1 .. SPDX-License-Identifier: GPL-2.0
3 =========================================
4 IAA Compression Accelerator Crypto Driver
5 =========================================
7 Tom Zanussi <tom.zanussi@linux.intel.com>
9 The IAA crypto driver supports compression/decompression compatible
10 with the DEFLATE compression standard described in RFC 1951, which is
11 the compression/decompression algorithm exported by this module.
13 The IAA hardware spec can be found here:
15 https://cdrdv2.intel.com/v1/dl/getContent/721858
17 The iaa_crypto driver is designed to work as a layer underneath
18 higher-level compression devices such as zswap.
20 Users can select IAA compress/decompress acceleration by specifying
21 one of the supported IAA compression algorithms in whatever facility
22 allows compression algorithms to be selected.
24 For example, a zswap device can select the IAA 'fixed' mode
25 represented by selecting the 'deflate-iaa' crypto compression
28 # echo deflate-iaa > /sys/module/zswap/parameters/compressor
30 This will tell zswap to use the IAA 'fixed' compression mode for all
31 compresses and decompresses.
33 Currently, there is only one compression modes available, 'fixed'
36 The 'fixed' compression mode implements the compression scheme
37 specified by RFC 1951 and is given the crypto algorithm name
38 'deflate-iaa'. (Because the IAA hardware has a 4k history-window
39 limitation, only buffers <= 4k, or that have been compressed using a
40 <= 4k history window, are technically compliant with the deflate spec,
41 which allows for a window of up to 32k. Because of this limitation,
42 the IAA fixed mode deflate algorithm is given its own algorithm name
43 rather than simply 'deflate').
46 Config options and other setup
47 ==============================
49 The IAA crypto driver is available via menuconfig using the following
52 Cryptographic API -> Hardware crypto devices -> Support for Intel(R) IAA Compression Accelerator
54 In the configuration file the option called CONFIG_CRYPTO_DEV_IAA_CRYPTO.
56 The IAA crypto driver also supports statistics, which are available
57 via menuconfig using the following path::
59 Cryptographic API -> Hardware crypto devices -> Support for Intel(R) IAA Compression -> Enable Intel(R) IAA Compression Accelerator Statistics
61 In the configuration file the option called CONFIG_CRYPTO_DEV_IAA_CRYPTO_STATS.
63 The following config options should also be enabled::
67 CONFIG_INTEL_IOMMU_SVM=y
72 CONFIG_INTEL_IDXD_SVM=y
74 IAA is one of the first Intel accelerator IPs that can work in
75 conjunction with the Intel IOMMU. There are multiple modes that exist
76 for testing. Based on IOMMU configuration, there are 3 modes::
86 Scalable mode supports Shared Virtual Memory (SVM or SVA). It is
87 entered when using the kernel boot commandline::
91 with VT-d turned on in BIOS.
93 With scalable mode, both shared and dedicated workqueues are available
96 For scalable mode, the following BIOS settings should be enabled::
98 Socket Configuration > IIO Configuration > Intel VT for Directed I/O (VT-d) > Intel VT for Directed I/O
100 Socket Configuration > IIO Configuration > PCIe ENQCMD > ENQCMDS
106 Legacy mode is entered when using the kernel boot commandline::
110 or VT-d is not turned on in BIOS.
112 If you have booted into Linux and not sure if VT-d is on, do a "dmesg
113 | grep -i dmar". If you don't see a number of DMAR devices enumerated,
114 most likely VT-d is not on.
116 With legacy mode, only dedicated workqueues are available for use.
122 No IOMMU mode is entered when using the kernel boot commandline::
126 With no IOMMU mode, only dedicated workqueues are available for use.
135 When loaded, the iaa_crypto driver automatically creates a default
136 configuration and enables it, and assigns default driver attributes.
137 If a different configuration or set of driver attributes is required,
138 the user must first disable the IAA devices and workqueues, reset the
139 configuration, and then re-register the deflate-iaa algorithm with the
140 crypto subsystem by removing and reinserting the iaa_crypto module.
142 The :ref:`iaa_disable_script` in the 'Use Cases'
143 section below can be used to disable the default configuration.
145 See :ref:`iaa_default_config` below for details of the default
148 More likely than not, however, and because of the complexity and
149 configurability of the accelerator devices, the user will want to
150 configure the device and manually enable the desired devices and
153 The userspace tool to help doing that is called accel-config. Using
154 accel-config to configure device or loading a previously saved config
155 is highly recommended. The device can be controlled via sysfs
156 directly but comes with the warning that you should do this ONLY if
157 you know exactly what you are doing. The following sections will not
158 cover the sysfs interface but assumes you will be using accel-config.
160 The :ref:`iaa_sysfs_config` section in the appendix below can be
161 consulted for the sysfs interface details if interested.
163 The accel-config tool along with instructions for building it can be
166 https://github.com/intel/idxd-config/#readme
171 In order for the iaa_crypto module to actually do any
172 compression/decompression work on behalf of a facility, one or more
173 IAA workqueues need to be bound to the iaa_crypto driver.
175 For instance, here's an example of configuring an IAA workqueue and
176 binding it to the iaa_crypto driver (note that device names are
177 specified as 'iax' rather than 'iaa' - this is because upstream still
178 has the old 'iax' device naming in place) ::
182 accel-config config-wq --group-id=0 --mode=dedicated --type=kernel --priority=10 --name="iaa_crypto" --driver-name="crypto" iax1/wq1.0
184 accel-config config-engine iax1/engine1.0 --group-id=0
186 # enable IAA device iax1
188 accel-config enable-device iax1
190 # enable wq1.0 on IAX device iax1
192 accel-config enable-wq iax1/wq1.0
194 Whenever a new workqueue is bound to or unbound from the iaa_crypto
195 driver, the available workqueues are 'rebalanced' such that work
196 submitted from a particular CPU is given to the most appropriate
197 workqueue available. Current best practice is to configure and bind
198 at least one workqueue for each IAA device, but as long as there is at
199 least one workqueue configured and bound to any IAA device in the
200 system, the iaa_crypto driver will work, albeit most likely not as
203 The IAA crypto algorigthms is operational and compression and
204 decompression operations are fully enabled following the successful
205 binding of the first IAA workqueue to the iaa_crypto driver.
207 Similarly, the IAA crypto algorithm is not operational and compression
208 and decompression operations are disabled following the unbinding of
209 the last IAA worqueue to the iaa_crypto driver.
211 As a result, the IAA crypto algorithms and thus the IAA hardware are
212 only available when one or more workques are bound to the iaa_crypto
215 When there are no IAA workqueues bound to the driver, the IAA crypto
216 algorithms can be unregistered by removing the module.
222 There are a couple user-configurable driver attributes that can be
223 used to configure various modes of operation. They're listed below,
224 along with their default values. To set any of these attributes, echo
225 the appropriate values to the attribute file located under
226 /sys/bus/dsa/drivers/crypto/
228 The attribute settings at the time the IAA algorithms are registered
229 are captured in each algorithm's crypto_ctx and used for all compresses
230 and decompresses when using that algorithm.
232 The available attributes are:
236 Toggle compression verification. If set, each compress will be
237 internally decompressed and the contents verified, returning error
238 codes if unsuccessful. This can be toggled with 0/1::
240 echo 0 > /sys/bus/dsa/drivers/crypto/verify_compress
242 The default setting is '1' - verify all compresses.
246 Select mode to be used to wait for completion of each compresses
247 and decompress operation.
249 The crypto async interface support implemented by iaa_crypto
250 provides an implementation that satisfies the interface but does
251 so in a synchronous manner - it fills and submits the IDXD
252 descriptor and then loops around waiting for it to complete before
253 returning. This isn't a problem at the moment, since all existing
254 callers (e.g. zswap) wrap any asynchronous callees in a
255 synchronous wrapper anyway.
257 The iaa_crypto driver does however provide true asynchronous
258 support for callers that can make use of it. In this mode, it
259 fills and submits the IDXD descriptor, then returns immediately
260 with -EINPROGRESS. The caller can then either poll for completion
261 itself, which requires specific code in the caller which currently
262 nothing in the upstream kernel implements, or go to sleep and wait
263 for an interrupt signaling completion. This latter mode is
264 supported by current users in the kernel such as zswap via
265 synchronous wrappers. Although it is supported this mode is
266 significantly slower than the synchronous mode that does the
267 polling in the iaa_crypto driver previously mentioned.
269 This mode can be enabled by writing 'async_irq' to the sync_mode
270 iaa_crypto driver attribute::
272 echo async_irq > /sys/bus/dsa/drivers/crypto/sync_mode
274 Async mode without interrupts (caller must poll) can be enabled by
275 writing 'async' to it::
277 echo async > /sys/bus/dsa/drivers/crypto/sync_mode
279 The mode that does the polling in the iaa_crypto driver can be
280 enabled by writing 'sync' to it::
282 echo sync > /sys/bus/dsa/drivers/crypto/sync_mode
284 The default mode is 'sync'.
286 .. _iaa_default_config:
288 IAA Default Configuration
289 -------------------------
291 When the iaa_crypto driver is loaded, each IAA device has a single
292 work queue configured for it, with the following attributes::
296 size Total WQ Size from WQCAP
303 The devices and workqueues are also enabled and therefore the driver
304 is ready to be used without any additional configuration.
306 The default driver attributes in effect when the driver is loaded are::
311 In order to change either the device/work queue or driver attributes,
312 the enabled devices and workqueues must first be disabled. In order
313 to have the new configuration applied to the deflate-iaa crypto
314 algorithm, it needs to be re-registered by removing and reinserting
315 the iaa_crypto module. The :ref:`iaa_disable_script` in the 'Use
316 Cases' section below can be used to disable the default configuration.
321 If the optional debugfs statistics support is enabled, the IAA crypto
322 driver will generate statistics which can be accessed in debugfs at::
324 # ls -al /sys/kernel/debug/iaa-crypto/
326 drwxr-xr-x 2 root root 0 Mar 3 07:55 .
327 drwx------ 53 root root 0 Mar 3 07:55 ..
328 -rw-r--r-- 1 root root 0 Mar 3 07:55 global_stats
329 -rw-r--r-- 1 root root 0 Mar 3 07:55 stats_reset
330 -rw-r--r-- 1 root root 0 Mar 3 07:55 wq_stats
332 The global_stats file shows a set of global statistics collected since
333 the driver has been loaded or reset::
337 total_comp_calls: 4300
338 total_decomp_calls: 4164
339 total_sw_decomp_calls: 0
340 total_comp_bytes_out: 5993989
341 total_decomp_bytes_in: 5993989
342 total_completion_einval_errors: 0
343 total_completion_timeout_errors: 0
344 total_completion_comp_buf_overflow_errors: 136
346 The wq_stats file shows per-wq stats, a set for each iaa device and wq
347 in addition to some global stats::
398 decomp_bytes: 5993989
404 decomp_bytes: 5993989
407 Writing to 'stats_reset' resets all the stats, including the
408 per-device and per-wq stats::
410 # echo 1 > stats_reset
414 total_decomp_calls: 0
415 total_comp_bytes_out: 0
416 total_decomp_bytes_in: 0
417 total_completion_einval_errors: 0
418 total_completion_timeout_errors: 0
419 total_completion_comp_buf_overflow_errors: 0
429 For this example, the kernel should be configured according to the
430 dedicated mode options described above, and zswap should be enabled as
435 This is a simple test that uses iaa_compress as the compressor for a
436 swap (zswap) device. It sets up the zswap device and then uses the
437 memory_memadvise program listed below to forcibly swap out and in a
438 specified number of pages, demonstrating both compress and decompress.
440 The zswap test expects the work queues for each IAA device on the
441 system to be configured properly as a kernel workqueue with a
442 workqueue driver_name of "crypto".
444 The first step is to make sure the iaa_crypto module is loaded::
448 If the IAA devices and workqueues haven't previously been disabled and
449 reconfigured, then the default configuration should be in place and no
450 further IAA configuration is necessary. See :ref:`iaa_default_config`
451 below for details of the default configuration.
453 If the default configuration is in place, you should see the iaa
454 devices and wq0s enabled::
456 # cat /sys/bus/dsa/devices/iax1/state
458 # cat /sys/bus/dsa/devices/iax1/wq1.0/state
461 To demonstrate that the following steps work as expected, these
462 commands can be used to enable debug output::
464 # echo -n 'module iaa_crypto +p' > /sys/kernel/debug/dynamic_debug/control
465 # echo -n 'module idxd +p' > /sys/kernel/debug/dynamic_debug/control
467 Use the following commands to enable zswap::
469 # echo 0 > /sys/module/zswap/parameters/enabled
470 # echo 50 > /sys/module/zswap/parameters/max_pool_percent
471 # echo deflate-iaa > /sys/module/zswap/parameters/compressor
472 # echo zsmalloc > /sys/module/zswap/parameters/zpool
473 # echo 1 > /sys/module/zswap/parameters/enabled
474 # echo 100 > /proc/sys/vm/swappiness
475 # echo never > /sys/kernel/mm/transparent_hugepage/enabled
476 # echo 1 > /proc/sys/vm/overcommit_memory
478 Now you can now run the zswap workload you want to measure. For
479 example, using the memory_memadvise code below, the following command
480 will swap in and out 100 pages::
484 Allocating 100 pages to swap in/out
485 Swapping out 100 pages
486 Swapping in 100 pages
487 Swapped out and in 100 pages
489 You should see something like the following in the dmesg output::
491 [ 404.202972] idxd 0000:e7:02.0: iaa_comp_acompress: dma_map_sg, src_addr 223925c000, nr_sgs 1, req->src 00000000ee7cb5e6, req->slen 4096, sg_dma_len(sg) 4096
492 [ 404.202973] idxd 0000:e7:02.0: iaa_comp_acompress: dma_map_sg, dst_addr 21dadf8000, nr_sgs 1, req->dst 000000008d6acea8, req->dlen 4096, sg_dma_len(sg) 8192
493 [ 404.202975] idxd 0000:e7:02.0: iaa_compress: desc->src1_addr 223925c000, desc->src1_size 4096, desc->dst_addr 21dadf8000, desc->max_dst_size 4096, desc->src2_addr 2203543000, desc->src2_size 1568
494 [ 404.202981] idxd 0000:e7:02.0: iaa_compress_verify: (verify) desc->src1_addr 21dadf8000, desc->src1_size 228, desc->dst_addr 223925c000, desc->max_dst_size 4096, desc->src2_addr 0, desc->src2_size 0
497 Now that basic functionality has been demonstrated, the defaults can
498 be erased and replaced with a different configuration. To do that,
499 first disable zswap::
501 # echo lzo > /sys/module/zswap/parameters/compressor
503 # echo 0 > /sys/module/zswap/parameters/accept_threshold_percent
504 # echo 0 > /sys/module/zswap/parameters/max_pool_percent
505 # echo 0 > /sys/module/zswap/parameters/enabled
506 # echo 0 > /sys/module/zswap/parameters/enabled
508 Then run the :ref:`iaa_disable_script` in the 'Use Cases' section
509 below to disable the default configuration.
511 Finally turn swap back on::
515 Following all that the IAA device(s) can now be re-configured and
516 enabled as desired for further testing. Below is one example.
518 The zswap test expects the work queues for each IAA device on the
519 system to be configured properly as a kernel workqueue with a
520 workqueue driver_name of "crypto".
522 The below script automatically does that::
528 echo "# IAA devices:"
529 lspci -d:0cfe | wc -l
532 # count iaa instances
535 num_iaa=$(lspci -d:${iaa_dev_id} | wc -l)
536 echo "Found ${num_iaa} IAA instances"
539 # disable iaa wqs and devices
543 for ((i = 1; i < ${num_iaa} * 2; i += 2)); do
544 echo disable wq iax${i}/wq${i}.0
545 accel-config disable-wq iax${i}/wq${i}.0
546 echo disable iaa iax${i}
547 accel-config disable-device iax${i}
550 echo "End Disable IAA"
552 echo "Reload iaa_crypto module"
557 echo "End Reload iaa_crypto module"
560 # configure iaa wqs and devices
563 for ((i = 1; i < ${num_iaa} * 2; i += 2)); do
564 accel-config config-wq --group-id=0 --mode=dedicated --wq-size=128 --priority=10 --type=kernel --name="iaa_crypto" --driver-name="crypto" iax${i}/wq${i}.0
565 accel-config config-engine iax${i}/engine${i}.0 --group-id=0
568 echo "End Configure IAA"
571 # enable iaa wqs and devices
575 for ((i = 1; i < ${num_iaa} * 2; i += 2)); do
576 echo enable iaa iax${i}
577 accel-config enable-device iax${i}
578 echo enable wq iax${i}/wq${i}.0
579 accel-config enable-wq iax${i}/wq${i}.0
582 echo "End Enable IAA"
584 When the workqueues are bound to the iaa_crypto driver, you should
585 see something similar to the following in dmesg output if you've
586 enabled debug output (echo -n 'module iaa_crypto +p' >
587 /sys/kernel/debug/dynamic_debug/control)::
589 [ 60.752344] idxd 0000:f6:02.0: add_iaa_wq: added wq 000000004068d14d to iaa 00000000c9585ba2, n_wq 1
590 [ 60.752346] iaa_crypto: rebalance_wq_table: nr_nodes=2, nr_cpus 160, nr_iaa 8, cpus_per_iaa 20
591 [ 60.752347] iaa_crypto: rebalance_wq_table: iaa=0
592 [ 60.752349] idxd 0000:6a:02.0: request_iaa_wq: getting wq from iaa_device 0000000042d7bc52 (0)
593 [ 60.752350] idxd 0000:6a:02.0: request_iaa_wq: returning unused wq 00000000c8bb4452 (0) from iaa device 0000000042d7bc52 (0)
594 [ 60.752352] iaa_crypto: rebalance_wq_table: assigned wq for cpu=0, node=0 = wq 00000000c8bb4452
595 [ 60.752354] iaa_crypto: rebalance_wq_table: iaa=0
596 [ 60.752355] idxd 0000:6a:02.0: request_iaa_wq: getting wq from iaa_device 0000000042d7bc52 (0)
597 [ 60.752356] idxd 0000:6a:02.0: request_iaa_wq: returning unused wq 00000000c8bb4452 (0) from iaa device 0000000042d7bc52 (0)
598 [ 60.752358] iaa_crypto: rebalance_wq_table: assigned wq for cpu=1, node=0 = wq 00000000c8bb4452
599 [ 60.752359] iaa_crypto: rebalance_wq_table: iaa=0
600 [ 60.752360] idxd 0000:6a:02.0: request_iaa_wq: getting wq from iaa_device 0000000042d7bc52 (0)
601 [ 60.752361] idxd 0000:6a:02.0: request_iaa_wq: returning unused wq 00000000c8bb4452 (0) from iaa device 0000000042d7bc52 (0)
602 [ 60.752362] iaa_crypto: rebalance_wq_table: assigned wq for cpu=2, node=0 = wq 00000000c8bb4452
603 [ 60.752364] iaa_crypto: rebalance_wq_table: iaa=0
608 Once the workqueues and devices have been enabled, the IAA crypto
609 algorithms are enabled and available. When the IAA crypto algorithms
610 have been successfully enabled, you should see the following dmesg
613 [ 64.893759] iaa_crypto: iaa_crypto_enable: iaa_crypto now ENABLED
615 Now run the following zswap-specific setup commands to have zswap use
616 the 'fixed' compression mode::
618 echo 0 > /sys/module/zswap/parameters/enabled
619 echo 50 > /sys/module/zswap/parameters/max_pool_percent
620 echo deflate-iaa > /sys/module/zswap/parameters/compressor
621 echo zsmalloc > /sys/module/zswap/parameters/zpool
622 echo 1 > /sys/module/zswap/parameters/enabled
624 echo 100 > /proc/sys/vm/swappiness
625 echo never > /sys/kernel/mm/transparent_hugepage/enabled
626 echo 1 > /proc/sys/vm/overcommit_memory
628 Finally, you can now run the zswap workload you want to measure. For
629 example, using the code below, the following command will swap in and
634 Allocating 100 pages to swap in/out
635 Swapping out 100 pages
636 Swapping in 100 pages
637 Swapped out and in 100 pages
639 You should see something like the following in the dmesg output if
640 you've enabled debug output (echo -n 'module iaa_crypto +p' >
641 /sys/kernel/debug/dynamic_debug/control)::
643 [ 404.202972] idxd 0000:e7:02.0: iaa_comp_acompress: dma_map_sg, src_addr 223925c000, nr_sgs 1, req->src 00000000ee7cb5e6, req->slen 4096, sg_dma_len(sg) 4096
644 [ 404.202973] idxd 0000:e7:02.0: iaa_comp_acompress: dma_map_sg, dst_addr 21dadf8000, nr_sgs 1, req->dst 000000008d6acea8, req->dlen 4096, sg_dma_len(sg) 8192
645 [ 404.202975] idxd 0000:e7:02.0: iaa_compress: desc->src1_addr 223925c000, desc->src1_size 4096, desc->dst_addr 21dadf8000, desc->max_dst_size 4096, desc->src2_addr 2203543000, desc->src2_size 1568
646 [ 404.202981] idxd 0000:e7:02.0: iaa_compress_verify: (verify) desc->src1_addr 21dadf8000, desc->src1_size 228, desc->dst_addr 223925c000, desc->max_dst_size 4096, desc->src2_addr 0, desc->src2_size 0
647 [ 409.203227] idxd 0000:e7:02.0: iaa_comp_adecompress: dma_map_sg, src_addr 21ddd8b100, nr_sgs 1, req->src 0000000084adab64, req->slen 228, sg_dma_len(sg) 228
648 [ 409.203235] idxd 0000:e7:02.0: iaa_comp_adecompress: dma_map_sg, dst_addr 21ee3dc000, nr_sgs 1, req->dst 000000004e2990d0, req->dlen 4096, sg_dma_len(sg) 4096
649 [ 409.203239] idxd 0000:e7:02.0: iaa_decompress: desc->src1_addr 21ddd8b100, desc->src1_size 228, desc->dst_addr 21ee3dc000, desc->max_dst_size 4096, desc->src2_addr 0, desc->src2_size 0
650 [ 409.203254] idxd 0000:e7:02.0: iaa_comp_adecompress: dma_map_sg, src_addr 21ddd8b100, nr_sgs 1, req->src 0000000084adab64, req->slen 228, sg_dma_len(sg) 228
651 [ 409.203256] idxd 0000:e7:02.0: iaa_comp_adecompress: dma_map_sg, dst_addr 21f1551000, nr_sgs 1, req->dst 000000004e2990d0, req->dlen 4096, sg_dma_len(sg) 4096
652 [ 409.203257] idxd 0000:e7:02.0: iaa_decompress: desc->src1_addr 21ddd8b100, desc->src1_size 228, desc->dst_addr 21f1551000, desc->max_dst_size 4096, desc->src2_addr 0, desc->src2_size 0
654 In order to unregister the IAA crypto algorithms, and register new
655 ones using different parameters, any users of the current algorithm
656 should be stopped and the IAA workqueues and devices disabled.
658 In the case of zswap, remove the IAA crypto algorithm as the
659 compressor and turn off swap (to remove all references to
662 echo lzo > /sys/module/zswap/parameters/compressor
665 echo 0 > /sys/module/zswap/parameters/accept_threshold_percent
666 echo 0 > /sys/module/zswap/parameters/max_pool_percent
667 echo 0 > /sys/module/zswap/parameters/enabled
669 Once zswap is disabled and no longer using iaa_crypto, the IAA wqs and
670 devices can be disabled.
672 .. _iaa_disable_script:
677 The below script automatically does that::
683 echo "# IAA devices:"
684 lspci -d:0cfe | wc -l
687 # count iaa instances
690 num_iaa=$(lspci -d:${iaa_dev_id} | wc -l)
691 echo "Found ${num_iaa} IAA instances"
694 # disable iaa wqs and devices
698 for ((i = 1; i < ${num_iaa} * 2; i += 2)); do
699 echo disable wq iax${i}/wq${i}.0
700 accel-config disable-wq iax${i}/wq${i}.0
701 echo disable iaa iax${i}
702 accel-config disable-device iax${i}
705 echo "End Disable IAA"
707 Finally, at this point the iaa_crypto module can be removed, which
708 will unregister the current IAA crypto algorithms::
713 memory_madvise.c (gcc -o memory_memadvise memory_madvise.c)::
719 #include <sys/mman.h>
720 #include <linux/mman.h>
723 #define MADV_PAGEOUT 21 /* force pages out immediately */
728 int main(int argc, char **argv)
736 nr_pages = atoi(argv[1]);
738 printf("Allocating %d pages to swap in/out\n", nr_pages);
741 addr = mmap(NULL, nr_pages * PG_SZ, PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0);
744 /* initialize data in page to all '*' chars */
745 memset(addr, '*', nr_pages * PG_SZ);
747 printf("Swapping out %d pages\n", nr_pages);
749 /* Tell kernel to swap it out */
750 madvise(addr, nr_pages * PG_SZ, MADV_PAGEOUT);
753 /* Wait for swap out to finish */
758 printf("Swapping in %d pages\n", nr_pages);
760 /* Access the page ... this will swap it back in again */
761 for (i = 0; i < nr_pages; i++) {
763 printf("Bad data from decompress!!!!!\n");
765 dump_ptr = (int64_t *)a;
766 for (int j = 0; j < 100; j++) {
767 printf(" page %d data: %#llx\n", i, *dump_ptr);
778 printf("Swapped out and in %d pages\n", nr_pages);
783 .. _iaa_sysfs_config:
785 IAA sysfs config interface
786 --------------------------
788 Below is a description of the IAA sysfs interface, which as mentioned
789 in the main document, should only be used if you know exactly what you
790 are doing. Even then, there's no compelling reason to use it directly
791 since accel-config can do everything the sysfs interface can and in
792 fact accel-config is based on it under the covers.
794 The 'IAA config path' is /sys/bus/dsa/devices and contains
795 subdirectories representing each IAA device, workqueue, engine, and
796 group. Note that in the sysfs interface, the IAA devices are actually
797 named using iax e.g. iax1, iax3, etc. (Note that IAA devices are the
798 odd-numbered devices; the even-numbered devices are DSA devices and
799 can be ignored for IAA).
801 The 'IAA device bind path' is /sys/bus/dsa/drivers/idxd/bind and is
802 the file that is written to enable an IAA device.
804 The 'IAA workqueue bind path' is /sys/bus/dsa/drivers/crypto/bind and
805 is the file that is written to enable an IAA workqueue.
807 Similarly /sys/bus/dsa/drivers/idxd/unbind and
808 /sys/bus/dsa/drivers/crypto/unbind are used to disable IAA devices and
811 The basic sequence of commands needed to set up the IAA devices and
815 1) Disable any workqueues enabled on the device. For example to
816 disable workques 0 and 1 on IAA device 3::
818 # echo wq3.0 > /sys/bus/dsa/drivers/crypto/unbind
819 # echo wq3.1 > /sys/bus/dsa/drivers/crypto/unbind
821 2) Disable the device. For example to disable IAA device 3::
823 # echo iax3 > /sys/bus/dsa/drivers/idxd/unbind
825 3) configure the desired workqueues. For example, to configure
826 workqueue 3 on IAA device 3::
828 # echo dedicated > /sys/bus/dsa/devices/iax3/wq3.3/mode
829 # echo 128 > /sys/bus/dsa/devices/iax3/wq3.3/size
830 # echo 0 > /sys/bus/dsa/devices/iax3/wq3.3/group_id
831 # echo 10 > /sys/bus/dsa/devices/iax3/wq3.3/priority
832 # echo "kernel" > /sys/bus/dsa/devices/iax3/wq3.3/type
833 # echo "iaa_crypto" > /sys/bus/dsa/devices/iax3/wq3.3/name
834 # echo "crypto" > /sys/bus/dsa/devices/iax3/wq3.3/driver_name
836 4) Enable the device. For example to enable IAA device 3::
838 # echo iax3 > /sys/bus/dsa/drivers/idxd/bind
840 5) Enable the desired workqueues on the device. For example to
841 enable workques 0 and 1 on IAA device 3::
843 # echo wq3.0 > /sys/bus/dsa/drivers/crypto/bind
844 # echo wq3.1 > /sys/bus/dsa/drivers/crypto/bind