Commit Graph

1630 Commits

Author SHA1 Message Date
Roman Sudarikov
5712088a55 Add support for VTune
Change-Id: Ica7f76a5354b27e6e5fd7114de1fdba46434b073
Signed-off-by: Roman Sudarikov <roman.sudarikov@intel.com>
Reviewed-on: https://review.gerrithub.io/362994
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>
2017-06-06 16:24:14 -04:00
Daniel Verkamp
887683858b nvmf: add NVMe I/O passthru for virtual mode
Change-Id: I338f47000a1c898309a7421fe52f148bae9f2e6b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/364116
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-06 13:07:32 -04:00
Daniel Verkamp
eba5c996c8 bdev: add NVMe I/O passthru command
Change-Id: I4d281be99553629563026cc4f9ab890d0a97986c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/364115
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-06 13:07:32 -04:00
Daniel Verkamp
bac4c02bc4 bdev: change name and product_name to char *
Dynamically allocate bdev names to remove the arbitrary 16-character
name length limit.

All of the existing product_names are constant strings, so those can
just use string literals instead of a copy per bdev.

Change-Id: I3280da67a4fcf2e4ec8ee8193362ca1b96a9c0cb
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363601
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-06 12:37:39 -04:00
Ziye Yang
7bc1a7d1a6 blockdev_nvme: free name in nvme_ctrlr
Need to free this variable.

Change-Id: I78a7d5b312db6487ed65b9d314590a28408da761
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/363477
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-06 12:37:26 -04:00
Maciej Szwed
61119d7862 json: Added support for 64 bit unsigned value converter in json
This patch prepares support for 64 bit values that will be used in lvol bdev.

Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Ib5c8a438135e246c69887c4775c9ea6f0fa6eb1b
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/363344
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Piotr Pelpliński <piotr.pelplinski@intel.com>
2017-06-06 12:00:25 -04:00
Daniel Verkamp
7c1d827f78 json_util: remove spdk_json_number_to_double()
This function is no longer used for integer parsing, and it does not
work in locales with decimal separators other than '.', which is
required to be the decimal separator for JSON numbers.

Change-Id: I8e97e15cc2699e647652f83b71676c11d32d29ce
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/364134
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-06 12:00:25 -04:00
Daniel Verkamp
7089d58244 json_util: parse integer numbers directly
Avoid a trip through floating-point numbers when parsing JSON numbers as
integers.

This will allow parsing 64-bit integers that cannot be represented in
double precision floating point.

Change-Id: Ic428c9f12e44e6dbee72f39a91ecb56ab30b365f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/364132
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-06-06 12:00:25 -04:00
Ben Walker
d7134439f6 event: Remove use of rte_lcore
Change-Id: Ia20284118077e49f2988f30469c3e3f4100ebdf0
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/363607
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-05 18:43:43 -04:00
Edward Yang
54e16f96b3 bdev: Add spdk_bdev_nvme_admin_passthru
Support passthru for NVMe admin commands.

Change-Id: If926f2ecabb078a553158f544c10a92452dbdb39
Signed-off-by: Edward Yang <eyang@us.fujitsu.com>
Reviewed-on: https://review.gerrithub.io/363294
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>
2017-06-05 18:07:10 -04:00
Ben Walker
aec9d3dcb5 bdev: Add I/O stat tracking
Change-Id: I05bb8294471bac3f9ecc4744e28d7d40e57e5364
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362622
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-06-05 16:38:40 -04:00
Ziye Yang
5b89e4a18f iscsi, param: fix memory leak related issue
issue: ASAN reported that param_ut free already
freed memory.

Reason: spdk_iscsi_negotiate_params does some
modifcation on params and may free old params
and create new params. To solve this issue,
we need to input **params but not *params in
spdk_iscsi_negotiate_params

Change-Id: I68658fd8e08f317343753620692f04e7b0b57577
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/363670
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>
2017-06-05 13:34:40 -04:00
Changpeng Liu
b6a9493bb1 nvme: enable CMB for submission queue by default
Also provide an option in perf tool let users to
disable it.

Change-Id: If4952513d77cecaa4f9403fbea811d86916ee87c
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/363311
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>
2017-06-05 13:03:48 -04:00
Ben Walker
ce8de311a4 vhost: Fix reset handling
Resets were not correctly aborting all I/O.

Change-Id: I03fd5f7094b3aa53f070eb731c30e5f9bfbfea16
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/363626
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-05 12:56:26 -04:00
Daniel Verkamp
a0cb1377f6 Update existing fallthrough comments for GCC 7
The stricter warning levels (-Wimplicit-fallthrough=4) require all-caps
FALLTHROUGH, so update the existing comments to match.

Change-Id: I5f8608101cad31d8ea8e84d48604397f98400e87
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363491
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-02 19:48:49 -04:00
Daniel Verkamp
aab3b1def3 scsi: add GCC 7 -Wimplicit-fallthrough annotations
Mark the intentional switch/case fallthrough locations with comments
recognized by -Wimplicit-fallthrough=4 in GCC 7.

Change-Id: Ibf2b077d9432e7f23a3833f98edd4768e5e8cfe9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363490
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-02 19:48:49 -04:00
Daniel Verkamp
adf3cc2bc8 vhost: add fallback if linux/vfio.h is missing
On older kernels that don't have VFIO, stub out the vhost IOMMU
functions.

Fixes build on older distributions (e.g. CentOS 6).

Change-Id: Icda9998e494c24d252511a4da93ecb56502fe5a7
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363473
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-02 19:47:09 -04:00
Ben Walker
40217741c0 bdev: Move pending queues to mgmt channel
Instead of an array of queues per core, allocate
the queues per management channel.

Change-Id: I4ace5bd13362a549a45aba62e978dabbbe1a333d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362617
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: John Meneghini <johnm@netapp.com>
2017-06-02 13:30:37 -04:00
Changpeng Liu
0a49531fab vhost: remove direct DPDK API call from vhost library
Change-Id: I5bad83197824dfb9042540c573134962a917b8bd
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/363152
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2017-06-02 12:04:15 -04:00
Ben Walker
151de2c479 bdev: Add robust, thread-safe reset handling
Reset requests will now pass messages to all channels,
correctly abort all queued I/O, wait for that to complete,
and then pass the reset request down to the next lower
layer in the stack.

Change-Id: I167cc5f424d3d0fc52b041bda63ee176f6acea29
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362616
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-02 12:02:53 -04:00
Ben Walker
4fe9536de0 bdev: Remove reset_type
There is now just one type of reset, which is equivalent
to a HARD reset previously.

Change-Id: I955b219cbc5c25793d97de1cc003b30ae99313ac
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362615
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-06-02 12:02:53 -04:00
Daniel Verkamp
7b2a6b05d0 nvmf: add public API to get subsystem serial number
Change-Id: I61b6579e8698d16e5a8ab74d304af9ea53f9dce4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363307
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Daniel Verkamp
bbfcb0943b nvmf_tgt: use public API to set serial number
Also move the validation of serial number length into the library.

Change-Id: Ibc9384fc1fccb87bd15b75da7f3942570900bd96
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363304
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Daniel Verkamp
77db92cf38 nvmf: remove unnecessary subsystem num_hosts field
Change-Id: Ic1b7b1b6497dbae5ceda9bb91f9d4d2037b7243f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363301
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Daniel Verkamp
806644a05f nvmf: remove spdk_nvmf_subsystem_exists()
spdk_nvmf_subsystem_exists(nqn) is exactly equivalent to
spdk_nvmf_find_subsystem(nqn) != NULL.

Change-Id: Icafa43166c99d1364b49cef30a5ade4b6625bcd5
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363299
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Daniel Verkamp
1c7b6f8dab nvmf: add spdk_ prefix to nvmf_find_subsystem()
Change-Id: I02a44c3790830b3918dca418c6bb85e82ddac273
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363298
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Daniel Verkamp
81344a66ef nvmf: make spdk_nvmf_ctrlr_ops contents private
This structure is not part of the public API and shouldn't be in
spdk/nvmf.h.

Change-Id: I29fbb269d3ab77cd9fa9908726abc7c7a4c6d8c1
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/363296
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 19:43:45 -04:00
Dariusz Stojaczyk
7d42767817 vhost: removed redundant check in get_scsi_lun()
Each call to get_scsi_lun is preceeded with the same check.

Change-Id: I4597674f0024e923908d85bbfbd5404c500fddb1
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/363172
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-06-01 14:15:50 -04:00
Jim Harris
59ed2aa9f6 blobfs: lock accesses to sync fs_request list
Synchronous blobfs channels may allocate requests from
the synchronous thread, but free the request from the
async thread - especially for code that is shared between
sync and async modes.  So add a lock to the request list
for sync channels only.

Note this only affects metadata path (i.e. sync, delete, open)
and not I/O path.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I932c500807c2f459c697fab2ffd91a88b88b0c87

Reviewed-on: https://review.gerrithub.io/362964
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>
2017-06-01 13:18:26 -04:00
Tomasz Zawadzki
5e132b6b7a scsi: Set proper transport protocol id
This patch adds property of transport protocol id to spdk_scsi_dev.
This allows to change it depending on which appliation is using
the device. Previously only iSCSI was used for all.

Setting protocol id is done at the time when device is added to
vhost controller (vhost) or target node (iSCSI).

Please note that for SPDK vhost SAS protocol id is used,
as that is what kernel vhost reports as well.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I455a856c5d7796a749b6650fee0218d526e094ed
Reviewed-on: https://review.gerrithub.io/362864
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>
2017-06-01 12:10:22 -04:00
Pawel Wodkowski
0888dd4cc6 vhost: add device type field
Change-Id: I848c8c5dcf837c11e052bb03cba962a843f28778
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/363088
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>
2017-05-31 20:25:45 -04:00
Dariusz Stojaczyk
3671543a5b vhost: changed dev_construct()/destruct() to dev_load()/unload()
Since vhost_scsi::new_device() would look mostly the same in vhost_blk,
most of it's logic has been moved into spdk_vhost_dev_construct(),
which is now renamed to spdk_vhost_dev_load().

Change-Id: Ic6db93273f916bd021c2ff68946706d055a514d6
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/362547
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>
2017-05-31 20:25:45 -04:00
Pawel Wodkowski
d7054e23b7 vhost: move VM memory register/unregister to vhost.c
It is not vhost SCSI specific.


Change-Id: I7eff272fcd1aa47135de7178d0763b1793b298ec
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/363087
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>
2017-05-31 20:25:45 -04:00
Pawel Wodkowski
c190dcd767 vhost: move gpa_to_vva to vhost.c
This function is not vhost SCSI specific and can be reused. Also change
type of returned value to 'void *' so no explicit cast is needed.

Change-Id: I33a37856903dacdddae9e73e695cc66465ade089
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/363086
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>
2017-05-31 20:25:45 -04:00
Dariusz Stojaczyk
7a9af3963e vhost: moved vqueue/vring management to vhost.c
vqueue/vring management shouldn't differ between vhost-blk and
vhost-scsi, so it has been moved to vhost.c/vhost_internal.h

Change-Id: I5ecf2f7a34ec28deb94c7abef4225038913edf0d
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/362505
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>
2017-05-31 20:25:45 -04:00
Jim Harris
1285f823e4 blobfs: relax locking for __rw_send_from_file calls
The file's lock does not need to be held across
calls to this function as the file is not dereferenced
at all.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Idf0a52f2528521f12e45963bb0ab1f414b37380f
Reviewed-on: https://review.gerrithub.io/363138
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-05-31 17:49:46 -04:00
Ziye Yang
2b2ab24477 reactor: handle start_fn event after subsystem init is ready
SPDK subysystem initialization now is async,
so we need to guarantee the initialization is done,
then call event composed of start_fn passed by user
in spdk_app_start.

Change-Id: Icc790cbb3da04c1063204938b79140c4218986e4
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/362654
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-05-31 15:30:58 -04:00
Ziye Yang
9104fe73ce bdev: make (v)bdev init in async manner
Change-Id: I93684a004e2ae276734edbb4767b5ba1bac3dd48
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/362111
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>
2017-05-31 15:30:58 -04:00
Ziye Yang
6e0d1dcdfa subsystem: make subsystem init in async manner
The next patch will make bdev modules init
in the async manner.

Change-Id: I4909c80510d786daf54003b99a5925428cf37373
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/362110
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-31 15:30:58 -04:00
John Meneghini
8a44220b1a env: Rename spdk_malloc/zmalloc/realloc/free to spdk_dma_(func)
- rename spdk_malloc_socket to spdk_dma_malloc_socket
  - rename spdk_malloc to spdk_dma_malloc
  - rename spdk_zmalloc to spdk_dma_zmalloc
  - rename spdk_realloc to spdk_dma_realloc
  - rename spdk_free to spdk_dma_free

Change-Id: I52a11b7a4243281f9c56f503e826fd7c4a1fd883
Signed-off-by: John Meneghini <johnm@netapp.com>
Reviewed-on: https://review.gerrithub.io/362604
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-31 15:30:27 -04:00
Seth Howell
7883a289ec json_util: fixed off by 1 error.
Also add tests to ensure against regressions on this fix.

Change-Id: Ic617f75f1865f17121896f7622290e64020bb2d6
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/362275
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>
2017-05-30 23:55:36 -04:00
Daniel Verkamp
cd88c7c53b bdev: add hook for bdev registration
This allows virtual blockdevs to inspect newly-added bdevs and
potentially insert themselves automatically.

Change-Id: If567a950d753e5f08861a5de22a2e1350376e50f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362077
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>
2017-05-30 08:53:29 -04:00
Ben Walker
f198553530 blob: Coalesce unmaps
Previously, each freed cluster was unmapped individually.
Instead, coalesce unmaps for contiguous clusters to reduce
the volume of commands sent.

Change-Id: I6ea1d2e1235e3c030cd2826c97e57aca571bd2ae
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362773
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-05-26 17:59:38 -04:00
Ben Walker
7058f85318 bdev: When a channel is destroyed, abort all queued I/O
Change-Id: I03d042cbb6de08f7e07b8c0ccc8af44d7b16ae3b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362614
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 17:41:22 -04:00
Ben Walker
35ab754449 bdev: Make spdk_bdev_mgr an I/O device
Currently this is just a placeholder, but eventually
it will be used to replace the per-core queues.

Change-Id: Iefeb90711bcf001a383e36cd4eaadf813af66506
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362613
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 17:41:22 -04:00
Ben Walker
ff87d29cc3 io_channel: Add mechanism to call a function on each channel
For a given I/O device, call a user function on the correct
thread for each channel.

Change-Id: I568a443184ac834c80c5e36b80aa9b6f8bb0ac99
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362256
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 17:41:22 -04:00
Ben Walker
a30ffb986e io_channel: No longer use thread-local storage
This serves two purposes. First, some older compiler
chains don't support thread local storage. Second,
we're going to need a way to iterate the list of
threads in the future, so keep them in a list.

Change-Id: I80e709f4665afb03cf4bcf0db19ef8ea353acdc1
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362255
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 17:41:22 -04:00
Ben Walker
666bf939f1 io_channel: Rename io_channel_create_cb_t to spdk_io_channel_create_cb
Add spdk, remove _t

Change-Id: Ic05b405439020b5ec5f2098ab04d37a4dd3cb5b6
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362254
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 17:41:22 -04:00
Ben Walker
2ef5722f3c io_channel: Add a message passing callback
When a thread is registered, the user must provide
a function pointer that can pass a message to that thread.

Change-Id: I743b5e0d6e3b5118c0a68d2fcedbccdd6fb237f9
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362067
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
2017-05-26 13:42:19 -04:00
Ben Walker
f1c5344b32 io_channel: Add a thread identifier
This will be used in the future to pass a message
to any given thread.

Change-Id: I3be5fe66244e360b7667427647fd8fdede110930
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362066
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 13:42:19 -04:00
Ben Walker
d969ac445a io_channel: Remove per-channel priority
This wasn't used anywhere and we currently believe there
are superior software-only techniques for controlling
quality of service.

Change-Id: Icdadd5870ed0629b338c307d2619bbc242c3e7a3
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362065
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-26 13:42:19 -04:00
Dariusz Stojaczyk
d5b6b1880e vhost: reordered functions in vhost_scsi.c
General cleanup.

Change-Id: I31083a7210634b8fc9d1336aff08f9793b249930
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/362296
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>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
d237aba3b9 vhost: separated scsi layer from vhost.c
Implemented vhost abstraction layer in preparations for vhost-blk.
Generalized vhost interface and separated vhost-scsi code.

Code placement:
vhost.h
 * vhost_startup() & shutdown()
 * public generic controller API

vhost_internal.h
 * includes rte_vhost.h from DPDK
 * struct spdk_vhost_dev (it has to be inherited by vhost_scsi)
 * vhost_dev management
   * register/unregister() to be used in new_device()/destroy_device()
   * construct/destruct() to be used on permanent creation/destruction.
 * reactor core management

vhost_scsi.h
 * spdk_vhost_scsi_dev management
   * reading VhostScsi section of vhost config file
   * creating/removing controllers
   * adding/removing devices to controllers

vhost_scsi.c
 * virtio scsi logic
 * new_device(), destroy_device()

Change-Id: Iaf661b14eba034593bef24853c525935d9b18e46
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361963
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: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
fb4b23b50b vhost: introduced spdk_vhost_free_reactor()
g_num_ctrlrs can now be fully encapsulated in vhost.c

Change-Id: I812c9ec0a21fec37025665104172ff5ac71ff3bc
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361951
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: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
aa72cafd8b vhost: introduced spdk_vhost_dev_unregister()
Separated new function spdk_vhost_dev_unregister() out of
spdk_vhost_scsi_dev_remove().

Change-Id: Ic604f10233ca3f7998774595c10d6c120dd43e2e
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361948
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>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
2f1cfc7be0 vhost: introduced spdk_vhost_dev_backend struct
Added new struct and separated new function spdk_vhost_dev_register()
out of spdk_vhost_scsi_dev_construct().

Change-Id: I97c303e40759426fa68a097c7ba47c62fbc97bd0
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361882
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: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
53a452d228 vhost: removed dpdk_vid_mapping
As a consequence MAX_VHOST_DEVICE define was removed, which was a
copy of a private DPDK define. Using it was more likely a hack.
Also renamed MAX_SCSI_CTRLRS to MAX_VHOST_DEVICES.

Instead of dpdk_vid_mapping, a new function
spdk_vhost_dev_find_by_vid is used. It simply iterates through
all the controllers.

Since this patch required to move the definition of spdk_vhost_ctrlrs
array, it has been renamed to g_spdk_vhost_devices.

Change-Id: I9b70ccb97ffaa90a22c6a1c90850b3205a9b9b27
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361881
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
c6745bc2b8 vhost: generalized spdk_vhost_dev struct
Moved controller-related variables from spdk_vhost_scsi_controller to
spdk_vhost_dev. Generalized vhost to operate on non scsi-specific
controller struct.

Now spdk_vhost_scsi_dev contains direct spdk_vhost_dev field instead
of a pointer. This means there is no more allocation/deallocation
on poller start/stop.

Change-Id: I99739d58833ee023a75cb530cc80294e02f28090
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/361880
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-25 18:30:44 -04:00
Dariusz Stojaczyk
78360d5646 vhost: extracted alloc/free out of vhost_dev constructor/destructor
* spdk_vhost_dev_create() operates on **existing** vhost_devs, rather
than allocates and returns them.
 * spdk_vhost_dev_destruct() no longer deallocates devices

This is a preparation for making vhost_dev a direct field of
vhost_scsi_dev (no longer a pointer). That patch would be too massive
to contain this one.

Change-Id: Idec84cee22795ad88e8499eb933664c4b8c7d101
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/362187
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>
2017-05-25 18:30:44 -04:00
Jim Harris
e045a02ca0 blobfs: cleanup cache pool
Keep a global counter of the number of opened
blobfs instances.  Allocate the cache pool when the
first instance is opened, and free the cache pool
when the last instance closes.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I99db2b587b738badcd2a54c9f5dc6ec8370ce22c

Reviewed-on: https://review.gerrithub.io/362606
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-25 16:55:03 -04:00
Jim Harris
8933ba2119 bdev/error: do not submit an I/O that will be failed
If error injection is enabled and we choose an I/O to
fail, do not submit the I/O and fail it when completed -
just fail it immediately.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ib58513f66e0df22c36137c0adb273fc31066c983

Reviewed-on: https://review.gerrithub.io/362386
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-25 13:53:06 -04:00
Pawel Wodkowski
5fafd5c277 env: remove element size from spdk_ring_create
It is unused an misleading.

Change-Id: Idb2e3c1b244dfb4aa99bcecee8a3dba6efbc33aa
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/362335
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
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>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-25 13:52:49 -04:00
Pawel Wodkowski
8557b1c2c5 env: add single-producer, single-consumer ring type
This case will be used in incoming vhost patches.

Change-Id: I62d569c01e9b8bf25c9293dff827d1b4406cea04
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/362334
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
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: Ben Walker <benjamin.walker@intel.com>
2017-05-25 13:52:49 -04:00
Jim Harris
24e0f710cf bdev/nvme: only enable hotplug in primary processes
Only DPDK primary processes can initiate device probe.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia9f966a65fc98ad92b48814dbd6f36f78905162f

Reviewed-on: https://review.gerrithub.io/362452
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-25 13:52:09 -04:00
Jim Harris
41d477cfa3 iscsi: free mempools on exit
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia3b3afc1a5331fe2b89d75eeddfd6e4430567cff

Reviewed-on: https://review.gerrithub.io/362450
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-25 13:52:09 -04:00
Tomasz Zawadzki
216901dbd7 vhost: close remaining memory region FDs during backend cleanup
After changes done in commit f325e71c, closing old FD is delayed until
VHOST_USER_SET_VRING_ADDR. If VM is closed before this call, original FDs
remain during vhost_backend_cleanup. This resolves issue #162.

This patch closes second set of FDs during vhost backend cleanup.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: Ieb9d123c987009ac451b6214bb74d2720d852781
Reviewed-on: https://review.gerrithub.io/361787
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Piotr Pelpliński <piotr.pelplinski@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-24 11:53:47 -04:00
Changpeng Liu
2e28ef9d02 scsi: fix scsi compliance issue for inquiry VPD NAA identifier
NAA locally assinged designator(3h) format was defined in SPC4,
as SPDK claimed can only support SPC3, so we used NAA IEEE
extended identifier format instead.

Change-Id: Ia4c153263c6b89ae8160488deaf16fd0e5cac8dc
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/361904
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>
2017-05-24 11:51:00 -04:00
Pawel Wodkowski
5b392ecdd1 vhost: fix deadlock on VFIO registration failure
Move mutext locking to API functions to prevent deadlock when memory
registration fail. Fixes #163.

Change-Id: I4804f671ed9d82ede87bc9439c36368b94e2d636
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/361934
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>
2017-05-24 11:49:11 -04:00
Jim Harris
2e3f07ac40 bdev: add spdk_io_channel parameter to spdk_bdev_reset
This ensures that all spdk_bdev_io structures now have
an attached channel, simplifying some future work around
things like counting the number of outstanding IOs for
a given channel (which otherwise would have had to
account specially for resets).

Reset semantics are still that they affect the entire bdev
and not just the channel it was submitted on.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8aad21a88faacecfd94bdba350059528eb62c390

Reviewed-on: https://review.gerrithub.io/362251
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>
2017-05-24 11:42:15 -04:00
Jim Harris
1a2cd90c98 scsi: simplify lun task mgmt path
1) Consolidate the checks for a null LUN
2) Eliminate a bunch of static functions that are no
   longer needed due to #1

This prepares for some upcoming bdev reset changes.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I38475581b69ba4a034fc7b9e133cfcdfc535a5d3

Reviewed-on: https://review.gerrithub.io/362262
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>
2017-05-24 11:42:15 -04:00
Jim Harris
fa1429ba02 bdev/nvme: clean up return codes
This prepares for translating -ENOMEM from the NVMe
driver into an associated bdev status code.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0250d5d48e2131da31b71cf9695d12fee67b0fbb

Reviewed-on: https://review.gerrithub.io/362234
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>
2017-05-24 11:42:15 -04:00
Daniel Verkamp
a02c0062bd nvme/host: use spec-compliant PSDT value 01b
Resolves GitHub issue #165.

Change-Id: If214bdfbe5b92faf2a6611d25e36c3c8876e3190
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362239
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>
2017-05-23 17:10:50 -04:00
Daniel Verkamp
b9bcc3531e bdev: make enum spdk_bdev_io_status private
The user should not see the bdev_io status directly; the NVMe and SCSI
error code wrappers provide the ability to translate to the desired
format regardless of what kind of error is stored inside the bdev_io.

Replace the spdk_bdev_io_completion_cb status parameter with a bool
simply indiciating whether the I/O completed successfully.

Change-Id: Iad18c2dac4374112c41b7a656154ed3ae1a68569
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362047
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-23 16:30:10 -04:00
Daniel Verkamp
54cfee2e84 iscsi: replace magic numbers with SCSI constants
Change-Id: I9b33b038a019576d162c868e2021db1c6589b4f4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/362061
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2017-05-23 16:29:34 -04:00
GangCao
afbb5ceec2 nvme: remove NVME_IO_TRACKERS to use hardware and user specified size
Change-Id: Ie6946343314a62e4416d8cd7dadd98c20a1c1e3a
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/361700
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-23 16:01:52 -04:00
Ben Walker
305cb239d2 io_channel: Remove unique flag
This is no longer used anywhere. For the places where we previously
used it, we've since found alternate solutions that do not
require it.

Change-Id: I738a80b95ef50348ce1c14969a3812b0a625b3fd
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/362064
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Ben Walker
267a4e1ebd blobfs: Add separate I/O target for md
This isn't strictly necessary, but follow the
pattern for sync and I/O devices.

Change-Id: I34b1bd6f9da06008e419f55ee9810022f5fed3d5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361674
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Ben Walker
a3ab7610f6 blobfs: Use a separate io_device for sync channels
This allows us to no longer use the unique=true flag as well
as assign sync channels different queue depths.

Change-Id: I0b3129a3139bf39d745c2a4dc1d9f402c0a8faa8
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361673
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Ben Walker
3b683c188e blobfs: Use a separate io_device for io channels
This allows us to have a different queue depth
on I/O channels than on metadata/sync channels

Change-Id: I30dfc95af550082efffec4b12ed6c4cd4b8c11cf
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361672
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Ben Walker
345f3d7ddb blobfs: No longer use unique_ctx when creating channels
This temporarily increases the size of the sync channel
to 512 (like all of the other channels). This will be
fixed in a later patch.

Change-Id: I679f1cb7c9f7afdad0acf9a3a26f9632f60cc510
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361670
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Ben Walker
60e8fb4979 blob: Use a separate md and I/O target
This allows us to allocate different size channels and
not require the unique flag.

Change-Id: I4b1ffd244b60e9e9ab06f9ab4da8161ab57e1169
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/361668
Tested-by: <sys_sgsw@intel.com>
Reviewed-by: Piotr Pelpliński <piotr.pelplinski@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
2017-05-22 19:27:17 -04:00
Daniel Verkamp
5f3841f0e3 nvmf: allow NQN to be 223 bytes + null terminator
An ECN to the NVMe 1.3 specification has clarified that the NQN may
contain 223 bytes before the null terminator.  Make all of our NQN
length checks consistently enforce this behavior.

Change-Id: Iebfd57d11abea64964c7a6ad9d886e40efa243c3
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-18 14:36:39 -07:00
Ben Walker
4eda4fd245 Fix incorrect free in blob/blobfs
The channel memory isn't allocated by these
libraries, so they can't free it.

Change-Id: I30909fa4e77bc5a41b45230f04ba5fe75b172dbf
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-18 14:36:11 -07:00
Ben Walker
4a3182b891 blob: Add a separate config parameter for channel ops
Separate the maximum metadata operations from the
maximum channel operations.

Change-Id: I1bbd440ab094a2a2e19c9a5b71724ac91ba88e42
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-18 14:36:11 -07:00
Ben Walker
bea2e2308f blob: Remove per-channel queue size configuration
This will need to be configured globally for all channels.

Change-Id: I773252f220373617f8d09d1f24243db8095cf8a4
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-18 14:36:11 -07:00
Daniel Verkamp
f66a0c9e34 scsi: pass management function as a parameter
Specify the function as a parameter to spdk_scsi_dev_queue_mgmt_task.
This makes the API clearer by making it explicit that the management
function is required for management tasks.

Change-Id: I92d893aadb6faebea81dd79729894d2c2fddf088
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-18 13:47:42 -07:00
Daniel Verkamp
1e7e6a618b scsi: automatically set spdk_scsi_task::type field
The type of a task is known due to the function used to submit it:
- spdk_scsi_dev_queue_task() for normal SCSI command tasks
- spdk_scsi_dev_queue_mgmt_task() for management tasks

Change-Id: I183a1f89ab85f3fce1de2491e77d95d4b147fd72
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-18 13:47:42 -07:00
Daniel Verkamp
de5ffac4ec scsi: factor out read/write LBA range check
The SCSI read and write functions use the same LBA range check, so move
it up to spdk_bdev_scsi_readwrite().

spdk_bdev_scsi_read() previously passed task->transfer_len / blen rather
than xfer_len to the LBA range checking function, but this should
actually be using the transfer length field from the CDB (which is the
xfer_len parameter).

Change-Id: I8285abf936a18a0baf7cc25709945637e4e6b87d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-18 13:47:42 -07:00
Jim Harris
af781ee003 bdev/nvme: print errno on spdk_nvme_ns_cmd submission failure
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I785bfd663a2921f5e4c0630bce7906a928f26376
2017-05-18 10:23:43 -07:00
Cunyin Chang
32b4ab71a2 bdev/error: Add new bdev/error to inject errors in bdev layer.
Change-Id: I3c68cb5911b3662dbb35f551dc291d5ebdd84bf6
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
2017-05-18 07:56:14 -07:00
Dariusz Stojaczyk
679e2831bd vhost: added rpc commands to remove vhost controllers and devices
Added new rpc commands together with underlying vhost API and tests.

Change-Id: Ib9c6a530d0909193ea5115aaac4920c44f39613c
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-18 01:01:46 -07:00
Pawel Wodkowski
82d26c4f20 vhost: track VFIO mappings
If more than one controller is added to guest the same memory table is
set multiple times making VFIO registration/unregistration failures

Change-Id: Ib55c38e292495e549c070beb0234e73f269e63d5
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-18 00:47:55 -07:00
Dariusz Stojaczyk
709d5453a3 vhost: obey VIRTIO_F_NOTIFY_ON_EMPTY
It'll trigger interrupts even if VRING_AVAIL_F_NO_INTERRUPT is present.
Also delegate notifying check to separate function vq_should_notify. It
will be expanded even more once we implement VIRTIO_RING_F_EVENT_IDX
support.

Change-Id: Ic41fd18c9c005c6266b109b0eda4a6dfaef5f06a
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-17 10:24:48 -07:00
Dariusz Stojaczyk
7463bf6d2a vhost: add negotiated_features field in spdk_vhost_dev
Added new struct field and an rte_vhost_get_negotiated_features() call.

Change-Id: I5937026720b5341c7377c27fee682996b51e836e
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-17 10:24:48 -07:00
Daniel Verkamp
a591161cb2 bdev: make struct spdk_bdev contents private
Change-Id: If203e82f8cd10d5998a565ad490ef11e2916687f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-17 09:45:42 -07:00
Daniel Verkamp
9d1e05d2f2 bdev: add getter for write cache enable status
Change-Id: I93656deda8956c77e9565ee1f271795642e62ac9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-17 09:45:42 -07:00
Daniel Verkamp
0d56499d3a scsi_bdev: use max unmap descriptor getter
Change-Id: Id67b158c36b2b879849b18909bc9bb9a5273f8c7
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-17 09:45:42 -07:00
Daniel Verkamp
76e67baf28 blob_bdev: use bdev block size getter
Change-Id: Id22f904fdb71798ef2369415d38f468b06d7ac07
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-17 09:45:42 -07:00
Daniel Verkamp
133dc1e6ae iscsi: clean up iSCSI/SCSI task conversions
Add a helper function to find the iSCSI task given its nested SCSI task
structure, and use it to remove all casting between spdk_scsi_task and
spdk_iscsi_task.

Change-Id: Idc7c6d0a3b4d53041916d25a1bdecedfb56b94f9
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-16 16:45:22 -07:00
GangCao
b347d551e8 nvme: detach the pci device with calling process's own devhandle
Change-Id: I2693b4bd29e0500379d5e399723aec7e44bffca3
Signed-off-by: GangCao <gang.cao@intel.com>
2017-05-16 10:35:29 -07:00
Ziye Yang
644678258f blobfs: move the location of next buffer check.
This patch address the issue:
https://github.com/spdk/spdk/issues/151.

For cache_append_no_cache in cache_ut testcase,
there is resource contention for buffer among two
threads in the following two functions.
Thread 0: cache_free_buffers
Thread1: __file_flush_done

When the thread1 execuctes __file_flush_done,
it calls the call back: __sem_post defined in
following statement in spdk_file_sync

_file_sync(file, channel, __sem_post, &channel->sem);

Thus Thread 0 will execute next function
cache_buffers, and it frees the buffer.

Then Thread 1 continues executing the remaining statements
in __file_flush_done with the assert function, and touches
the space already freed.

So it will be safe to move ahead the next buffer check.

Change-Id: Ic007b3481f4e3a17d47eeca5c9c802001949a5ab
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-05-16 10:51:26 +08:00
Jim Harris
4eafea0360 bdev: add a null bdev module
Also change the discovery/nvmf.sh test to use it.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I56bce9a84bd46f13b6d4f34da81abf23413f2598
2017-05-15 14:00:37 -07:00
Daniel Verkamp
7455910d1b scsi_bdev: simplify scsi_readwrite helper function
Factor out the common data direction Check Condition logic from read
and write.

Change-Id: Ia2fbd5f69eefd826ebb5ab2a8e99aeb380f1b7ea
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-15 13:35:18 -07:00
Daniel Verkamp
8e379851ab scsi_bdev: clarify block count variable names
What was previously called "maxlba" was actually the block count
of the device, which is the maximum LBA plus one.

Change-Id: I5e141a3eb0c79b2aa70006739c5586cb50744d3d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-15 13:35:18 -07:00
Daniel Verkamp
19a9871495 scsi: remove spdk_scsi_task::owner_task_ctr
The SCSI library already provides a callback when the task is released
(free_fn), so the user can update their own task counter.

Change-Id: I7fb13f6fff66dbba2315fd03fb06e49f793be123
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-15 10:12:42 -07:00
Daniel Verkamp
7145cf62c6 scsi: set free_fn in spdk_scsi_task_construct()
The task free callback function is required, so make its assignment part
of task construction.

Change-Id: I2f5fdf73b064653ee85b4e7961cb1653a0a4107d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-15 10:12:42 -07:00
Daniel Verkamp
1aea6c5b6f bdev: add getter for maximum unmap descriptors
Change-Id: Ifbdd29e2081600bf0d860582d80696546107cf1b
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-15 10:11:23 -07:00
wenzhong wu
8651c2eb47 nvmf: call function spdk_nvmf_valid_nqn to check the host_nqn.
Change-Id: I035fe2538b46d0a19eb00fd829fae9e735c1ded9
Signed-off-by: wenzhong wu <wenzhongx.wu@intel.com>
2017-05-16 03:19:54 +08:00
Daniel Verkamp
326786a943 bdev: add getters for block size and block count
Change-Id: I6fad28da43c163ea4e2c4a04ced356b67d63652f
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-12 10:34:58 -07:00
Daniel Verkamp
1bcf22cde1 bdev: add getters for bdev name and product name
Change-Id: I6a75fada94fa845ecedd4cd8afc78f4259df14fe
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-12 09:36:30 -07:00
Dariusz Stojaczyk
6973898164 vhost: wrap vhost_user_read_cb with vhost_user mutex
This fixes spontaneous vhost hangs on SIGINT shutdown.
Apperently during vhost_destroy_device(conn->vid) from
line #284 another QEMU message might arrive, causing
vsocket->conn_mutex deadlock. (line #286)

Change-Id: I4f1c31a52facffd1eb1e1192591095f00da55031
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-12 08:05:37 -07:00
Pawel Wodkowski
66894635fa vhost: obey VRING_AVAIL_F_NO_INTERRUPT
If VRING_AVAIL_F_NO_INTERRUPT flag for queue is negotiated, guest can
set this flag in avail ring to suppress interrupts from host.

Change-Id: I2043067de33da009a974be5915ae81b1343fc424
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-12 07:53:59 -07:00
Paul Luse
bf31637789 nvme: add timeout value to multi-process driver init
Unit tests will be added as part of a separate patch updating all UT for
nvme.c.  Global used for timeout value so it can be easily overwritten
by the upcoming unit tests for this function.

Change-Id: I7fc15aab91601ac57c94cae266b212c0998d2495
Signed-off-by: paul luse <paul.e.luse@intel.com>
2017-05-12 07:53:48 -07:00
Changpeng Liu
c7f6f553b9 env/dpdk: use the new *rte_pci* API for DPDK 17.05
Change-Id: I1091a9556bbd42349dd8d26f238a54a5b55610a1
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-12 07:53:42 -07:00
Changpeng Liu
6c0d4c552b env/dpdk: return value changed of API rte_ring_mp_enqueue_bulk for DPDK 17.05
Change-Id: I44f1c1fc75fd343d30c602ca77940ebde3ed184a
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-11 12:45:31 -07:00
Ben Walker
58afa58a00 bdev: Simplify spdk_bdev_get_io_buf
This was implemented as two functions, but it
is much simpler as one. Also, the public function
was way at the bottom of the file instead of near
spdk_bdev_put_io_buf.

Change-Id: I3a90688910b0542cc77b6333bab15132cf514eeb
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-11 10:48:58 -07:00
Ben Walker
812ffaf1a3 bdev: Simplify spdk_bdev_finish
This was implemented as 3 separate functions but
it is simpler as 1.

Also, this wasn't previously freeing the buffer pools.

Change-Id: Ic1b2b3a0596e745a223099cb2a79bea6ef5c69cc
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-11 10:48:58 -07:00
Ben Walker
f2132bfd8a bdev: Simplify spdk_bdev_initialize
This was broken into three functions, but it is
a lot simpler as one.

Change-Id: If58ad50fe7d4f65c598b62f24e9e1ce7a64fdd8e
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-11 10:48:58 -07:00
Ben Walker
ba17fd1bf2 bdev: Move all globals into a single struct
This is better organizationally, but also will serve as
an io_device in the future.

Change-Id: I6d65cf39df59e874d13f5fccc5a489720e86c48f
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-11 10:48:58 -07:00
Xiaofan Yang
d8369fee70 iscsi/conn: Fix the bug when iterates the read command subtasks.
Change-Id: Ib11e804977b6057bcb9cc6d3a2d21ec43b45a99a
Signed-off-by: Xiaofan Yang <xiaofanx.yang@intel.com>
2017-05-11 08:09:23 +08:00
Daniel Verkamp
6a8cd333a9 bdev: add API function to get buffer alignment
Change-Id: I4c7eff4c903411994fc6bb9fae98f967eb14669c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-10 13:16:27 -07:00
Ben Walker
68be2e2842 bdev: Fix formatting issues
Return types should be on a separate line for definitions.

Change-Id: Iaa38dd00042359fc6640fc67053bd69ebbb7af03
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-10 13:12:18 -07:00
Ben Walker
58734f5222 bdev: Generalize buffer allocation
Make the buffer allocation work for all types of
commands, not just read.

Change-Id: I72d8f67a724566630e7c4a74759fcb08449f7de4
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-10 13:12:18 -07:00
Changpeng Liu
6de5c05556 scsi/bdev: fix wrong condition check for inquiry VPD page
Change-Id: I435825dbd6a7f990d702db9e0965a489f452cdc3
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-11 15:01:03 +08:00
Ben Walker
42491fb8fc env: Add wrappers for a lockless ring
Change-Id: I9679e4bcfc10e38672d1851f7e9f16b6adf7de9b
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 14:02:15 -07:00
Ben Walker
1ad673a9dd copy: Replace rte_memcpy with regular memcpy
Eliminate rte_memcpy dependency by replacing it with
regular memcpy. This may impact performance, but the only
use of rte_memcpy was in the malloc bdev which is for
testing only.

Change-Id: I3e8592cb08262272518ec3d29ea165b4e8f48a5c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 13:11:39 -07:00
Ed Rodriguez
2d2fde0d7f log: Use SPDK_ERRLOG in lieu of fprintf(stderr
Change-Id: Ic87d62516324b9c388a932b268714255b15a9a57
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 11:28:28 -07:00
John Meneghini
70f8a8e2f4 bdev: Use env mempool wrappers
Change-Id: I22f4f7894462c576c447346684a780bbc34f7ba3
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 11:26:34 -07:00
Daniel Verkamp
22b6098965 bdev: remove spdk_bdev thin_provisioning flag
Blockdevs already indicate support for unmap via
spdk_bdev_io_type_supported(bdev, SPDK_BDEV_IO_TYPE_UNMAP).

Change-Id: I634f27a281fd900bb3a6da2e4ff8a74e43579578
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-09 08:42:10 -07:00
Ziye Yang
4c9645b555 blobstore: Add the size check of length field in desc_xattr
Change-Id: I522b29b3a603b1e83c866486d036137ee4fafcbd
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-05-09 08:41:53 -07:00
Ben Walker
9ed75e4dab bdev: Rename rbuf to just buf
We plan to use these buffers for more than just reads.

Change-Id: I8fa6cb432a6cfe4406fbf240cd3aa2ae4ab5f3d5
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 08:14:31 -07:00
Ben Walker
ed159eae1b bdev: Eliminate spdk_bdev_io::ctx
The user can get there via the bdev, so this didn't
have a purpose.

Change-Id: I7f85bb71d5ee238d37ba3624d0ac68a161c95e49
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 08:14:31 -07:00
Ben Walker
3415ce1227 bdev: Pass correct channel to bdev modules
Change-Id: I38911e70303f66f479c1495d4dbe02b2205cab8a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-09 08:14:31 -07:00
Daniel Verkamp
6d3a4cd2a4 vhost: define VIRTIO_F_VERSION_1 if it is missing
Older kernel headers don't have the definition of this macro, so define
it if necessary.

This is the same workaround as used in rte_vhost/vhost.h.

Change-Id: I01e0661db05de517adf8e24a47c63d32853cd385
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-08 20:12:19 -07:00
Pawel Wodkowski
8d7acdaaef vhost: upgrade SPDK vhost code to DPDK 17.05
Also replace the internal DPDK v17.02-based rte_vhost
library with the patched DPDK v17.05-based version.

Change-Id: Ibec0b0746592a1a3911c31642a945ab65495e33e
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-08 17:04:29 -07:00
Pawel Wodkowski
d391647bd0 vhost/rte_vhost: fix scan build and comment formatting issues
vhost_net.c file is not needed and fail scan build so remove it.

Change-Id: I5817201373f7253cc8bc1a9bdc5884197e166a14
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-05-08 17:04:29 -07:00
Daniel Verkamp
c2683a8e42 vhost_scsi: access VhostUsrMsg via packed struct
Fixes unaligned access to fields.

Change-Id: I43cff0c1cca7829da8f0d90774970e5feaa95515
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
0cf6cc7b54 vhost: free virtio_net::guest_pages in vhost_backend_cleanup()
guest_pages is being allocated in vhost_setup_mem_table(), reallocated
in add_one_guest_page(), but never freed. This patch fixes a memory
leak.

Change-Id: Ie381c43bafea5cdea2ac9f057c0282044a340dce
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
13657c7d75 vhost: fix malloc in rte_vhost_get_mem_table
Amount of allocated memory was too small, causing buffer overflow.

Change-Id: Ib43e0a9040f594fed0a8c5660a45aeb07e4400c7
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
2542779301 vhost: close callfd on VHOST_USER_GET_VRING_BASE message
This prevents from destroying & recreating user device in "incomplete"
vring state. virtio_is_ready() was returning true for devices with
vrings which did not have valid callfd (their VHOST_USER_SET_VRING_CALL
hasn't arrived yet)

Change-Id: Idc4b41efd544ff5c6b093a5a48798b41c55bbe06
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
52b10970c4 vhost: added API for getting/setting last descriptor indices of vrings
vhost-net devices might keep track of last descriptors indices by
themselves, and assuming they initially start at 0, but that is not the
case for vhost-scsi. Initial last descriptor indices are set via
VHOST_USER_SET_VRING_BASE message, and we cannot possibly predict what
will they be. Setting these to vqueue->used->idx is also not an option,
because there might be some yet unprocessed requests between these and
the actual last_idx. This patch adds API for getting/setting last
descriptor indices of vrings, so that they can be synchronized between
user-device and rte_vhost.

The last_idx flow could be as following:
 * vhost start
 * received SET_VRING_BASE msg, last_idx is set on rte_vhost side
 * created user-device, last_idx pulled from rte_vhost
 * requests are being processed by user-device, last_idx changes
 * destroyed user-device, last_idx pushed to rte_vhost
 * *at this point, vrings could be recreated and another SET_VRING_BASE
 message could arrive, so last_idx would be set*
 * recreated user-device, last_idx pulled from rte_vhost

Change-Id: I247ba4e461a2a2b524ccade364f5b7bf260f7538
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
8457b98cf2 vhost: defer setting up new mem table
There is an issue when QEMU sets new memory table just after guest OS
starts booting. Then, if guest OS tries to issue any I/O to device (e.g.
using BIOS INT13h - EDD) it will get stuck because previous addresses of
mmaped memory might change.

To fix this issue, defer using the new mem table until after we receive
the first SET_VRING_ADDR message. SET_VRING_ADDR will be sent by QEMU
when guest OS virtio (e.g. virtio-scsi) driver starts initialization.
At this point it is safe to invalidate the old mem tables because there
will be no more outstanding IO at this point.

Change-Id: I24772be87a8b6c8781868b9b7773317761499748
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Dariusz Stojaczyk
fab374a7c6 vhost: destroy vhost device before updating public vring data
For now DPDK assumes that callfd, kickfd and last_idx are being set just
once during vring initialization and device cannot be running while DPDK
receives SET_VRING_KICK, SET_VRING_CALL and SET_VRING_BASE messages.
However, that assumption is wrong. For Vhost SCSI messages might arrive
at any point of time, possibly multiple times, one after another.

QEMU issues SET_VRING_CALL once during device initialization, then again
during device start. The second message will close previous callfd,
which is still being used by the user-implementation of vhost device.
This results in writing to invalid (closed) callfd.

This patch destroys vhost device before setting callfd, kickfd and last
vring indices. It will be recreated right after (with updated vring
data).

Change-Id: I293bd91106f53f6c2f65d8b8a41f47ae7548cddc
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 17:04:29 -07:00
Jim Harris
a191eedb19 vhost: import copy of dpdk rte_vhost v17.05
This will be decoupled from the build to start.  Next
patches will modify this code to prepare it for use with
SPDK vhost-scsi.  The final patch will replace the existing
v17.02-based code with this version, and make the necessary
SPDK vhost changes to use it.

This enables to better track the differences between upstream
DPDK and our internal copy, while not breaking the build at
any point in the git history.

While here, expand the POSIX include file check to exclude
any directory starting with lib/vhost/rte_vhost (which would
include this new directory).

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Icf1202c1b7a898edff12aa226943a08b578cf962
2017-05-08 14:39:55 -07:00
Ben Walker
1b2764421c build: Automatically detect use of POSIX includes
Scan the source for POSIX includes outside of the
allowed locations in check_format.sh. This only
tests for POSIX headers - not Linux Standards Base.

Also, fix one bug that was caught by this addition.

Change-Id: Ib0ca93fe6ac552dc49d95b27b4803e40282027e8
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-08 13:20:37 -07:00
Ben Walker
b961d9cc12 include: Move the remainder of the code base to stdinc.h
Change-Id: I6a142feeaad3117bd3c75e7c5cb7231a1cfa78ae
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-08 13:20:36 -07:00
Ben Walker
ba1e1d5308 vhost: Rename LOG_DEBUG to VHOST_LOG_DEBUG
LOG_DEBUG is a symbol defined by POSIX, so if sys/log.h
is included the symbols conflict.

We'll need to push this patch to upstream DPDK too.

Change-Id: Ib263731864aca4791226ea6e3abb5ddfe42e97d8
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-08 13:19:20 -07:00
Dariusz Stojaczyk
328bc32f31 env/dpdk: allow compiling with DPDK 17.05-rc0 and rc1
FOREACH_DEVICE_ON_PCIBUS macro has been defined since rc2.

Change-Id: Iad61401520735dfde4e5715c32e74a54a2dff7da
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-05-08 13:01:53 -07:00
Vishal Verma
578c0d550d env: Adding socket variants for malloc and zmalloc
Signed-off-by: Vishal Verma <vishal4.verma@intel.com>
Change-Id: I2399842cf7bb80aefb6c511e167157502ba9018a
2017-05-08 09:41:47 -07:00
Ziye Yang
e48e569d83 nvmf/rdma: Replace the queue empty check
Replace with it with check the returned req
via spdk_unlikely macro

Change-Id: I1202b3955af9a68496d8ced7cf66c20cf26f7fff
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-05-08 09:40:15 -07:00
Daniel Verkamp
730a63d02b scsi: remove unreachable read iovec copy path
The SCSI layer always passes task->iovs to spdk_bdev_readv(), so there
is no way for task->iovs != bdev_io->u.read.iovs to be true.

Change-Id: I4c0a2075c6e50e4304d62707a29bededa37b4e5c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-05 09:39:49 -07:00
Daniel Verkamp
15fcb1020c bdev: make struct spdk_bdev_io contents private
Change-Id: I24ff52a7375d54df06b9769bc53ddf84691e1cb1
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-05 09:39:49 -07:00
Daniel Verkamp
825379870b bdev: add accessor function to get an I/O's iovec
Change-Id: I09321abe93303d6225fbcce4065b0024dd541b19
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-05 09:39:49 -07:00
Daniel Verkamp
91b13bd105 scsi: don't change bdev_io status in spdk_scsi_task_put()
The SCSI task bdev I/O should never be pending when spdk_scsi_task_put()
is called, and just setting the status to failed is not correct (when
the bdev eventually completes the I/O, it will write into the now-freed
bdev_io, which may be reused by someone else).

Change-Id: Iaad6ce9ab41539652abc40147fed47c5012109dc
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 14:59:12 -07:00
Changpeng Liu
fa511620c4 vhost: eliminate loop log print when no available requests
Change-Id: I3cd72317392b15bad888391b7bc7e1bc5e69385c
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-04 13:21:27 -07:00
Daniel Verkamp
2621c7bdc4 scsi: move bytes_completed to iSCSI task
Change-Id: Iff345f555f94a34fa4a6f86bc11efea704e39735
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 10:10:47 -07:00
Daniel Verkamp
7e7ed7ec10 scsi: move data_out_cnt to iSCSI task
Change-Id: I550de37d5088e2273b9d53ab8a2d6134c486fba0
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 10:10:47 -07:00
Daniel Verkamp
d4da788057 scsi: move desired_data_transfer_length to iSCSI task
Change-Id: I425f9cc7c572b4d819d0aae370f7a96c57d6eac1
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 10:10:47 -07:00
Daniel Verkamp
414b754579 scsi: remove SCSI task id and add iSCSI task tag
The SCSI layer was not using the task ID for anything; the iSCSI layer
was using it to store the task tag, so move it there and rename it to
"tag" to make its purpose clear.

Change-Id: Ibda4f4e215056116b9be4a3a0264f98bc4c29535
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 10:10:47 -07:00
Daniel Verkamp
4ccf74ab3b scsi: move subtask_list to spdk_iscsi_task
The SCSI layer doesn't use subtasks; these are an iSCSI layer concept.

Change-Id: I83871f02362f10fd4ecd4b2a1544eb76bfa53595
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 10:10:47 -07:00
Daniel Verkamp
7fb6b33488 bdev: convert set_scsi_status into a bdev_io completion function
In the pattern set by spdk_bdev_io_complete_nvme_status(), allow
blockdev modules to complete a bdev_io with a SCSI status code.

Also move it to the internal bdev header file, since only bdev modules
should be setting bdev_io status codes.

Change-Id: I8b6afad2c02d7c010c5e60f06a7c7e0785eb87ca
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 09:11:42 -07:00
Daniel Verkamp
dc9e11163e bdev: add API to translate to SCSI status
Move the scsi_nvme translation code from the SCSI library into bdev, and
provide a generic way to translate any bdev_io status into a SCSI
status.

Change-Id: Ib61a6209387c24543e31574e2b5ca249e2ac8b74
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 09:11:42 -07:00
Changpeng Liu
95d9ffc26a vhost/lib: remove dependency on DPDK vhost library
Since we keep a copy of DPDK vhost library, the header file don't
have dependency on DPDK vhost library.

Change-Id: I14d48e10227633547231e4f429e7375ffa76128d
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-04 09:11:31 -07:00
Ziye Yang
08e6f94226 blobfs: separate cache tree related definitions into tree.h
The code will locks clear to put those definitions into tree.h header

Change-Id: Ib1a34f19d9849acd7ea979eb0a6e153b0e8e39de
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-05-04 09:11:28 -07:00
Daniel Verkamp
2990f869a1 scsi: make spdk_scsi_port definition private
Change-Id: Ib2c17a4dd4ce680161be92f76b831df792f9ff4d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-04 09:05:53 -07:00
Changpeng Liu
2eb9a35323 env/dpdk: remove rte_eal_device_remove API call since DPDK 17.05
Since DPDK 17.05 API rte_eal_device_insert is only used for
virtual device scan and initialization, for PCI devices
which use Domain:Bus:Dev:Function, this API is no longer
valid.

Change-Id: I1ab63dfc3af188d01836e67cd8db745e035fc450
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-05 09:40:31 +08:00
Ben Walker
3d62b39b0d bdev: The bdev layer now generically creates io_channels
These channels can handle generic bdev context.

Change-Id: I61f41884ddf4cf86fa156e9051421b354bbb349d
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-05-03 15:16:02 -07:00
Jim Harris
5343c22f8d iscsi: do not free io channel refs if login failed
When destroying a connection, we need to check if we got to
full feature phase before freeing any io channels.  This is because
the io channels are only allocated as part of a successful login.

The Calsoft iSCSI test suite has tests which will fail login.
Since the test system was just using a malloc backend with memcpy,
so even though a channel was NULL in some cases, it was never used
since the memcpy engine doesn't need it.

This prepares for some future patches which extend the use
of io channels in the bdev layer.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2fb7b18a781caa0aadca319aa1e61a6ccf2c55fd
2017-05-03 14:44:39 -07:00
Daniel Verkamp
a3738d9031 scsi: make spdk_scsi_dev definition private
Change-Id: I62b36a22e11e845045f190886ae00aa644f96ec6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-03 12:34:57 -07:00
Daniel Verkamp
46af047ae7 json/util: simplify if (p) free(p) -> free(p)
Change-Id: I874b9893afcf2bdf333b14557feedcacad47d355
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-03 10:57:51 -07:00
Changpeng Liu
e9bfc11043 env/dpdk: additional library required for DPDK 17.05 ring based mempool
Change-Id: I2ef4a330a5628c84657f516cb2e4fad609b6c6d8
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-05-02 15:18:15 -07:00
Daniel Verkamp
fa81bdb3fe iscsi: drop redundant parentheses in rsph casts
This allows astyle to format the cast of address-of operations
correctly.

Change-Id: I9c8a4545c44601e769acc712ec7acf3a96f45ebb
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-02 15:15:39 -07:00
Daniel Verkamp
cce22074a7 iscsi: fix ISID interpretation in login requests
The ISID field is a 6-byte field in network (big-endian) byte order.

The previous code was casting the uint8_t isid[6] value into uint64_t,
which was actually casting the address of the first byte of isid (not
the contents of the array), and it was also not correctly converting
byte order.

Change-Id: Idd114e06d30040cf28931d7da7ffdc8d6c45e82a
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-02 15:15:39 -07:00
Liang Yan
b0091cfe5c bdev/nvme: Increase maximum controllers to 1024
Change-Id: I5a7a5f40b0d3c8c21e26f5358bc84f538a62bff4
Signed-off-by: Liang Yan <liang.z.yan@intel.com>
2017-05-02 13:51:39 -07:00
Daniel Verkamp
6b4da26449 bdev/nvme: dynamically allocate probe_ctx
Avoid allocating a large amount of stack space when increasing
NVME_MAX_CONTROLLERS.

Change-Id: I7017e5ed9f4d4f5c860dac608c3e5ae3c35864e7
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-02 13:51:39 -07:00
Daniel Verkamp
d0290b6525 scsi_internal.h: remove duplicated min/max macros
Use the ones from spdk/util.h instead.

Change-Id: Icd8416d1b3cb0cde35ac46ea8b1ba7a1ea8994c6
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-02 09:41:43 -07:00
Daniel Verkamp
12965bb6ce scsi: make spdk_scsi_lun definition private
The contents of struct spdk_scsi_lun don't need to be part of the public
API.

Change-Id: I101b77871054557380610fd901ab38bada463202
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-05-02 09:41:43 -07:00
Jim Harris
13f8cf1536 nvme: add NVME_QUIRK_DELAY_AFTER_QUEUE_ALLOC
The VirtualBox emulated NVMe device will intermittently
hang on the first read/write command after an I/O
qpair has been allocated.  The frequency of the hang
diminishes if a delay is added after allocating the I/O
qpair - until it disappears completely with a 100us delay.
So add a quirk to insert this delay.

Note - the 100us delay was tested by running
the hello_world example app 50000 times.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I237e31b1b8a1a1e28262851ae0a21cd7345f0f1a
2017-05-01 10:22:18 -07:00
Changpeng Liu
80e4f5ae8b env/dpdk: make SPDK can compile with DPDK 17.05
Change-Id: Id69a162b6f2abfdcd4d41b368cfcde52665eb904
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
2017-04-28 09:22:09 -07:00
Pawel Wodkowski
1bc2c5ab57 rpc: add bool-type decoder
Allow passing booleans in JSON.

Change-Id: I0b8f6c1579d8382b5b19a987ef5d913b4423c954
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-04-26 21:52:11 -07:00
Daniel Verkamp
50cc397d6f nvme: simplify nvme_ctrlr_destruct qpair loop
Fixes a scan-build warning about using qpairs after they have been
freed.

Change-Id: I263eabd6b784acf540c66136965f7705ef110a78
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-26 06:40:17 -07:00
Daniel Verkamp
5639b965ac scripts/check_format.sh: catch misaligned comments
Fix up the existing comment blocks misaligned in the first column.

Also add line numbers to the comment checks.

Change-Id: I9d28c365271df36e7013d74cbb02d0023ab4f581
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-26 06:39:58 -07:00
Tomasz Zawadzki
9879b50769 scsi: fix page control page field in MODE SENSE
This patch assigns correct value to page control.
Now that page control value is correctly taken from CDB,
error via sense data is reported when processing "saved values".
"Changeable values" are not supported, so all parameters
are reported as not changeable when requested.

Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I41378c96b1e8c716b5d0ce4b72777065fb122228
2017-04-24 15:26:39 -07:00
Cunyin Chang
bee8c04a4b nvmf: Move the check of NVMe remove from session layer to subsystem layer.
Change-Id: Icf2f7122c9af0cd6420384c93813d2e79ed9c69e
Signed-off-by: Cunyin Chang <cunyin.chang@intel.com>
2017-04-24 13:57:25 -07:00
Jim Harris
95487138e3 event: add spdk_trace instructions if event group mask specified
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I022bdccf502c6e63e6e8f5a80c5b5ec493262db9
2017-04-24 13:57:06 -07:00
Daniel Verkamp
6b91943209 scripts/check_format.sh: check spaces in comments
Fix up all existing spacing errors in comments and add an automated
check for patterns like /*comment*/.

Change-Id: I28f61c93612dc0f8aed66bd509da78e91ea9737e
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-24 13:37:41 -07:00
Pawel Wodkowski
9aaccfe3d7 vhost: enable VFIO
Vhost needs to register memory given by guest in VFIO container to be
able to do any DMA using this memory.

Currently DPDK doesn't provide any interface to handle guest memory, so
for now lets find container fd in /proc/self/fd/ directory and provide
some VFIO internal API that finally should extend DPDK API.

Change-Id: Iee9d496367ccd61219068fc0eadc17e786ff0731
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-04-24 12:20:04 -07:00
Ziye Yang
e15bd00711 nvme: support new format in spdk_pci_addr_parse
The new format is: domain.bus.device.function

For this format, since we use '.' as separator,
to avoid misusing, we only support the following:

1 domain.bus.device.function ( 4 values provided)
2 bus.device.function  (3 values provoided with domain = 0)
3 bus.device  (2 values provided with domain = 0, function = 0)

Change-Id: Ide03db38b4ac7802cf36f0e536e8b997101d6cd3
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-04-21 09:38:45 -07:00
Ziye Yang
e02d0bbd5f nvme: add the '=' separator support in spdk_nvme_transport_id_parse
Thus, we can support both ':' and '='.

Change-Id: I4400ba770912a2b20e49432c2e6e2a1bd45c28ed
Signed-off-by: Ziye Yang <ziye.yang@intel.com>
2017-04-21 09:38:45 -07:00
Dariusz Stojaczyk
154323ab51 scsi: Include null terminator in Unit Serial Number VPD Page
According to the scsi standard, all ASCII data fields "may be
terminated with one or more ASCII null (00h) characters"
[7.6.10, 4.4.1]. Windows SCSI Compliance tests expect a null terminator
there, so let's include it.

Change-Id: I18fa35295233a163cea711a5c4ff8e3d3e80c4f1
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
2017-04-21 08:12:04 -07:00
Piotr Pelplinski
c2c995c5b0 vhost: make SPDK compile with dpdk-17.05 with vhost patches
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Change-Id: Iff659bb3df9df92c4b6aa9aeaef7244bdf5b3211
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
2017-04-19 20:46:36 +02:00
Jim Harris
b5b9410da3 nvme: remove DELAY_BEFORE_ENABLE quirk and always wait 100us
A 100us is so small that applying the quirk to the specific
SSDs that require the delay is more trouble than it is worth.
So remove the quirk and always wait 100us before re-enabling
the NVMe SSD during initialization.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Id6a8cc6e35d103fffdf135580301fc3e5b27e722
2017-04-17 11:56:36 -07:00
Jim Harris
fd5d6901f2 nvme: use spdk_likely/unlikely macros in I/O path
Also avoid an spdk_get_ticks() call in the default
case where a timeout_cb_fn is not defined.

On my Intel(R) Xeon(R) E5-2699 v3 system with an
Intel(R) P3700 SSD, these modifications reduce software
overhead per I/O by 3-5% (as measured by the SPDK
overhead tool).

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I5bb5a87747b15d9e27655fabcd2bc1a40b0b990e
2017-04-17 11:56:36 -07:00
Daniel Verkamp
5742e9b9e7 nvme: allocate requests on a per-queue basis
Change-Id: I7bec816e518a0a6f2e9fb719128c83d4b908d46c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-12 21:05:24 -07:00
Daniel Verkamp
cd13f280f4 nvme: track qpair within nvme_request
Change-Id: Ia40a1e79db6327c4693731e9bb7a57810795429d
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
2017-04-12 21:05:23 -07:00
Ben Walker
15f910ece7 bdev/nvme: Improve names of fields in config file
Remove the "Nvme" from several field names. The parser
will still accept the old name for backward compatibility.

Change-Id: I6fa86ec359b23fb63960d0aa479a845b36a0977a
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-04-12 13:11:01 -07:00
Ben Walker
acd0b4573d bdev/nvme: Allow the user to control the I/O timeout behavior
The user can now not only specify an optional timeout for
commands, but also the action to take when a timeout is
detected.

Change-Id: I7d7cdd846d580e0b3a5f733d398ee9b19d6fe034
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-04-12 13:11:01 -07:00
Ben Walker
193f4f8392 nvme: Queue aborts beyond the abort command limit
Queue aborts that would exceed the abort command limit
in software as a convenience for the user.

Change-Id: I8c1f0380984cc6c0cdb453db961939a7f571b336
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-04-12 13:11:01 -07:00
Ben Walker
ee460db71f nvme: Don't pass admin qpair through timeout callback
Instead, pass NULL when an ADMIN command times out.
We don't expose the admin queue to the user.

Change-Id: If0768d329a689f6f7c3734c9d419e680d7378ed1
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-04-12 13:11:01 -07:00
Ben Walker
b9feeb03c1 nvme: Call the timeout callback only a single time
For each command that times out, call the timeout
callback one time if the user registered one.

Change-Id: Iaad39a886468e89bef63fe292c5cad1dce97a57c
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
2017-04-12 13:11:01 -07:00