| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20250926033502.7486-1-kanchana.p.sridhar@intel.com>
Date: Thu, 25 Sep 2025 20:34:39 -0700
From: Kanchana P Sridhar <kanchana.p.sridhar@...el.com>
To: linux-kernel@...r.kernel.org,
linux-mm@...ck.org,
hannes@...xchg.org,
yosry.ahmed@...ux.dev,
nphamcs@...il.com,
chengming.zhou@...ux.dev,
usamaarif642@...il.com,
ryan.roberts@....com,
21cnbao@...il.com,
ying.huang@...ux.alibaba.com,
akpm@...ux-foundation.org,
senozhatsky@...omium.org,
sj@...nel.org,
kasong@...cent.com,
linux-crypto@...r.kernel.org,
herbert@...dor.apana.org.au,
davem@...emloft.net,
clabbe@...libre.com,
ardb@...nel.org,
ebiggers@...gle.com,
surenb@...gle.com,
kristen.c.accardi@...el.com,
vinicius.gomes@...el.com
Cc: wajdi.k.feghali@...el.com,
vinodh.gopal@...el.com,
kanchana.p.sridhar@...el.com
Subject: [PATCH v12 00/23] zswap compression batching with optimized iaa_crypto driver
v12: zswap compression batching with optimized iaa_crypto driver
Based on Herbert's suggestions, the most significant change in v12 is, the
batching interfaces from zswap to crypto, from crypto to iaa_crypto, and
the batching implementation within iaa_crypto now use the folio directly as the
source (sg_page_iter for retrieving pages) and destination SG lists. A unit_size
has been added to struct acomp_req, with kernel users such as zswap using the
new acomp_request_set_unit_size() API to set the unit size to use while breaking
down the request's src/dst scatterlists. zswap sets the unit-size to PAGE_SIZE.
Following Andrew's suggestion, the next two paragraphs emphasize generality and
alignment with current kernel efforts.
Architectural considerations for the zswap batching framework:
==============================================================
We have designed the zswap batching framework to be
hardware-agnostic. It has no dependencies on Intel-specific features and
can be leveraged by any hardware accelerator or software-based
compressor. In other words, the framework is open and inclusive by
design.
Other ongoing work that can use batching:
=========================================
This patch-series demonstrates the performance benefits of compress
batching when used in zswap_store() of large folios. shrink_folio_list()
"reclaim batching" of any-order folios is the next major work that uses
this zswap compress batching framework: our testing of kernel_compilation
with writeback and the zswap shrinker indicates 10X fewer pages get
written back when we reclaim 32 folios as a batch, as compared to one
folio at a time: this is with deflate-iaa and with zstd. We expect to
submit a patch-series with this data and the resulting performance
improvements shortly. Reclaim batching relieves memory pressure faster
than reclaiming one folio at a time, hence alleviates the need to scan
slab memory for writeback.
Many thanks to Nhat for suggesting ideas on using batching with the
ongoing kcompressd work, as well as beneficially using decompression
batching & block IO batching to improve zswap writeback efficiency.
Experiments with kernel compilation benchmark (allmod config) that
combine zswap compress batching, reclaim batching, swapin_readahead()
decompression batching of prefetched pages, and writeback batching show
that 0 pages are written back to disk with deflate-iaa and zstd. For
comparison, the baselines for these compressors see 200K-800K pages
written to disk.
To summarize, these are future clients of the batching framework:
- shrink_folio_list() reclaim batching of multiple folios:
Implemented, will submit patch-series.
- zswap writeback with decompress batching:
Implemented, will submit patch-series.
- zram:
Implemented, will submit patch-series.
- kcompressd:
Not yet implemented.
- file systems:
Not yet implemented.
- swapin_readahead() decompression batching of prefetched pages:
Implemented, will submit patch-series.
iaa_crypto Driver Rearchitecting and Optimizations:
===================================================
The most significant highlight of v11 is a new, lightweight and highly
optimized iaa_crypto driver, resulting directly in the latency and
throughput improvements noted later in this cover letter.
1) Better stability, more functionally versatile to support zswap and
zram with better performance on different Intel platforms.
a) Patches 0002, 0005 and 0010 together resolve a race condition in
mainline v6.15, reported from internal validation, when IAA
wqs/devices are disabled while workloads are using IAA.
b) Patch 0002 introduces a new architecture for mapping cores to
IAAs based on packages instead of NUMA nodes, and generalizing
how WQs are used: as package level shared resources for all
same-package cores (default for compress WQs), or dedicated to
mapped cores (default for decompress WQs). Further, users are
able to configure multiple WQs and specify how many of those are
for compress jobs only vs. decompress jobs only. sysfs iaa_crypto
driver parameters can be used to change the default settings for
performance tuning.
c) idxd descriptor allocation moved from blocking to non-blocking
with retry limits and mitigations if limits are exceeded.
d) Code cleanup for readability and clearer code flow.
e) Fixes IAA re-registration errors upon disabling/enabling IAA wqs
and devices that exists in the mainline v6.15.
f) Rearchitecting iaa_crypto to be independent of crypto_acomp to
enable a zram/zcomp backend_deflate_iaa.c, while fully supporting
the crypto_acomp interfaces for zswap. A new
include/linux/iaa_comp.h is added.
g) New Dynamic compression mode for Granite Rapids to get better
compression ratio by echo-ing 'deflate-iaa-dynamic' as the zswap
compressor.
h) New crypto_acomp API crypto_acomp_batch_size() that will return
the driver's max batch size if the driver has registered a batch_size
that's greater than 1; or 1 if there is no driver specific definition of
batch_size.
Accordingly, iaa_crypto sets the acomp_alg batch_size to its internal
IAA_CRYPTO_MAX_BATCH_SIZE for fixed and dynamic modes.
i) A versatile set of interfaces independent of crypto_acomp for use
in developing a zram zcomp backend for iaa_crypto.
2) Performance optimizations (please refer to the latency data per
optimization in the commit logs):
a) Distributing [de]compress jobs in round-robin manner to available
IAAs on package.
b) Replacing the compute-intensive iaa_wq_get()/iaa_wq_put() with a
percpu_ref in struct iaa_wq, thereby eliminating acquiring a
spinlock in the fast path, while using a combination of the
iaa_crypto_enabled atomic with spinlocks in the slow path to
ensure the compress/decompress code sees a consistent state of the
wq tables.
c) Directly call movdir64b for non-irq use cases, i.e., the most
common usage. Avoid the overhead of irq-specific computes in
idxd_submit_desc() to gain latency.
d) Batching of compressions/decompressions using async submit-poll
mechanism to derive the benefits of hardware parallelism.
e) Batching compressors need to manage their own "requests"
abstraction, and remove this driver-specific aspect from being
managed by kernel users such as zswap. iaa_crypto maintains
per-CPU "struct iaa_req **reqs" to submit multiple jobs to the
hardware accelerator to run in parallel.
f) Modifies the iaa_crypto batching API and their implementation to expect a
src SG list that contains the batch's pages and a dst SG list that has
multiple scatterlists for the batch's output buffers.
g) Submit the two largest data buffers first for decompression
batching, so that the longest running jobs get a head start,
reducing latency for the batch.
Main Changes in Zswap Compression Batching:
===========================================
Note to zswap maintainers:
--------------------------
Patches 20 and 21 can be reviewed and improved/merged independently
of this series, since they are zswap centric. These 2 patches help
batching but the crypto_acomp_batch_size() from the iaa_crypto commits
in this series is not a requirement, unlike patches 22-23.
1) v12 preserves the pool acomp_ctx resources creation/deletion
simplification of v11, namely, lasting from pool creation-deletion,
persisting through CPU hot[un]plug operations. Further, zswap no
longer needs to create multiple "struct acomp_req" in the per-CPU
acomp_ctx. zswap only needs to manage multiple "u8 **buffers".
2) We store the compressor's batch-size (@pool->compr_batch_size) and
the batch-size to use during compression batching
(@pool->store_batch_size) directly in struct zswap_pool for quick
retrieval in the zswap_store() fast path.
3) Optimizations to not cause regressions in software compressors with
the introduction of the new unified zswap_compress() framework that
implements compression batching for all compressors. These optimizations
help recover the performance for non-batching compressors:
a) kmem_cache_alloc_bulk(), kmem_cache_free_bulk() to allocate/free
batch zswap_entry-s. These kmem_cache API allow allocator
optimizations with internal locks for multiple allocations.
b) Writes to the zswap_entry right after it is allocated without
modifying the publishing order. This avoids different code blocks
in zswap_store_pages() having to bring the zswap_entries to the
cache for writing, potentially evicting other working set
structures, impacting performance.
c) ZSWAP_MAX_BATCH_SIZE is used as the batch-size for software
compressors, since this gives the best performance with zstd.
d) More likely()/unlikely() annotations to try and minimize branch
mis-predicts.
4) During pool creation, these key additions are allocated as part of the
per-CPU acomp_ctx so as to recover performance with the new, generalized SG
lists based zswap_compress() batching interface:
a) An sg_table "acomp_ctx->sg_outputs" is allocated to contain the
compressor's batch-size number of SG lists that will contain the
destination buffers/lengths after batch compression.
b) The per-CPU destination buffers are mapped to the per-CPU SG lists: this
needs to be done only once, and optimizes performance.
c) Only for batching compressors, extra memory is allocated in the
acomp_req, so as to avail of struct acomp_req's "void *__ctx[]" to store
pool->compr_batch_size number of int pointers to the
acomp_ctx->sg_outputs-sgl's component scatterlists. This is also added
to optimize/recover performance by avoiding traversing the sg_outputs per
batch compress call. Without this optimization, it is difficult to
realize the benefits of batching within large folios.
5) Finally, zswap_compress() has been re-written to incorporate Herbert's
suggestions to use source folios and output SG lists for batching. The new
zswap_compress() code has been made as generic to software and batching
compressors as possible, so that it is easy to read and maintain. The
recent changes related to PAGE_SIZE dst buffers, zsmalloc and incompressible
pages have been incorporated into the batched zswap_compress() as well! To
resolve regressions with zstd, I took the liberty of not explicitly checking
for dlen == 0 and dlen > PAGE_SIZE (as is assumed in the mainline); instead,
expecting that a negative err value will be returned by the software
compressor in such cases.
Compression Batching:
=====================
This patch-series introduces batch compression of pages in large folios to
improve zswap swapout latency. It preserves the existing zswap protocols
for non-batching software compressors by calling crypto_acomp sequentially
per page in the batch. Additionally, in support of hardware accelerators
that can process a batch as an integral unit, the patch-series allows
zswap to call crypto_acomp without API changes, for compressors
that intrinsically support batching. The zswap_compress() code has very minimal
special casing for software/batching compressors.
The patch series provides a proof point by using the Intel Analytics
Accelerator (IAA) for implementing the compress/decompress batching API
using hardware parallelism in the iaa_crypto driver and another proof point
with a sequential software compressor, zstd.
SUMMARY:
========
The first proof point is to test with IAA using a sequential call (fully
synchronous, compress one page at a time) vs. a batching call (fully
asynchronous, submit a batch to IAA for parallel compression, then poll for
completion statuses).
The performance testing data with 30 usemem processes/64K folios
shows 49% throughput gains and 25% elapsed/sys time reductions with
deflate-iaa; and 1% sys time reduction with zstd for a small
throughput increase. For PMD folios, a 55% throughput gain and 15%
elapsed/sys time reduction is seen.
Kernel compilation test with 64K folios using 32 threads and the
zswap shrinker_enabled set to "N", demonstrates similar
improvements: zswap_store() large folios using IAA compress batching
improves the workload performance by 6.1% and reduces sys time by
1% as compared to IAA sequential. For zstd, compress batching
improves workload performance by 7.8% and reduces sys time by
2.8% as compared to sequentially calling zswap_compress() per page
in a folio.
The main takeaway from usemem, a workload that is mostly compression
dominated (very few swapins) is that the higher the number of batches,
such as, with larger folios, the more the benefit of batching cost
amortization, as shown by the PMD usemem data. This aligns well with the
future direction for batching.
The second proof point is to make sure that software algorithms such as
zstd do not regress. The data indicates that for sequential software
algorithms a performance gain is achieved.
With the performance optimizations implemented in patches 22-23 of v12:
* zstd usemem metrics with 64K folios are within range of variation
with a slight sys time improvement. zstd usemem30 throughput with PMD
folios increases by 1%, workload performance improves by 6% and sys time
reduces by 8%.
* With kernel compilation, I used zstd without the zswap shrinker to enable
more direct comparisons with the changes in this series. Subsequent patch
series I expect to submit in collaboration with Nhat, will enable the
zswap shrinker to quantify the benefits of decompression batching during
writeback. With this series' compression batching within large folios, we
get a 1%-2.8% reduction in sys time, a 6.1%-7.8% improvement in workload
performance with 64K folios for deflate-iaa/zstd respectively.
These optimizations pertain to ensuring common code paths, removing
redundant branches/computes, selectively annotating branches with
likely()/unlikely() compiler directives to minimize branch
mis-prediction penalty. Additionally, using the batching code for
non-batching compressors to sequentially compress/store batches of up
to ZSWAP_MAX_BATCH_SIZE pages seems to help, most likely due to
cache locality of working set structures such as the array of
zswap_entry-s for the batch.
Our internal validation of zstd with the batching interface vs. IAA with
the batching interface on Emerald Rapids has shown that IAA
compress/decompress batching gives 21.3% more memory savings as compared
to zstd, for 5% performance loss as compared to the baseline without any
memory pressure. IAA batching demonstrates more than 2X the memory
savings obtained by zstd at this 95% performance KPI.
The compression ratio with IAA is 2.23, and with zstd 2.96. Even with
this compression ratio deficit for IAA, batching is extremely
beneficial. As we improve the compression ratio of the IAA accelerator,
we expect to see even better memory savings with IAA as compared to
software compressors.
Batching Roadmap:
=================
1) Compression batching within large folios (this series).
2) zswap writeback decompression batching:
This is being co-developed with Nhat Pham, and shows promising
results. We plan to submit an RFC shortly.
3) Reclaim batching of hybrid folios:
We can expect to see even more significant performance and throughput
improvements if we use the parallelism offered by IAA to do reclaim
batching of 4K/large folios (really any-order folios), and using the
zswap_store() high throughput compression pipeline to batch-compress
pages comprising these folios, not just batching within large
folios. This is the reclaim batching patch 13 in v1, which we expect
to submit in a separate patch-series. As mentioned earlier, reclaim
batching reduces the # of writeback pages by 10X for zstd and
deflate-iaa.
4) swapin_readahead() decompression batching:
We have developed a zswap load batching interface to be used
for parallel decompression batching, using swapin_readahead().
These capabilities are architected so as to be useful to zswap and
zram. We have integrated these components with zram and expect to submit an
RFC soon.
v12 Performance Summary:
========================
This is a performance testing summary of results with usemem30
(30 usemem processes running in a cgroup limited at 150G, each trying to
allocate 10G).
usemem30 with 64K folios:
=========================
zswap shrinker_enabled = N.
-----------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-----------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-----------------------------------------------------------------------
Total throughput (KB/s) 7,191,949 10,702,115 49%
Average throughput (KB/s) 239,731 356,737 49%
elapsed time (sec) 93.21 69.98 -25%
sys time (sec) 2,190.52 1,651.51 -25%
-----------------------------------------------------------------------
-----------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-----------------------------------------------------------------------
zswap compressor zstd zstd v12 zstd
improvement
-----------------------------------------------------------------------
Total throughput (KB/s) 6,258,312 6,269,511 0.2%
Average throughput (KB/s) 208,610 208,983 0.2%
elapsed time (sec) 100.01 100.50 0.5%
sys time (sec) 2,505.14 2,490.00 -1%
-----------------------------------------------------------------------
usemem30 with 2M folios:
========================
-----------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-----------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-----------------------------------------------------------------------
Total throughput (KB/s) 7,237,676 11,228,928 55%
Average throughput (KB/s) 241,255 374,297 55%
elapsed time (sec) 82.26 69.30 -16%
sys time (sec) 1,901.90 1,634.78 -14%
-----------------------------------------------------------------------
-----------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-----------------------------------------------------------------------
zswap compressor zstd zstd v12 zstd
improvement
-----------------------------------------------------------------------
Total throughput (KB/s) 6,796,376 6,865,781 1.0%
Average throughput (KB/s) 226,545 228,859 1.0%
elapsed time (sec) 94.07 88.80 -6%
sys time (sec) 2,261.67 2,082.91 -8%
-----------------------------------------------------------------------
This is a performance testing summary of results with
kernel_compilation test (allmod config, 32 cores, cgroup limited to 2G).
zswap shrinker_enabled = N.
kernel_compilation with 64K folios:
===================================
--------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
--------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
--------------------------------------------------------------------------
real_sec 874.74 821.59 -6.1%
sys_sec 3,834.35 3,791.12 -1%
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
--------------------------------------------------------------------------
zswap compressor zstd zstd Improvement
--------------------------------------------------------------------------
real_sec 925.08 853.14 -7.8%
sys_sec 5,318.65 5,172.23 -2.8%
--------------------------------------------------------------------------
kernel_compilation with PMD folios:
===================================
--------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
--------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
--------------------------------------------------------------------------
real_sec 808.10 794.85 -1.6%
sys_sec 4,351.01 4,266.95 -2%
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
--------------------------------------------------------------------------
zswap compressor zstd zstd Improvement
--------------------------------------------------------------------------
real_sec 848.06 845.42 -0.3%
sys_sec 5,898.58 5,741.31 -2.7%
--------------------------------------------------------------------------
DETAILS:
========
(A) From zswap's perspective, the most significant changes are:
===============================================================
1) A unified zswap_compress() API is added to compress multiple
pages:
- If the compressor has multiple acomp requests, i.e., internally
supports batching, crypto_acomp_batch_compress() is called. If all
pages are successfully compressed, or, if folio writeback is enabled, the
batch is stored in zsmalloc, preserving latest mm-stable behavior.
- If the compressor can only compress one page at a time, each page
is compressed and stored sequentially.
Many thanks to Yosry for this suggestion, because it is an essential
component of unifying common code paths between sequential/batching
compressions.
Overall, the likely()/unlikely() annotations prevent regressions with
software compressors like zstd, and generally improve non-batching
compressors' performance with the batching code by 1% to 8%.
2) A new zswap_store_pages() is added, that stores multiple pages in a
folio in a range of indices. This is an extension of the earlier
zswap_store_page(), except it operates on a batch of pages.
3) zswap_store() is modified to store the folio's pages in batches
by calling zswap_store_pages(). If the compressor supports batching,
the folio will be compressed in batches of
"pool->compr_batch_size". If the compressor does not support
batching, the folio will be compressed in batches of
ZSWAP_MAX_BATCH_SIZE pages, where each page in the batch is
compressed sequentially. We see better performance by processing
the folio in batches of ZSWAP_MAX_BATCH_SIZE, due to cache locality
of working set structures such as the array of zswap_entry-s for the
batch.
Many thanks to Yosry and Johannes for steering towards a common
design and code paths for sequential and batched compressions (i.e.,
for software compressors and hardware accelerators such as IAA). As per
Yosry's suggestion in v8, the "batch_size" is an attribute of the
compressor/pool, and hence is stored in struct zswap_pool instead of
in struct crypto_acomp_ctx.
4) Simplifications to the acomp_ctx resources allocation/deletion
vis-a-vis CPU hot[un]plug. This further improves upon v8 of this
patch-series based on the discussion with Yosry, and formalizes the
lifetime of these resources from pool creation to pool
deletion. zswap does not register a CPU hotplug teardown
callback. The acomp_ctx resources will persist through CPU
online/offline transitions. The main changes made to avoid UAF/race
conditions, and correctly handle process migration, are:
a) No acomp_ctx mutex locking in zswap_cpu_comp_prepare().
b) No CPU hotplug teardown callback, no acomp_ctx resources deleted.
c) New acomp_ctx_dealloc() procedure that cleans up the acomp_ctx
resources, and is shared by
zswap_cpu_comp_prepare()/zswap_pool_create() error handling and
zswap_pool_destroy().
d) The zswap_pool node list instance is removed right after the node
list add function in zswap_pool_create().
e) We directly call mutex_[un]lock(&acomp_ctx->mutex) in
zswap_[de]compress(). acomp_ctx_get_cpu_lock()/acomp_ctx_put_unlock()
are deleted.
The commit log of patch 0020 has a more detailed analysis.
(B) Main changes in crypto_acomp and iaa_crypto:
================================================
1) A new architecture is introduced for IAA device WQs' usage as:
- compress only
- decompress only
- generic, i.e., both compress/decompress.
Further, IAA devices/wqs are assigned to cores based on packages
instead of NUMA nodes.
The WQ rebalancing algorithm that is invoked as WQs are
discovered/deleted has been made very general and flexible so that
the user can control exactly how IAA WQs are used. In addition to the
user being able to specify a WQ type as comp/decomp/generic, the user
can also configure if WQs need to be shared among all same-package
cores, or, whether the cores should be divided up amongst the
available IAA devices.
If distribute_[de]comps is enabled, from a given core's perspective,
the iaa_crypto driver will distribute comp/decomp jobs among all
devices' WQs in round-robin manner. This improves batching latency
and can improve compression/decompression throughput for workloads
that see a lot of swap activity.
The commit log of patch 0002 provides more details on new iaa_crypto
driver parameters added, along with recommended settings (defaults
are optimal settings).
2) Compress/decompress batching are implemented using
crypto_acomp_[de]compress() with SG lists used as the batching interface.
(C) The patch-series is organized as follows:
=============================================
1) crypto acomp & iaa_crypto driver enablers for batching: Relevant
patches are tagged with "crypto:" in the subject:
Patch 1) Reorganizes the iaa_crypto driver code into logically related
sections and avoids forward declarations, in order to facilitate
subsequent iaa_crypto patches. This patch makes no
functional changes.
Patch 2) Makes an infrastructure change in the iaa_crypto driver
to map IAA devices/work-queues to cores based on packages
instead of NUMA nodes. This doesn't impact performance on
the Sapphire Rapids system used for performance
testing. However, this change fixes functional problems we
found on Granite Rapids during internal validation, where the
number of NUMA nodes is greater than the number of packages,
which was resulting in over-utilization of some IAA devices
and non-usage of other IAA devices as per the current NUMA
based mapping infrastructure.
This patch also develops a new architecture that
generalizes how IAA device WQs are used. It enables
designating IAA device WQs as either compress-only or
decompress-only or generic. Once IAA device WQ types are
thus defined, it also allows the configuration of whether
device WQs will be shared by all cores on the package, or
used only by "mapped cores" obtained by a simple allocation
of available IAAs to cores on the package.
As a result of the overhaul of wq_table definition,
allocation and rebalancing, this patch eliminates
duplication of device WQs in per-CPU wq_tables, thereby
saving 140MiB on a 384 cores dual socket Granite Rapids server
with 8 IAAs.
Regardless of how the user has configured the WQs' usage,
the next WQ to use is obtained through a direct look-up in
per-CPU "cpu_comp_wqs" and "cpu_decomp_wqs" structures so
as to minimize latency in the critical path driver compress
and decompress routines.
Patch 3) Code cleanup, consistency of function parameters.
Patch 4) Makes a change to iaa_crypto driver's descriptor allocation,
from blocking to non-blocking with retries/timeouts and
mitigations in case of timeouts during compress/decompress
ops. This prevents tasks getting blocked indefinitely, which
was observed when testing 30 cores running workloads, with
only 1 IAA enabled on Sapphire Rapids (out of 4). These
timeouts are typically only encountered, and associated
mitigations exercised, only in configurations with 1 IAA
device shared by 30+ cores.
Patch 5) Optimize iaa_wq refcounts using a percpu_ref instead of
spinlocks and "int refcount".
Patch 6) Code simplification and restructuring for understandability
in core iaa_compress() and iaa_decompress() routines.
Patch 7) Refactor hardware descriptor setup to their own procedures
to reduce code clutter.
Patch 8) Simplify and optimize (i.e. reduce computes) job submission
for the most commonly used non-irq async mode by directly
calling movdir64b.
Patch 9) Deprecate exporting symbols for adding IAA compression
modes.
Patch 10) All dma_map_sg() calls will pass in 1 for the nents instead of
sg_nents(), for these main reasons: performance; no existing
iaa_crypto use cases that allow multiple SG lists to be mapped for
dma at once; facilitates new SG lists batching interface through
crypto.
Patch 11) Rearchitect iaa_crypto to be agnostic of crypto_acomp for
it be usable in both zswap and zram. crypto_acomp interfaces are
maintained as earlier, for use in zswap.
Patch 12) Descriptor submit and polling mechanisms, enablers for batching.
Patch 13) Define a unit_size in struct acomp_req to enable batching, and
provides acomp_request_set_unit_size() for use by kernel modules.
Patch 14) Implement IAA batching of compressions and decompressions
for deriving hardware parallelism.
Patch 15) Enables the "async" mode, sets it as the default.
Patch 16) Disables verify_compress by default.
Patch 17) Decompress batching optimization: Find the two largest
buffers in the batch and submit them first.
Patch 18) Add a new Dynamic compression mode that can be used on
Granite Rapids.
Patch 19) Add a batch_size data member to structs acomp_alg/crypto_acomp and
a crypto_acomp_batch_size() API that returns the compressor's
batch-size, if it has defined one; 1 otherwise. Add the definition
of batch_size to iaa_crypto acomp_algs.
2) zswap modifications to enable compress batching in zswap_store()
of large folios (including pmd-mappable folios):
Patch 20) Simplifies the zswap_pool's per-CPU acomp_ctx resource
management and lifetime to be from pool creation to pool
deletion.
Patch 21) Uses IS_ERR_OR_NULL() in zswap_cpu_comp_prepare() to check for
valid acomp/req, thereby making it consistent with the resource
de-allocation code.
Patch 22) Defines a zswap-specific ZSWAP_MAX_BATCH_SIZE (currently set
as 8U) to denote the maximum number of acomp_ctx batching
resources to allocate, thus limiting the amount of extra
memory used for batching. Further, the "struct
crypto_acomp_ctx" is modified to contain multiple buffers.
New "u8 compr_batch_size" and "u8 store_batch_size" members are
added to "struct zswap_pool" to track the number of dst
buffers associated with the compressor (more than 1 if
the compressor supports batching) and the unit for storing
large folios using compression batching respectively.
Modifies zswap_store() to store the folio in batches of
pool->store_batch_size by calling a new zswap_store_pages() that
takes a range of indices in the folio to be stored.
zswap_store_pages() pre-allocates zswap entries for the batch,
calls zswap_compress() for each page in this range, and stores
the entries in xarray/LRU.
Patch 23) Introduces a new unified batching implementation of
zswap_compress() for compressors that do and do not support
batching. This eliminates code duplication and facilitates
maintainability of the code with the introduction of compress
batching. Further, there are many optimizations to this common
code that result in workload throughput and performance
improvements with software compressors and hardware accelerators
such as IAA.
zstd performance is better or on par with mm-unstable. We
see impressive throughput/performance improvements with
IAA and zstd batching vs. no-batching.
With v12 of this patch series, the IAA compress batching feature will be
enabled seamlessly on Intel platforms that have IAA by selecting
'deflate-iaa' as the zswap compressor, and using the iaa_crypto 'async'
sync_mode driver attribute (the default).
System setup for testing:
=========================
Testing of this patch-series was done with mm-unstable as of 9-18-2025,
commit 1f98191f08b4, without and with this patch-series. Data was
gathered on an Intel Sapphire Rapids (SPR) server, dual-socket 56 cores
per socket, 4 IAA devices per socket, each IAA has total 128 WQ entries,
503 GiB RAM and 525G SSD disk partition swap. Core frequency was fixed
at 2500MHz.
Other kernel configuration parameters:
zswap compressor : zstd, deflate-iaa
zswap allocator : zsmalloc
vm.page-cluster : 0
IAA "compression verification" is disabled and IAA is run in the async
mode (the defaults with this series).
I ran experiments with these workloads:
1) usemem 30 processes with zswap shrinker_enabled=N. Two sets of
experiments, one with 64K folios, another with PMD folios.
2) Kernel compilation allmodconfig with 2G max memory, 32 threads, with
zswap shrinker_enabled=N to test batching performance impact in
isolation. Two sets of experiments, one with 64K folios, another with PMD
folios.
IAA configuration is done by a CLI: script is included at the end of the
cover letter.
Performance testing (usemem30):
===============================
The vm-scalability "usemem" test was run in a cgroup whose memory.high
was fixed at 150G. There is no swap limit set for the cgroup. 30 usemem
processes were run, each allocating and writing 10G of memory, and
sleeping for 10 sec before exiting:
usemem --init-time -w -O -b 1 -s 10 -n 30 10g
echo 0 > /sys/module/zswap/parameters/shrinker_enabled
IAA WQ Configuration (script is iincluded at the end of the cover
letter):
./enable_iaa.sh -d 4 -q 1
This enables all 4 IAAs on the socket, and configures 1 WQ per IAA
device, each containing 128 entries. The driver distributes compress
jobs from each core to wqX.0 of all same-package IAAs in a
round-robin manner. Decompress jobs are send to the wqX.0 of the
mapped IAA device.
Since usemem has significantly more swapouts than swapins, this
configuration is the most optimal.
64K folios: usemem30: deflate-iaa:
==================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-------------------------------------------------------------------------------
Total throughput (KB/s) 7,191,949 10,702,115 49%
Avg throughput (KB/s) 239,731 356,737 49%
elapsed time (sec) 93.21 69.98 -25%
sys time (sec) 2,190.52 1,651.51 -25%
-------------------------------------------------------------------------------
memcg_high 1,056,505 1,320,579
memcg_swap_fail 3,966 321
64kB_swpout_fallback 3,965 321
zswpout 61,577,440 71,448,317
zswpin 488 141
pswpout 0 0
pswpin 0 0
ZSWPOUT-64kB 3,844,593 4,465,129
SWPOUT-64kB 0 0
pgmajfault 2,786 2,454
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
IAA incompressible pages 0 0
-------------------------------------------------------------------------------
2M folios: usemem30: deflate-iaa:
=================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-------------------------------------------------------------------------------
Total throughput (KB/s) 7,237,676 11,228,928 55%
Avg throughput (KB/s) 241,255 374,297 55%
elapsed time (sec) 82.26 69.30 -16%
sys time (sec) 1,901.90 1,634.78 -14%
-------------------------------------------------------------------------------
memcg_high 100,141 119,644
memcg_swap_fail 356 12
thp_swpout_fallback 356 12
zswpout 63,470,927 71,773,688
zswpin 455 169
pswpout 0 0
pswpin 0 0
ZSWPOUT-2048kB 123,610 140,169
thp_swpout 0 0
pgmajfault 2,743 2,471
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
IAA incompressible pages 0 0
-------------------------------------------------------------------------------
64K folios: usemem30: zstd:
===========================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor zstd zstd v12 zstd
improvement
-------------------------------------------------------------------------------
Total throughput (KB/s) 6,258,312 6,269,511 0.2%
Avg throughput (KB/s) 208,610 208,983 0.2%
elapsed time (sec) 100.01 100.50 0.5%
sys time (sec) 2,505.14 2,490.00 -1%
-------------------------------------------------------------------------------
memcg_high 1,071,250 1,043,886
memcg_swap_fail 2,251 73
64kB_swpout_fallback 2,251 73
zswpout 49,491,137 49,693,220
zswpin 455 179
pswpout 0 0
pswpin 0 0
ZSWPOUT-64kB 3,090,904 3,105,690
SWPOUT-64kB 0 0
pgmajfault 2,780 2,456
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
-------------------------------------------------------------------------------
2M folios: usemem30: zstd:
==========================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor zstd zstd v12 zstd
improvement
-------------------------------------------------------------------------------
Total throughput (KB/s) 6,796,376 6,865,781 1.0%
Avg throughput (KB/s) 226,545 228,859 1.0%
elapsed time (sec) 94.07 88.80 -6%
sys time (sec) 2,261.67 2,082.91 -8%
-------------------------------------------------------------------------------
memcg_high 89,582 88,103
memcg_swap_fail 14 56
thp_swpout_fallback 14 56
zswpout 48,846,660 48,000,106
zswpin 58 266
pswpout 0 0
pswpin 0 0
ZSWPOUT-2048kB 95,388 93,692
thp_swpout 0 0
pgmajfault 2,348 2,557
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
-------------------------------------------------------------------------------
Performance testing (Kernel compilation, allmodconfig):
=======================================================
The experiments with kernel compilation test use 32 threads and build
the "allmodconfig" that takes ~14 minutes, and has considerable
swapout/swapin activity. The cgroup's memory.max is set to 2G. zswap
writeback is not enabled so as to isolate the performance impact of only large
folio batch compression.
echo 0 > /sys/module/zswap/parameters/shrinker_enabled
IAA WQ Configuration (script is at the end of the cover letter):
./enable_iaa.sh -d 4 -q 2
This enables all 4 IAAs on the socket, and configures 2 WQs per IAA,
each containing 64 entries. The driver sends decompresses to wqX.0 of
the mapped IAA device, and distributes compresses to wqX.1 of all
same-package IAAs in a round-robin manner.
64K folios: Kernel compilation/allmodconfig: deflate-iaa:
=========================================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-------------------------------------------------------------------------------
real_sec 874.74 821.59 -6.1%
user_sec 15,701.49 15,718.29
sys_sec 3,834.35 3,791.12 -1%
-------------------------------------------------------------------------------
Max_Res_Set_Size_KB 1,871,264 1,872,780
-------------------------------------------------------------------------------
memcg_high 0 0
memcg_swap_fail 0 0
64kB_swpout_fallback 0 0
zswpout 91,614,674 103,057,250
zswpin 27,028,688 30,946,488
pswpout 0 0
pswpin 0 0
ZSWPOUT-64kB 3,007,705 3,337,682
SWPOUT-64kB 0 0
pgmajfault 28,697,315 32,778,092
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
IAA incompressible pages 440 239
-------------------------------------------------------------------------------
2M folios: Kernel compilation/allmodconfig: deflate-iaa:
========================================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor deflate-iaa deflate-iaa IAA Batching
vs.
IAA Sequential
-------------------------------------------------------------------------------
real_sec 808.10 794.85 -1.6%
user_sec 15,807.59 15,794.49
sys_sec 4,351.01 4,266.95 -2%
-------------------------------------------------------------------------------
Max_Res_Set_Size_KB 1,871,100 1,874,684
-------------------------------------------------------------------------------
memcg_high 0 0
memcg_swap_fail 0 0
thp_swpout_fallback 0 0
zswpout 109,563,983 111,586,692
zswpin 38,396,482 39,089,000
pswpout 0 0
pswpin 0 0
ZSWPOUT-2048kB 16,661 17,020
thp_swpout 0 0
pgmajfault 39,727,295 40,433,559
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
IAA incompressible pages 531 317
-------------------------------------------------------------------------------
With the iaa_crypto driver changes for non-blocking descriptor allocations,
no timeouts-with-mitigations were seen in compress/decompress jobs, for all
of the above experiments.
64K folios: Kernel compilation/allmodconfig: zstd:
==================================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor zstd zstd Improvement
-------------------------------------------------------------------------------
real_sec 925.08 853.14 -7.8%
user_sec 15,708.03 15,703.87
sys_sec 5,318.65 5,172.23 -2.8%
-------------------------------------------------------------------------------
Max_Res_Set_Size_KB 1,874,060 1,874,276
-------------------------------------------------------------------------------
memcg_high 0 0
memcg_swap_fail 0 0
64kB_swpout_fallback 0 0
zswpout 80,735,528 78,500,218
zswpin 24,155,464 22,898,395
pswpout 0 0
pswpin 0 0
ZSWPOUT-64kB 2,580,722 2,489,796
SWPOUT-64kB 0 0
pgmajfault 25,564,063 24,223,218
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
-------------------------------------------------------------------------------
2M folios: Kernel compilation/allmodconfig: zstd:
=================================================
-------------------------------------------------------------------------------
mm-unstable-9-18-2025 v12
-------------------------------------------------------------------------------
zswap compressor zstd zstd Improvement
-------------------------------------------------------------------------------
real_sec 848.06 845.42 -0.3%
user_sec 15,822.19 15,826.68
sys_sec 5,898.58 5,741.31 -2.7%
-------------------------------------------------------------------------------
Max_Res_Set_Size_KB 1,872,888 1,874,684
-------------------------------------------------------------------------------
memcg_high 0 0
memcg_swap_fail 0 0
thp_swpout_fallback 0 0
zswpout 91,399,677 87,856,850
zswpin 29,871,078 28,452,891
pswpout 0 0
pswpin 0 0
ZSWPOUT-2048kB 12,130 11,968
thp_swpout 0 0
pgmajfault 30,723,070 29,308,735
zswap_reject_compress_fail 0 0
zswap_reject_reclaim_fail 0 0
-------------------------------------------------------------------------------
IAA configuration script "enable_iaa.sh":
=========================================
Acknowledgements: Binuraj Ravindran and Rakib Al-Fahad.
Usage:
------
./enable_iaa.sh -d <num_IAAs> -q <num_WQs_per_IAA>
#---------------------------------<cut here>----------------------------------
#!/usr/bin/env bash
#SPDX-License-Identifier: BSD-3-Clause
#Copyright (c) 2025, Intel Corporation
#Description: Configure IAA devices
VERIFY_COMPRESS_PATH="/sys/bus/dsa/drivers/crypto/verify_compress"
iax_dev_id="0cfe"
num_iaa=$(lspci -d:${iax_dev_id} | wc -l)
sockets=$(lscpu | grep Socket | awk '{print $2}')
echo "Found ${num_iaa} instances in ${sockets} sockets(s)"
# The same number of devices will be configured in each socket, if there
# are more than one socket.
# Normalize with respect to the number of sockets.
device_num_per_socket=$(( num_iaa/sockets ))
num_iaa_per_socket=$(( num_iaa / sockets ))
iaa_wqs=2
verbose=0
iaa_engines=8
mode="dedicated"
wq_type="kernel"
iaa_crypto_mode="async"
verify_compress=0
# Function to handle errors
handle_error() {
echo "Error: $1"
exit 1
}
# Process arguments
while getopts "d:hm:q:vD" opt; do
case $opt in
d)
device_num_per_socket=$OPTARG
;;
m)
iaa_crypto_mode=$OPTARG
;;
q)
iaa_wqs=$OPTARG
;;
D)
verbose=1
;;
v)
verify_compress=1
;;
h)
echo "Usage: $0 [-d <device_count>][-q <wq_per_device>][-v]"
echo " -d - number of devices"
echo " -q - number of WQs per device"
echo " -v - verbose mode"
echo " -h - help"
exit
;;
\?)
echo "Invalid option: -$OPTARG" >&2
exit
;;
esac
done
LOG="configure_iaa.log"
# Update wq_size based on number of wqs
wq_size=$(( 128 / iaa_wqs ))
# Take care of the enumeration, if DSA is enabled.
dsa=`lspci | grep -c 0b25`
# set first,step counters to correctly enumerate iax devices based on
# whether running on guest or host with or without dsa
first=0
step=1
[[ $dsa -gt 0 && -d /sys/bus/dsa/devices/dsa0 ]] && first=1 && step=2
echo "first index: ${first}, step: ${step}"
#
# Switch to software compressors and disable IAAs to have a clean start
#
COMPRESSOR=/sys/module/zswap/parameters/compressor
last_comp=`cat ${COMPRESSOR}`
echo lzo > ${COMPRESSOR}
echo "Disable IAA devices before configuring"
for ((i = ${first}; i < ${step} * ${num_iaa}; i += ${step})); do
for ((j = 0; j < ${iaa_wqs}; j += 1)); do
cmd="accel-config disable-wq iax${i}/wq${i}.${j} >& /dev/null"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
done
cmd="accel-config disable-device iax${i} >& /dev/null"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
done
rmmod iaa_crypto
modprobe iaa_crypto
# apply crypto parameters
echo $verify_compress > ${VERIFY_COMPRESS_PATH} || handle_error "did not change verify_compress"
# Note: This is a temporary solution for during the kernel transition.
if [ -f /sys/bus/dsa/drivers/crypto/g_comp_wqs_per_iaa ];then
echo 1 > /sys/bus/dsa/drivers/crypto/g_comp_wqs_per_iaa || handle_error "did not set g_comp_wqs_per_iaa"
elif [ -f /sys/bus/dsa/drivers/crypto/g_wqs_per_iaa ];then
echo 1 > /sys/bus/dsa/drivers/crypto/g_wqs_per_iaa || handle_error "did not set g_wqs_per_iaa"
fi
if [ -f /sys/bus/dsa/drivers/crypto/g_consec_descs_per_gwq ];then
echo 1 > /sys/bus/dsa/drivers/crypto/g_consec_descs_per_gwq || handle_error "did not set g_consec_descs_per_gwq"
fi
echo ${iaa_crypto_mode} > /sys/bus/dsa/drivers/crypto/sync_mode || handle_error "could not set sync_mode"
echo "Configuring ${device_num_per_socket} device(s) out of $num_iaa_per_socket per socket"
if [ "${device_num_per_socket}" -le "${num_iaa_per_socket}" ]; then
echo "Configuring all devices"
start=${first}
end=$(( ${step} * ${device_num_per_socket} ))
else
echo "ERROR: Not enough devices"
exit
fi
#
# enable all iax devices and wqs
#
for (( socket = 0; socket < ${sockets}; socket += 1 )); do
for ((i = ${start}; i < ${end}; i += ${step})); do
echo "Configuring iaa$i on socket ${socket}"
for ((j = 0; j < ${iaa_engines}; j += 1)); do
cmd="accel-config config-engine iax${i}/engine${i}.${j} --group-id=0"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
done
# Config WQs
for ((j = 0; j < ${iaa_wqs}; j += 1)); do
# Config WQ: group 0, priority=10, mode=shared, type = kernel name=kernel, driver_name=crypto
cmd="accel-config config-wq iax${i}/wq${i}.${j} -g 0 -s ${wq_size} -p 10 -m ${mode} -y ${wq_type} -n iaa_crypto${i}${j} -d crypto"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
done
# Enable Device and WQs
cmd="accel-config enable-device iax${i}"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
for ((j = 0; j < ${iaa_wqs}; j += 1)); do
cmd="accel-config enable-wq iax${i}/wq${i}.${j}"
[[ $verbose == 1 ]] && echo $cmd; eval $cmd
done
done
start=$(( start + ${step} * ${num_iaa_per_socket} ))
end=$(( start + (${step} * ${device_num_per_socket}) ))
done
# Restore the last compressor
echo "$last_comp" > ${COMPRESSOR}
# Check if the configuration is correct
echo "Configured IAA devices:"
accel-config list | grep iax
#---------------------------------<cut here>----------------------------------
Changes since v11:
==================
1) Rebased to mm-unstable as of 9-18-2025, commit 1f98191f08b4.
2) Incorporated Herbert's suggestions on submitting the folio as the source and
SG lists for the destination to create the compress batching interface from
zswap to crypto.
3) As per Herbert's suggestion, added a new unit_size member to struct
acomp_req, along with a acomp_request_set_unit_size() API for kernel modules
to set the unit size to use while breaking down the request's src/dst
scatterlists.
4) Implemented iaa_crypto batching using the new SG lists based architecture and
crypto interfaces.
5) To make the SG lists based approach functional and performant for IAA, I have
changed all the calls to dma_map_sg() to use nents of 1. This should not be a
concern, since it eliminates redundant computes to scan an SG list with only
one scatterlist for existing kernel users, i.e. zswap with the
zswap_compress() modifications in this series. This will continue to hold
true with the zram IAA batching support I am developing. There are no kernel
use cases for the iaa_crypto driver that will break this assumption.
6) Addressed Herbert's comment about batch_size being a statically defined data
member in struct acomp_alg and struct crypto_acomp.
7) Addressed Nhat's comment about VM_WARN_ON_ONCE(nr_pages >
ZSWAP_MAX_BATCH_SIZE) in zswap_store_pages().
8) Nhat's comment about deleting struct swap_batch_decomp_data is automatically
addressed by the SG lists based rewrite of the crypto batching interface.
9) Addressed Barry's comment about renaming pool->batch_size to
pool->store_batch_size.
10) Incorporated Barry's suggestion to merge patches that introduce data members
to structures and/or API and their usage.
11) Added performance data to patch 0023's commit log, as suggested by Barry.
Changes since v10:
==================
1) Rebased to mm-unstable as of 7-30-2025, commit 01da54f10fdd.
2) Added change logging in patch 0024 on there being no Intel-specific
dependencies in the batching framework, as suggested by
Andrew Morton. Thanks Andrew!
3) Added change logging in patch 0024 on other ongoing work that can use
batching, as per Andrew's suggestion. Thanks Andrew!
4) Added the IAA configuration script in the cover letter, as suggested
by Nhat Pham. Thanks Nhat!
5) As suggested by Nhat, dropped patch 0020 from v10, that moves CPU
hotplug procedures to pool functions.
6) Gathered kernel_compilation 'allmod' config performance data with
writeback and zswap shrinker_enabled=Y.
7) Changed the pool->batch_size for software compressors to be
ZSWAP_MAX_BATCH_SIZE since this gives better performance with the zswap
shrinker enabled.
8) Was unable to replicate in v11 the issue seen in v10 with higher
memcg_swap_fail than in the baseline, with usemem30/zstd.
Changes since v9:
=================
1) Rebased to mm-unstable as of 6-24-2025, commit 23b9c0472ea3.
2) iaa_crypto rearchitecting, mainline race condition fix, performance
optimizations, code cleanup.
3) Addressed Herbert's comments in v9 patch 10, that an array based
crypto_acomp interface is not acceptable.
4) Optimized the implementation of the batching zswap_compress() and
zswap_store_pages() added in v9, to recover performance when
integrated with the changes in commit 56e5a103a721 ("zsmalloc: prefer
the the original page's node for compressed data").
Changes since v8:
=================
1) Rebased to mm-unstable as of 4-21-2025, commit 2c01d9f3c611.
2) Backported commits for reverting request chaining, since these are
in cryptodev-2.6 but not yet in mm-unstable: without these backports,
deflate-iaa is non-functional in mm-unstable:
commit 64929fe8c0a4 ("crypto: acomp - Remove request chaining")
commit 5976fe19e240 ("Revert "crypto: testmgr - Add multibuffer acomp
testing"")
Backported this hotfix as well:
commit 002ba346e3d7 ("crypto: scomp - Fix off-by-one bug when
calculating last page").
3) crypto_acomp_[de]compress() restored to non-request chained
implementations since request chaining has been removed from acomp in
commit 64929fe8c0a4 ("crypto: acomp - Remove request chaining").
4) New IAA WQ architecture to denote WQ type and whether or not a WQ
should be shared among all package cores, or only to the "mapped"
ones from an even cores-to-IAA distribution scheme.
5) Compress/decompress batching are implemented in iaa_crypto using new
crypto_acomp_batch_compress()/crypto_acomp_batch_decompress() API.
6) Defines a "void *data" in struct acomp_req, based on Herbert advising
against using req->base.data in the driver. This is needed for async
submit-poll to work.
7) In zswap.c, moved the CPU hotplug callbacks to reside in "pool
functions", per Yosry's suggestion to move procedures in a distinct
patch before refactoring patches.
8) A new "u8 nr_reqs" member is added to "struct zswap_pool" to track
the number of requests/buffers associated with the per-cpu acomp_ctx,
as per Yosry's suggestion.
9) Simplifications to the acomp_ctx resources allocation, deletion,
locking, and for these to exist from pool creation to pool deletion,
based on v8 code review discussions with Yosry.
10) Use IS_ERR_OR_NULL() consistently in zswap_cpu_comp_prepare() and
acomp_ctx_dealloc(), as per Yosry's v8 comment.
11) zswap_store_folio() is deleted, and instead, the loop over
zswap_store_pages() is moved inline in zswap_store(), per Yosry's
suggestion.
12) Better structure in zswap_compress(), unified procedure that
compresses/stores a batch of pages for both, non-batching and
batching compressors. Renamed from zswap_batch_compress() to
zswap_compress(): Thanks Yosry for these suggestions.
Changes since v7:
=================
1) Rebased to mm-unstable as of 3-3-2025, commit 5f089a9aa987.
2) Changed the acomp_ctx->nr_reqs to be u8 since ZSWAP_MAX_BATCH_SIZE is
defined as 8U, for saving memory in this per-cpu structure.
3) Fixed a typo in code comments in acomp_ctx_get_cpu_lock():
acomp_ctx->initialized to acomp_ctx->__online.
4) Incorporated suggestions from Yosry, Chengming, Nhat and Johannes,
thanks to all!
a) zswap_batch_compress() replaces zswap_compress(). Thanks Yosry
for this suggestion!
b) Process the folio in sub-batches of ZSWAP_MAX_BATCH_SIZE, regardless
of whether or not the compressor supports batching. This gets rid of
the kmalloc(entries), and allows us to allocate an array of
ZSWAP_MAX_BATCH_SIZE entries on the stack. This is implemented in
zswap_store_pages().
c) Use of a common structure and code paths for compressing a folio in
batches, either as a request chain (in parallel in IAA hardware) or
sequentially. No code duplication since zswap_compress() has been
replaced with zswap_batch_compress(), simplifying maintainability.
5) A key difference between compressors that support batching and
those that do not, is that for the latter, the acomp_ctx mutex is
locked/unlocked per ZSWAP_MAX_BATCH_SIZE batch, so that decompressions
to handle page-faults can make progress. This fixes the zstd kernel
compilation regression seen in v7. For compressors that support
batching, for e.g. IAA, the mutex is locked/released once for storing
the folio.
6) Used likely/unlikely compiler directives and prefetchw to restore
performance with the common code paths.
Changes since v6:
=================
1) Rebased to mm-unstable as of 2-27-2025, commit d58172d128ac.
2) Deleted crypto_acomp_batch_compress() and
crypto_acomp_batch_decompress() interfaces, as per Herbert's
suggestion. Batching is instead enabled by chaining the requests. For
non-batching compressors, there is no request chaining involved. Both,
batching and non-batching compressions are accomplished by zswap by
calling:
crypto_wait_req(crypto_acomp_compress(acomp_ctx->reqs[0]), &acomp_ctx->wait);
3) iaa_crypto implementation of batch compressions/decompressions using
request chaining, as per Herbert's suggestions.
4) Simplification of the acomp_ctx resource allocation/deletion with
respect to CPU hot[un]plug, to address Yosry's suggestions to explore the
mutex options in zswap_cpu_comp_prepare(). Yosry, please let me know if
the per-cpu memory cost of this proposed change is acceptable (IAA:
64.8KB, Software compressors: 8.2KB). On the positive side, I believe
restarting reclaim on a CPU after it has been through an offline-online
transition, will be much faster by not deleting the acomp_ctx resources
when the CPU gets offlined.
5) Use of lockdep assertions rather than comments for internal locking
rules, as per Yosry's suggestion.
6) No specific references to IAA in zswap.c, as suggested by Yosry.
7) Explored various solutions other than the v6 zswap_store_folio()
implementation, to fix the zstd regression seen in v5, to attempt to
unify common code paths, and to allocate smaller arrays for the zswap
entries on the stack. All these options were found to cause usemem30
latency regression with zstd. The v6 version of zswap_store_folio() is
the only implementation that does not cause zstd regression, confirmed
by 10 consecutive runs, each giving quite consistent latency
numbers. Hence, the v6 implementation is carried forward to v7, with
changes for branching for batching vs. sequential compression API
calls.
Changes since v5:
=================
1) Rebased to mm-unstable as of 2-1-2025, commit 7de6fd8ab650.
Several improvements, regression fixes and bug fixes, based on Yosry's
v5 comments (Thanks Yosry!):
2) Fix for zstd performance regression in v5.
3) Performance debug and fix for marginal improvements with IAA batching
vs. sequential.
4) Performance testing data compares IAA with and without batching, instead
of IAA batching against zstd.
5) Commit logs/zswap comments not mentioning crypto_acomp implementation
details.
6) Delete the pr_info_once() when batching resources are allocated in
zswap_cpu_comp_prepare().
7) Use kcalloc_node() for the multiple acomp_ctx buffers/reqs in
zswap_cpu_comp_prepare().
8) Simplify and consolidate error handling cleanup code in
zswap_cpu_comp_prepare().
9) Introduce zswap_compress_folio() in a separate patch.
10) Bug fix in zswap_store_folio() when xa_store() failure can cause all
compressed objects and entries to be freed, and UAF when zswap_store()
tries to free the entries that were already added to the xarray prior
to the failure.
11) Deleting compressed_bytes/bytes. zswap_store_folio() also comprehends
the recent fixes in commit bf5eaaaf7941 ("mm/zswap: fix inconsistency
when zswap_store_page() fails") by Hyeonggon Yoo.
iaa_crypto improvements/fixes/changes:
12) Enables asynchronous mode and makes it the default. With commit
4ebd9a5ca478 ("crypto: iaa - Fix IAA disabling that occurs when
sync_mode is set to 'async'"), async mode was previously just sync. We
now have true async support.
13) Change idxd descriptor allocations from blocking to non-blocking with
timeouts, and mitigations for compress/decompress ops that fail to
obtain a descriptor. This is a fix for tasks blocked errors seen in
configurations where 30+ cores are running workloads under high memory
pressure, and sending comps/decomps to 1 IAA device.
14) Fixes a bug with unprotected access of "deflate_generic_tfm" in
deflate_generic_decompress(), which can cause data corruption and
zswap_decompress() kernel crash.
15) zswap uses crypto_acomp_batch_compress() with async polling instead of
request chaining for slightly better latency. However, the request
chaining framework itself is unchanged, preserved from v5.
Changes since v4:
=================
1) Rebased to mm-unstable as of 12-20-2024, commit 5555a83c82d6.
2) Added acomp request chaining, as suggested by Herbert. Thanks Herbert!
3) Implemented IAA compress batching using request chaining.
4) zswap_store() batching simplifications suggested by Chengming, Yosry and
Nhat, thanks to all!
- New zswap_compress_folio() that is called by zswap_store().
- Move the loop over folio's pages out of zswap_store() and into a
zswap_store_folio() that stores all pages.
- Allocate all zswap entries for the folio upfront.
- Added zswap_batch_compress().
- Branch to call zswap_compress() or zswap_batch_compress() inside
zswap_compress_folio().
- All iterations over pages kept in same function level.
- No helpers other than the newly added zswap_store_folio() and
zswap_compress_folio().
Changes since v3:
=================
1) Rebased to mm-unstable as of 11-18-2024, commit 5a7056135bb6.
2) Major re-write of iaa_crypto driver's mapping of IAA devices to cores,
based on packages instead of NUMA nodes.
3) Added acomp_has_async_batching() API to crypto acomp, that allows
zswap/zram to query if a crypto_acomp has registered batch_compress and
batch_decompress interfaces.
4) Clear the poll bits on the acomp_reqs passed to
iaa_comp_a[de]compress_batch() so that a module like zswap can be
confident about the acomp_reqs[0] not having the poll bit set before
calling the fully synchronous API crypto_acomp_[de]compress().
Herbert, I would appreciate it if you can review changes 2-4; in patches
1-8 in v4. I did not want to introduce too many iaa_crypto changes in
v4, given that patch 7 is already making a major change. I plan to work
on incorporating the request chaining using the ahash interface in v5
(I need to understand the basic crypto ahash better). Thanks Herbert!
5) Incorporated Johannes' suggestion to not have a sysctl to enable
compress batching.
6) Incorporated Yosry's suggestion to allocate batching resources in the
cpu hotplug onlining code, since there is no longer a sysctl to control
batching. Thanks Yosry!
7) Incorporated Johannes' suggestions related to making the overall
sequence of events between zswap_store() and zswap_batch_store() similar
as much as possible for readability and control flow, better naming of
procedures, avoiding forward declarations, not inlining error path
procedures, deleting zswap internal details from zswap.h, etc. Thanks
Johannes, really appreciate the direction!
I have tried to explain the minimal future-proofing in terms of the
zswap_batch_store() signature and the definition of "struct
zswap_batch_store_sub_batch" in the comments for this struct. I hope the
new code explains the control flow a bit better.
Changes since v2:
=================
1) Rebased to mm-unstable as of 11-5-2024, commit 7994b7ea6ac8.
2) Fixed an issue in zswap_create_acomp_ctx() with checking for NULL
returned by kmalloc_node() for acomp_ctx->buffers and for
acomp_ctx->reqs.
3) Fixed a bug in zswap_pool_can_batch() for returning true if
pool->can_batch_comp is found to be equal to BATCH_COMP_ENABLED, and if
the per-cpu acomp_batch_ctx tests true for batching resources having
been allocated on this cpu. Also, changed from per_cpu_ptr() to
raw_cpu_ptr().
4) Incorporated the zswap_store_propagate_errors() compilation warning fix
suggested by Dan Carpenter. Thanks Dan!
5) Replaced the references to SWAP_CRYPTO_SUB_BATCH_SIZE in comments in
zswap.h, with SWAP_CRYPTO_BATCH_SIZE.
Changes since v1:
=================
1) Rebased to mm-unstable as of 11-1-2024, commit 5c4cf96cd702.
2) Incorporated Herbert's suggestions to use an acomp_req flag to indicate
async/poll mode, and to encapsulate the polling functionality in the
iaa_crypto driver. Thanks Herbert!
3) Incorporated Herbert's and Yosry's suggestions to implement the batching
API in iaa_crypto and to make its use seamless from zswap's
perspective. Thanks Herbert and Yosry!
4) Incorporated Yosry's suggestion to make it more convenient for the user
to enable compress batching, while minimizing the memory footprint
cost. Thanks Yosry!
5) Incorporated Yosry's suggestion to de-couple the shrink_folio_list()
reclaim batching patch from this series, since it requires a broader
discussion.
I would greatly appreciate code review comments for the iaa_crypto driver
and mm patches included in this series!
Thanks,
Kanchana
Kanchana P Sridhar (23):
crypto: iaa - Reorganize the iaa_crypto driver code.
crypto: iaa - New architecture for IAA device WQ comp/decomp usage &
core mapping.
crypto: iaa - Simplify, consistency of function parameters, minor
stats bug fix.
crypto: iaa - Descriptor allocation timeouts with mitigations.
crypto: iaa - iaa_wq uses percpu_refs for get/put reference counting.
crypto: iaa - Simplify the code flow in iaa_compress() and
iaa_decompress().
crypto: iaa - Refactor hardware descriptor setup into separate
procedures.
crypto: iaa - Simplified, efficient job submissions for non-irq mode.
crypto: iaa - Deprecate exporting add/remove IAA compression modes.
crypto: iaa - Expect a single scatterlist for a [de]compress request's
src/dst.
crypto: iaa - Rearchitect the iaa_crypto driver to be usable by zswap
and zram.
crypto: iaa - Enablers for submitting descriptors then polling for
completion.
crypto: acomp - Define a unit_size in struct acomp_req to enable
batching.
crypto: iaa - IAA Batching for parallel compressions/decompressions.
crypto: iaa - Enable async mode and make it the default.
crypto: iaa - Disable iaa_verify_compress by default.
crypto: iaa - Submit the two largest source buffers first in
decompress batching.
crypto: iaa - Add deflate-iaa-dynamic compression mode.
crypto: acomp - Add crypto_acomp_batch_size() to get an algorithm's
batch-size.
mm: zswap: Per-CPU acomp_ctx resources exist from pool creation to
deletion.
mm: zswap: Consistently use IS_ERR_OR_NULL() to check acomp_ctx
resources.
mm: zswap: zswap_store() will process a large folio in batches.
mm: zswap: Batched zswap_compress() with compress batching of large
folios.
.../driver-api/crypto/iaa/iaa-crypto.rst | 168 +-
crypto/acompress.c | 1 +
crypto/testmgr.c | 10 +
crypto/testmgr.h | 74 +
drivers/crypto/intel/iaa/Makefile | 4 +-
drivers/crypto/intel/iaa/iaa_crypto.h | 60 +-
.../intel/iaa/iaa_crypto_comp_dynamic.c | 22 +
drivers/crypto/intel/iaa/iaa_crypto_main.c | 2936 ++++++++++++-----
drivers/crypto/intel/iaa/iaa_crypto_stats.c | 8 +
drivers/crypto/intel/iaa/iaa_crypto_stats.h | 2 +
include/crypto/acompress.h | 63 +
include/crypto/internal/acompress.h | 3 +
include/linux/iaa_comp.h | 122 +
mm/zswap.c | 766 +++--
14 files changed, 3181 insertions(+), 1058 deletions(-)
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_comp_dynamic.c
create mode 100644 include/linux/iaa_comp.h
--
2.27.0
Powered by blists - more mailing lists