Word-align the address of polymul_asm #37
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Let's just include this in V2 and re-benchmark it there. |
mkannwischer
added a commit
that referenced
this pull request
May 1, 2019
* Add mupq submodule * Word-align the address of polymul_asm (#37) * added hal.h * hal.h * Rewrite utils to use interface.py Fix dev references * Actually register submodule * add stm32f4 hal.c * Refactor Makefile; build single binaries * kick-out schemes that did not make it to round2. Resolves #40 * Shrink RAM from 128 KiB to 112 KiB (#73) By not using SRAM2 we get more consistent benchmarks since SRAM2 seems to slow down memory accesses. This lead to schemes that are wasteful with stack (i.e., overflowing the 16 KiB SRAM2 into SRAM1) were faster. * Move common sources and headers to mupq * Use a pattern rule for various tests in Makefile * replaced cortexm4 specific function calls by hal calls * Adhere to slightly changed randombytes API * integrate hal for stm32f4 * fix variable renaming * update mupq * use hal.h from mupq * fix crypto sign benchmarks * adds hashing benchmarks. Closes #29 * Update README.md * Update README.md * Big refactor of python files * Include project into makefile targets * fix frodo builds * add hashing benchmarks * benchmarks * Refactor makefile As a bonus, includes build_everything.py * fix sike build * support sign in type detection * Add mupq note * suppress output of st-flash * removed stuff that now resides in mupq from pqm4, resolved symlinks * updated submodule * Use context manager to handle device * fix builds for kyber and frodo * add mupq implementations to builds * allow partial benchmarks and tests; small bug fixes * fix build everything and allow skipping stack benchmarks * fix recovery after flashing fails * remove duplicate code * CSV Benchmarks (#75) * hashing benchmarks * some more refactoring * sample benchmarks * sample benchmarks * new benchmarks * fix kyber1024 benchmarks * PQClean Integration (#76) * hashing benchmarks * some more refactoring * sample benchmarks * sample benchmarks * namespacing for pqclean * namespacing in testvectors on host * Integrate PQClean's AES into pqm4 * Fix path for host sources in Makefile * Update mupq * Fix bug in testvectors AES-128 * Add optimized AES-192/256 implementations. Round-2-Kyber will use this. Everything is put into separate sections such that the linker with gc-sections will succesfully throw out an AES instance (or all of them) whenever it is not used by a scheme. * Update mupq * fix incremental keccak implementation * added benchmarks for ntruhps2048509 and sphincs-shake256-128f-simple * make frodo scheme naming consistent with PQClean (breaks testvectors) * rename frodo in benchmark results * Measure code size (#77) * code-size in benchmarks * update README.md * Fix minor typo's in README.md * Add instructions for using AES to README * Update README.md * Some more polishing of README.md
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This imports the outcome of mupq/polymul-z2mx-m4#1 into pqm4.
I haven't re-benchmarked and edited
README.mdyet as we might want to make changes in #36 that affect cycle counts, so let's leave this open for a bit.