Add initial support for Armv8.1-M+MVE targets #1220

willieyz · 2025-10-07T08:18:33Z

This PR porting following changes in following PR from mldsa-native to mlkem-native, adds initial support for baremetal targets.
- Add initial support for Armv8.1-M targets mldsa-native#462
General build system changes:
- Add separate LDFLAGS (previously only CFLAGS for both compilation and linking)
- Add EXTRA_MAKEFILE to include custom makefiles with platform-specific configuration
- Add EXTRA_SOURCES to compile and link additional platform files
- Fix ACVP tests to respect EXEC_WRAPPER
Baremetal support for MPS3 AN547 (Arm Cortex-M55):
- Add nix shell (arm-embedded) with arm-none-eabi-gcc, qemu, and platform support files from pqmx
- Add test/baremetal/platform/m55-an547/platform.mk with compiler flags, and build configuration
- Add exec_wrapper.py to pass test arguments via QEMU
- Add CI workflow for functional, KAT, and ACVP tests on AN547
Note that: unlike mldsa-native, mlkem-native auto-detects the host architecture (AArch64) during cross-compilation, incorrectly setting MLK_FORCE_AARCH64 and breaking Cortex-M55 baremetal builds, we added extra_args: --no-auto in .github/workflows/baremetal.yml to disable auto-detection in baremetal build + test.

oqs-bot

Mac Mini (M1, 2020) benchmarks

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`12312` cycles	`12313` cycles	`1.00`
`ML-KEM-512 encaps`	`14889` cycles	`14889` cycles	`1`
`ML-KEM-512 decaps`	`19258` cycles	`19262` cycles	`1.00`
`ML-KEM-768 keypair`	`21361` cycles	`21361` cycles	`1`
`ML-KEM-768 encaps`	`23569` cycles	`23571` cycles	`1.00`
`ML-KEM-768 decaps`	`30141` cycles	`30140` cycles	`1.00`
`ML-KEM-1024 keypair`	`30354` cycles	`30355` cycles	`1.00`
`ML-KEM-1024 encaps`	`34546` cycles	`34546` cycles	`1`
`ML-KEM-1024 decaps`	`44333` cycles	`44333` cycles	`1`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Intel Xeon 4th gen (c7i)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`9599` cycles	`9599` cycles	`1`
`ML-KEM-512 encaps`	`11056` cycles	`11078` cycles	`1.00`
`ML-KEM-512 decaps`	`15037` cycles	`15136` cycles	`0.99`
`ML-KEM-768 keypair`	`16564` cycles	`16584` cycles	`1.00`
`ML-KEM-768 encaps`	`17726` cycles	`17763` cycles	`1.00`
`ML-KEM-768 decaps`	`23536` cycles	`23518` cycles	`1.00`
`ML-KEM-1024 keypair`	`22431` cycles	`22408` cycles	`1.00`
`ML-KEM-1024 encaps`	`24284` cycles	`24255` cycles	`1.00`
`ML-KEM-1024 decaps`	`31918` cycles	`31828` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Intel Xeon 4th gen (c7i) (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`28853` cycles	`28872` cycles	`1.00`
`ML-KEM-512 encaps`	`35068` cycles	`35191` cycles	`1.00`
`ML-KEM-512 decaps`	`44971` cycles	`45071` cycles	`1.00`
`ML-KEM-768 keypair`	`47758` cycles	`47783` cycles	`1.00`
`ML-KEM-768 encaps`	`57028` cycles	`56916` cycles	`1.00`
`ML-KEM-768 decaps`	`69712` cycles	`69812` cycles	`1.00`
`ML-KEM-1024 keypair`	`71032` cycles	`71072` cycles	`1.00`
`ML-KEM-1024 encaps`	`82575` cycles	`82659` cycles	`1.00`
`ML-KEM-1024 decaps`	`99707` cycles	`99677` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`51992` cycles	`51873` cycles	`1.00`
`ML-KEM-512 encaps`	`59533` cycles	`59604` cycles	`1.00`
`ML-KEM-512 decaps`	`75957` cycles	`77137` cycles	`0.98`
`ML-KEM-768 keypair`	`89009` cycles	`88992` cycles	`1.00`
`ML-KEM-768 encaps`	`96749` cycles	`96188` cycles	`1.01`
`ML-KEM-768 decaps`	`122388` cycles	`120429` cycles	`1.02`
`ML-KEM-1024 keypair`	`134104` cycles	`133034` cycles	`1.01`
`ML-KEM-1024 encaps`	`146966` cycles	`145051` cycles	`1.01`
`ML-KEM-1024 decaps`	`179710` cycles	`177770` cycles	`1.01`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

AMD EPYC 4th gen (c7a)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`11867` cycles	`11914` cycles	`1.00`
`ML-KEM-512 encaps`	`13398` cycles	`13411` cycles	`1.00`
`ML-KEM-512 decaps`	`18290` cycles	`18287` cycles	`1.00`
`ML-KEM-768 keypair`	`20578` cycles	`20566` cycles	`1.00`
`ML-KEM-768 encaps`	`21619` cycles	`21640` cycles	`1.00`
`ML-KEM-768 decaps`	`28763` cycles	`28851` cycles	`1.00`
`ML-KEM-1024 keypair`	`27262` cycles	`27156` cycles	`1.00`
`ML-KEM-1024 encaps`	`29247` cycles	`29227` cycles	`1.00`
`ML-KEM-1024 decaps`	`38736` cycles	`38757` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

AMD EPYC 3rd gen (c6a)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`16815` cycles	`16718` cycles	`1.01`
`ML-KEM-512 encaps`	`18576` cycles	`18508` cycles	`1.00`
`ML-KEM-512 decaps`	`23972` cycles	`23818` cycles	`1.01`
`ML-KEM-768 keypair`	`28513` cycles	`28595` cycles	`1.00`
`ML-KEM-768 encaps`	`29807` cycles	`29849` cycles	`1.00`
`ML-KEM-768 decaps`	`37651` cycles	`37779` cycles	`1.00`
`ML-KEM-1024 keypair`	`41645` cycles	`41720` cycles	`1.00`
`ML-KEM-1024 encaps`	`44044` cycles	`44071` cycles	`1.00`
`ML-KEM-1024 decaps`	`54485` cycles	`54432` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton2

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`28836` cycles	`28844` cycles	`1.00`
`ML-KEM-512 encaps`	`34044` cycles	`34072` cycles	`1.00`
`ML-KEM-512 decaps`	`44607` cycles	`44639` cycles	`1.00`
`ML-KEM-768 keypair`	`49282` cycles	`49202` cycles	`1.00`
`ML-KEM-768 encaps`	`54298` cycles	`54271` cycles	`1.00`
`ML-KEM-768 decaps`	`69160` cycles	`69152` cycles	`1.00`
`ML-KEM-1024 keypair`	`71544` cycles	`71610` cycles	`1.00`
`ML-KEM-1024 encaps`	`79791` cycles	`79933` cycles	`1.00`
`ML-KEM-1024 decaps`	`100007` cycles	`99967` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton4

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`17850` cycles	`17875` cycles	`1.00`
`ML-KEM-512 encaps`	`20825` cycles	`20822` cycles	`1.00`
`ML-KEM-512 decaps`	`27416` cycles	`27409` cycles	`1.00`
`ML-KEM-768 keypair`	`30567` cycles	`30575` cycles	`1.00`
`ML-KEM-768 encaps`	`33262` cycles	`33232` cycles	`1.00`
`ML-KEM-768 decaps`	`42694` cycles	`42678` cycles	`1.00`
`ML-KEM-1024 keypair`	`44344` cycles	`44375` cycles	`1.00`
`ML-KEM-1024 encaps`	`49260` cycles	`49307` cycles	`1.00`
`ML-KEM-1024 decaps`	`62141` cycles	`62170` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

AMD EPYC 4th gen (c7a) (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`36459` cycles	`36407` cycles	`1.00`
`ML-KEM-512 encaps`	`42718` cycles	`42673` cycles	`1.00`
`ML-KEM-512 decaps`	`55658` cycles	`55632` cycles	`1.00`
`ML-KEM-768 keypair`	`59893` cycles	`59940` cycles	`1.00`
`ML-KEM-768 encaps`	`67654` cycles	`67645` cycles	`1.00`
`ML-KEM-768 decaps`	`85008` cycles	`84960` cycles	`1.00`
`ML-KEM-1024 keypair`	`87567` cycles	`87570` cycles	`1.00`
`ML-KEM-1024 encaps`	`98046` cycles	`97952` cycles	`1.00`
`ML-KEM-1024 decaps`	`119516` cycles	`119591` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

AMD EPYC 3rd gen (c6a) (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`38517` cycles	`38463` cycles	`1.00`
`ML-KEM-512 encaps`	`47488` cycles	`47455` cycles	`1.00`
`ML-KEM-512 decaps`	`60959` cycles	`60907` cycles	`1.00`
`ML-KEM-768 keypair`	`63922` cycles	`63951` cycles	`1.00`
`ML-KEM-768 encaps`	`74768` cycles	`74840` cycles	`1.00`
`ML-KEM-768 decaps`	`92800` cycles	`92765` cycles	`1.00`
`ML-KEM-1024 keypair`	`94377` cycles	`94414` cycles	`1.00`
`ML-KEM-1024 encaps`	`108266` cycles	`108243` cycles	`1.00`
`ML-KEM-1024 decaps`	`131336` cycles	`130955` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Intel Xeon 3rd gen (c6i)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`16236` cycles	`16212` cycles	`1.00`
`ML-KEM-512 encaps`	`18344` cycles	`18368` cycles	`1.00`
`ML-KEM-512 decaps`	`24947` cycles	`24901` cycles	`1.00`
`ML-KEM-768 keypair`	`28784` cycles	`30050` cycles	`0.96`
`ML-KEM-768 encaps`	`29975` cycles	`29915` cycles	`1.00`
`ML-KEM-768 decaps`	`39355` cycles	`39427` cycles	`1.00`
`ML-KEM-1024 keypair`	`37643` cycles	`37639` cycles	`1.00`
`ML-KEM-1024 encaps`	`40275` cycles	`40307` cycles	`1.00`
`ML-KEM-1024 decaps`	`55062` cycles	`55626` cycles	`0.99`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton4 (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`35323` cycles	`35363` cycles	`1.00`
`ML-KEM-512 encaps`	`40756` cycles	`40772` cycles	`1.00`
`ML-KEM-512 decaps`	`51505` cycles	`51497` cycles	`1.00`
`ML-KEM-768 keypair`	`59456` cycles	`58817` cycles	`1.01`
`ML-KEM-768 encaps`	`65597` cycles	`66151` cycles	`0.99`
`ML-KEM-768 decaps`	`80185` cycles	`80264` cycles	`1.00`
`ML-KEM-1024 keypair`	`87747` cycles	`87750` cycles	`1.00`
`ML-KEM-1024 encaps`	`97043` cycles	`97061` cycles	`1.00`
`ML-KEM-1024 decaps`	`116486` cycles	`116461` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton3

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`19069` cycles	`19075` cycles	`1.00`
`ML-KEM-512 encaps`	`22342` cycles	`22330` cycles	`1.00`
`ML-KEM-512 decaps`	`29573` cycles	`29567` cycles	`1.00`
`ML-KEM-768 keypair`	`32610` cycles	`32624` cycles	`1.00`
`ML-KEM-768 encaps`	`35689` cycles	`35668` cycles	`1.00`
`ML-KEM-768 decaps`	`46018` cycles	`45989` cycles	`1.00`
`ML-KEM-1024 keypair`	`46927` cycles	`46958` cycles	`1.00`
`ML-KEM-1024 encaps`	`52065` cycles	`52083` cycles	`1.00`
`ML-KEM-1024 decaps`	`65935` cycles	`66034` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton2 (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`59350` cycles	`59424` cycles	`1.00`
`ML-KEM-512 encaps`	`68403` cycles	`68473` cycles	`1.00`
`ML-KEM-512 decaps`	`87175` cycles	`87282` cycles	`1.00`
`ML-KEM-768 keypair`	`99672` cycles	`99249` cycles	`1.00`
`ML-KEM-768 encaps`	`110659` cycles	`110750` cycles	`1.00`
`ML-KEM-768 decaps`	`135488` cycles	`135526` cycles	`1.00`
`ML-KEM-1024 keypair`	`148952` cycles	`148667` cycles	`1.00`
`ML-KEM-1024 encaps`	`165121` cycles	`164945` cycles	`1.00`
`ML-KEM-1024 decaps`	`196900` cycles	`196150` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Intel Xeon 3rd gen (c6i) (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`46166` cycles	`46219` cycles	`1.00`
`ML-KEM-512 encaps`	`54760` cycles	`54786` cycles	`1.00`
`ML-KEM-512 decaps`	`69951` cycles	`69990` cycles	`1.00`
`ML-KEM-768 keypair`	`76093` cycles	`76153` cycles	`1.00`
`ML-KEM-768 encaps`	`86760` cycles	`86769` cycles	`1.00`
`ML-KEM-768 decaps`	`106757` cycles	`106726` cycles	`1.00`
`ML-KEM-1024 keypair`	`110809` cycles	`110926` cycles	`1.00`
`ML-KEM-1024 encaps`	`124910` cycles	`125022` cycles	`1.00`
`ML-KEM-1024 decaps`	`150464` cycles	`150695` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Graviton3 (no-opt)

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`38836` cycles	`38859` cycles	`1.00`
`ML-KEM-512 encaps`	`45035` cycles	`45053` cycles	`1.00`
`ML-KEM-512 decaps`	`56840` cycles	`56859` cycles	`1.00`
`ML-KEM-768 keypair`	`65770` cycles	`64363` cycles	`1.02`
`ML-KEM-768 encaps`	`72122` cycles	`72783` cycles	`0.99`
`ML-KEM-768 decaps`	`88022` cycles	`88098` cycles	`1.00`
`ML-KEM-1024 keypair`	`95831` cycles	`95820` cycles	`1.00`
`ML-KEM-1024 encaps`	`106272` cycles	`106291` cycles	`1.00`
`ML-KEM-1024 decaps`	`127035` cycles	`126975` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Arm Cortex-A55 (Snapdragon 888) benchmarks

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`59504` cycles	`59461` cycles	`1.00`
`ML-KEM-512 encaps`	`66745` cycles	`66700` cycles	`1.00`
`ML-KEM-512 decaps`	`85404` cycles	`85316` cycles	`1.00`
`ML-KEM-768 keypair`	`101503` cycles	`101409` cycles	`1.00`
`ML-KEM-768 encaps`	`112747` cycles	`112460` cycles	`1.00`
`ML-KEM-768 decaps`	`139359` cycles	`139417` cycles	`1.00`
`ML-KEM-1024 keypair`	`154103` cycles	`153673` cycles	`1.00`
`ML-KEM-1024 encaps`	`172616` cycles	`171514` cycles	`1.01`
`ML-KEM-1024 decaps`	`207563` cycles	`207185` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`28872` cycles	`28854` cycles	`1.00`
`ML-KEM-512 encaps`	`34105` cycles	`33992` cycles	`1.00`
`ML-KEM-512 decaps`	`44546` cycles	`44559` cycles	`1.00`
`ML-KEM-768 keypair`	`49207` cycles	`49258` cycles	`1.00`
`ML-KEM-768 encaps`	`54273` cycles	`54259` cycles	`1.00`
`ML-KEM-768 decaps`	`69153` cycles	`69165` cycles	`1.00`
`ML-KEM-1024 keypair`	`71597` cycles	`71661` cycles	`1.00`
`ML-KEM-1024 encaps`	`79839` cycles	`79817` cycles	`1.00`
`ML-KEM-1024 decaps`	`100000` cycles	`99833` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

oqs-bot

SpacemiT K1 8 (Banana Pi F3) benchmarks

Benchmark suite	Current: `298d67f`	Previous: `d5eea7c`	Ratio
`ML-KEM-512 keypair`	`226745` cycles	`226622` cycles	`1.00`
`ML-KEM-512 encaps`	`270848` cycles	`270707` cycles	`1.00`
`ML-KEM-512 decaps`	`345086` cycles	`344944` cycles	`1.00`
`ML-KEM-768 keypair`	`376031` cycles	`375797` cycles	`1.00`
`ML-KEM-768 encaps`	`433145` cycles	`432781` cycles	`1.00`
`ML-KEM-768 decaps`	`529734` cycles	`529290` cycles	`1.00`
`ML-KEM-1024 keypair`	`554792` cycles	`554517` cycles	`1.00`
`ML-KEM-1024 encaps`	`631241` cycles	`630901` cycles	`1.00`
`ML-KEM-1024 decaps`	`753040` cycles	`752352` cycles	`1.00`

This comment was automatically generated by workflow using github-action-benchmark.

hanno-becker

Note that: unlike mldsa-native, mlkem-native auto-detects the host architecture (AArch64) during cross-compilation, incorrectly setting MLK_FORCE_AARCH64 and breaking Cortex-M55 baremetal builds, we added extra_args: --no-auto in .github/workflows/baremetal.yml to disable auto-detection in baremetal build + test.

Why don't we force AUTO=0 (in the makefile) if a cross-prefix is set?

willieyz · 2025-10-09T09:52:03Z

Note that: unlike mldsa-native, mlkem-native auto-detects the host architecture (AArch64) during cross-compilation, incorrectly setting MLK_FORCE_AARCH64 and breaking Cortex-M55 baremetal builds, we added extra_args: --no-auto in .github/workflows/baremetal.yml to disable auto-detection in baremetal build + test.

Why don't we force AUTO=0 (in the makefile) if a cross-prefix is set?

Hello, @hanno-becker, thank you for your review!
Now I add the following in config.mk instead of using the extra_args:

mlkem-native/test/mk/config.mk

Lines 80 to 85 in 0ce6879

    
           # Force AUTO=0 when cross-compiling to avoid host architecture detection 
        
           ifneq ($(CROSS_PREFIX),) 
        
           override AUTO:=0 
        
           else 
        
           AUTO ?= 1 
        
           endif

Signed-off-by: willieyz <[email protected]>

- When EXTRA_MAKEFILE is passed, it will be included at the start of the main Makefile. This can be used to configure builds requiring a large number of flags (e.g., baremetal builds). - `EXTRA_SOURCES` can be passed to compile and link additional sources, e.g., platform files for baremetal builds. Signed-off-by: willieyz <[email protected]>

- This commit remove `additional_packages` to get down to max 10 Github limits the number of inputs for a workflow_dispatch event to 10. - We currently have 11 for bench_ec2_any.yml. This removes the additional_packages inputs that is never used in our CI. Signed-off-by: willieyz <[email protected]>

`run_acvp` currently does not respect the `EXEC_WRAPPER` preventing running ACVP tests, e.g., on baremetal targets. Signed-off-by: willieyz <[email protected]>

- This commits adds a new shell: arm-embedded for baremetal targets. - It includes arm-none-eabi-gcc, qemu, and platform support files from pqmx (currently limited to the MPS3 AN547 Cortex-M55). Signed-off-by: willieyz <[email protected]>

- This adds support for testing on the MPS3 AN547 platform in qemu. - Tests can be excuted using `EXTRA_MAKEFILE=test/baremetal/platform/m55-an547/platform.mk tests all --no-auto` - Platform sources (hal, ld script, semihosting) are taken from pqmx. - `baremetal/platform/m55-an547/platform.mk` sets the required flags and other build configuration. - `test/baremetal/platform/m55-an547/exec_wrapper.py` handles passing inputs to the tests (e.g., required for the ACVP tests). Signed-off-by: willieyz <[email protected]>

- Adds CI to exercise the MPS3 AN547 (Cortex-M55) baremetal builds. - It includes functional tests, KAT tests, and ACVP tests. Signed-off-by: willieyz <[email protected]>

hanno-becker · 2025-10-12T03:53:19Z

@mkannwischer Do you agree with forcing AUTO=0 for cross-compilation?

hanno-becker

LGTM, and tested successfully on my M1. Thanks @willieyz

willieyz changed the title ~~Build: Add separate LDFLAGS~~ Add initial support for Armv8.1-M targets Oct 7, 2025

willieyz force-pushed the armv8.1-m-porting branch 5 times, most recently from 0e2ef4b to 298d67f Compare October 8, 2025 06:34

mkannwischer added the benchmark this PR should be benchmarked in CI label Oct 8, 2025

oqs-bot reviewed Oct 8, 2025

View reviewed changes

willieyz force-pushed the armv8.1-m-porting branch from 298d67f to 6705b2e Compare October 8, 2025 06:54

oqs-bot reviewed Oct 8, 2025

View reviewed changes

willieyz marked this pull request as ready for review October 8, 2025 08:42

willieyz requested a review from a team as a code owner October 8, 2025 08:42

hanno-becker requested a review from mkannwischer October 9, 2025 03:45

hanno-becker requested changes Oct 9, 2025

View reviewed changes

willieyz force-pushed the armv8.1-m-porting branch 2 times, most recently from 418e185 to 0ce6879 Compare October 9, 2025 09:28

bremoran mentioned this pull request Oct 9, 2025

Add FIPS202 backend for Armv8.1-M+MVE #1224

Open

hanno-becker added enhancement New feature or request MVE Relating to Armv8.1-M+MVE architecture labels Oct 10, 2025

hanno-becker changed the title ~~Add initial support for Armv8.1-M targets~~ Add initial support for Armv8.1-M+MVE targets Oct 10, 2025

willieyz added 7 commits October 12, 2025 04:53

Build: Add separate LDFLAGS

c9b8f95

Signed-off-by: willieyz <[email protected]>

ACVP tests: Add missing EXEC_WRAPPER

38c9c39

`run_acvp` currently does not respect the `EXEC_WRAPPER` preventing running ACVP tests, e.g., on baremetal targets. Signed-off-by: willieyz <[email protected]>

Nix: Add a baremetal shell (arm-embedded)

6f10b9f

- This commits adds a new shell: arm-embedded for baremetal targets. - It includes arm-none-eabi-gcc, qemu, and platform support files from pqmx (currently limited to the MPS3 AN547 Cortex-M55). Signed-off-by: willieyz <[email protected]>

CI: Add Baremetal CI

6493d16

- Adds CI to exercise the MPS3 AN547 (Cortex-M55) baremetal builds. - It includes functional tests, KAT tests, and ACVP tests. Signed-off-by: willieyz <[email protected]>

hanno-becker force-pushed the armv8.1-m-porting branch from 0ce6879 to 6493d16 Compare October 12, 2025 03:53

hanno-becker approved these changes Oct 12, 2025

View reviewed changes

Add initial support for Armv8.1-M+MVE targets #1220

Are you sure you want to change the base?

Add initial support for Armv8.1-M+MVE targets #1220

Uh oh!

Conversation

willieyz commented Oct 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Mac Mini (M1, 2020) benchmarks

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 4th gen (c7i)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 4th gen (c7i) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 4th gen (c7a)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 3rd gen (c6a)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton2

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton4

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 4th gen (c7a) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 3rd gen (c6a) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 3rd gen (c6i)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton4 (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton3

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton2 (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 3rd gen (c6i) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton3 (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Arm Cortex-A55 (Snapdragon 888) benchmarks

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

Uh oh!

willieyz commented Oct 7, 2025 •

edited

Loading

hanno-becker left a comment •

edited

Loading