xref: /dpdk/doc/guides/cryptodevs/qat.rst (revision 97b914f4e715565d53d38ac6e04815b9be5e58a9)
1..  SPDX-License-Identifier: BSD-3-Clause
2    Copyright(c) 2015-2019 Intel Corporation.
3
4Intel(R) QuickAssist (QAT) Crypto Poll Mode Driver
5==================================================
6
7QAT documentation consists of three parts:
8
9* Details of the symmetric and asymmetric crypto services below.
10* Details of the :doc:`compression service <../compressdevs/qat_comp>`
11  in the compressdev drivers section.
12* Details of building the common QAT infrastructure and the PMDs to support the
13  above services. See :ref:`building_qat` below.
14
15
16Symmetric Crypto Service on QAT
17-------------------------------
18
19The QAT symmetric crypto PMD (hereafter referred to as `QAT SYM [PMD]`) provides
20poll mode crypto driver support for the following hardware accelerator devices:
21
22* ``Intel QuickAssist Technology DH895xCC``
23* ``Intel QuickAssist Technology C62x``
24* ``Intel QuickAssist Technology C3xxx``
25* ``Intel QuickAssist Technology 200xx``
26* ``Intel QuickAssist Technology D15xx``
27* ``Intel QuickAssist Technology C4xxx``
28* ``Intel QuickAssist Technology 4xxx``
29
30
31Features
32~~~~~~~~
33
34The QAT SYM PMD has support for:
35
36Cipher algorithms:
37
38* ``RTE_CRYPTO_CIPHER_3DES_CBC``
39* ``RTE_CRYPTO_CIPHER_3DES_CTR``
40* ``RTE_CRYPTO_CIPHER_AES128_CBC``
41* ``RTE_CRYPTO_CIPHER_AES192_CBC``
42* ``RTE_CRYPTO_CIPHER_AES256_CBC``
43* ``RTE_CRYPTO_CIPHER_AES128_CTR``
44* ``RTE_CRYPTO_CIPHER_AES192_CTR``
45* ``RTE_CRYPTO_CIPHER_AES256_CTR``
46* ``RTE_CRYPTO_CIPHER_AES_XTS``
47* ``RTE_CRYPTO_CIPHER_SNOW3G_UEA2``
48* ``RTE_CRYPTO_CIPHER_NULL``
49* ``RTE_CRYPTO_CIPHER_KASUMI_F8``
50* ``RTE_CRYPTO_CIPHER_DES_CBC``
51* ``RTE_CRYPTO_CIPHER_AES_DOCSISBPI``
52* ``RTE_CRYPTO_CIPHER_DES_DOCSISBPI``
53* ``RTE_CRYPTO_CIPHER_ZUC_EEA3``
54
55Hash algorithms:
56
57* ``RTE_CRYPTO_AUTH_SHA1``
58* ``RTE_CRYPTO_AUTH_SHA1_HMAC``
59* ``RTE_CRYPTO_AUTH_SHA224``
60* ``RTE_CRYPTO_AUTH_SHA224_HMAC``
61* ``RTE_CRYPTO_AUTH_SHA256``
62* ``RTE_CRYPTO_AUTH_SHA256_HMAC``
63* ``RTE_CRYPTO_AUTH_SHA384``
64* ``RTE_CRYPTO_AUTH_SHA384_HMAC``
65* ``RTE_CRYPTO_AUTH_SHA512``
66* ``RTE_CRYPTO_AUTH_SHA512_HMAC``
67* ``RTE_CRYPTO_AUTH_AES_XCBC_MAC``
68* ``RTE_CRYPTO_AUTH_SNOW3G_UIA2``
69* ``RTE_CRYPTO_AUTH_MD5_HMAC``
70* ``RTE_CRYPTO_AUTH_NULL``
71* ``RTE_CRYPTO_AUTH_KASUMI_F9``
72* ``RTE_CRYPTO_AUTH_AES_GMAC``
73* ``RTE_CRYPTO_AUTH_ZUC_EIA3``
74* ``RTE_CRYPTO_AUTH_AES_CMAC``
75
76Supported AEAD algorithms:
77
78* ``RTE_CRYPTO_AEAD_AES_GCM``
79* ``RTE_CRYPTO_AEAD_AES_CCM``
80* ``RTE_CRYPTO_AEAD_CHACHA20_POLY1305``
81
82Protocol offloads:
83
84* ``RTE_SECURITY_PROTOCOL_DOCSIS``
85
86Supported Chains
87~~~~~~~~~~~~~~~~
88
89All the usual chains are supported and also some mixed chains:
90
91.. table:: Supported hash-cipher chains for wireless digest-encrypted cases
92
93   +------------------+-----------+-------------+----------+----------+
94   | Cipher algorithm | NULL AUTH | SNOW3G UIA2 | ZUC EIA3 | AES CMAC |
95   +==================+===========+=============+==========+==========+
96   | NULL CIPHER      | Y         | 2&3         | 2&3      | Y        |
97   +------------------+-----------+-------------+----------+----------+
98   | SNOW3G UEA2      | 2&3       | 1&2&3       | 2&3      | 2&3      |
99   +------------------+-----------+-------------+----------+----------+
100   | ZUC EEA3         | 2&3       | 2&3         | 2&3      | 2&3      |
101   +------------------+-----------+-------------+----------+----------+
102   | AES CTR          | 1&2&3     | 2&3         | 2&3      | Y        |
103   +------------------+-----------+-------------+----------+----------+
104
105* The combinations marked as "Y" are supported on all QAT hardware versions.
106* The combinations marked as "2&3" are supported on GEN2 and GEN3 QAT hardware only.
107* The combinations marked as "1&2&3" are supported on GEN1, GEN2 and GEN3 QAT hardware only.
108
109
110Limitations
111~~~~~~~~~~~
112
113* Only supports the session-oriented API implementation (session-less APIs are not supported).
114* SNOW 3G (UEA2), KASUMI (F8) and ZUC (EEA3) supported only if cipher length and offset fields are byte-multiple.
115* SNOW 3G (UIA2) and ZUC (EIA3) supported only if hash length and offset fields are byte-multiple.
116* No BSD support as BSD QAT kernel driver not available.
117* ZUC EEA3/EIA3 is not supported by dh895xcc devices
118* Maximum additional authenticated data (AAD) for GCM is 240 bytes long and must be passed to the device in a buffer rounded up to the nearest block-size multiple (x16) and padded with zeros.
119* Queue-pairs are thread-safe on Intel CPUs but Queues are not (that is, within a single
120  queue-pair all enqueues to the TX queue must be done from one thread and all dequeues
121  from the RX queue must be done from one thread, but enqueues and dequeues may be done
122  in different threads.)
123* A GCM limitation exists, but only in the case where there are multiple
124  generations of QAT devices on a single platform.
125  To optimise performance, the GCM crypto session should be initialised for the
126  device generation to which the ops will be enqueued. Specifically if a GCM
127  session is initialised on a GEN2 device, but then attached to an op enqueued
128  to a GEN3 device, it will work but cannot take advantage of hardware
129  optimisations in the GEN3 device. And if a GCM session is initialised on a
130  GEN3 device, then attached to an op sent to a GEN1/GEN2 device, it will not be
131  enqueued to the device and will be marked as failed. The simplest way to
132  mitigate this is to use the PCI allowlist to avoid mixing devices of different
133  generations in the same process if planning to use for GCM.
134* The mixed algo feature on GEN2 is not supported by all kernel drivers. Check
135  the notes under the Available Kernel Drivers table below for specific details.
136* Out-of-place is not supported for combined Crypto-CRC DOCSIS security
137  protocol.
138* ``RTE_CRYPTO_CIPHER_DES_DOCSISBPI`` is not supported for combined Crypto-CRC
139  DOCSIS security protocol.
140* Multi-segment buffers are not supported for combined Crypto-CRC DOCSIS
141  security protocol.
142
143Extra notes on KASUMI F9
144~~~~~~~~~~~~~~~~~~~~~~~~
145
146When using KASUMI F9 authentication algorithm, the input buffer must be
147constructed according to the
148`3GPP KASUMI specification <http://cryptome.org/3gpp/35201-900.pdf>`_
149(section 4.4, page 13). The input buffer has to have COUNT (4 bytes),
150FRESH (4 bytes), MESSAGE and DIRECTION (1 bit) concatenated. After the DIRECTION
151bit, a single '1' bit is appended, followed by between 0 and 7 '0' bits, so that
152the total length of the buffer is multiple of 8 bits. Note that the actual
153message can be any length, specified in bits.
154
155Once this buffer is passed this way, when creating the crypto operation,
156length of data to authenticate "op.sym.auth.data.length" must be the length
157of all the items described above, including the padding at the end.
158Also, offset of data to authenticate "op.sym.auth.data.offset"
159must be such that points at the start of the COUNT bytes.
160
161Asymmetric Crypto Service on QAT
162--------------------------------
163
164The QAT asymmetric crypto PMD (hereafter referred to as `QAT ASYM [PMD]`) provides
165poll mode crypto driver support for the following hardware accelerator devices:
166
167* ``Intel QuickAssist Technology DH895xCC``
168* ``Intel QuickAssist Technology C62x``
169* ``Intel QuickAssist Technology C3xxx``
170* ``Intel QuickAssist Technology D15xx``
171* ``Intel QuickAssist Technology C4xxx``
172
173The QAT ASYM PMD has support for:
174
175* ``RTE_CRYPTO_ASYM_XFORM_MODEX``
176* ``RTE_CRYPTO_ASYM_XFORM_MODINV``
177* ``RTE_CRYPTO_ASYM_XFORM_RSA``
178* ``RTE_CRYPTO_ASYM_XFORM_ECDSA``
179* ``RTE_CRYPTO_ASYM_XFORM_ECPM``
180
181Limitations
182~~~~~~~~~~~
183
184* Big integers longer than 4096 bits are not supported.
185* Queue-pairs are thread-safe on Intel CPUs but Queues are not (that is, within a single
186  queue-pair all enqueues to the TX queue must be done from one thread and all dequeues
187  from the RX queue must be done from one thread, but enqueues and dequeues may be done
188  in different threads.)
189* RSA-2560, RSA-3584 are not supported
190
191.. _building_qat:
192
193Building PMDs on QAT
194--------------------
195
196A QAT device can host multiple acceleration services:
197
198* symmetric cryptography
199* data compression
200* asymmetric cryptography
201
202These services are provided to DPDK applications via PMDs which register to
203implement the corresponding cryptodev and compressdev APIs. The PMDs use
204common QAT driver code which manages the QAT PCI device. They also depend on a
205QAT kernel driver being installed on the platform, see :ref:`qat_kernel` below.
206
207
208Configuring and Building the DPDK QAT PMDs
209~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
210
211
212Further information on configuring, building and installing DPDK is described
213:doc:`here <../linux_gsg/build_dpdk>`.
214
215.. _building_qat_config:
216
217Build Configuration
218~~~~~~~~~~~~~~~~~~~
219
220These are the build configuration options affecting QAT, and their default values:
221
222.. code-block:: console
223
224	RTE_PMD_QAT_MAX_PCI_DEVICES=48
225	RTE_PMD_QAT_COMP_IM_BUFFER_SIZE=65536
226
227Both QAT SYM PMD and QAT ASYM PMD have an external dependency on libcrypto, so are not
228built by default.
229
230The QAT compressdev PMD has no external dependencies, so is built by default.
231
232The number of VFs per PF varies - see table below. If multiple QAT packages are
233installed on a platform then RTE_PMD_QAT_MAX_PCI_DEVICES should be
234adjusted to the number of VFs which the QAT common code will need to handle.
235
236.. Note::
237
238        There are separate config items (not QAT-specific) for max cryptodevs
239        RTE_CRYPTO_MAX_DEVS and max compressdevs RTE_COMPRESS_MAX_DEVS,
240        if necessary these should be adjusted to handle the total of QAT and other
241        devices which the process will use. In particular for crypto, where each
242        QAT VF may expose two crypto devices, sym and asym, it may happen that the
243        number of devices will be bigger than MAX_DEVS and the process will show an error
244        during PMD initialisation. To avoid this problem RTE_CRYPTO_MAX_DEVS may be
245        increased or -a, allow domain:bus:devid:func option may be used.
246
247
248QAT compression PMD needs intermediate buffers to support Deflate compression
249with Dynamic Huffman encoding. RTE_PMD_QAT_COMP_IM_BUFFER_SIZE
250specifies the size of a single buffer, the PMD will allocate a multiple of these,
251plus some extra space for associated meta-data. For GEN2 devices, 20 buffers are
252allocated while for GEN1 devices, 12 buffers are allocated, plus 1472 bytes overhead.
253
254.. Note::
255
256	If the compressed output of a Deflate operation using Dynamic Huffman
257	Encoding is too big to fit in an intermediate buffer, then the
258	operation will be split into smaller operations and their results will
259	be merged afterwards.
260	This is not possible if any checksum calculation was requested - in such
261	case the code falls back to fixed compression.
262	To avoid this less performant case, applications should configure
263	the intermediate buffer size to be larger than the expected input data size
264	(compressed output size is usually unknown, so the only option is to make
265	larger than the input size).
266
267
268Running QAT PMD with minimum threshold for burst size
269~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
270
271If only a small number or packets can be enqueued. Each enqueue causes an expensive MMIO write.
272These MMIO write occurrences can be optimised by setting any of the following parameters:
273
274- qat_sym_enq_threshold
275- qat_asym_enq_threshold
276- qat_comp_enq_threshold
277
278When any of these parameters is set rte_cryptodev_enqueue_burst function will
279return 0 (thereby avoiding an MMIO) if the device is congested and number of packets
280possible to enqueue is smaller.
281To use this feature the user must set the parameter on process start as a device additional parameter::
282
283  -a 03:01.1,qat_sym_enq_threshold=32,qat_comp_enq_threshold=16
284
285All parameters can be used with the same device regardless of order. Parameters are separated
286by comma. When the same parameter is used more than once first occurrence of the parameter
287is used.
288Maximum threshold that can be set is 32.
289
290
291Device and driver naming
292~~~~~~~~~~~~~~~~~~~~~~~~
293
294* The qat cryptodev symmetric crypto driver name is "crypto_qat".
295* The qat cryptodev asymmetric crypto driver name is "crypto_qat_asym".
296
297The "rte_cryptodev_devices_get()" returns the devices exposed by either of these drivers.
298
299* Each qat sym crypto device has a unique name, in format
300  "<pci bdf>_<service>", e.g. "0000:41:01.0_qat_sym".
301* Each qat asym crypto device has a unique name, in format
302  "<pci bdf>_<service>", e.g. "0000:41:01.0_qat_asym".
303  This name can be passed to "rte_cryptodev_get_dev_id()" to get the device_id.
304
305.. Note::
306
307	The cryptodev driver name is passed to the dpdk-test-crypto-perf tool in the "-devtype" parameter.
308
309	The qat crypto device name is in the format of the worker parameter passed to the crypto scheduler.
310
311* The qat compressdev driver name is "compress_qat".
312  The rte_compressdev_devices_get() returns the devices exposed by this driver.
313
314* Each qat compression device has a unique name, in format
315  <pci bdf>_<service>, e.g. "0000:41:01.0_qat_comp".
316  This name can be passed to rte_compressdev_get_dev_id() to get the device_id.
317
318.. _qat_kernel:
319
320Dependency on the QAT kernel driver
321~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
322
323To use QAT an SRIOV-enabled QAT kernel driver is required. The VF
324devices created and initialised by this driver will be used by the QAT PMDs.
325
326Instructions for installation are below, but first an explanation of the
327relationships between the PF/VF devices and the PMDs visible to
328DPDK applications.
329
330Each QuickAssist PF device exposes a number of VF devices. Each VF device can
331enable one symmetric cryptodev PMD and/or one asymmetric cryptodev PMD and/or
332one compressdev PMD.
333These QAT PMDs share the same underlying device and pci-mgmt code, but are
334enumerated independently on their respective APIs and appear as independent
335devices to applications.
336
337.. Note::
338
339   Each VF can only be used by one DPDK process. It is not possible to share
340   the same VF across multiple processes, even if these processes are using
341   different acceleration services.
342
343   Conversely one DPDK process can use one or more QAT VFs and can expose both
344   cryptodev and compressdev instances on each of those VFs.
345
346
347Available kernel drivers
348~~~~~~~~~~~~~~~~~~~~~~~~
349
350Kernel drivers for each device for each service are listed in the following table. (Scroll right
351to see the full table)
352
353
354.. _table_qat_pmds_drivers:
355
356.. table:: QAT device generations, devices and drivers
357
358   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
359   | S   | A   | C   | Gen | Device   | Driver/ver    | Kernel Module | Pci Driver | PF Did | #PFs | VF Did | VFs/PF |
360   +=====+=====+=====+=====+==========+===============+===============+============+========+======+========+========+
361   | Yes | No  | No  | 1   | DH895xCC | linux/4.4+    | qat_dh895xcc  | dh895xcc   | 435    | 1    | 443    | 32     |
362   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
363   | Yes | Yes | No  | "   | "        | 01.org/4.2.0+ | "             | "          | "      | "    | "      | "      |
364   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
365   | Yes | Yes | Yes | "   | "        | 01.org/4.3.0+ | "             | "          | "      | "    | "      | "      |
366   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
367   | Yes | No  | No  | 2   | C62x     | linux/4.5+    | qat_c62x      | c6xx       | 37c8   | 3    | 37c9   | 16     |
368   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
369   | Yes | Yes | Yes | "   | "        | 01.org/4.2.0+ | "             | "          | "      | "    | "      | "      |
370   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
371   | Yes | No  | No  | 2   | C3xxx    | linux/4.5+    | qat_c3xxx     | c3xxx      | 19e2   | 1    | 19e3   | 16     |
372   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
373   | Yes | Yes | Yes | "   | "        | 01.org/4.2.0+ | "             | "          | "      | "    | "      | "      |
374   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
375   | Yes | No  | No  | 2   | 200xx    | p             | qat_200xx     | 200xx      | 18ee   | 1    | 18ef   | 16     |
376   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
377   | Yes | No  | No  | 2   | D15xx    | 01.org/4.2.0+ | qat_d15xx     | d15xx      | 6f54   | 1    | 6f55   | 16     |
378   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
379   | Yes | No  | No  | 3   | C4xxx    | p             | qat_c4xxx     | c4xxx      | 18a0   | 1    | 18a1   | 128    |
380   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
381   | Yes | No  | No  | 4   | 4xxx     | N/A           | qat_4xxx      | 4xxx       | 4940   | 4    | 4941   | 16     |
382   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
383   | Yes | No  | No  | 4   | 401xxx   | N/A           | qat_401xxx    | 4xxx       | 4942   | 2    | 4943   | 16     |
384   +-----+-----+-----+-----+----------+---------------+---------------+------------+--------+------+--------+--------+
385
386* Note: Symmetric mixed crypto algorithms feature on Gen 2 works only with 01.org driver version 4.9.0+
387
388The first 3 columns indicate the service:
389
390* S = Symmetric crypto service (via cryptodev API)
391* A = Asymmetric crypto service  (via cryptodev API)
392* C = Compression service (via compressdev API)
393
394The ``Driver`` column indicates either the Linux kernel version in which
395support for this device was introduced or a driver available on Intel's 01.org
396website. There are both linux in-tree and 01.org kernel drivers available for some
397devices. p = release pending.
398
399If you are running on a kernel which includes a driver for your device, see
400`Installation using kernel.org driver`_ below. Otherwise see
401`Installation using 01.org QAT driver`_.
402
403
404Installation using kernel.org driver
405~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
406
407The examples below are based on the C62x device, if you have a different device
408use the corresponding values in the above table.
409
410In BIOS ensure that SRIOV is enabled and either:
411
412* Disable VT-d or
413* Enable VT-d and set ``"intel_iommu=on iommu=pt"`` in the grub file.
414
415Check that the QAT driver is loaded on your system, by executing::
416
417    lsmod | grep qa
418
419You should see the kernel module for your device listed, e.g.::
420
421    qat_c62x               5626  0
422    intel_qat              82336  1 qat_c62x
423
424Next, you need to expose the Virtual Functions (VFs) using the sysfs file system.
425
426First find the BDFs (Bus-Device-Function) of the physical functions (PFs) of
427your device, e.g.::
428
429    lspci -d:37c8
430
431You should see output similar to::
432
433    1a:00.0 Co-processor: Intel Corporation Device 37c8
434    3d:00.0 Co-processor: Intel Corporation Device 37c8
435    3f:00.0 Co-processor: Intel Corporation Device 37c8
436
437Enable the VFs for each PF by echoing the number of VFs per PF to the pci driver::
438
439     echo 16 > /sys/bus/pci/drivers/c6xx/0000:1a:00.0/sriov_numvfs
440     echo 16 > /sys/bus/pci/drivers/c6xx/0000:3d:00.0/sriov_numvfs
441     echo 16 > /sys/bus/pci/drivers/c6xx/0000:3f:00.0/sriov_numvfs
442
443Check that the VFs are available for use. For example ``lspci -d:37c9`` should
444list 48 VF devices available for a ``C62x`` device.
445
446To complete the installation follow the instructions in
447`Binding the available VFs to the vfio-pci driver`_.
448
449.. Note::
450
451   If the QAT kernel modules are not loaded and you see an error like ``Failed
452   to load MMP firmware qat_895xcc_mmp.bin`` in kernel logs, this may be as a
453   result of not using a distribution, but just updating the kernel directly.
454
455   Download firmware from the `kernel firmware repo
456   <http://git.kernel.org/cgit/linux/kernel/git/firmware/linux-firmware.git/tree/>`_.
457
458   Copy qat binaries to ``/lib/firmware``::
459
460      cp qat_895xcc.bin /lib/firmware
461      cp qat_895xcc_mmp.bin /lib/firmware
462
463   Change to your linux source root directory and start the qat kernel modules::
464
465      insmod ./drivers/crypto/qat/qat_common/intel_qat.ko
466      insmod ./drivers/crypto/qat/qat_dh895xcc/qat_dh895xcc.ko
467
468.. Note::
469
470   If you see the following warning in ``/var/log/messages`` it can be ignored:
471   ``IOMMU should be enabled for SR-IOV to work correctly``.
472
473
474Installation using 01.org QAT driver
475~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
476
477Download the latest QuickAssist Technology Driver from `01.org
478<https://01.org/packet-processing/intel%C2%AE-quickassist-technology-drivers-and-patches>`_.
479Consult the *Getting Started Guide* at the same URL for further information.
480
481The steps below assume you are:
482
483* Building on a platform with one ``C62x`` device.
484* Using package ``qat1.7.l.4.2.0-000xx.tar.gz``.
485* On Fedora26 kernel ``4.11.11-300.fc26.x86_64``.
486
487In the BIOS ensure that SRIOV is enabled and VT-d is disabled.
488
489Uninstall any existing QAT driver, for example by running:
490
491* ``./installer.sh uninstall`` in the directory where originally installed.
492
493
494Build and install the SRIOV-enabled QAT driver::
495
496    mkdir /QAT
497    cd /QAT
498
499    # Copy the package to this location and unpack
500    tar zxof qat1.7.l.4.2.0-000xx.tar.gz
501
502    ./configure --enable-icp-sriov=host
503    make install
504
505You can use ``cat /sys/kernel/debug/qat<your device type and bdf>/version/fw`` to confirm the driver is correctly installed and is using firmware version 4.2.0.
506You can use ``lspci -d:37c9`` to confirm the presence of the 16 VF devices available per ``C62x`` PF.
507
508Confirm the driver is correctly installed and is using firmware version 4.2.0::
509
510    cat /sys/kernel/debug/qat<your device type and bdf>/version/fw
511
512
513Confirm the presence of 48 VF devices - 16 per PF::
514
515    lspci -d:37c9
516
517
518To complete the installation - follow instructions in
519`Binding the available VFs to the vfio-pci driver`_.
520
521.. Note::
522
523   If using a later kernel and the build fails with an error relating to
524   ``strict_stroul`` not being available apply the following patch:
525
526   .. code-block:: diff
527
528      /QAT/QAT1.6/quickassist/utilities/downloader/Target_CoreLibs/uclo/include/linux/uclo_platform.h
529      + #if LINUX_VERSION_CODE >= KERNEL_VERSION(3,18,5)
530      + #define STR_TO_64(str, base, num, endPtr) {endPtr=NULL; if (kstrtoul((str), (base), (num))) printk("Error strtoull convert %s\n", str); }
531      + #else
532      #if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,38)
533      #define STR_TO_64(str, base, num, endPtr) {endPtr=NULL; if (strict_strtoull((str), (base), (num))) printk("Error strtoull convert %s\n", str); }
534      #else
535      #if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,25)
536      #define STR_TO_64(str, base, num, endPtr) {endPtr=NULL; strict_strtoll((str), (base), (num));}
537      #else
538      #define STR_TO_64(str, base, num, endPtr)                                 \
539           do {                                                               \
540                 if (str[0] == '-')                                           \
541                 {                                                            \
542                      *(num) = -(simple_strtoull((str+1), &(endPtr), (base))); \
543                 }else {                                                      \
544                      *(num) = simple_strtoull((str), &(endPtr), (base));      \
545                 }                                                            \
546           } while(0)
547      + #endif
548      #endif
549      #endif
550
551
552.. Note::
553
554   If the build fails due to missing header files you may need to do following::
555
556      sudo yum install zlib-devel
557      sudo yum install openssl-devel
558      sudo yum install libudev-devel
559
560.. Note::
561
562   If the build or install fails due to mismatching kernel sources you may need to do the following::
563
564      sudo yum install kernel-headers-`uname -r`
565      sudo yum install kernel-src-`uname -r`
566      sudo yum install kernel-devel-`uname -r`
567
568
569Binding the available VFs to the vfio-pci driver
570~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
571
572Note:
573
574* Please note that due to security issues, the usage of older DPDK igb_uio
575  driver is not recommended. This document shows how to use the more secure
576  vfio-pci driver.
577* If QAT fails to bind to vfio-pci on Linux kernel 5.9+, please see the
578  QATE-39220 and QATE-7495 issues in
579  `01.org doc <https://01.org/sites/default/files/downloads/336211-015-qatsoftwareforlinux-rn-hwv1.7-final.pdf>`_
580  which details the constraint about trusted guests and add `disable_denylist=1`
581  to the vfio-pci params to use QAT. See also `this patch description <https://lkml.org/lkml/2020/7/23/1155>`_.
582
583Unbind the VFs from the stock driver so they can be bound to the vfio-pci driver.
584
585For an Intel(R) QuickAssist Technology DH895xCC device
586^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
587
588The unbind command below assumes ``BDFs`` of ``03:01.00-03:04.07``, if your
589VFs are different adjust the unbind command below::
590
591    cd to the top-level DPDK directory
592    for device in $(seq 1 4); do \
593        for fn in $(seq 0 7); do \
594            usertools/dpdk-devbind.py -u 0000:03:0${device}.${fn}; \
595        done; \
596    done
597
598For an Intel(R) QuickAssist Technology C62x device
599^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
600
601The unbind command below assumes ``BDFs`` of ``1a:01.00-1a:02.07``,
602``3d:01.00-3d:02.07`` and ``3f:01.00-3f:02.07``, if your VFs are different
603adjust the unbind command below::
604
605    cd to the top-level DPDK directory
606    for device in $(seq 1 2); do \
607        for fn in $(seq 0 7); do \
608            usertools/dpdk-devbind.py -u 0000:1a:0${device}.${fn}; \
609            usertools/dpdk-devbind.py -u 0000:3d:0${device}.${fn}; \
610            usertools/dpdk-devbind.py -u 0000:3f:0${device}.${fn}; \
611        done; \
612    done
613
614For Intel(R) QuickAssist Technology C3xxx or 200xx or D15xx device
615^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
616
617The unbind command below assumes ``BDFs`` of ``01:01.00-01:02.07``, if your
618VFs are different adjust the unbind command below::
619
620    cd to the top-level DPDK directory
621    for device in $(seq 1 2); do \
622        for fn in $(seq 0 7); do \
623            usertools/dpdk-devbind.py -u 0000:01:0${device}.${fn}; \
624        done; \
625    done
626
627Bind to the vfio-pci driver
628^^^^^^^^^^^^^^^^^^^^^^^^^^^
629
630Load the vfio-pci driver, bind the VF PCI Device id to it using the
631``dpdk-devbind.py`` script then use the ``--status`` option
632to confirm the VF devices are now in use by vfio-pci kernel driver,
633e.g. for the C62x device::
634
635    cd to the top-level DPDK directory
636    modprobe vfio-pci
637    usertools/dpdk-devbind.py -b vfio-pci 0000:03:01.1
638    usertools/dpdk-devbind.py --status
639
640Use ``modprobe vfio-pci disable_denylist=1`` from kernel 5.9 onwards.
641See note in the section `Binding the available VFs to the vfio-pci driver`_
642above.
643
644Testing
645~~~~~~~
646
647QAT SYM crypto PMD can be tested by running the test application::
648
649    cd ./<build_dir>/app/test
650    ./dpdk-test -l1 -n1 -a <your qat bdf>
651    RTE>>cryptodev_qat_autotest
652
653QAT ASYM crypto PMD can be tested by running the test application::
654
655    cd ./<build_dir>/app/test
656    ./dpdk-test -l1 -n1 -a <your qat bdf>
657    RTE>>cryptodev_qat_asym_autotest
658
659QAT compression PMD can be tested by running the test application::
660
661    cd ./<build_dir>/app/test
662    ./dpdk-test -l1 -n1 -a <your qat bdf>
663    RTE>>compressdev_autotest
664
665
666Debugging
667~~~~~~~~~
668
669There are 2 sets of trace available via the dynamic logging feature:
670
671* pmd.qat.dp exposes trace on the data-path.
672* pmd.qat.general exposes all other trace.
673
674pmd.qat exposes both sets of traces.
675They can be enabled using the log-level option (where 8=maximum log level) on
676the process cmdline, e.g. using any of the following::
677
678    --log-level="pmd.qat.general,8"
679    --log-level="pmd.qat.dp,8"
680    --log-level="pmd.qat,8"
681
682.. Note::
683
684    The global RTE_LOG_DP_LEVEL overrides data-path trace so must be set to
685    RTE_LOG_DEBUG to see all the trace. This variable is in config/rte_config.h
686    for meson build.
687    Also the dynamic global log level overrides both sets of trace, so e.g. no
688    QAT trace would display in this case::
689
690	--log-level="7" --log-level="pmd.qat.general,8"
691