SPDK applications. After this change, all of the
xxx_MODULES_LINKER_ARGS variables are just a list
of SPDK libraries wrapped in --whole-archive.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0cf069c2f74e8998d42dea9598db55f333c861b7
Reviewed-on: https://review.gerrithub.io/434146
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This allows us to remove $(ENV_LINKER_ARGS) from all of the
application Makefiles.
Modify spdk.unittest.mk to explicitly set ENV_LINKER_ARGS
to nothing, to avoid DPDK libraries from getting linked
into unit tests. This seems better than creating separate
LINK_C rules for unit tests. Provide a UNIT_TEST_LINK_ENV
flag to allow select unit tests to override this behavior -
it's needed for some of the env unit tests.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I647f2c848cf434023cf5a1f3a45869bbc4fc8102
Reviewed-on: https://review.gerrithub.io/434145
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
SPDK uses constructor functions for registration functionality
in several different areas - RPCs, bdev modules, event tracing,
etc. Since these functions are not explicitly referenced, the
linker will remove these constructor functions from the linked
application unless their library is included using the
--whole-archive linker option.
SPDK has tried to minimize the number of libraries linked with
--whole-archive to only those that are necessary. But there
are still a lot of exceptions that we need to work around. The
benefit to trying to restrict it is minimal - for example,
an nvmf_tgt binary built before this patch is 4746320 bytes
on my system, but linking all of the libraries with
--whole-archive only increases it to 4784080 bytes - a difference
of less than 1%.
Note that DPDK also links all libraries with --whole-archive by
default.
This will also simplify some upcoming changes around shared
library dependencies - we will be able to blindly use
--whole-archive when linking the shared library as well.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Iab0dbbace06a3d15c835491e55683ab4cf9e53f0
Reviewed-on: https://review.gerrithub.io/434107
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
The various *_MODULES_LIST are intended for the modules
that implement the specific interface - not for the
associated core library itself.
For example, BLOCKDEV_MODULES_LIST doesn't contain bdev,
it contains all of the bdev modules. Similar for
COPY_MODULES_LIST and copy.
So make SOCK_MODULES_LIST consistent. This is a real
pain because we have to add sock explicitly to a bunch
of individual Makefiles, but I have plans to clean that
up in some future patches.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0d293372cfb38bfa46c8478a8d9ca9ce23924747
Reviewed-on: https://review.gerrithub.io/434125
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
There was an implicit merge conflict between this patch
and e956be96e which added NVMe/TCP support including adding
"sock" as an nvme.libtest.mk depdendency.
This reverts commit 7962ce312d.
Change-Id: Ie7e58e89128f28080a986e12cd014de7df4752e5
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Reviewed-on: https://review.gerrithub.io/434103
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This matches the lib/ut_mock directory name, and also
avoids the final library being named spdk_spdk_mock.a.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I80ec09568d0455d1f131e45694b6e2d536e6ca91
Reviewed-on: https://review.gerrithub.io/433909
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
SPDK uses constructor functions for registration functionality
in several different areas - RPCs, bdev modules, event tracing,
etc. Since these functions are not explicitly referenced, the
linker will remove these constructor functions from the linked
application unless their library is included using the
--whole-archive linker option.
SPDK has tried to minimize the number of libraries linked with
--whole-archive to only those that are necessary. But there
are still a lot of exceptions that we need to work around. The
benefit to trying to restrict it is minimal - for example,
an nvmf_tgt binary built before this patch is 4746320 bytes
on my system, but linking all of the libraries with
--whole-archive only increases it to 4784080 bytes - a difference
of less than 1%.
Note that DPDK also links all libraries with --whole-archive by
default.
This will also simplify some upcoming changes around shared
library dependencies - we will be able to blindly use
--whole-archive when linking the shared library as well.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Iab0dbbace06a3d15c835491e55683ab4cf9e53fe
Reviewed-on: https://review.gerrithub.io/432917
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
It is the first patch to follow the NVMe over fabrics
spec and implmenent the NVMe/TCP transport. It can be
divided into work in the host and target sides:
Host side: Add the TCP/IP transport in nvme lib (lib/nvme).
Target side: Add the TCP/IP transport in nvmf lib (lib/nvmf).
Change-Id: Idc4f93750df676354f6c2ea8ecdb234e3638fd44
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/425191
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>
blob and blob_bdev are already included via the lvol bdev module
dependencies in mk/spdk.modules.mk. This patch adds blobfs there
too.
These changes are needed for some upcoming patches which will
apply --whole-archive to all libraries linked into the application.
That patch will require that we never include the same library
twice.
Shared library builds *require* that blobfs be listed before
blob on the link line, hence adding blobfs here even though it's
not technically associated with a bdev module.
We can also simplify spdk.modules.mk while we're here - just
filter out bdev_lvol to disable lvol for applications like
blob_cli that need to disable it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I01e5e6b91d947463749f942f54c12f48229bbe5a
Reviewed-on: https://review.gerrithub.io/432916
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: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Let's just make them consistent and all with a
bdev_* format.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia3dec7f4be2ecc765b4aa095e6b675f13d326063
Reviewed-on: https://review.gerrithub.io/429289
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@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>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Currently only the iscsi => scsi dependency is defined to
demonstrate how the framework behaves.
Eventually, lib/Makefile will be populated with more lines
of the form:
DEPDIRS-iscsi : scsi log conf event
This line will ensure that the iscsi subdirectory is built
after the specified dependency directories.
Then the lib/iscsi/Makefile also defines:
DEP_LIBNAMES = scsi log conf event
This line adds dependencies from the iscsi.so to the specified
libraries.
Ideally we could avoid this duplication by passing the DEPDIRS
value to the lower level Makefile, but that breaks down when we
start considering event/subsystems/* libraries. So for now we'll
need to duplicate it. Incidentally, DPDK also duplicates the
dependency information like this.
While here, remove use of $(MAKESUBDIRFLAGS) in spdk.subdirs.mk -
it was no longer used anywhere in the tree.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If63cbc10ffa18ac6361e4210bd6196ffe88bec43
Reviewed-on: https://review.gerrithub.io/429287
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: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
We have hairy configuration method. First configure script uses CONFIG,
environment and arguments to produce CONFIG.local and config.h then
makefile uses CONFIG, environment and CONFIG.local to produce config.h
again. First time config.h can't be generated reliably as config.h
target in Makefile is passing MAKEFLAGS and they are unknown at
configuration phase.
Lets remove this mess by producing final mk/config.mk file and get rid
of CONFIG.local. This will also generate full build configuration in one
file instead scattering them in CONFIG and CONFIG.local.
In next patch scripts/genconfig.py can be removed.
Change-Id: I5696813312edff7e493207f7aa8569234a365ec0
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/426364
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>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
While here, change spdk_lib_list_to_files to
spdk_lib_list_to_static_libs to differentiate it from
the new spdk_lib_list_to_shared_libs.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I6e5913addfbdd556fae2451d4e2b2c43feaf33ab
Reviewed-on: https://review.gerrithub.io/429286
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Lance Hartmann <lance.hartmann@oracle.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>
This prevents the SPDK build from requirng unnecessary dependencies when
not compiling certain features. Also, fixes github issue #434
Change-Id: I7d0520474f3656ae32670313f2290e6b741c5ca8
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/426131
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This will allow us to specify more granular system dependencies for each
.so file.
Change-Id: I759876e09ffb36380a503962b4be8fc9d7c71b63
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/426129
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Also fix options that doesn't start with CONFIG_ prefix.
Change-Id: I32a24373328bab8f513d56c0c09b5a86589f690c
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/427767
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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Each file that need to check SPDK_CONFIG_* options need to include
spdk/config.h explicitly.
Change-Id: If9f2a91ac4c2b1a300dcf88ec3e2a12714ad344a
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/427221
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>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
We need to include it in devel package.
Change-Id: I823200632e8bcb9fdb86c8cb5fbf3a651a710b78
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/426839
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>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
As a side effect, if SPDK_LOG_DISABLED is used in spdk_log() as log
level then no message will be printed.
Change-Id: I2d57b60a5a310a9ef2a1187a81088d0acf828742
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/425105
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
We kept this disabled by default originally because it lacked
multiple iov support needed for vhost and nvmf in-capsule data.
Now that support has been added, there is no reason to not
enable it automatically - it has no extra package dependencies,
so requiring users to specify --with-raid to configure is just
an extra unneeded step.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia9796370e12a6f93993e70083de36114a37851bd
Reviewed-on: https://review.gerrithub.io/426003
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add dependency $(ENV_LIBS) for test/env/*
as part of new variable $(ADDITIONAL_LIBS) which is direct dependency
of $(APP) in mk/spdk.unittest.mk
This solves the problem of
make not triggering LINKing for test/env/*
after changes made to lib/env_dpdk/*.c
Issue was that Makefiles in test/env/*/ classified tests as unittests
by including mk/spdk.unittest.mk
But unittests are not using env_dpdk which means that
unittests don't relink on env_dpdk update
Change-Id: I91b6690dc92d68bc69bbf11ac5336d3a7f56e4e4
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
Reviewed-on: https://review.gerrithub.io/425665
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Initial support for softare AESNI_MB DPDK driver only.
Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices
with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from
1 to 512 in powers of 2 for 30 seconds each run.
QAT can be included in the code (but not makefile) and marked as experimental
until we are ready to test in CI. It works well on 2 systems but is a big PITA to get
the hardware setup and configured for use with DPDK (IMHO).
Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc
Signed-off-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/403107
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>
This linker flag is really more important when dealing with external SO
files such as libuuid. It shoudl follow those linked options around
instead of all of our internal dependencies.
This series is aimed at fixing github issue 434
Change-Id: Iee65470632b5ab0adeb361005edc894e05c43ea4
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/425441
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>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Instead of adding the linker flags just before application compilation
on linux, move that linking to earlier in the process. This will also be
useful for the shared library build.
This series is aimed at fixing github issue 434
Change-Id: I84bcc6bd922ddb6e48adb2dfe0b154700773cf27
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/425435
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: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This allows us to better distinguish between external and internal libs.
This series is aimed at fixing github issue 434
Change-Id: I2ed141f909e7c4a800df02061007b0d23da25380
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/425434
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: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Unit tests define mock functions in the unit test code
itself. This links fine with static libraries, but
doesn't work if trying to link with shared libraries.
So change the unit tests to always explicitly link
with the static libraries.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3a8a60162fcab6c73a216c4bb364175767fdc62c
Reviewed-on: https://review.gerrithub.io/422441
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>
When we build the SPDK_FILTER_LIB_LIST, we should
maintain the dependencies between libraries. Primary
culprit is vhost depending on scsi. So add vhost
to SPDK_FILTER_LIB_LIST before scsi, and then stop
sorting the list unnecessarily before using it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib22a5016aa492c30a389a498ae0360fbcec7c12b
Reviewed-on: https://review.gerrithub.io/422439
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>
Updates build and install of shared lib to conform with
standard naming conventions for shared libraries, including
the setting of the lib's soname.
Change-Id: Ib46d298ab5f77fd8c32b7c04022b546446189428
Signed-off-by: Lance Hartmann <lance.hartmann@oracle.com>
Reviewed-on: https://review.gerrithub.io/421936
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>
Raid module:
============
- SPDK raid bdev module is a new bdev module which is
responsible for striping various NVMe devices and expose the raid bdev
to bdev layer which would enhance the performance and capacity.
- It can support theoretically 256 base devices (currently it is being
tested max upto 8 base devices)
- Multiple strip sizes like 32KB, 64KB, 128KB, 256KB, 512KB etc is
supported. Most of the current testing is focused on 64KB strip size.
- New RPC commands like "create raid bdev", "destroy raid bdev" and "get raid bdevs"
are introduced to configure raid bdev dynamically in a running
SPDK system.
- Currently raid bdev configuration parameters are persisted in the
current SPDK configuration file for across reboot support. DDF will be
introduced later.
High level testing done:
=======================
- Raid bdev is created with 8 base NVMe devices via configuration
file and is exposed to initiator via existing methods. Initiator is
able to see a single NVMe namespace with capacity equal to sum of the
minimum capacities of 8 devices. Initiator was able to run raw
read/write workload, file system workload etc (tested with XFS file
system workload).
- Multiple raid bdevs are also created and exposed to initiator and
tested with file system and other workloads for read/write IO.
- LVS / LVOL are created over raid bdev and exposed to initiator.
Testing was done for raw read/write workloads and XFS file system
workloads.
- RPC testing is done where on the running SPDK system raid bdevs
are created out of NVMe base devices. These raid bdevs (and LVOLs
over raid bdevs) are then exposed to initiator and IO workload was
tested for raw read/write and XFS file system workload.
- RPC testing is done for delete raid bdevs where all raid bdevs
are deleted in running SPDK system.
- RPC testing is done for get raid bdevs where existing list of
raid bdev names is printed (it can be all raid bdevs or only
online or only configuring or only offline).
- RPC testing is done where raid bdevs and underlying NVMe devices
relationship was returned in JSON RPC commands
Change-Id: I10ae1266f8f2cca3c106e4df8c1c0993ddf435d8
Signed-off-by: Kunal Sablok <kunal.sablok@intel.com>
Reviewed-on: https://review.gerrithub.io/410484
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>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
This will be necessary when the NVMe/TCP transport is added, which will
use the sock abstraction library.
Change-Id: I6561ccdad1ff575db2afa418eab2f2cd24806cb9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/416477
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Also remove libspdk_net from the list and add it to each app that
originally linked it via NET_MODULES.
This will allow the use of SOCK_MODULES in application Makefiles without
pulling in the RPC libraries required by spdk_net.
Change-Id: Iad68ddd25bea24bba53a681c78d41e6e703f866d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/416476
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This separates the network interface management code (interface.c and
net_rpc.c) from the socket abstraction layer, which allows users that
only want the socket abstraction to avoid pulling in the JSON RPC
libraries.
Change-Id: I9b00285a70bac0c74c73353cfa900d4f3b2e465f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/416475
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This is already required to build the shared library and iSCSI
components; simplify these Makefiles by adding -lcrypto to the central
SYS_LIBS in mk/spdk.common.mk.
Change-Id: I8fdaffbccc1294e24e32559387b0ed99d1deb0ce
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/415351
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 makes more sense as a first class library.
Change-Id: Ibd5c578f8708bd8c7d83fe1629e97c9a3316b56b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/414698
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>
Reviewed-by: Madhu Pai <mpai@netapp.com>
Change-Id: I7d929e7520007f29511299442f791038f7a4ce1d
Signed-off-by: John Meneghini <johnm@netapp.com>
Reviewed-on: https://review.gerrithub.io/409539
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 patch integrates VPP userspace TCP/IP stack,
as optional network framework to be used in SPDK.
Support is done via VPP Communications Library.
Change-Id: I4c2945c76878dbf24ff2f8612e0a21615f7d87e6
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/389566
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
At this time blobcli makes use of all the bdev modules
during compilation. Yet vbdev_lvol does examine,
that already creates bs_dev on a bdev and loads
blobstore for all devices where lvol store exists.
Since bs_dev cannot be created on device twice,
same as blobstore cannot be loaded twice.
It prevented blobcli from displaying information
for blobstore underlying lvol store.
This patch remove vbdev_lvol from list of bdev modules
for blobcli.
Change-Id: I144ba22ec28f7f315e768745f5dd0cbbe3bbad86
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/404832
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>
NVML (Non Volatile Memory Library) changed its name to
PMDK (Persistent Memory Development Kit), so make the
necessary changes to the SPDK repository.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id07a87eafb4e9a3099603030c58c0e927bafa608
Reviewed-on: https://review.gerrithub.io/406256
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This uses libiscsi to implement an iSCSI initiator bdev
module for SPDK. Still a lots of work to do on this - posting
it in case anyone is interested in working on this further.
A number of todo items are listed in a README in the lib/bdev/iscsi
directory.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Change-Id: I060e33de0cd6796246789bf0e1bb4f2df59d8f71
Reviewed-on: https://review.gerrithub.io/390313
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Example showing the basics of a vbdev via a module that implements
the main functions, and a few extras, needed for a vbdev. The tests
also demonstrate how to configure an application to use a vbdev and
an underlying bdev
Change-Id: I67bf542143eb6532465cdb6af63c0f2ab7084c0e
Signed-off-by: paul luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/399768
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>
Enable dynamic compilation of SPDK with VPP.
Change-Id: I9f5f7223625cbebec0d6209b5a2ce6bdcc1a04bf
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/401825
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id370631770cb2e090c3efbe95cd86db4f6168c5e
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/403702
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
The unit test mock library contains wrappers that aren't directly
referenced but still need to be kept to allow the final link to work
when they are used via --wrap. Add the spdk_mock library to the
whole-archive list to prevent it from getting removed.
Fixes build with LTO enabled.
Change-Id: I2c70be7cb517a373178d34026f9980396727f8df
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/401074
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>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
in gmake, -e gives precedence to variables defined in the environment.
this argument disables appending to CFLAGS or CXXFLAGS when they are set
in the environment and breaks our make configuration.
Change-Id: Id395e7edc4b243bd3805d7f5fdb3456d94893952
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/400863
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>
It is not ideal to have to add this in a bunch of
different Makefiles, but further consolidation of
Makefiles is going to be a more arduous process.
With this change, rebuilding SPDK after no changes
will result in no output - all of the "Nothing to
be done for 'all'" messages have been removed. Note
that DPDK build output still remains - this can be
suppressed by either using an out-of-tree DPDK, or
using SKIP_DPDK_BUILD=1 when using the in-tree DPDK
submodule.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib1d45432f2baaa1d9c1b9509906ee1e77966d3ca
Reviewed-on: https://review.gerrithub.io/399918
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This suppresses:
make[3]: Nothing to be done for 'all'.
when nothing needs to be rebuilt.
Still more work will be needed, specifically in
examples and apps directories. But this will cover
most of lib/ and anything in unit/test that is using
the new spdk.unittest.mk.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1c0c2b88e70ec37629740aa47f612c9d46fc88fb
Reviewed-on: https://review.gerrithub.io/399725
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This includes removing the nvmf.unittest.mk that these
unit tests were using previously.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3b53db0a565aee910587826b37e8b43d6a324e75
Reviewed-on: https://review.gerrithub.io/399724
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This includes removing the json.unittest.mk that these
unit tests were using previously.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Icfa2e5a5554b686002407d1ef5328a0aecb442f1
Reviewed-on: https://review.gerrithub.io/399723
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I53a04187638a8ea4569b18bb4959b0bcf7f3d860
Reviewed-on: https://review.gerrithub.io/399717
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This is the first of a long line of patches which will
reduce Makefile rule duplication in all of our SPDK
unit tests.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1cd6e9e1e9cabab0a15dbcb901e4db782d73b1a4
Reviewed-on: https://review.gerrithub.io/399716
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
These aren't needed - the common %.o rule is sufficient.
As part of this patch, remove OTHER_FILES references from
nvme and nvmf unit tests, and just include these .c files
directly.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I76ab20da2f560c7997e978fcbfe28c2a4907f759
Reviewed-on: https://review.gerrithub.io/399715
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This can make profiling more challenging, but should not
affect debugging since we still have unwind tables.
Keep the C_OPT in COMMON_CFLAGS to allow a way to
still add compile flags, even though C_OPT is now
not set by default.
This results in a 11% performance improvement on
a bdevperf microbenchmark using null bdevs.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If16fef6e94a73dfca6c10d3b7f158a12dfc70b53
Reviewed-on: https://review.gerrithub.io/393823
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>
Only Makefiles for libraries that directly depend on DPDK (rather than
the SPDK env abstraction) should add $(ENV_CFLAGS).
Change-Id: Ifdf44d3ef8c42bbf7f20edd524b330d00658235b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/392818
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>
Exported lib/bdev/rte_virtio as a separate
library not dependent on bdev.
Virtio is now accessible via spdk_internal/virtio.h.
The header is marked `internal`, as it's
not meant to be used by end users. It's
not handy to handle all backend-specific
(e.g. Virtio-SCSI) logic in a user code.
For now the Virtio interface is publicly
exposed only via bdev_virtio module. We
might want to consider adding a separate,
public Virtio-SCSI library in the future.
Note: this patch doesn't do any changes
to the virtio code. Everything is
moved 1:1.
Change-Id: I805e5d12d265d82b0bc5784c89fbadb81abdb278
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/388166
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>
For now, this only installs libraries and headers; we will need to
consider which binaries should be installed and what they should be
named before we add them to the install rule.
Change-Id: I78dc8631f793d0df88cd884b0ac66406df9e4427
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/387637
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>
Running the 'touch' command unconditionally causes the compile step to
return 0 even if the compilation failed. Instead, only run it if the
compiler succeeded.
Fixes: 4f16344356 ("make: Update build artifact timestamps for clang
support")
Change-Id: I46907851fde8bb9e3f73549276e489b6c30ed1fd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/388324
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: <ed.rodriguez@netapp.com>
The tab inside the define ends up in the output when called, which leads
to problems if the output is quoted (the tab will be treated as part of
the filename).
Change-Id: I1038e1650a74861660fa4164dbde065217bc7b99
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/387631
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>
Change-Id: Ibc165386b074d82969610467f4da73c73d8f4ea0
Signed-off-by: Ed rodriguez <edwinr@netapp.com>
Reviewed-on: https://review.gerrithub.io/386164
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>
Check the target machine type explicitly before turning on -march=native
rather than assuming any non-PowerPC machine is x86.
Change-Id: I651b1852dc5ac6a3d42c60d639ec48f399f6b3bd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/386348
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Similar to commit 98f84e6255 ("build: detect OS via cc -dumpmachine"),
we can use the C compiler's -dumpmachine output to detect the target
CPU architecture rather than relying on the build machine's uname
output. This helps enable correct CPU detection when $(CC) is a cross
compiler.
Change-Id: I72c34294a1ff7dd9df3aa45dfc319e5b81d51f85
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/385709
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Rather than running 'uname', which returns information about the build
system, use the C compiler's -dumpmachine option, which outputs the
target triplet, which looks something like "x86_64-unknown-linux-gnu".
This enables us to detect the correct target OS when
the configured $(CC) is a cross compiler.
Change-Id: I5659dd5b091094cc8e408a6c6f4e7f64a65e9070
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/385166
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>
ppc64le does not support compiler option march=native
but instead uses mcpu
Change-Id: I2dd14b1acf003e8229e9c9392c4c0606e2c15af8
Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>
Reviewed-on: https://review.gerrithub.io/383725
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use SPDK bdev subsystem to allow basic IO to pmem pool
using pmemblk_* calls.
New pmem bdev can be constructed using RPC call
construct_pmem_bdev PMEM_FILE
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: If5ca94161fe53644b8fccd3b77de5479da1b2e55
Reviewed-on: https://review.gerrithub.io/376973
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 be used to simulate multi-threading for
bdev unit tests.
While here, alphabetize the existing calls - calloc
had been inserted out of order in a couple of places.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I559cd1fd79e78f03ebdac313e0bbedbcdde4a8c1
Reviewed-on: https://review.gerrithub.io/377968
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>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
This should be the default already on x86-64 compilers.
Note that the code may still not necessarily compile or work properly on
other architectures, but this is at least a start toward portability.
Change-Id: I9c842aa329dd761277bf90669f2c905e436e0af4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/378365
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>
Change-Id: Ia96ae78ff9530d953181ac5f7255a38f3c8ec430
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Reviewed-on: https://review.gerrithub.io/375392
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>
Most of the work here revolves around having to split
an I/O that spans a cluster boundary. In this case
we need to allocate a separate iov array, and then
issue each sub-I/O serially, copying the relevant
subset of the original iov array.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0d46b3f832245900d109ee6c78cc6d49cf96428b
Reviewed-on: https://review.gerrithub.io/374880
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>
Supports both PCI mode (for usage in guest VMs) and
vhost-user mode (for usage in host processes). The rte_virtio
subdirectory contains a lot of code lifted from the DPDK
virtio-net driver. Most of the PCI and vhost-user code is
reused almost exactly as-is, but the virtio code is drastically
rewritten as the DPDK code was very network specific.
Has been lightly tested with both the bdevio and bdevperf
applications in both PCI and vhost-user modes.
Still quite a bit of work needed - a list of todo
items is included in a README in the module's directory.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I85989d3de9ea89a87b719ececdb6d2ac16b77f53
Reviewed-on: https://review.gerrithub.io/374519
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>
Change-Id: I6a138e1c1d775e8c8bfeede9600f8b0f799ecdad
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/362445
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>
Much like bdev modules inside the bdev directory,
add a subsystems directory inside of event. The subsystem
specific code for the bdev library is placed in to
a separate library in that directory, breaking the
strict dependency of the bdev library on the event subsystem
code.
Change-Id: I255941b823a9ec3e2d62f22a586414949d8ff5ad
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/365055
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This patch shows how to use the spdk_mock library which now only
consists of --wrap capability for a few functions.
Next will be a patch that provides a generic mock capability to
use either the --wrap feature or a combination of globals and
stubs to make adding more UT easier wrt mocking. After that is
in place nvme_ut.c will be updated to use the new library for
all mocking.
Change-Id: I1a6ffb722da043bb70bd4cfe1afa7c5e39a0fccb
Signed-off-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/365074
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>
This patch doesn't use it yet but sets up the ability to so that
upcoming unit test pathces can. The next patch will include use
of wrap in one nvme UT and then following that the wrap and
today's use of globally controlled stubs will be morphed into a
simple mock type capability for spdk unit tests.
Change-Id: I252160266f0b5c76e50fd213cb3e1686d48b9d0e
Signed-off-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-on: https://review.gerrithub.io/363441
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>
Change-Id: Ibeab0bc70bb59da4ebb8bd175642f5f549f1d273
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362425
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Also change the discovery/nvmf.sh test to use it.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I56bce9a84bd46f13b6d4f34da81abf23413f2598
Change-Id: I9546d715b8d2ca3ebf46183bdbaa58e8aa921d95
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Add a configure script in the root of the repository
that looks and feels like one that would be generated
by autotools. This script simply generates a CONFIG
file for you, which used to be done by hand.
Now to build SPDK you can do the following:
./configure --with-dpdk=path
make
Change-Id: I44ebb91f0cb1468b86da4c0033ac1406595d4967
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
No functional change, just rearranging code.
Change-Id: I28328dfefd7de269d326834c484f2c2fca4e6c1f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This patch adds a library, application and test scripts for extending
SPDK to present virtio-scsi controllers to QEMU-based VMs and
process I/O submitted to devices attached to those controllers.
This functionality is dependent on QEMU patches to enable
vhost-scsi in userspace - those patches are currently working their
way through the QEMU mailing list, but temporary patches to enable
this functionality in QEMU will be made available shortly through the
SPDK github repository.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
Signed-off-by: Michal Kosciowski <michal.kosciowski@intel.com>
Signed-off-by: Karol Latecki <karolx.latecki@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Signed-off-by: Krzysztof Jakimiak <krzysztof.jakimiak@intel.com>
Change-Id: I138e4021f0ac4b1cd9a6e4041783cdf06e6f0efb
This flag is on by default for compilers that support it, and it
prevents compilation on older compilers that do not recognize it.
Change-Id: I862cff6297ecd4fc40f2f8761bf398d524771921
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This virtual block device takes an underlying block device and splits it
into several smaller equal-sized block devices.
Change-Id: I6f6e686c1177b2e4885f7e88809ad329caae55bd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This flag prevents the compiler from merging multiple tenative global
variable declarations (e.g. 'int i;' without an assignment) into a
single copy later. In the default (-fcommon) mode, this allows multiple
definitions of an uninitialized global variable to be linked into a
single program without any warnings.
Change-Id: I0339012a7164dccd311ab1f62d11111d24529fbd
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The unit test Makefiles should already declare exactly which files to
include.
Change-Id: Icae8a2c862db164d72d8e9ed4064d777de43742b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is shorter and matches the more commonly-used name.
Change-Id: I6af862de90e5cf8a7c27161f1ee9bb5a1164bec6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
If a library rule is being built, one of the files it contains was
rebuilt, so we need to regenerate the whole library anyway.
Remove the previous copy of the library to make sure nothing is
left from the previous build.
Change-Id: I835f721d5bf7cc908c343f47de78bd1f44be58b8
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
-Wformat-nonliteral needs to be disabled since clang triggers it on the
call to vsnprintf() now that it is nested two calls deep.
Change-Id: I228b9d099cfc2b65181941cbb4798b7f8eae3baa
Signed-off-by: Daniel Verkamp <daniel.verkamp@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>
This requires a couple of related changes:
- I/O queue IDs are now allocated by using a bit array of free queue IDs
instead of keeping an array of pre-initialized qpair structures.
- The "create I/O qpair" function has been split into two: one to create
the queue pair at startup, and one to reinitialize an existing qpair
structure after a reset.
Change-Id: I4ff3bf79b40130044428516f233b07c839d1b548
Signed-off-by: Daniel Verkamp <daniel.verkamp@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>
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>
This patch will add a new bdev module, rbd.
It can make ceph rbd as the backend of iSCSI
target.
Change-Id: Id5eb3b159ee607052e3c33a2e59d721739fd9977
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Fix the existing cases (all missing void in parameter lists) and enable
the warning to prevent new ones from being introduced.
Change-Id: Ieaf00b3dfd5daf1e21fcbefb124514882e8996c9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The V= syntax is more familiar from other build systems, so allow both.
Change-Id: I024cf4a338c2fe10bef482311a1703bf5fb2a38e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Drop the special-case preprocessor definition for PCI access library now
that config.h is available with an equivalent SPDK_CONFIG_PCIACCESS
define.
Change-Id: I4891d0f2fd7d3eea51b767df9e594555b36265ea
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The upcoming DPDK version adds objects that have no externally
referenced symbols but that must be linked in (specifically, the
default mempool handlers are now assigned by a constructor function).
Link all of the DPDK libraries with -Wl,--whole-archive to fix linking
against a static library build of DPDK.
Change-Id: Ia54b100d81aa7be30a8f4bcad65cc182e91dcc65
Reported-by: Jay Sternberg <jay.e.sternberg@intel.com>
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The NVMf target is being refactored to split the RDMA transport-specific
code into its own file. Once this is complete, we should be able to
plug in other transports and build the NVMf target without any RDMA
dependency if desired.
To enable this, change the CONFIG option to RDMA; it still controls
whether the whole NVMf target is built for now, but once the RDMA
dependency is actually made optional, we will be able to build the
generic NVMf target code without libibverbs installed.
Change-Id: I8cd90a9aaa85dcefcc9b0f8f2e7b6af21958b2a8
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Linking this library into an application makes the log
configurable via RPC. This only works if the rpc subsystem
is also loaded.
Change-Id: If1340cf2a845ef159290232c26f341150c98fb9a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
On some platforms, librt may be a static library, so linker command line
order matters. Create a new Makefile variable, SYS_LIBS, which is added
to the end of all link commands and move -lrt there.
Change-Id: Ie28ebd91a255097bf33cbcf9f0f74f7177a0b142
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is the standard way to add multithreading support and handles
platform-specific requirements for pthread library name, #defines, etc.
Also include -pthread in CFLAGS, since it affects preprocessor
definitions.
Change-Id: I94b436c7ef482ae3320b7937f4f64b51e5d57032
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>
All source files will now depend on config.h, which is generated based
on the defaults in CONFIG and the command line arguments passed to make.
This ensures that any configuration changes, including on the command
line, cause a full rebuild.
Change-Id: I6b6fa3290941200dbcf32297c66df8dc5ee18e94
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Resolve relative paths before using them to clean up command lines.
This should also help shorten the overall command line length that gets
embedded in the binary and used when locating the executable from a
coredump.
Change-Id: Ibff9849ede198bb04313496c8b7131485ffaf14f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
1) -Wstrict-prototypes applies to C code only
2) add CXXFLAGS definition
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I440702d989c4e89bd297956a58725f62d2d72831
Allow end users to build the code even if, for example, warnings are
added with new compilers versions.
-Werror is still enforced for automated builds.
Change-Id: I567009d55cf5b941c452bd41d09f75d03b037209
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Since the individual build steps include the path now, the subdirectory
printouts are less useful (especially with parallel make, where output
from multiple subdirs gets mixed together).
Change-Id: Ied71d39b360b847ca32ccd3547def751ddd9434c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
For example, if the source code is located in a directory called "spdk",
don't print the "spdk/" prefix on all of the directories.
Change-Id: I9adc650b13e013773450c721c8434c6e5afeecc2
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This makes it somewhat clearer what is happening during parallel
builds (make -j).
Change-Id: Ibbe8862da7600e784cba64cd333086f1473455ac
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
With CONFIG_PCIACCESS=y in CONFIG file, we can
use libpciaccess library; With CONFIG_PCIACCESS=n
in CONFIG file, we use pciaccess functions provided
in DPDK.
Change-Id: I786c5589b8e7909ba2e59d222938dd5ba45bf92d
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
Also add a space between Copyright and (c).
The copyright year can be determined using git metadata.
Also remove the duplicated "All rights reserved." - every instance of
this line already has a corresponding "All rights reserved" immediately
below it, except for examples/ioat/kperf/kmod/dma_perf.c, where I have
added it manually.
Performed using this command:
git ls-files | xargs sed -i -e 's/Copyright(c) \(.*\) Intel Corporation. All rights reserved./Copyright (c) Intel Corporation./'
Change-Id: I3779f404966800709024eb1eb66a50068af2716c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The normal CU_ASSERT_FATAL macro calls a function that is not marked as
noreturn, so static analyzers (e.g. scan-buid) can't figure out that
fatal asserts are really fatal.
Add a wrapper macro that calls abort(), which the analyzer can determine
does not return.
Change-Id: I0c087bf9c8d3c272bf88120caa70e87dab6d9546
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Some compilers automatically define _FORTIFY_SOURCE to a different
value, so undefine it before setting it on the compiler command line.
Change-Id: Id662953bd4466b99d0fe9bd5bb96f9fc335ff9a8
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
The ioat driver supports DMA engine copy offload hardware available on
Intel Xeon platforms.
Change-Id: Ida0b17b25816576948ddb1b0443587e0f09574d4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Use $(CLEAN_C) throughout the Makefiles to clean up a consistent set of
generated files.
This also adds coverage files to the list of cleaned files.
Change-Id: Iceb922935a45c9eecbf2f3443bd0ee4f5c966825
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This is very convenient. It also works from subdirectories, as
long as you adjust the relative path you pass appropriately.
Change-Id: I8cfeac380ccb5fad4389ee1feb838f986a837f11
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
librte_malloc has been deprecated by DPDK, but to allow for backwards
compability, check for its existence and link with it if it exists.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3e05a3295b6768bd6804a8ec86c301e22040df37
Replace $(UT_APP) in nvme.unittest.mk with $(APP); there is no actual
difference in the way they are used.
Change-Id: I039c2488ee998ee8b1a67af129045b66a97bc29c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Add common $(COMPILE_C), $(LINK_C), and $(LIB_C) variables that contain
the commands to build a .o from a .c, an app from objects and libraries,
and a library from objects, respectively.
Change-Id: Ie2eaa13156b8bf3db7a4ffa66161382d829aef07
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Modern GCC and Clang versions can generate .d files without needing the
complex sequence of sed scripts that we currently use.
Also adjust the .o rule so the .d file is regenerated if it gets
deleted.
Also make the .o depend on all of the Makefile fragments in
$(MAKEFILE_LIST) so that it is rebuilt if the Makefile is changed.
Change-Id: I44e186c0f34dabfa0cf35db1ffa66b84d9dae87d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
We don't use any C11-specific features, and switching to C99 will allow
us to support older versions of GCC.
In particular, the version of GCC used by Travis CI is too old to
support -std=gnu11.
Change-Id: Id5bd8351fdd83e7f4b6703048a921c78c3f06ec6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This partially reverts commit fd93c1b1ef.
Older versions of GCC don't consider the empty initializer to be
acceptable when this warning is turned on.
Change-Id: I5c1c22ca988f464990c68eb56e4dd4d51cb99410
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This catches functions with empty parameter lists, which are not the
same as void parameter lists in C, as well as other (less-likely)
K&R-style parameter lists.
Change-Id: I18b09e01a60b3669d1cc5d8d96eec20cd872c497
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This helps weed out functions that should be static, functions that are
not declared in public header files, and .c files that don't include
their .h interface headers.
Change-Id: Ie39f83ad4b320847e4a938bd1d4d0b4fa21c2ffa
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
-Wold-style-declaration, -Wsign-compare, and
-Wmissing-field-initializers are included in -Wextra.
-Wunused-parameter is not very useful, so disable that explicitly, but
enable the rest of -Wextra.
Change-Id: I034b368d16306544e0eadafab9ec1ccf27f21645
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Replace {0} initializers with {}, since it is equivalent and simpler
(and avoids the warning).
Change-Id: Id2014a352e2c627598cd3b4f1ca04dba35f2dc3a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Collect coverage information for use with gcov.
Coverage is disabled on FreeBSD because the current version of clang
provided by FreeBSD can't successfully link with -ftest-coverage enabled
(the compiler-rt support libs are too old).
Change-Id: Icc444936caa852bfb9a02b37223209319a27a770
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This allows comparing the Linux kernel driver's performance to the SPDK
user-mode NVMe driver.
Change-Id: I71c70163a4133c2f237c8c57b3c698ec261455f5
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This allows the same objects to be linked into static and shared
libraries and allows the creation of position-independent executables
with the static libraries.
Change-Id: I119949c3644c02a83e414227615dcc2d8f896286
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Some of these are covered under -Wall, but add them all just to be clear.
Change-Id: I63104be4bf58becd19a4c30a4d275d2c24c6761d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Modern toolchains already enable noexecstack by default, but set it
explicitly just to be sure.
Change-Id: I2866220db5a5ed0bb6e78a4e83763839ea3b93b0
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This forces the ELF loader to resolve all relocations immediately at
load time rather than allowing lazy binding.
Therefore the loader can make the GOT (global offset table) read-only
once the relocations are complete during the loading process.
Change-Id: I89a878d1060ce79d58c5be7700d8b122a33a8ceb
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Move dependency includes into a new spdk.deps.mk file,
then include it at the end of Makefiles that build
source files.
Also add a test to autobuild.sh to confirm that
binaries are regenerated if we make after touching a
header file.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If6a1905706a840f92cbdf3ace7fbdb27fe2de213
Set -DNDEBUG to make assert() a no-op. Since nvme_assert() is just an
alias for assert(), this controls compilation of nvme_assert as well.
Change-Id: Ie9d0c7fea007eff030179c1b54865adc3dcaf6b6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
This will add different flags, src files, etc. for FreeBSD.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I4f2d664cf42d01745318f241ce9558dccf148b7d
Inherit the CFLAGS and other common setup from spdk.common.mk instead of
redefining it in nvme.unittest.mk.
In particular, this enables warnings for the unit test code that were
not enabled before.
Change-Id: Id0fbf0ee0b0d63f9de0c362d1d23e2bd036a7145
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
