Commit Graph

330 Commits

Author SHA1 Message Date
Seth Howell
bc83adc2db lib: move trace_rpc into lib/trace
Combinining these two libraries removes some directory complexity. It
also helps us to align on a common practice for including the rpc code
with a given module. This is how all of our bdev modules already do it.
The nbd, net, and scsi modules also follow this pattern.

Change-Id: I5e4c99d7f0facacc6dfe30b2274b60e0d151f8d8
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/465210
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-08-19 06:10:30 +00:00
Ben Walker
fb641c4b54 iscsi: Assign connections to poll groups instead of lcores.
This patch binds poll groups to SPDK thread through IO channel and
assigns connections to poll groups instead of cores.

iSCSI subsystem registers iSCSI global object as an IO device, and create
poll groups as context of IO channels of the IO device.

Each portal get and hold portal group on which the corresponding acceptor is
running.

When a connection is constructed, iSCSI subsystem assigns a poll group
to the connection by getting it from the corresponding portal.

When a connection enters full-feature phase, iSCSI subsystem schedules
the connection to a poll group by round-robin.

Then, each connection can know its running SPDK thread directly and can
use SPDK message passing infrastructure instead of SPDK event framework.

By this change, iSCSI connections are binded to SPDK thread, and not
binded to processor core anymore.

Some other changes in this patch are
- core ID is removed from the output of get_iscsi_connections RPC. The
  upcoming patches will change the RPC to use spdk_for_each_channel and can
  access SPDK thread safely, and add SPDK thread ID instead.
- utilize UT multithread framework added by the last patch to test
  iSCSI poll groups by UT.

Change-Id: Iec73c778aa413bcabdb63141cc41d4160911ea0e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/463359
Reviewed-by: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-08-12 20:49:36 +00:00
Anil Veerabhadrappa
ed56a3d482 NVMe-oF Target: Add FC transport.
- New files and updates to existing SPDK files to add the NVMf-FC transport.
  - Depends on an existing low level driver library. This driver is not part of SPDK repository.
  - Makefile updates to build FC transport (using CONFIG_FC)
  - Update configure script for FC build.
  - New FC unit test for FC-LS commands.
  - Update unittest.sh to run FC unit test (when built).

Signed-off-by: John Barnard <john.barnard@broadcom.com>
Signed-off-by: Anil Veerabhadrappa <anil.veerabhadrappa@broadcom.com>
Change-Id: If31d4d25feab76c2dbe90a7faf71d465c2c3a354
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/450077
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-07-26 22:17:17 +00:00
Darek Stojaczyk
27b920a2a4 configure: unset CONFIG_VHOST on BSD
Remove linux #ifdefs from the vhost code and just
implicitly disable CONFIG_VHOST for BSD systems.

This serves as cleanup.

Change-Id: I8b0e0e8f80478f50ca8586cc974f7afcee2566f0
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/460562
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-07-16 11:06:03 +00:00
Wojciech Malikowski
8ec04f81c9 app/spdk_lspci: Mark VMD and device behind VMD
Added information that enumerated device
is NVMe disk behind VMD or VMD itself.

Change-Id: I0265995ac11859f8f503bb6f6021667db4d0f2a8
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/459638
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>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-07-09 04:04:16 +00:00
Wojciech Malikowski
1ee1acfea1 app/iscsi_tgt: Optional VMD enumeration
VMD section with Enable flag set to true
need to be defined in config file to
enumerate devices behind VMD.

Change-Id: Ie1cb67f7e58d393783eeac6b0e437ee36bdeb0d9
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/459643
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: Paul Luse <paul.e.luse@intel.com>
2019-07-01 03:53:06 +00:00
Wojciech Malikowski
fd508c14ca app/vhost: Optional VMD enumeration
VMD section with Enable flag set to true
need to be defined in config file to
enumerate devices behind VMD.

Change-Id: Id22e0b940e789a32507e582efda564adae5beb28
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/459642
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-07-01 03:53:06 +00:00
Wojciech Malikowski
54020194c6 app/nvmf_tgt: Optional VMD enumeration
VMD section with Enable flag set to true
need to be defined in config file to
enumerate devices behind VMD.

Change-Id: Idbed36e57e3987486d8f0fb5a6c6336b4679344a
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/459641
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-07-01 03:53:06 +00:00
Wojciech Malikowski
c3abcfefda app/spdk_lspci: App for printing available SPDK PCI devices
This application enumerates all PCI devices
in system (including VMD) and prints their
address, vendor ID and device ID.

Change-Id: I2e73bc62c0a282b673486fa4527ff4b8319d47a2
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/450925
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-06-27 12:20:35 +00:00
Wojciech Malikowski
5ef24d5256 app/spdk_tgt: Optional VMD enumeration
VMD section with Enable flag set to true
need to be defined in config file to
enumerate devices behind VMD.

Change-Id: Ic725c8e750e6505e5cee8e8889f3cc8c0342eb51
Signed-off-by: Wojciech Malikowski <wojciech.malikowski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/458451
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>
2019-06-26 06:35:07 +00:00
Piotr Pelplinski
c4d9daeb7b Makefile: Add possibility to uninstall spdk.
Add uninstall target to makefiles to be able to perform
reverse of install target.
Fixes #464

This patch adds 'uninstall' target to makefile.
'make uninstall' will remove spdk_tgt app, headers, libraries
and shared libraries from system directories defined by $DESTDIR.
Additionaly, if there will be any empty directories left after
this operation, they will be removed as well.

Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: I7b07fb4b81081d3914ff09165991fbe3a26b9067
Signed-off-by: Pawel Kaminski <pawelx.kaminski@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/431471
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>
2019-05-16 20:56:18 +00:00
Jim Harris
c199f56d99 app/trace: suppress TSC output by default
Number of microseconds is sufficient in almost all cases,
so don't print out the TSC offset unless user specifies
-t option.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3bbf55204bf58d94f43161ff41c565ca92e218f7

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/452737
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-05-02 08:41:56 +00:00
Jim Harris
617184be3b trace: remove short_name
This was not used by any of the trace register descriptions.
Let's remove it rather keeping it around if we don't need it.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Idda809e2911db5be555ff6aa13695484a14bf665

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/452734
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
2019-05-02 08:41:56 +00:00
Jim Harris
dd1c38cc68 trace: allow arg1 to represent a string
This requires changing arg1_is_ptr to arg1_type.
We will use this to print the first 8 characters of
a blobfs filename when collecting event trace data.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I1b321d99145e82b42dcf6d901ce9d6298158edae

Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/452259
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-05-01 17:10:43 +00:00
Xiaodong Liu
1468f8dd3b app/trace: check num_entries before referring
For trace file which is generated by spdk trace_record,
if one lcore recorded no entries, there is no space for
it to have idle entries to referring.

Change-Id: I1cd89ec934407fc805bda66c8c87f76cb181679e
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/449840
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-04-04 01:43:44 +00:00
Xiaodong Liu
0c41a7ce95 app/trace: remove redundant memory/pointer
Previously, copying trace-file from disk or page-cache
to memory aims to accelerate disk-read in one large
sequential read.
But it shows no value, since trace-data are originally
retrieved sequentially, and page-cache can handle it
well.

Change-Id: I1c328286686ed05e83f3b7a1bef10dc353c5b653
Signed-off-by: Xiaodong Liu <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/449839
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-04-04 01:43:44 +00:00
Piotr Pelplinski
de5f53220c notify: add notification library
This patch adds library that will provide ability to notify about
events. On one side each spdk library can specify what kind of
notification it can produce and on the other side libraries can fetch
those notifications.

Example is bdev module, which lib notify about added or removed bdev.

Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Change-Id: Ia95e564a8a43400b2745d9de8217b9cc84cd1b16
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/431920
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2019-03-29 21:15:14 +00:00
Ben Walker
deb8ee5c33 event: spdk_app_start now takes spdk_msg_fn
This callback type takes 1 argument instead of 2.

Change-Id: Ic3710fafb2828f08fc064f7658849b3d20521092
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/446997
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>
2019-03-19 15:54:34 +00:00
Darek Stojaczyk
2b846acc49 configure: add option not to use the internal rte_vhost copy
It's disabled by default, so no functionality is changed yet.
The intention is to use the upstream rte_vhost from DPDK,
which - starting from DPDK 19.05 - is finally capable of
running with storage device backends.

SPDK still requires a lot of changes in order to support
that upstream version, but the most fundamental change is
dropping vhost-nvme support. It'll remain usable only with
the internal rte_vhost copy and with the upstream rte_vhost
it simply won't be compiled. This allows us at least to
compile with that upstream rte_vhost, where we can pursue
adding the full integration.

Change-Id: Ic8bc5497c4d77bfef77c57f3d5a1f8681ffb6d1f
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/446082
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-03-13 14:26:20 +00:00
Ben Walker
362879570a event: Remove arg2 from spdk_app_start()
We never used this anywhere, and I need to move to a model where
the start up action is a thread message instead

Change-Id: I6b21ba9afb93a3245aceca2fe24713ffd16d0933
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/446986
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-03-05 08:43:12 +00:00
Jim Harris
20fba4d886 test, nvmf: remove last usages of max_delay_us
nvmf was just setting it to 0 (which is the default).
The reactor test was setting it to 1000, which wasn't
actually used since there are always pollers registered.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I673d859584c404d9b2746fbc8cd4f00fa38df5e5

Reviewed-on: https://review.gerrithub.io/c/444307
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-02-13 16:43:18 +00:00
Liu Xiaodong
a206234d09 lib/trace: add trace_record tool
trace_record is used to poll the spdk trace shm file
and store new entries from it to another specified trace file.
This could help retain the trace_entires from the overlay of
trace circular buffer

Note:
* trace_record reads the input tracefile into a process-local
memory and writes trace entries to the output file only at shutdown.
* trace_record can be shut down on SIGINT or SIGTERM signal.

A usage sample is:
./spdk_trace_record -s bdev_svc -p <spdk app pid> -f trace.tmp -q

Change-Id: If073a05022ec9c1b45923c38ba407a873be8741b
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/433385
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>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
2019-01-30 06:36:25 +00:00
Jim Harris
b4c8e45632 app/trace: print error messages when input file is corrupted
Found while trying to process a file that wasn't completely
written yet - it would SIGBUS instead of printing an error
message.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I56d60519f0608900697102c06310ebb92ec4d39c

Reviewed-on: https://review.gerrithub.io/c/441988
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2019-01-28 02:33:50 +00:00
Jim Harris
20476c6521 app/trace: simplify error messages when printing file name
Just have file_name point to shm_name when we're processing
shared memory.  This will be more helpful with the next
commit that adds more error messages.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I7c50ea0c16c6ced0a1aed59b225cb29d0f7b7345

Reviewed-on: https://review.gerrithub.io/c/441987
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
2019-01-28 02:33:50 +00:00
Arshad Hussain
8f0d7e57fa app: Fix spelling error in iscsi_tgt.c
This patch fixes spelling error while printing
error message using SPDK_ERRLOG under file
app/iscsi_tgt/iscsi_tgt.c

Change-Id: I7e052822358dc72a65c8ca8a8ac908a7f74c8f4b
Signed-off-by: Arshad Hussain <arshad.super@gmail.com>
Reviewed-on: https://review.gerrithub.io/c/441246
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2019-01-24 10:52:31 +00:00
Chunyang Hui
26979c50c7 app/parse: add default return value for app parse function
Change-Id: Icae46bb96502c7edf3a453e0fff1d8560241c5ab
Signed-off-by: Chunyang Hui <chunyang.hui@intel.com>
Reviewed-on: https://review.gerrithub.io/c/441277
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2019-01-22 01:03:39 +00:00
Chunyang Hui
01e5610da8 parse: add return value to app_parse callback
Need to check user input and return status of parsing
to prevent app or target from crashing.
Input checking function will be added in the future.

Change-Id: I8167ac13306ae4f81e2cacb80edd9dcf9382c374
Signed-off-by: Chunyang Hui <chunyang.hui@intel.com>
Reviewed-on: https://review.gerrithub.io/c/439479
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
2019-01-18 10:08:09 +00:00
Liu Xiaodong
ae0aae1532 lib/trace: add 3 RPC method for tpoint_group_mask
Now trace facility can be enabled/disabled at runtime,
besides app starting parameters

Change-Id: I086564665ba4375b946a5339d5921d6584e9bfda
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/435470
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-12-14 21:29:12 +00:00
Jim Harris
5b16c18d17 build: move app rules to spdk.app.mk
This removes a ton of duplicate from a lot of
Makefiles.  Note that spdk.unittest.mk still keeps
its own app rules, and now just includes
spdk.app_vars.mk instead.  Fixing Makefiles that
include this .mk file requires more work that is
deferred for now.  We do need to add the
"install: all" rule explicitly to this .mk files.

Note that Makefiles need to include spdk.app.mk
now at the end of the file, so that things like
APP and SPDK_LIB_LIST are defined before including
it.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If6823cda135180c37690ffc96c91660b7f0718e3

Reviewed-on: https://review.gerrithub.io/434283
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: Ben Walker <benjamin.walker@intel.com>
2018-12-12 22:35:21 +00:00
Jim Harris
20c69bbc30 env: remove default pre-reserved memory size
Much of this was originally applied by Darek in
6f5840d22 but then had to be reverted due to NVMe-oF
MR issues in a817ccf5.  Now that the NVMe-oF MR issues
have been resolved, we can reapply these changes.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Change-Id: I58d603523901c78dbc81279078a09b2a24c091e6

Reviewed-on: https://review.gerrithub.io/436547
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>
2018-12-12 11:00:57 +00:00
Jim Harris
cceaeedb1f build: add mk/spdk.app_cxx.mk
This is just a copy of spdk.app.mk for now.  But later
patches will modify spdk.app.mk in a "C-specific" way,
so add this cxx variant to use now in the C++ apps.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0d4d28c6119364d4c3cc7fa5274598b69f4ee6b1

Reviewed-on: https://review.gerrithub.io/434279
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: Ben Walker <benjamin.walker@intel.com>
2018-12-11 18:07:53 +00:00
Ziye Yang
2893aa6117 nvmf tgt: update the makefile to add net lib
Change-Id: Iac3d1a4be406b8e4981946de51ff93f18ead3679
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/435593
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: Changpeng Liu <changpeng.liu@intel.com>
2018-12-04 01:55:57 +00:00
Tomasz Zawadzki
6b8ca50e84 app/trace: don't use shm_name when using file_name during remaping trace
Change-Id: I9449d40ff31039f3f59bc60c956592e0e565437a
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/435114
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Xiaodong Liu <xiaodong.liu@intel.com>
2018-12-03 20:24:29 +00:00
Jim Harris
5704be8987 build: remove xx_MODULES_LINKER_ARGS and xx_MODULES_FILES
Application Makefiles can now just add xx_MODULES_LIB_LIST
to SPDK_LIB_LIST.  This is possible now since all
SPDK libraries are linked with --whole-archive, so there
is no need to differentiate between "modules" libraries
and other SPDK libraries.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Iabf81a388b72d3b2a2f48287a8491ddc977722ac

Reviewed-on: https://review.gerrithub.io/434277
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>
2018-11-28 16:43:01 +00:00
Jim Harris
0cc8f76225 build: remove unused LINKER_MODULES
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I944a2cd3fb646ce96c58412c72fdd299651ca148

Reviewed-on: https://review.gerrithub.io/434276
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-11-28 16:43:01 +00:00
Jim Harris
e7d0bd8c17 build: remove unused SPDK_WHOLE_LIBS
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I11f6391b14de0c0be09d387f9d41768b429f855f

Reviewed-on: https://review.gerrithub.io/434275
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>
2018-11-28 16:43:01 +00:00
Jim Harris
42b6771f40 build: add ALL_MODULES_* variables to spdk.modules.mk
This allows a lot of simplification to SPDK application
makefiles.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I5fa463f6369834b84a8d92e79fa7768082209d7a

Reviewed-on: https://review.gerrithub.io/434274
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-11-28 10:21:45 +00:00
Jim Harris
3da5847023 build: add ENV_LINKER_ARGS to LINK_C and LINK_CXX
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>
2018-11-28 10:21:45 +00:00
Jim Harris
0b5fa082db build: remove sock from SOCK_MODULES_LIST
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>
2018-11-28 08:13:46 +00:00
Liu Xiaodong
9d2f39ab0b lib/trace: records num-trace-entries for lcores
With the trace_size records for each lcore, spdk_trace
can read trace_file in which each lcore has different
number of trace entries.
Offset of each trace_history from the beginning of this
data structure.

Change-Id: I06afaba129812fe40ed000265fc66b02c5d9e3d9
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/433503
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-11-26 18:34:18 +00:00
Liu Xiaodong
fd4c75721e lib/trace: set num-trace-entries by app param
Number of trace entries in circular buffer per lcore can be
assigned by the boot parameter of SPDK app with
"--num-trace-entries <NUM>"

Change-Id: I855ce6b4f14a716dcdd9078913da5ea8e577af3a
Signed-off-by: Liu Xiaodong <xiaodong.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/433099
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-11-26 18:34:18 +00:00
Tomasz Zawadzki
a7e049e722 app/trace: do not print shm_name when using file_name
Change-Id: Ibaeac2d85f3e29df00f64d3e5c1d17d663e32278
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/434068
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>
2018-11-20 22:40:39 +00:00
Tomasz Zawadzki
c3c6922941 app/trace: fix app_name dereference on init
In case when only file_name was provided,
app_name was still dereferenced.

Change-Id: Ica948e072ef02a8daadf303b3e2a004640d19000
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/433609
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
2018-11-19 15:34:26 +00:00
Jim Harris
e0b669312a spdk_tgt, vhost: move vhost libraries earlier in SPDK_LIB_LIST
Upcoming changes will use --whole-archive on all libraries.  This
introduces some issues with the shared library builds, although
it's not entirely clear why these issues don't occur today.

vhost depends on both scsi and bdev - but for the spdk_tgt and
vhost application Makefiles, vhost is listed later than scsi
and bdev.  It's best practice to list the libraries in order
of dependence - at least until we add dependencies to the shared
libraries themselves.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ibacfb7781dad60cf3ce6906deaaa2e5c0f5e9571

Reviewed-on: https://review.gerrithub.io/433356
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@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>
2018-11-16 08:57:16 +00:00
Ben Walker
4e3cc6e1de nvmf: Change max_delay_us to 0
This sets the maximum tolerated additional latency
on an I/O, which potentially allows a thread to
sleep and save power. However, today if a poller
is registered on the thread it will never sleep,
which is always true for the nvmf_tgt.

Further, if this were fixed such that the thread could
sleep as long as the pollers all reported idle, this
max latency ends up breaking the QoS test because the
service provided is more than 10% away from the target
number.

Change-Id: Ic9decbd92ad4e0c19eb81eea18a59207d6028367
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/432497
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: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Piotr Pelpliński <piotr.pelplinski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-11-13 14:26:37 +00:00
Darek Stojaczyk
a817ccf571 Revert "env: remove default pre-reserved memory size"
Apparently this patch revealed an issue somewhere in SGL
handling in NVMf initiator which is causing our CI to fail.
Let's revert it while we work on a proper fix.

This reverts commit 6f5840d225.
This is not a full revert, some app.c changes to the usage text
were kept unchanged.

Change-Id: Iddea5c2b9df50bd12ef8f6226165883f6622ab33
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/432576
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: Ben Walker <benjamin.walker@intel.com>
2018-11-09 18:31:45 +00:00
Darek Stojaczyk
6f5840d225 env: remove default pre-reserved memory size
Now that we utilize DPDK dynamic memory allocation, we
no longer need to set the default pre-reserved memory size.
It'll now be 0, which implies all memory will be allocated
at runtime.

The option to pre-reserve all hugepages on the system is
now only available on BSD, so also clarify that bit in the
app usage text.

Change-Id: I5a8a1d9bf14ad6d938532d7e6254a45e4a81bb92
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/432204
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Seth Howell <seth.howell5141@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2018-11-08 23:07:32 +00:00
Pawel Wodkowski
c4fee1e970 mk: don't use '-include spdk/config.h'
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>
2018-10-02 23:13:32 +00:00
Jim Harris
b6ee6f9241 iscsi, nvmf, vhost: remove default config file pathnames
Config file pathnames have been possible on the command line
for a very long time, and all of the SPDK testing infrastructure
uses it.  The default pathnames were added a very long time ago,
and now can cause weird issues like conflicting with --wait-for-rpc
when an INI config file is not passed on the command line but is
pulled in by the default pathname instead.

So remove the default pathnames.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia914ff1c9321235f79f8cc9a054ed7f099340151

Reviewed-on: https://review.gerrithub.io/424606
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>
2018-09-05 19:46:17 +00:00
Jim Harris
3f229f0e46 app/trace: calculate per-core overlaps correctly
Remove the calculation of g_last_tsc.  This was erroneously
cutting out valid data - for example, when reactor X has
an event that happened very early on, then reactor Y fills
its buffer before reactor X has another event.  All of reactor
Y's data would be ignored until reactor X processed another
event.

While here, remove some code that should have been removed
as part of the fudge_factor removal.  Also use SPDK_COUNTOF
to simplify calculation of the number of entries per
reactor buffer.

Reported-by: John Kariuki
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I16c7f80cbfc8b4bb5f90a195d30df69e07b4ff85

Reviewed-on: https://review.gerrithub.io/424615
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: Ben Walker <benjamin.walker@intel.com>
2018-09-05 18:03:43 +00:00