This is a preparation to abort split I/Os. To abort any I/O by
bio_cb_arg, they have to be managed by the submitted I/O list.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I9459d9d2d3511aad0325dcc20d88610444a4ea04
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2231
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add spdk_bdev_abort function as a new public API.
This goes all the way down to the bdev driver module and attempts to
abort all I/Os which has bio_cb_arg as its callback argument.
We can separate when only a single I/O has bio_cb_arg and when multiple
I/Os have bio_cb_arg, but unify both by using parent - children I/O
relationship. To avoid confusion, return matched_ios by _bdev_abort() and
store it into split_outstanding by the caller.
Exclude any I/O submitted after this abort command because the same cb_arg
may be used by all I/Os and abort may never complete.
bdev_io needs to have both bio_cb_arg and bio_to_abort because bio_cb_arg
is used to continue abort processing when it is stopped due to the capacity
of bdev_io pool, and bio_to_abort is used to pass it to the underlying
bdev module at submission. Parent I/O is not submitted directly, and is
only used in the generic bdev layer, and parent I/O's bdev_io uses bio_cb_arg.
Hence add bio_cb_arg to bdev structure and add bio_to_abort to abort structure.
In the meantime of abort operation, target I/Os may be completed. Hence
check if the target I/O still exists at completion, and set the completion
status to false only if it still exists.
Upon completion of this, i.e., this returned zero, the status
SPDK_BDEV_IO_STATUS_SUCCESS indicates all I/Os were successfully aborted,
or the status SPDK_BDEV_IO_STATUS_FAILED indicates any I/O was failed to
abort by any reason.
spdk_bdev_abort() does not support aborting abort or reset request
due to the complexity for now.
Following patches will support I/O split case.
Add unit tests together to cover the basic paths.
Besides, ABI compatibility check required us to bump up SO version of
a few libraries or modules. Bump up SO version of blob bdev module simply
because it does not have any out-of-tree consumer, and suppress bumping
up SO version of lvol library because the affected struct spdk_lvol
is not part of public APIs.
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I515da688503557615d491bf0bfb36322ce37df08
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/2014
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Mellanox Build Bot
Community-CI: Broadcom CI
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
CUnit provides a helper macro CU_ADD_TEST() that
simplifies usage of CU_add_test() function.
Test name no longer needs to be provided,
as it will be inherited from function name.
This is a follow up to
https://review.spdk.io/gerrit/c/spdk/spdk/+/931
Change-Id: I8078f02e08b14f12328ae022d7090ba13fbd64e4
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1239
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
This patch changes the way that unittests initialize.
The new way is shorter and simpler.
It assumes that error during initialization is a fatal error,
but in our cases it always is.
This patch is a followp up to
https://review.spdk.io/gerrit/c/spdk/spdk/+/930
lvol_ut.c is the only test that skipped because
it runs same tests multiple times which is not allowed
by new method.
Change-Id: I0baf7ad09a35d5fca2dc4a03b4453c12c14f38a7
Signed-off-by: Vitaliy Mysak <vitaliy.mysak@intel.com>
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/1238
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
These are unit tests for non emulated version of
spdk_bdev_comparev_blocks function.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I459ac614e1a03e42b9d584aaa907a1448a13d16e
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478969
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Upcoming patch will use most of the comparev test code
so let's reuse it.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Ibc6230d38da224bfcf25f76fe1d834d48200bfd9
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/479832
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This patch adds test for spdk_bdev_comparev_blocks
function. For now it tests only emulated compare
operation. For native compare call will be created
separate patch.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Icf2297575f59ab4ee46ba1b40eff6729b3e95b1a
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478948
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
We will only support a vectored variant of
compare-and-write for now.
This does no locking for now. Ii will be added
in a separate patch.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I5bd075c912de60090e19cf8fced19c4879fcc900
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475941
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
We can't allow overlapped locked ranges - otherwise
two different channels could be deadlocked.
So add a pending_locked_ranges to the bdev. When we
start a lock operation, check if the new range overlaps
one that's already locked. If so, put it on the pending
list. When an unlock operation completes, we will
check if any pending ranges can now be locked.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2e3113216a195887b954533495ff200df14fadc1
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478537
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Add an io_locked TAILQ to each channel, which hold
IO that will write to a currently locked region.
Also add a new step to the locking process per channel.
Each channel needs to wait until all existing outstanding
writes to the newly locked range have been completed.
Only the channel that locked an LBA range may submit
write I/O to it. It must use the same cb_arg for the
write I/O as the cb_arg used when locking the LBA range.
This ensures that only the specific I/O operations needing
the lock will bypass the lock.
When a range is unlocked, we will just blindly try to
resubmit all IO in the io_locked tailq. This could be
made more efficient in the future, but we don't expect
this path to occur very often, so going for simplicity
in the first pass.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ibdc992144dfaffe7c05471a5b3c020cedd8cdfc3
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478226
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This adds new internal APIs bdev_lock_lba_range and
bdev_unlock_lba_range. To start, these APIs will
manage dissemination of lock/unlock requests to all
existing channels for a given bdev. This does not
yet interact at all with any I/O sent to the channel.
Future patches will check new I/O to see if they
are trying to write to a range that is locked. Future
patches will also ensure we do not have overlapping
ranges active at once.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I6d5b1cc84b41a7adc2a3c5791c766bb77376581f
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478225
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This will be used by upcoming patches for implementing
LBA range locks.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ifa4ad8dcc0d09ccf20d35f010fcae19dcc17abc9
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478224
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Rename stub_submit_request_aligned_buffer function
to stub_submit_request_get_buf and its callback function
to stub_submit_request_get_buf_cb.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: Ie8cde43d30c5aa2d353bb33b023d5318948bee44
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478944
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
These buffers weren't actually used, but "buf" is an array,
so we should just pass buf, not &buf.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I9d73a73b138d2bd9c14172c58b1825cf83532b4f
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478119
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add the unit test to verify the submitted list,
timeout IO poller and timeout IO check.
Change-Id: Ie7832cc29e19a624328a1a96cffe313bf9f18a69
Signed-off-by: Jin Yu <jin.yu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/476051
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>
Community-CI: SPDK CI Jenkins <sys_sgci@intel.com>
The _ was supposed to indicate an internal function,
but really leaving off the spdk in the function
is the standard way to denote an internal function.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I74facafb67b793502838b9c1b5f90aec2c88c69b
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475033
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
We try to reserve the spdk_ prefix for functions that
are part of the public SPDK API.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I0a95ba0f9db04b588a2e948427f8a9f53a7b5740
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/475032
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Alexey Marchuk <alexeymar@mellanox.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
When splitting bdev's, if we run out of child IOVs, we have to
make sure that we end block aligned. That may require us to
shorten or eliminate one or more child IOVs to be picked up
on the next split. If we eliminate enough such that there
are no IOVs for this split IO, just continue and the next
split completion will kick off another split to pick up
the remaining data to be transferred.
Fixes issue #981
Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: If2bfbe5f0862295e1d74cbea00692890a2178967
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471313
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Community-CI: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
When we ran out of child_iov space, ensure the iovs to be aligned
with block size. However the calculation was wrong.
(to_next_boundary_bytes % blocklen) meant not to_last_block but to_next_block.
So calculate to_last_block_size by reducing to_last_block_size from blocklen.
The data was collected when the issue occured. So add unit test
by using the data.
Fixes#979
Reported-by: Geoffrey McRae <geoff@hostfission.com>
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Change-Id: I62a50bada450288ea7c60aec0e557c2a53cd8916
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/470806
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Next commits in the series will need these callbacks in other tests
that are closer to the beginning of file.
Signed-off-by: Evgeniy Kochetov <evgeniik@mellanox.com>
Signed-off-by: Sasha Kotchubievsky <sashakot@mellanox.com>
Signed-off-by: Alexey Marchuk <alexeymar@mellanox.com>
Change-Id: I55e346f273f169c7894f662dcfa1f57a02024577
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/469708
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
This patch adds new interface for opening bdev and
implements new style remove event. With that changes
user can be notified about different types of events
that occur in regards to bdev. spdk_bdev_open_ext
function uses bdev name as an argument instead of bdev
structure to remove race condition where user gets
the bdev structure and bdev is removed after getting
that structure and before open function is called.
spdk_bdev_open is now deprecated.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I44ebeb988bc6a2f441fc6a0c38a30668aad999ad
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/455647
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Scan-build on Fedora29/30 reports error, which this patch fixes:
bdev_ut.c:1436:17: warning: The left operand of '!=' is a garbage value
CU_ASSERT(desc != NULL);
~~~~ ^
/usr/include/CUnit/CUnit.h:123:30: note: expanded from macro 'CU_ASSERT'
{ CU_assertImplementation((value), LINE, #value, FILE, "", CU_FALSE); }
^~~~~
bdev_ut.c:1654:17: warning: The left operand of '!=' is a garbage value
CU_ASSERT(desc != NULL);
~~~~ ^
/usr/include/CUnit/CUnit.h:123:30: note: expanded from macro 'CU_ASSERT'
{ CU_assertImplementation((value), LINE, #value, FILE, "", CU_FALSE); }
Fixes#924
Change-Id: Ia7dc6ce066ac80fc45f022eefd43a224d2f11503
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/466110
Reviewed-by: Broadcom SPDK FC-NVMe CI <spdk-ci.pdl@broadcom.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
bdev_ut.c:1300:17: warning: The left operand of '!=' is a garbage value
CU_ASSERT(desc != NULL);
~~~~ ^
Change-Id: Ie87cd4d7218a380e61616a83f3a9bceccc0798cb
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/463258
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
bdev_ut.c:1863:2: warning: Potential leak of memory pointed to by
'histogram'
poll_threads();
^~~~~~~~~~~~~~
We used to free g_histogram instead of histogram. Those
should be the same thing, but scan-build gets confused.
Change the code to free histogram - this should also make
it slightly easier to read.
Change-Id: Ifaad1c2d9c7f9cc1a106f6edf3d22b54e42fe867
Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/463256
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This fixes error on scan-build on Fedora 30, which assumes spdk_bdev_open()
is called from non-SPDK thread - failing before allocating spdk_bdev_desc:
bdev_ut.c:1766:17: warning: The left operand of '!=' is a garbage value
CU_ASSERT(desc != NULL);
This is related to issue #822.
Change-Id: I8ee2112bb4d71aafc93d63bddca083c009ec11f3
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/459755
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Here is the an example to describe existing issue:
There is a Write request with 64KiB data length, and this IO is cross the IO
boundary. We assume that the parent IO will have 2 children requests, one is
33KiB length, the other one is 31KiB. Here is the view of parent iovs, the
first 33KiB length data has 33 iovs:
iov.[0].iov_length = 1024;
.
.
iov.[31].iov_length = 256;
iov.[32].iov_length = 768;
.
.
iov.[64].iov_length = 1024;
In function _spdk_bdev_io_split(), then you can see that for the 33KiB length
child request, exiting code will run out of child child_iov space and return
error due to last one data buffer is not block size aligned.
Here we can rewind the existing offset to last block size aligned buffer to
avoid the error case, for backend which need aligned data buffer such as
AIO backend, the request will go through spdk_bdev_io_get_buf() again to
do the data copy, otherwise for those backend devices such as NVMe with
hardware SGL support, 256 data segment is fine for them.
Change-Id: I96ebdf29829d86f9b38fab28a7406eedc9fa44ef
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/453604
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>
If bdev is configured to use separate buffers for metadata transfer,
but it doesn't support the write zeroes command, pass empty zeroed
buffer to make sure the metadata is cleared out as well.
Change-Id: If6f024266067e5764a28a276296f651d31da4792
Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/457628
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>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Existing code in spdk_bdev_write_zeroes_blocks() will call spdk_bdev_free_io()
for the error case, which will cause assertion because the bdev_io isn't
submitted to the backend yet, so we will check the condtion first to
avoid the error case.
Change-Id: If27d78217f709a3315e74c00869d345abd6b9a69
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/453491
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
When parent IO was splitted into several children requests, SPDK
may return parent completion callback with error status before
all the children requests are finished.
Change-Id: I63221a0ae1a5925a7fcd9744b4f5d8079c641252
Signed-off-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/453611
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>
Test multi vector command that needs to be split by strip and then
needs to be split further due to the capacity of child iovs.
Add a case that was not tested before. In this case, the length of the
rest of iovec array with an I/O boundary is the multiple of block size.
Expect the rest of iovec array to be submitted in the completion of
previous iovec array.
Change-Id: I5b95b1f1884a73b31709b2fd9187a8a9e9b2cd0b
Signed-off-by: lorneli <lorneli@163.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/452414
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
This patch changes UT so that bdev modules have async init.
It demonstrates issue with triggering assert in init path,
when bdev modules are reinitialized multiple times.
Change-Id: I9b2b16d8ac53bac6a929f6929ceedb70b250c500
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/454618
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>
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>
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>
This API had good intentions, but as more complicated
use cases came up where base bdevs could come and go,
we've realized that the bdev layer will need another
mechanism to query bdev modules on these types of
relationships between a virtual bdev and its base
bdevs. We removed all code related to tracking
the array of base bdevs a long time ago.
Change all existing callers to use spdk_bdev_register.
Document spdk_vbdev_register as deprecated for now,
and change its implementation to just call
spdk_bdev_register for simplicity sake.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I3b40ed96480c0fa7184db42953a9f4e4c167fed1
Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/450076
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
When the specified buffer size to spdk_bdev_io_get_buf() is greater
than the permitted maximum, spdk_bdev_io_get_buf() asserts simply and
doesn't call the specified callback function.
SPDK SCSI library doesn't allocate read buffer and specifies
expected read buffer size, and expects that it is allocated by
spdk_bdev_io_get_buf().
Bdev perf tool also doesn't allocate read buffer and specifies
expected read buffer size, and expects that it is allocated by
spdk_bdev_io_get_buf().
When we support DIF insert and strip in iSCSI target, the read
buffer size iSCSI initiator requests and the read buffer size iSCSI target
requests will become different.
Even after that, iSCSI initiator and iSCSI target will negotiate correctly
not to cause buffer overflow in spdk_bdev_io_get_buf(), but if iSCSI
initiator ignores the result of negotiation, iSCSI initiator can request
read buffer size larger than the permitted maximum, and can cause
failure in iSCSI target. This is very flagile and should be avoided.
This patch do the following
- Add the completion status of spdk_bdev_io_get_buf() to
spdk_bdev_io_get_buf_cb(),
- spdk_bdev_io_get_buf() calls spdk_bdev_io_get_buf_cb() by setting
success to false, and return.
- spdk_bdev_io_get_buf_cb() in each bdev module calls assert if success
is false.
Subsequent patches will process the case that success is false
in spdk_bdev_io_get_buf_cb().
Change-Id: I76429a86e18a69aa085a353ac94743296d270b82
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/c/446045
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ziye Yang <ziye.yang@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Currently, the SPDK_BDEV_REGISTER_MODULE() macro uses __LINE__
to generate functions like spdk_bdev_module_register_187().
Typically, this is not a problem as these functions are not called directly
rather, they are only used as constructor functions to load the bdevs during
system startup.
There are languages however, (e.g rust) that require these functions to be
referenced explicitly to prevent them from being removed during the linking phase.
In order to reference them, having the names predictable (and potentially
changed per commit) makes things easier.
Change-Id: I15947ed9136912cfe2368db7e5bba833f1d94b15
Signed-off-by: gila <jeffry.molanus@gmail.com>
Reviewed-on: https://review.gerrithub.io/c/443536
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>
Since division is more expensive than right shift operator, in
function spdk_bdev_bytes_to_blocks, use right shift instead of
division if the blocklen of bdev is a power of two.
Change-Id: Ib3dbc792e86582bba30b3dc028efbd12c69075ba
Signed-off-by: lorneli <lorneli@163.com>
Reviewed-on: https://review.gerrithub.io/438318
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This is the callback type used for message passing. The
old name is easy to confuse with the callback to pass
a message to a thread or the upcoming callback to
spawn a thread.
Change-Id: I5fd63b57c4be2a4262a197850e6de4901be03ee7
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/435941
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: Jim Harris <james.r.harris@intel.com>
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: Idcfb328a26fc85f3c144da9c27086a71c00d0d46
Reviewed-on: https://review.gerrithub.io/427830
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Large read I/O will be typical in some use cases such as
web stream services. On the other hand, large write I/O
may not be typical but will be sufficiently probable.
Currently when large I/O is submitted to the RAID bdev,
the I/O will be divided by the strip size of it and then
divided I/Os are submitted sequentially.
This patch tries to improve the performance of the RAID bdev
in large I/Os. Besides, when the RAID bdev supports higher
levels of RAID (such as RAID5), it should issue multiple
I/Os to multiple base bdevs by batch fasion in the parity
update. Having experience in batched I/O will be helpful
in the future case too.
In this patch, submit split I/Os by batch until all child IOVs
are consumed or all data are submitted. If all child IOVs are
consumed before all data are submitted, wait until all batched
split I/Os complete and then submit again.
In this patch, test code is added too.
Change-Id: If6cd81cc0c306e3875a93c39dbe4288723b78937
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/424770
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Splitting IO in the bdev layer is a little complex and increasing
the coverage in unit tests will be good.
Change-Id: I9d42cedc795b2804a472bdc310e936135c424fd3
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/427948
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Expected values are managed by linked list now. Hence we can set
up all expected values before starting tested split I/Os.
This change will make the logic clearer.
Change-Id: I9226619468220132aebd55a5da36490a145c128a
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/427947
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: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Each file that need to check SPDK_CONFIG_* options need to include
spdk/config.h explicitly.
Change-Id: If9f2a91ac4c2b1a300dcf88ec3e2a12714ad344a
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/427221
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Current test code works only for sequential split I/Os. The next
patch will support submitting batchedsplit I/Os.
Prepare test code as a separate patch to make review easier.
Change-Id: I05b732c9a656ecab74a2594bf50ddf42eb41584e
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/425877
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
When a bdev IO is split, if iovec size in a strip is more than 32,
the IO will fail.
Remove the limitation by spliting the split IO further.
Change-Id: I962ad86dfe63ea1fcd86ffa52ead7452fb80e53d
Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-on: https://review.gerrithub.io/425876
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com>
This was added a long time back for tracking an rte_mbuf
whose buffer was a different rte_mbuf - all related to
a userspace TCP stack that is no longer in development.
The concept isn't useful now, so remove it to reduce
the complexity of the tracing code.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I310e492eba7f55df242bb29d82fb19f6daee1f51
Reviewed-on: https://review.gerrithub.io/424565
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
This is a string name used for debugging only.
Change-Id: I9827f0e6c83be7bc13951c7b5f0951ce6c2a1ece
Signed-off-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-on: https://review.gerrithub.io/424127
Reviewed-by: Jim Harris <james.r.harris@intel.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>
Splitting a 1TB unmap into individual 64KB unmap commands
(for a RAID volume with 64KB strip size) would be awful -
the RAID module can be much smarter about this.
So back out the changes for splitting I/O without payload.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I24fe6d911f4e3c9db4b2cb5d66c7236a5596e0d9
Reviewed-on: https://review.gerrithub.io/424103
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Fixes a use after free error in bdev/bdev_ut
Fixes an uninitialized argument value error in mt/bdev/bdev_ut
Adds extra checks in the spdk_bdev_part_construct code
Fixes a dereference of null pointer error in bdev/vbdev_lvol_ut
Change-Id: I1eaa6ebfec0a36a0d910504c70b16ccc651de22a
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/423923
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>
A number of modules (RAID, logical volumes) have logical
"stripes" that require splitting an I/O into several
child I/O. For example, on a RAID-0 with 128KB strip size,
an I/O that spans a 128KB boundary will require sending
one I/O for the portion that comes before the boundary to
one member disk, and another I/O for the portion that comes
after the boundary to another member disk. Logical volumes
are similar - data is allocated in clusters, so an I/O that
spans a cluster boundary may need to be split since the
clusters may not be contiguous on disk.
Putting the splitting logic in the common bdev layer ensures
bdev module authors don't have to always do this themselves.
This is especially helpful for cases like splitting an I/O
described by many iovs - we can simplify this a lot by
handling it in the common bdev layer.
Note that currently we will only submit one child I/O
at a time. This could be improved later to submit multiple
child I/O in parallel, but the complexity in the iov splitting
code also increases a lot.
Note: Some Intel NVMe SSDs have a similar characteristic.
We will not use this bdev stripe feature for NVMe though -
we want to primarily use the splitting functionality inside
of the NVMe driver itself to ensure it remains fully
functional. Many SPDK users use the NVMe driver without
the bdev layer.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ife804ecc56f6b2b55345a0d0ae9fda9e68632b3b
Reviewed-on: https://review.gerrithub.io/423024
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Paul Luse <paul.e.luse@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
A new unit test in an upcoming patch will initialize the bdev
layer separately, so modify the existing tests now so that
they properly cleanup the bdev layer after the unit test
that initialized it is complete.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ie00837a90391ca4d0dd578e978fb26483fe5c0cf
Reviewed-on: https://review.gerrithub.io/423023
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: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
So we don't need to allocate memory (maybe failed) just for free other
memory.
Change-Id: I2c83f6acc2aa6ed79455bff90f952a2e70b44d59
Signed-off-by: wuzhouhui <wuzhouhui@kingsoft.com>
Reviewed-on: https://review.gerrithub.io/422203
Chandler-Test-Pool: 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>
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>
The intents of these arrays was to keep track in the
bdev layer of all base<->virtual bdev relationships -
i.e. which member disk bdevs make up a RAID bdev,
which logical volume bdevs are associated with a
bdev that contains an lvolstore, etc.
Currently none of this is used however. And trying
to keep track in the bdev layer instead of asking
the bdev modules for the relationships has a number
of complications. Early one, we tried to do this
with TAILQs - but that doesn't work since this can't
be done with a single TAILQ_ENTRY in the bdev
structures. So we moved to arrays - that works a bit
better, but then the pointer arrays have to be
realloc'd which isn't ideal.
The biggest problem though with these arrays is that
they held bdev pointers - not bdev descriptor pointers.
It's not really valid to access bdevs without a
descriptor - the descriptors are what make sure active
references are accounted for when a bdev is hotplugged.
Of course the bdev layer knows when a bdev is getting
removed and could go and do the updates to these
arrays separately - but that just seems very convoluted.
So for now just remove these arrays completely. If
there is a future need for the bdev layer to
understand relationships between bdevs, we can add
module APIs so that the generic layer can ask
the modules about the relationships.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I99ef1068240bff1262f64f234260cf2fb44df51d
Reviewed-on: https://review.gerrithub.io/420932
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: GangCao <gang.cao@intel.com>
During spdk_bdev_init, examine_config is called.
This call can claim bdev synchronously, based on
configuration. On spdk_bdev_start if none module
claimed bdev, examine_disk is called and can
perform I/O before claiming bdev.
Signed-off-by: Piotr Pelplinski <piotr.pelplinski@intel.com>
Change-Id: I1448dd368cf3a24a5daccab387d7af7c3d231127
Reviewed-on: https://review.gerrithub.io/413913
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Maciej Szwed <maciej.szwed@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Specifically, the errors that occur in scan-build that uses clang-6.0 as
a backend.
Change-Id: I3922a4f65134cbd866fcb27ad9b7a75b6b7803bb
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/418229
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
This function is intended to be used when an spdk_bdev
I/O operation (such as spdk_bdev_write or spdk_bdev_write_blocks)
fails due to spdk_bdev_io buffer exhaustion. The caller
can queue an spdk_bdev_io_wait structure on the calling thread
which will be invoked when an spdk_bdev_io buffer is available.
While here, turn off error messages in bdev.c related to
spdk_bdev_io pool exhaustion, since we now have an API designed
to gracefully recover from it.
Also modify bdevperf as an example of how to use this new API.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ia55f6582dc5a8d6d8bcc74689fd846d742324510
Reviewed-on: https://review.gerrithub.io/415074
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Add a new function and its RPC caller. By using it, we can
get the statistics of all the bdevs or the specified bdev.
Meanwhile, with this patch, the open source tool 'sysstat/iostat'
can support for SPDK. The 'iostat' tool can call this function to
get the statistics of all the SPDK managed devices via the rpc
interface.
Change-Id: I135a7bbd49d923014bdf93720f78dd5a588d7afa
Signed-off-by: Yanbo Zhou <yanbo.zhou@intel.com>
Reviewed-on: https://review.gerrithub.io/393130
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
This thread is needed in any context when we send a message.
Change-Id: Ifcb5e3bf2e6ee60385b425a976ed380228570ae6
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/406427
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
For application like vhost/iSCSI target/NVMe-oF target/etc,
a new section as below can be added for the QoS rate limiting.
[QoS]
Limit_IOPS Malloc0 100000
Limit_IOPS Nvme0n1 500000
Also added a sample change at test/lib/bdev/bdev.conf.in
Change-Id: I7c7d951fbe1352ca2571f135c657bc4fa43b56c7
Signed-off-by: GangCao <gang.cao@intel.com>
Reviewed-on: https://review.gerrithub.io/393221
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
SPKD base bdev might be part of multiple vbdevs. The same is true in
reverse direction. So consider folowing scenario:
bdev3 bdev4 bdev5
| | |
+-+--+ + +--+--+
/ \ | / \
bdev0 bdev1 bdev2
In current implementation bdev0/1/2 will apear as base base for
bdev3/4/5 which is obviously wrong.
This patch try to address this issue.
Change-Id: Ic99c13c8656ceb597aba7e41ccb2fa8090b4f13b
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/405104
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
This file is accessed by files in multiple directories under test.
Change-Id: I634481fb58eab5c097aaece5289f88e531954fcb
Signed-off-by: Seth Howell <seth.howell@intel.com>
Reviewed-on: https://review.gerrithub.io/404976
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>
Write information needed to recreate each bdev.
Change-Id: I3d2b24fd4aaa8b98ec558d864cc28b55a899452f
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/401217
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
This better matches the style in the rest of SPDK.
No functional change - this is a pure find/replace of
spdk_bdev_module_if to spdk_bdev_module. Instances of this struct will
be renamed in another patch.
Change-Id: I3f6933c8a366e625fc3a1b6401aee26ee03ba69c
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/403368
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Currently SPDK_BDEV_MODULE_REGISTER() take many parameters. Extending it
(eg for incoming JSON configuration dump/load) is quite challenging and
error prone. As we are already here in next patches, rework this macro
to take one parameter - the pointer to struct spdk_bdev_module_if.
This patch also remove following macros:
SPDK_GET_BDEV_MODULE - this is not really needed, to find module outside
module translation unit use spdk_bdev_module_list_find()
SPDK_BDEV_MODULE_ASYNC_INIT and SPDK_BDEV_MODULE_ASYNC_FINI - replaced
by bool fields in spdk_bdev_module_if struct.
Change-Id: Ief88e023fbbaee7d5402c838dbecbdffd4dfb259
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Reviewed-on: https://review.gerrithub.io/402883
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>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I4b0704021bc8e9ccbc6ed6ce7ac8eb1efad4a561
Reviewed-on: https://review.gerrithub.io/399728
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Add api and unit test functions for
change number of blocks for provided block device.
Change-Id: I55d67c99375cb88bdaa79ce1a36d4298223beddc
Signed-off-by: Slawomir Mrozowicz <slawomirx.mrozowicz@intel.com>
Reviewed-on: https://review.gerrithub.io/390802
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>
spdk_vbdev_unregister() is part of internal bdev API,
yet bdev module that uses spdk_vbdev_register() directly
will not be removed correctly when using delete_bdev RPC.
spdk_vbdev_unregister() is now consolidated with
spdk_bdev_unregister().
This comes up when deleting lvol bdev, as it does not use
spdk_bdev_part_* functions.
base_bdev->vbdevs entry was not removed for bdev that lvs
is created on.
Additionally patch expands test to create lvol bdev,
after removing it using delete_bdev RPC.
With ASAN enabled this would report accessing
already freed memory previously.
Change-Id: I9547e83862e2daa50355d56a1c9f453aaa6cfdb8
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Reviewed-on: https://review.gerrithub.io/395711
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>
Added aliases list to bdev struct.
Added 2 API calls to add and remove aliases.
Added test for adding and removing aliases.
Change-Id: I1815aec8c02cfa398b2d1de41577197315665fdc
Signed-off-by: Sebastian Basierski <sebastianx.basierski@intel.com>
Reviewed-on: https://review.gerrithub.io/390200
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>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change the return type of spdk_bdev_register related
functions and try to handle the duplicated name
issue.
Change-Id: I23af11583cf2050579d1624508306a35394bffde
Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-on: https://review.gerrithub.io/388178
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>
First this change moves spdk_subsystem_fini() to trigger on
spdk_app_stop(). This ensures that spdk_subsystem_fini() is called
before reactors are stopped in spdk_reactors_stop().
Finish paths for subsystems, bdevs and copy engine is now
asynchronous.
Each of those three mentioned have to make sure they are
asynchronous as well.
Only bdev that currently has requirement for asynchronous finish
are logical volume.
Thus the change in vbdev_lvol.c making it move to next bdev module
only after all lvol stores were unloaded.
Fio_plugin finish of bdev and copy_engine was removed for now.
Next patch in series adds it back with async support.
Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
Change-Id: I80ee2d084f3d82c50bf1329e08996604ae61b1b3
Reviewed-on: https://review.gerrithub.io/381536
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>
Currently deleting bdev does not support asynchronous delete
operations. Because of that results are returned before device
is actually deleted and some operation can be peformed on that
device after removal of this device started.
Signed-off-by: Maciej Szwed <maciej.szwed@intel.com>
Change-Id: I305c302d8abd5d7c2c0f947fca70c58396872132
Reviewed-on: https://review.gerrithub.io/383732
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>
Also add a unit test that reproduces the original issue and passes
with the one line fix.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I120d42ba7b6cfa4a2cb11e6cc08885d95316fe81
Reviewed-on: https://review.gerrithub.io/380703
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Changpeng Liu <changpeng.liu@intel.com>
A later patch in this series fails the test pool because
scan-build says we might be using garbage for one of the
descriptor pointer comparisons. That later patch does not
touch this unit test file at all, and the bdev-related changes
do not touch the code path used around the bug report. But
the report is valid - it's possible that with the additional
bdev.c changes that scan-build's ability to detect this
bug changed for the better.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I2bf165fe4d91a13d481010176a0339b591b16014
Reviewed-on: https://review.gerrithub.io/378825
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>
Add new versions of all of the I/O calls that take parameters in blocks
instead of bytes. These are intended to replace the old APIs, but
we'll keep them for now to preserve compatibility.
Change-Id: I85ab665c653e8c697016c628837d49aa0c3bfcd0
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/376255
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: John Kariuki <John.K.Kariuki@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Change-Id: I33506d6b9ff09c45c057326f7339d742eebc45b4
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/372861
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I58fa306781fd2cd5cd2d63809cf900b03c132edc
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/372860
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Patch afe860ae deferred freeing the io_device. However, for nvme, the
io_device context (spdk_nvme_ctrlr) is still being destructed before
io_channels are destroyed, causing segfaults on hotremove.
This patch defers io_device context destruction and fixes nvme
hotremove.
Fixes: afe860aeb1 ("channel: Correctly defer unregisters if channels exist")
Fixes: 5533c3d208 ("util: defer put_io_channel")
Change-Id: I7af699174cac0c6c6a6faa2cc65418c47347eb9a
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Reviewed-on: https://review.gerrithub.io/370459
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>
We still will sort the bdev_module list so that modules
with an examine() callback are initialized first. This ensures
they have a chance to initialize before later modules start
registering physical block devices.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I792cfb41b0abe030fe2486a2c872cbf329735932
Reviewed-on: https://review.gerrithub.io/369486
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>
vbdev modules still open/close bdevs as normal, but
should open bdevs read only when tasting (i.e. reading
the GPT to see if there are SPDK partitions). When
a vbdev module is ready to claim the bdev for purposes
of creating virtual bdevs on top of it, it calls
spdk_vbdev_module_claim_bdev(). It can pass its
open descriptor as well to have it promoted to
write access (required for future vbdev modules like
logical volumes).
Note: error vbdev was changed to copy the base bdev
parameters one-by-one instead of a blind memcpy - we
do not want to copy the base bdev's vbdev_claim_module
into the new bdev!
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: If2ee67dc78daf96050343c473671aa3402991bb1
Reviewed-on: https://review.gerrithub.io/368628
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>
Change-Id: Iaed5e670bef4ef5e588c9df4858f4d1ff240df58
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-on: https://review.gerrithub.io/369273
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Reviewed-by: <roman.sudarikov@intel.com>
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Add a new spdk_vbdev_module_examine_done() API which
vbdev modules can use to notify the generic bdev layer
once a base bdev examination is complete.
This is especially required for asynchronous vbdevs
like GPT which must issue I/O to the base bdev.
As part of this patch, add examine callbacks
for both split and error, which for now only call
this new functions. Later patches will move code from
the init callback to the examine callback for these
modules. examine callbacks are now required for all
vbdev modules.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I49f2d012d1675b878bcd23afff427c740c6502c7
Reviewed-on: https://review.gerrithub.io/368831
Tested-by: SPDK Automated Test System <sys_sgsw@intel.com>
Reviewed-by: Daniel Verkamp <daniel.verkamp@intel.com>
Reviewed-by: Ziye Yang <optimistyzy@gmail.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic3051b63942770e45be22af0ae03a78a7c543f81
Reviewed-on: https://review.gerrithub.io/368597
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>
Retire the old claim/unclaim semantics in favor of
open/close. Clients must now open a bdev to get
an spdk_bdev_desc, then pass this desc to get an
I/O channel.
This allows multiple clients to open a bdev,
although only one may open a bdev with write
access.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I4d319f1278170124169a8a75fd791e926b3f7171
Reviewed-on: https://review.gerrithub.io/367611
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>
No actual tests in this patch - just getting the
framework in place including necessary stub
functions to get the bdev.c unit test file to link.
Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: Ic54001de814740a2049c7bbc6846d98c5a62a8b3
Reviewed-on: https://review.gerrithub.io/367625
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>
