ivampiresp/Spdk
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
13f97e6737
Spdk/test/unit/lib/bdev/crypto.c/crypto_ut.c

560 lines
19 KiB
C
Raw Normal View History

Replace most BSD 3-clause license text with SPDX identifier. Many open source projects have moved to using SPDX identifiers to specify license information, reducing the amount of boilerplate code in every source file. This patch replaces the bulk of SPDK .c, .cpp and Makefiles with the BSD-3-Clause identifier. Almost all of these files share the exact same license text, and this patch only modifies the files that contain the most common license text. There can be slight variations because the third clause contains company names - most say "Intel Corporation", but there are instances for Nvidia, Samsung, Eideticom and even "the copyright holder". Used a bash script to automate replacement of the license text with SPDX identifier which is checked into scripts/spdx.sh. Signed-off-by: Jim Harris <james.r.harris@intel.com> Change-Id: Iaa88ab5e92ea471691dc298cfe41ebfb5d169780 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12904 Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com> Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com> Reviewed-by: Changpeng Liu <changpeng.liu@intel.com> Reviewed-by: Dong Yi <dongx.yi@intel.com> Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: <qun.wan@intel.com>
2022-06-03 19:15:11 +00:00
/* SPDX-License-Identifier: BSD-3-Clause
update Intel copyright notices per Intel policy to include file commit date using git cmd below. The policy does not apply to non-Intel (C) notices. git log --follow -C90% --format=%ad --date default <file> | tail -1 and then pull just the 4 digit year from the result. Intel copyrights were not added to files where Intel either had no contribution ot the contribution lacked substance (ie license header updates, formatting changes, etc). Contribution date used "--follow -C95%" to get the most accurate date. Note that several files in this patch didn't end the license/(c) block with a blank comment line so these were added as the vast majority of files do have this last blank line. Simply there for consistency. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: Id5b7ce4f658fe87132f14139ead58d6e285c04d4 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15192 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> Community-CI: Mellanox Build Bot
2022-11-01 20:26:26 +00:00
* Copyright (C) 2018 Intel Corporation.
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
* Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES.
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
* All rights reserved.
*/
#include "spdk_cunit.h"
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
#include "common/lib/ut_multithread.c"
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
#include "spdk_internal/mock.h"
thread: Make the definition of struct spdk_io_channel private Move the definition of structure spdk_io_channel into lib/thread/thread_internal.h, so we don't have to update SO_VER for other libraries in future when we need to change the internal details on the structure. Signed-off-by: Jiewei Ke <jiewei@smartx.com> Change-Id: I3d2ca7a8737972e0b33ce92e464da42c48f89dec Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/8189 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ziye Yang <ziye.yang@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2021-06-05 12:50:15 +00:00
#include "thread/thread_internal.h"
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
#include "unit/lib/json_mock.c"
ut/crypto: remove internal rte_cryptodev headers Those were added to allow us mocking DPDK functions originaly defined as static inline. They are based on a fixed DPDK version and currently require a lot of effort to update for DPDK 19.02+. There's a different way of mocking them: > #define rte_crypto_op_bulk_alloc mock_rte_crypto_op_bulk_alloc > static inline unsigned mock_rte_crypto_op_bulk_alloc() { ... } This patch uses the above method to mock all static inline functions before including the crypto source file in crypto_ut. As a result we can get rid of the rte_ header copies from SPDK, which greatly reduces the effort required to make crypto_ut work with DPDK 19.02. Change-Id: I0f07a9ff4f1c63036e058dffd3fcf0c21e77bff3 Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Reviewed-on: https://review.gerrithub.io/c/443592 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>
2019-02-06 14:57:39 +00:00
#include <rte_crypto.h>
#include <rte_cryptodev.h>
bdev/crypto: do not track type of crypto session Starting with DPDK 22.11 the struct rte_cryptodev_sym_session is no longer part of public API. Instead the void * is used. There is no need for SPDK to track the type of session variable, so replace that with void * regardless of DPDK version. Change-Id: I29f82e87a593dd1886673fe2a56145da2dbe8354 Signed-off-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com> Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/15433 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-11-10 15:31:45 +00:00
#include <rte_version.h>
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
ut/crypto: reorder some functions In the next patch we'll e.g. move the vbdev_crypto.c include after some of those functions, so move them to the very top of the file now. Change-Id: I1abaf724ea10c8c6d461663ba3be0a5a37b34f4e Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Reviewed-on: https://review.gerrithub.io/c/443591 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>
2019-02-06 15:44:23 +00:00
unsigned ut_rte_crypto_op_bulk_alloc;
int ut_rte_crypto_op_attach_sym_session = 0;
module/crypto: load balance QAT qp assignment For the latest QAT devices there are 3 CPMs each with 16 VFs and 2 qp each VF. To load balance for multi-thread operations we want to assign each new queue pair (QP) on a per CPM (processing module) basis so this patch assigns the next QP, for QAT, by taking the last + 32 modulo the total number of QP. This will results in each new channel getting a QP on the next CPM. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: Iea608ada68517b6f2faecd45701c7aae6d23a2d0 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/477082 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>
2019-12-06 22:29:21 +00:00
#define MOCK_INFO_GET_1QP_AESNI 0
#define MOCK_INFO_GET_1QP_QAT 1
bdev/crypto: MLX5 AES_XTS general support - General MLX5 crypto support. - Unit-tests MLX5 crypto support. - Documentation update to list the MLX5 driver as supported, enumerate the cipher algorithms and provide some configuration hints. Signed-off-by: Yuriy Umanets <yumanets@nvidia.com> Change-Id: I0da1f49f4acd068d75a4d8633f84fe626d774431 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11630 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Community-CI: Mellanox Build Bot Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com>
2022-01-21 11:54:58 +00:00
#define MOCK_INFO_GET_1QP_MLX5 2
#define MOCK_INFO_GET_1QP_BOGUS_PMD 3
ut/crypto: reorder some functions In the next patch we'll e.g. move the vbdev_crypto.c include after some of those functions, so move them to the very top of the file now. Change-Id: I1abaf724ea10c8c6d461663ba3be0a5a37b34f4e Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Reviewed-on: https://review.gerrithub.io/c/443591 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>
2019-02-06 15:44:23 +00:00
int ut_rte_cryptodev_info_get = 0;
bool ut_rte_cryptodev_info_get_mocked = false;
ut/crypto: remove internal rte_cryptodev headers Those were added to allow us mocking DPDK functions originaly defined as static inline. They are based on a fixed DPDK version and currently require a lot of effort to update for DPDK 19.02+. There's a different way of mocking them: > #define rte_crypto_op_bulk_alloc mock_rte_crypto_op_bulk_alloc > static inline unsigned mock_rte_crypto_op_bulk_alloc() { ... } This patch uses the above method to mock all static inline functions before including the crypto source file in crypto_ut. As a result we can get rid of the rte_ header copies from SPDK, which greatly reduces the effort required to make crypto_ut work with DPDK 19.02. Change-Id: I0f07a9ff4f1c63036e058dffd3fcf0c21e77bff3 Signed-off-by: Darek Stojaczyk <dariusz.stojaczyk@intel.com> Reviewed-on: https://review.gerrithub.io/c/443592 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>
2019-02-06 14:57:39 +00:00
#include "bdev/crypto/vbdev_crypto.c"
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
/* SPDK stubs */
bdev/crypto: add IO queueing for out of mem condition via bdev layer Also made on the prints a DEBUG message instead and noticed the really name that was being registered by this component so updated it to make it look like the rest of SPDK. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I747a846cb365e7db49be50db941e83fb1b265ea0 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/460244 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-07-03 00:41:52 +00:00
DEFINE_STUB(spdk_bdev_queue_io_wait, int, (struct spdk_bdev *bdev, struct spdk_io_channel *ch,
struct spdk_bdev_io_wait_entry *entry), 0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
DEFINE_STUB_V(spdk_bdev_module_list_add, (struct spdk_bdev_module *bdev_module));
DEFINE_STUB_V(spdk_bdev_free_io, (struct spdk_bdev_io *g_bdev_io));
module/crypto: remove need to allocate a buffer for writes Use new bdev aux buf feature. Huge performance benefit for writes. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I5a27460a369ef5f13bf490a287603e566071be41 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478384 Community-CI: SPDK CI Jenkins <sys_sgci@intel.com> 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: Jim Harris <james.r.harris@intel.com>
2020-01-21 00:07:51 +00:00
DEFINE_STUB_V(spdk_bdev_io_put_aux_buf, (struct spdk_bdev_io *bdev_io, void *aux_buf));
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
DEFINE_STUB(spdk_bdev_io_type_supported, bool, (struct spdk_bdev *bdev,
enum spdk_bdev_io_type io_type), 0);
DEFINE_STUB_V(spdk_bdev_module_release_bdev, (struct spdk_bdev *bdev));
DEFINE_STUB_V(spdk_bdev_close, (struct spdk_bdev_desc *desc));
DEFINE_STUB(spdk_bdev_get_name, const char *, (const struct spdk_bdev *bdev), 0);
module/crypto: remove need to allocate a buffer for writes Use new bdev aux buf feature. Huge performance benefit for writes. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I5a27460a369ef5f13bf490a287603e566071be41 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478384 Community-CI: SPDK CI Jenkins <sys_sgci@intel.com> 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: Jim Harris <james.r.harris@intel.com>
2020-01-21 00:07:51 +00:00
DEFINE_STUB(spdk_bdev_get_buf_align, size_t, (const struct spdk_bdev *bdev), 64);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
DEFINE_STUB(spdk_bdev_get_io_channel, struct spdk_io_channel *, (struct spdk_bdev_desc *desc), 0);
DEFINE_STUB_V(spdk_bdev_unregister, (struct spdk_bdev *bdev, spdk_bdev_unregister_cb cb_fn,
void *cb_arg));
module/bdev: Use spdk_bdev_unregister_by_name() if possible Replace spdk_bdev_get_by_name() + spdk_bdev_unregister() by spdk_bdev_unregister_by_name() wherever possible. This simplifies the code and makes the code more reliable. Signed-off-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Change-Id: I91388c9d0b2e244cb745720a480803b03c42a226 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/12066 Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com> Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-03-29 05:55:53 +00:00
DEFINE_STUB(spdk_bdev_unregister_by_name, int, (const char *bdev_name,
struct spdk_bdev_module *module,
spdk_bdev_unregister_cb cb_fn, void *cb_arg), 0);
bdev/crypto: Open base bdev first using spdk_bdev_open_ext() vbdev_crypto_claim() gets bdev name instead of bdev pointer as a parameter, and open the corresponding base bdev first using spdk_bdev_open_ext(). The purpose is to fix the race condition due to the time gap between spdk_bdev_get_by_name() and spdk_bdev_open(). A bdev pointer is valid only while the bdev is opened. Resize event is not supported for now. Signed-off-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Change-Id: Ia0e1ce2ce696f431bb26af94729931c3ffb9a9d0 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/4588 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: Aleksey Marchuk <alexeymar@mellanox.com>
2020-10-10 11:29:25 +00:00
DEFINE_STUB(spdk_bdev_open_ext, int, (const char *bdev_name, bool write,
spdk_bdev_event_cb_t event_cb,
void *event_ctx, struct spdk_bdev_desc **_desc), 0);
DEFINE_STUB(spdk_bdev_desc_get_bdev, struct spdk_bdev *, (struct spdk_bdev_desc *desc), NULL);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
DEFINE_STUB(spdk_bdev_module_claim_bdev, int, (struct spdk_bdev *bdev, struct spdk_bdev_desc *desc,
struct spdk_bdev_module *module), 0);
DEFINE_STUB_V(spdk_bdev_module_examine_done, (struct spdk_bdev_module *module));
bdev: deprecate spdk_vbdev_register 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>
2019-04-03 12:59:05 +00:00
DEFINE_STUB(spdk_bdev_register, int, (struct spdk_bdev *vbdev), 0);
bdev/crypto: make sure that vbdev_crypto_destruct() returns 1 Make vbdev_crypto_destruct() return 1 to signal that program execution should wait for spdk_bdev_destruct_done() function, which is added inside _device_unregister_cb(). This change is related to _vdev_dev_get() not being able to find the devices, when called from _cryptodev_sym_session_free(), as it uses device driver name, which might already be freed. This occurs only during bdev module finish, when crypto bdevs are being unregistered and vbdev_crypto_finish() proceeds to call bdev name deletion without waiting for the unregister callbacks to complete, which ultimately results in reading freed pointers. This only happens when code execution takes path for DPDK 22.11+. Change-Id: Id9a43d07c90aef7a82867383fd77354ac521a3e7 Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com> Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16290 Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com> Reviewed-by: Konrad Sztyber <konrad.sztyber@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2023-01-13 10:53:54 +00:00
DEFINE_STUB_V(spdk_bdev_destruct_done, (struct spdk_bdev *bdev, int bdeverrno));
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_STUB(spdk_accel_crypto_key_destroy, int, (struct spdk_accel_crypto_key *key), 0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
/* global vars and setup/cleanup functions used for all test functions */
struct spdk_bdev_io *g_bdev_io;
struct crypto_bdev_io *g_io_ctx;
struct crypto_io_channel *g_crypto_ch;
struct spdk_io_channel *g_io_ch;
struct vbdev_crypto g_crypto_bdev;
bdev/crypto: Cleanup with crypto opts fields duplication - Fixed duplication of key, key2, drv_name, cipher, etc., fields in struct bdev_names and struct vbdev_crypto. Moved all of them into the new struct vbdev_crypto_opts, which is re-used by both structs. This aslo removes duplication in error handling and fininalization logic that checks the keys are zeroed out and properly freed. - Moved unhexlify into vbdev rpc code. All keys passed to vbdev already in the binary form. - Provide meaningful error messages in the rpc response on keys validation issues during setup of crypto vbdev. - Updated unit tests. Signed-off-by: Yuriy Umanets <yumanets@nvidia.com> Change-Id: I1fab8771bbbc0cd2f359f0d105fec28fb86893b3 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11631 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-01-20 08:25:05 +00:00
struct vbdev_crypto_opts g_crypto_bdev_opts;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
void
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
spdk_bdev_io_get_aux_buf(struct spdk_bdev_io *bdev_io, spdk_bdev_io_get_aux_buf_cb cb)
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
cb(g_io_ch, g_bdev_io, (void *)0xDEADBEEF);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
void
spdk_bdev_io_get_buf(struct spdk_bdev_io *bdev_io, spdk_bdev_io_get_buf_cb cb, uint64_t len)
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
cb(g_io_ch, g_bdev_io, true);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
struct ut_vbdev_crypto_bdev_cpl_args {
spdk_bdev_io_completion_cb cb_fn;
struct spdk_bdev_io *bdev_io;
void *cb_arg;
bool result;
};
static void
_ut_vbdev_crypto_bdev_cpl(void *arg)
module/crypto: remove need to allocate a buffer for writes Use new bdev aux buf feature. Huge performance benefit for writes. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I5a27460a369ef5f13bf490a287603e566071be41 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478384 Community-CI: SPDK CI Jenkins <sys_sgci@intel.com> 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: Jim Harris <james.r.harris@intel.com>
2020-01-21 00:07:51 +00:00
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
struct ut_vbdev_crypto_bdev_cpl_args *cpl_args = arg;
cpl_args->cb_fn(cpl_args->bdev_io, cpl_args->result, cpl_args->cb_arg);
free(cpl_args);
module/crypto: remove need to allocate a buffer for writes Use new bdev aux buf feature. Huge performance benefit for writes. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I5a27460a369ef5f13bf490a287603e566071be41 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478384 Community-CI: SPDK CI Jenkins <sys_sgci@intel.com> 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: Jim Harris <james.r.harris@intel.com>
2020-01-21 00:07:51 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
static void
ut_vbdev_crypto_bdev_cpl(spdk_bdev_io_completion_cb cb_fn, struct spdk_bdev_io *bdev_io,
bool result, void *cb_arg)
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
struct ut_vbdev_crypto_bdev_cpl_args *cpl_args = calloc(1, sizeof(*cpl_args));
SPDK_CU_ASSERT_FATAL(cpl_args);
cpl_args->cb_fn = cb_fn;
cpl_args->bdev_io = bdev_io;
cpl_args->result = result;
cpl_args->cb_arg = cb_arg;
spdk_thread_send_msg(spdk_get_thread(), _ut_vbdev_crypto_bdev_cpl, cpl_args);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
/* Mock these functions to call the callback and then return the value we require */
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_RETURN_MOCK(spdk_bdev_readv_blocks, int);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
int
spdk_bdev_readv_blocks(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
struct iovec *iov, int iovcnt,
uint64_t offset_blocks, uint64_t num_blocks,
spdk_bdev_io_completion_cb cb, void *cb_arg)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
HANDLE_RETURN_MOCK(spdk_bdev_readv_blocks);
ut_vbdev_crypto_bdev_cpl(cb, g_bdev_io, !ut_spdk_bdev_readv_blocks, cb_arg);
return 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_RETURN_MOCK(spdk_bdev_writev_blocks, int);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
int
spdk_bdev_writev_blocks(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
struct iovec *iov, int iovcnt,
uint64_t offset_blocks, uint64_t num_blocks,
spdk_bdev_io_completion_cb cb, void *cb_arg)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
HANDLE_RETURN_MOCK(spdk_bdev_writev_blocks);
ut_vbdev_crypto_bdev_cpl(cb, g_bdev_io, !ut_spdk_bdev_writev_blocks, cb_arg);
return 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_RETURN_MOCK(spdk_bdev_unmap_blocks, int);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
int
spdk_bdev_unmap_blocks(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
uint64_t offset_blocks, uint64_t num_blocks,
spdk_bdev_io_completion_cb cb, void *cb_arg)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
HANDLE_RETURN_MOCK(spdk_bdev_unmap_blocks);
ut_vbdev_crypto_bdev_cpl(cb, g_bdev_io, !ut_spdk_bdev_unmap_blocks, cb_arg);
return 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_RETURN_MOCK(spdk_bdev_flush_blocks, int);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
int
spdk_bdev_flush_blocks(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
uint64_t offset_blocks, uint64_t num_blocks, spdk_bdev_io_completion_cb cb,
void *cb_arg)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
HANDLE_RETURN_MOCK(spdk_bdev_flush_blocks);
ut_vbdev_crypto_bdev_cpl(cb, g_bdev_io, !ut_spdk_bdev_flush_blocks, cb_arg);
return 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
DEFINE_RETURN_MOCK(spdk_bdev_reset, int);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
int
spdk_bdev_reset(struct spdk_bdev_desc *desc, struct spdk_io_channel *ch,
spdk_bdev_io_completion_cb cb, void *cb_arg)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
HANDLE_RETURN_MOCK(spdk_bdev_reset);
ut_vbdev_crypto_bdev_cpl(cb, g_bdev_io, !ut_spdk_bdev_reset, cb_arg);
return 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
bool g_completion_called = false;
void
spdk_bdev_io_complete(struct spdk_bdev_io *bdev_io, enum spdk_bdev_io_status status)
{
bdev_io->internal.status = status;
g_completion_called = true;
}
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
struct ut_vbdev_crypto_accel_cpl_args {
spdk_accel_completion_cb cb_fn;
void *cb_arg;
int rc;
};
static void
_vbdev_crypto_ut_accel_cpl(void *arg)
{
struct ut_vbdev_crypto_accel_cpl_args *cpl_args = arg;
cpl_args->cb_fn(cpl_args->cb_arg, cpl_args->rc);
free(cpl_args);
}
static void
vbdev_crypto_ut_accel_cpl(spdk_accel_completion_cb cb_fn, void *cb_arg, int rc)
{
struct ut_vbdev_crypto_accel_cpl_args *cpl_args = calloc(1, sizeof(*cpl_args));
SPDK_CU_ASSERT_FATAL(cpl_args);
cpl_args->cb_fn = cb_fn;
cpl_args->cb_arg = cb_arg;
cpl_args->rc = rc;
spdk_thread_send_msg(spdk_get_thread(), _vbdev_crypto_ut_accel_cpl, cpl_args);
}
DEFINE_RETURN_MOCK(spdk_accel_submit_encrypt, int);
int ut_spdk_accel_submit_encrypt_cb_rc = 0;
int
spdk_accel_submit_encrypt(struct spdk_io_channel *ch, struct spdk_accel_crypto_key *key,
struct iovec *dst_iovs, uint32_t dst_iovcnt,
struct iovec *src_iovs, uint32_t src_iovcnt,
uint64_t iv, uint32_t block_size, int flags,
spdk_accel_completion_cb cb_fn, void *cb_arg)
{
HANDLE_RETURN_MOCK(spdk_accel_submit_encrypt);
/* We must not call cb_fn immediately */
vbdev_crypto_ut_accel_cpl(cb_fn, cb_arg, ut_spdk_accel_submit_encrypt_cb_rc);
return 0;
}
int ut_spdk_accel_submit_decrypt_cb_rc;
DEFINE_RETURN_MOCK(spdk_accel_submit_decrypt, int);
int
spdk_accel_submit_decrypt(struct spdk_io_channel *ch, struct spdk_accel_crypto_key *key,
struct iovec *dst_iovs, uint32_t dst_iovcnt,
struct iovec *src_iovs, uint32_t src_iovcnt,
uint64_t iv, uint32_t block_size, int flags,
spdk_accel_completion_cb cb_fn, void *cb_arg)
{
HANDLE_RETURN_MOCK(spdk_accel_submit_decrypt);
/* We must not call cb_fn immediately */
vbdev_crypto_ut_accel_cpl(cb_fn, cb_arg, ut_spdk_accel_submit_decrypt_cb_rc);
return 0;
}
struct spdk_io_channel *spdk_accel_get_io_channel(void)
{
return (struct spdk_io_channel *)0xfeedbeef;
}
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
/* Global setup for all tests that share a bunch of preparation... */
static int
test_setup(void)
{
/* Prepare essential variables for test routines */
g_bdev_io = calloc(1, sizeof(struct spdk_bdev_io) + sizeof(struct crypto_bdev_io));
g_bdev_io->u.bdev.iovs = calloc(1, sizeof(struct iovec) * 128);
g_bdev_io->bdev = &g_crypto_bdev.crypto_bdev;
g_io_ch = calloc(1, sizeof(struct spdk_io_channel) + sizeof(struct crypto_io_channel));
thread: Make the definition of struct spdk_io_channel private Move the definition of structure spdk_io_channel into lib/thread/thread_internal.h, so we don't have to update SO_VER for other libraries in future when we need to change the internal details on the structure. Signed-off-by: Jiewei Ke <jiewei@smartx.com> Change-Id: I3d2ca7a8737972e0b33ce92e464da42c48f89dec Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/8189 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Ziye Yang <ziye.yang@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2021-06-05 12:50:15 +00:00
g_crypto_ch = (struct crypto_io_channel *)spdk_io_channel_get_ctx(g_io_ch);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_io_ctx = (struct crypto_bdev_io *)g_bdev_io->driver_ctx;
memset(&g_crypto_bdev, 0, sizeof(struct vbdev_crypto));
bdev/crypto: Cleanup with crypto opts fields duplication - Fixed duplication of key, key2, drv_name, cipher, etc., fields in struct bdev_names and struct vbdev_crypto. Moved all of them into the new struct vbdev_crypto_opts, which is re-used by both structs. This aslo removes duplication in error handling and fininalization logic that checks the keys are zeroed out and properly freed. - Moved unhexlify into vbdev rpc code. All keys passed to vbdev already in the binary form. - Provide meaningful error messages in the rpc response on keys validation issues during setup of crypto vbdev. - Updated unit tests. Signed-off-by: Yuriy Umanets <yumanets@nvidia.com> Change-Id: I1fab8771bbbc0cd2f359f0d105fec28fb86893b3 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11631 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-01-20 08:25:05 +00:00
memset(&g_crypto_bdev_opts, 0, sizeof(struct vbdev_crypto_opts));
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
g_crypto_bdev.crypto_bdev.blocklen = 512;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_io_ctx->crypto_ch = g_crypto_ch;
g_io_ctx->crypto_bdev = &g_crypto_bdev;
bdev/crypto: Cleanup with crypto opts fields duplication - Fixed duplication of key, key2, drv_name, cipher, etc., fields in struct bdev_names and struct vbdev_crypto. Moved all of them into the new struct vbdev_crypto_opts, which is re-used by both structs. This aslo removes duplication in error handling and fininalization logic that checks the keys are zeroed out and properly freed. - Moved unhexlify into vbdev rpc code. All keys passed to vbdev already in the binary form. - Provide meaningful error messages in the rpc response on keys validation issues during setup of crypto vbdev. - Updated unit tests. Signed-off-by: Yuriy Umanets <yumanets@nvidia.com> Change-Id: I1fab8771bbbc0cd2f359f0d105fec28fb86893b3 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11631 Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-01-20 08:25:05 +00:00
g_io_ctx->crypto_bdev->opts = &g_crypto_bdev_opts;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
TAILQ_INIT(&g_crypto_ch->in_accel_fw);
modules/crypto/compress: changes to support DPDK mbuf changes Need to manage how we store IO context based on DPDK updates made in 19.11. Fixes issue #1671 Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: If1183808cd30987b6c999912f563949b7ade7fcb Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/5799 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Tomasz Zawadzki <tomasz.zawadzki@intel.com> Community-CI: Mellanox Build Bot
2021-01-06 21:49:42 +00:00
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
return 0;
}
/* Global teardown for all tests */
static int
test_cleanup(void)
{
free(g_bdev_io->u.bdev.iovs);
free(g_bdev_io);
free(g_io_ch);
return 0;
}
static void
test_error_paths(void)
{
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
g_crypto_bdev.crypto_bdev.blocklen = 512;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->u.bdev.iovcnt = 1;
g_bdev_io->u.bdev.num_blocks = 1;
g_bdev_io->u.bdev.iovs[0].iov_len = 512;
bdev/crypto: Switched to pkt_mbuf API - Switched to using rte_mempool for mbufs instead of spdk_mempool. This allows using rte pkt_mbuf API that properly handles mbuf fields we need for mlx5 and we don't have to do it manually when sending crypto ops. - Using rte_mempool *g_mbuf_mp in vbdev crypto ut and added the mocking API code. - crypto_ut update to follow pkt_mbuf API rules. Signed-off-by: Yuriy Umanets <yumanets@nvidia.com> Change-Id: Ia5576c672ac2eebb260bfdbb528ddb9edcd8f036 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/11623 Community-CI: Broadcom CI <spdk-ci.pdl@broadcom.com> Community-CI: Mellanox Build Bot Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Shuhei Matsumoto <smatsumoto@nvidia.com> Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
2022-01-21 08:02:42 +00:00
g_bdev_io->u.bdev.iovs[0].iov_base = (void *)0xDEADBEEF;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
/* test error returned by accel fw */
MOCK_SET(spdk_accel_submit_encrypt, -ENOMEM);
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
CU_ASSERT(g_io_ctx->bdev_io_wait.bdev == &g_crypto_bdev.crypto_bdev);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_fn == vbdev_crypto_resubmit_io);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_arg == g_bdev_io);
CU_ASSERT(g_io_ctx->resubmit_state == CRYPTO_IO_NEW);
memset(&g_io_ctx->bdev_io_wait, 0, sizeof(g_io_ctx->bdev_io_wait));
MOCK_SET(spdk_accel_submit_encrypt, -EINVAL);
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
MOCK_CLEAR(spdk_accel_submit_encrypt);
/* Test error returned in accel cpl cb */
ut_spdk_accel_submit_encrypt_cb_rc = -EINVAL;
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
poll_threads();
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
ut_spdk_accel_submit_encrypt_cb_rc = 0;
/* Test error returned from bdev */
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_writev_blocks, -ENOMEM);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
bdev/crypto: add IO queueing for out of mem condition via bdev layer Also made on the prints a DEBUG message instead and noticed the really name that was being registered by this component so updated it to make it look like the rest of SPDK. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I747a846cb365e7db49be50db941e83fb1b265ea0 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/460244 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2019-07-03 00:41:52 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
CU_ASSERT(g_io_ctx->bdev_io_wait.bdev == &g_crypto_bdev.crypto_bdev);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_fn == vbdev_crypto_resubmit_io);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_arg == g_bdev_io);
CU_ASSERT(g_io_ctx->resubmit_state == CRYPTO_IO_ENCRYPT_DONE);
memset(&g_io_ctx->bdev_io_wait, 0, sizeof(g_io_ctx->bdev_io_wait));
MOCK_CLEAR(spdk_bdev_readv_blocks);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_writev_blocks, -EINVAL);
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
poll_threads();
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
MOCK_CLEAR(spdk_bdev_writev_blocks);
/* Test error returned in bdev cpl */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
ut_spdk_bdev_writev_blocks = -EINVAL;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
ut_spdk_bdev_writev_blocks = 0;
/* the same for read path */
/* Test error returned from bdev */
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->type = SPDK_BDEV_IO_TYPE_READ;
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_readv_blocks, -ENOMEM);
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
CU_ASSERT(g_io_ctx->bdev_io_wait.bdev == &g_crypto_bdev.crypto_bdev);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_fn == vbdev_crypto_resubmit_io);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_arg == g_bdev_io);
CU_ASSERT(g_io_ctx->resubmit_state == CRYPTO_IO_NEW);
memset(&g_io_ctx->bdev_io_wait, 0, sizeof(g_io_ctx->bdev_io_wait));
MOCK_CLEAR(spdk_bdev_readv_blocks);
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
MOCK_SET(spdk_bdev_readv_blocks, -EINVAL);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_CLEAR(spdk_bdev_readv_blocks);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
/* Test error returned in bdev cpl */
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
ut_spdk_bdev_readv_blocks = -EINVAL;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
ut_spdk_bdev_readv_blocks = 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
/* test error returned by accel fw */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
MOCK_SET(spdk_accel_submit_decrypt, -ENOMEM);
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
poll_threads();
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
CU_ASSERT(g_io_ctx->bdev_io_wait.bdev == &g_crypto_bdev.crypto_bdev);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_fn == vbdev_crypto_resubmit_io);
CU_ASSERT(g_io_ctx->bdev_io_wait.cb_arg == g_bdev_io);
CU_ASSERT(g_io_ctx->resubmit_state == CRYPTO_IO_READ_DONE);
memset(&g_io_ctx->bdev_io_wait, 0, sizeof(g_io_ctx->bdev_io_wait));
MOCK_CLEAR(spdk_accel_submit_decrypt);
/* test error returned in accel cpl */
ut_spdk_accel_submit_decrypt_cb_rc = -EINVAL;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
ut_spdk_accel_submit_decrypt_cb_rc = 0;
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
static void
test_simple_write(void)
{
/* Single element block size write */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->u.bdev.iovcnt = 1;
g_bdev_io->u.bdev.num_blocks = 1;
g_bdev_io->u.bdev.offset_blocks = 0;
g_bdev_io->u.bdev.iovs[0].iov_len = 512;
g_bdev_io->u.bdev.iovs[0].iov_base = &test_simple_write;
g_crypto_bdev.crypto_bdev.blocklen = 512;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE;
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
/* 1st poll to trigger accel completions, 2nd for bdev */
poll_threads();
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
module/crypto: remove need to allocate a buffer for writes Use new bdev aux buf feature. Huge performance benefit for writes. Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I5a27460a369ef5f13bf490a287603e566071be41 Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/478384 Community-CI: SPDK CI Jenkins <sys_sgci@intel.com> 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: Jim Harris <james.r.harris@intel.com>
2020-01-21 00:07:51 +00:00
CU_ASSERT(g_io_ctx->aux_buf_iov.iov_len == 512);
CU_ASSERT(g_io_ctx->aux_buf_iov.iov_base != NULL);
CU_ASSERT(g_io_ctx->aux_offset_blocks == 0);
CU_ASSERT(g_io_ctx->aux_num_blocks == 1);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
}
static void
test_simple_read(void)
{
/* Single element block size read */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->u.bdev.iovcnt = 1;
g_bdev_io->u.bdev.num_blocks = 1;
g_bdev_io->u.bdev.iovs[0].iov_len = 512;
g_bdev_io->u.bdev.iovs[0].iov_base = &test_simple_read;
g_crypto_bdev.crypto_bdev.blocklen = 512;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_READ;
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
/* 1st poll to trigger dev completions, 2nd for accel */
poll_threads();
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
}
static void
test_passthru(void)
{
/* Make sure these follow our completion callback, test success & fail. */
g_bdev_io->type = SPDK_BDEV_IO_TYPE_UNMAP;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_CLEAR(spdk_bdev_unmap_blocks);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_unmap_blocks, -EINVAL);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
MOCK_CLEAR(spdk_bdev_unmap_blocks);
g_bdev_io->type = SPDK_BDEV_IO_TYPE_FLUSH;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_CLEAR(spdk_bdev_flush_blocks);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_flush_blocks, -EINVAL);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
MOCK_CLEAR(spdk_bdev_flush_blocks);
/* We should never get a WZ command, we report that we don't support it. */
g_bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE_ZEROES;
vbdev_crypto_submit_request(g_io_ch, g_bdev_io);
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
}
crypto: add proper reset handling Previously there was no consideration for IO that were outstanding to the crypto device when handling a reset. This patch makes sure that those IO are completed with FAIL status prior to completing the reset that we pass down the stack. It does so by sending down the reset first and in the completion using spdk_for_each_channel and the poller to quiesce each channel allowing the crypto side to complete all IOs before we finally complete the reset IO after the last channel is quiesced. Resets are tracked on a per bdev basis. Addresses github issue #449. Change-Id: Iadb07bada1fcaad33d9f224a60d983a7eb835236 Signed-off-by: paul luse <paul.e.luse@intel.com> Signed-off-by: Ben Walker <benjamin.walker@intel.com> Reviewed-on: https://review.gerrithub.io/428552 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Seth Howell <seth.howell5141@gmail.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-10-09 20:38:29 +00:00
static void
test_reset(void)
{
/* TODO: There are a few different ways to do this given that
* the code uses spdk_for_each_channel() to implement reset
module/crypto: fix some typos & clean up comments Signed-off-by: paul luse <paul.e.luse@intel.com> Change-Id: I21b6992d2a3a914e44c61d489633b5c3741e70cb Reviewed-on: https://review.gerrithub.io/c/spdk/spdk/+/471690 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: Tomasz Zawadzki <tomasz.zawadzki@intel.com>
2019-10-17 22:15:42 +00:00
* handling. Submitting w/o UT for this function for now and
crypto: add proper reset handling Previously there was no consideration for IO that were outstanding to the crypto device when handling a reset. This patch makes sure that those IO are completed with FAIL status prior to completing the reset that we pass down the stack. It does so by sending down the reset first and in the completion using spdk_for_each_channel and the poller to quiesce each channel allowing the crypto side to complete all IOs before we finally complete the reset IO after the last channel is quiesced. Resets are tracked on a per bdev basis. Addresses github issue #449. Change-Id: Iadb07bada1fcaad33d9f224a60d983a7eb835236 Signed-off-by: paul luse <paul.e.luse@intel.com> Signed-off-by: Ben Walker <benjamin.walker@intel.com> Reviewed-on: https://review.gerrithub.io/428552 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Seth Howell <seth.howell5141@gmail.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
2018-10-09 20:38:29 +00:00
* will follow up with something shortly.
*/
}
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
static void
test_crypto_op_complete(void)
{
/* Make sure completion code respects failure. */
g_completion_called = false;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
_crypto_operation_complete(g_bdev_io, -1);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
CU_ASSERT(g_completion_called == true);
/* Test read completion. */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_READ;
g_completion_called = false;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
_crypto_operation_complete(g_bdev_io, 0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
CU_ASSERT(g_completion_called == true);
/* Test write completion success. */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE;
g_completion_called = false;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_CLEAR(spdk_bdev_writev_blocks);
_crypto_operation_complete(g_bdev_io, 0);
poll_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_SUCCESS);
CU_ASSERT(g_completion_called == true);
/* Test write completion failed. */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_WRITE;
g_completion_called = false;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_SET(spdk_bdev_writev_blocks, -EINVAL);
_crypto_operation_complete(g_bdev_io, 0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
CU_ASSERT(g_completion_called == true);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
MOCK_CLEAR(spdk_bdev_writev_blocks);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
/* Test bogus type for this completion. */
g_bdev_io->internal.status = SPDK_BDEV_IO_STATUS_SUCCESS;
g_bdev_io->type = SPDK_BDEV_IO_TYPE_RESET;
g_completion_called = false;
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
_crypto_operation_complete(g_bdev_io, 0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_ASSERT(g_bdev_io->internal.status == SPDK_BDEV_IO_STATUS_FAILED);
CU_ASSERT(g_completion_called == true);
}
static void
test_supported_io(void)
{
void *ctx = NULL;
bool rc = true;
/* Make sure we always report false to WZ, we need the bdev layer to
* send real 0's so we can encrypt/decrypt them.
*/
rc = vbdev_crypto_io_type_supported(ctx, SPDK_BDEV_IO_TYPE_WRITE_ZEROES);
CU_ASSERT(rc == false);
}
int
main(int argc, char **argv)
{
CU_pSuite suite = NULL;
unsigned int num_failures;
ut: simplify error handling at initialization 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>
2020-03-11 17:59:24 +00:00
CU_set_error_action(CUEA_ABORT);
CU_initialize_registry();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
suite = CU_add_suite("crypto", test_setup, test_cleanup);
ut: make use of CUnit macro to reduce duplications 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>
2020-03-11 19:15:39 +00:00
CU_ADD_TEST(suite, test_error_paths);
CU_ADD_TEST(suite, test_simple_write);
CU_ADD_TEST(suite, test_simple_read);
CU_ADD_TEST(suite, test_passthru);
CU_ADD_TEST(suite, test_crypto_op_complete);
CU_ADD_TEST(suite, test_supported_io);
CU_ADD_TEST(suite, test_reset);
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
allocate_threads(1);
set_thread(0);
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
CU_basic_set_mode(CU_BRM_VERBOSE);
CU_basic_run_tests();
bdev/crypto: Use accel framework All DPDK related code is removed, handling of RESET command was sligthly updated. Handling of -ENOMEM was updated for cases when accel API returns -ENOMEM Crypto tests in blockdev.sh were extended with more crypto_bdevs to verify NOMEM cases - that failed with original vbdev_crypto implementation Signed-off-by: Alexey Marchuk <alexeymar@nvidia.com> Change-Id: If1feba2449bee852c6c4daca4b3406414db6fded Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/14860 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Paul Luse <paul.e.luse@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com>
2022-08-31 15:26:12 +00:00
free_threads();
bdev: Add crypto virtual bdev module Initial support for softare AESNI_MB DPDK driver only. Have tested (both aesni and QAT seprately and concurrrently) on underlying NVMe devices with bdevio and a bdevperf script that runs IOs from 512B to 128K each with Q depths from 1 to 512 in powers of 2 for 30 seconds each run. QAT can be included in the code (but not makefile) and marked as experimental until we are ready to test in CI. It works well on 2 systems but is a big PITA to get the hardware setup and configured for use with DPDK (IMHO). Change-Id: If518c3df8e74e00efa18afdf194824c5e69778fc Signed-off-by: Paul Luse <paul.e.luse@intel.com> Reviewed-on: https://review.gerrithub.io/403107 Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com>
2018-03-07 23:44:06 +00:00
num_failures = CU_get_number_of_failures();
CU_cleanup_registry();
return num_failures;
}
Reference in New Issue Copy Permalink