Nicolas Sarlin
b7fc208e40
chore(zk): match zkv2 hash impl with the description
...
- encode the position of bits proven to be 0 in the hashes
- hash the infinite norm instead of the euclidean one
- hash the value of k with the statement
2025-10-30 16:53:36 +01:00
Nicolas Sarlin
bcb1356b76
fix(versionable): handle #[default] in Versionize types
2025-10-30 16:53:36 +01:00
Mayeul@Zama
54626cab6d
refactor(shortint): use ShortintBootstrappingKey in DecompressionKey
2025-10-30 16:52:44 +01:00
Nicolas Sarlin
bc493a5641
fix(shortint): avoid to crash when thread engine is reused
2025-10-30 14:51:01 +01:00
David Testé
073cba10d1
chore(ci): print stddev divergence in regression report
2025-10-30 14:06:30 +01:00
David Testé
2a8885aa9f
chore(ci): run erc20 and dex throughput bench only on demand
...
Following the same pattern as other benchmarks.
2025-10-30 09:52:30 +01:00
David Testé
e17c481736
chore(ci): prefix regression ops results with layer name
...
This is done to avoid confusion for operations that might have the same between layer. For example, 'bitand' operation have the same name for shortint and integer layers
2025-10-30 09:51:44 +01:00
David Testé
2542ef38e6
chore(ci): add parameters filtering for data extractor
...
When doing regression generation, one can provide a global parameters set name pattern to filter head branch benchmark results.
This fixes the issue encountered when there are more than one parameters' set used to benchmark an operation, for example, in core_crypto or shortint tfhe-rs layer.
2025-10-30 09:51:44 +01:00
Enzo Di Maria
398c441c95
refactor(gpu): delete useless GPU params
2025-10-30 08:59:10 +01:00
Enzo Di Maria
026cc376ed
refactor(gpu): multibit decompression
2025-10-30 08:59:10 +01:00
Pedro Alves
867f8fb579
feat(gpu): implement re-randomization
...
- exposed to integer and HL API
- test on the HL API
- benchmarks for GPU and CPU implementation
2025-10-29 17:55:45 -03:00
David Testé
3c32b15d02
chore(ci): print change thresholds in regression reports
2025-10-29 15:33:33 +01:00
David Testé
1823321aad
chore(ci): skip regression operation with invalid data point
2025-10-29 15:33:33 +01:00
David Testé
67130646ad
chore(ci): support shortint layer name parsing in data extractor
2025-10-29 15:33:33 +01:00
David Testé
f768fd1cdd
chore(ci): set all operations for default cpu regression profile
2025-10-29 15:33:33 +01:00
Arthur Meyre
0223913aef
chore: make functions consistent to generate keyswitching keys
...
- so that normal and seeded variants have similar APIs
2025-10-29 15:31:22 +01:00
Arthur Meyre
a41cd47b9e
refactor(test): make modulus switch config system make more sense
...
- The config type can hold any type for the drift technique variant because
the bounds are too weird to set on the type, the functions making use of
the config type should properly declare the bounds
2025-10-29 15:31:22 +01:00
Arthur Meyre
d95b46cb9b
refactor(test): factorize the any modulus switch function for noise checks
2025-10-29 15:31:22 +01:00
Guillermo Oyarzun
0f0438c8cf
feat(gpu): add 1_1 classical pbs params for specialized version
2025-10-29 09:18:18 +01:00
Arthur Meyre
9d31e994aa
chore(docs): make difference between benchmarks stand out more
2025-10-28 10:35:23 +01:00
Nicolas Sarlin
95593b1ea9
fix(zk): missing compressed proof version
2025-10-28 09:50:00 +01:00
Agnes Leroy
231d0c5e50
chore(gpu): disable lto in gpu bench compilation
2025-10-28 09:37:14 +01:00
David Testé
1d0a5c96a4
chore(ci): add bench type selection to core_crypto bench workflow
2025-10-27 18:09:54 +01:00
David Testé
b0b49ae533
chore(bench): new parameters set to run core_crypto bench for docs
...
This creates extended parameters set to reflect what's displayed
in the documentation.
2025-10-27 17:25:41 +01:00
Pedro Alves
70773e442c
fix(gpu): fix 128-bit compression benchmark
2025-10-27 17:06:45 +01:00
dependabot[bot]
7b797b8af9
chore(deps): bump actions/upload-artifact from 4.6.2 to 5.0.0
...
Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact ) from 4.6.2 to 5.0.0.
- [Release notes](https://github.com/actions/upload-artifact/releases )
- [Commits](ea165f8d65...330a01c490support@github.com >
2025-10-27 16:08:28 +01:00
dependabot[bot]
b6efb109aa
chore(deps): bump actions/download-artifact from 5.0.0 to 6.0.0
...
Bumps [actions/download-artifact](https://github.com/actions/download-artifact ) from 5.0.0 to 6.0.0.
- [Release notes](https://github.com/actions/download-artifact/releases )
- [Commits](634f93cb29...018cc2cf5bsupport@github.com >
2025-10-27 16:08:19 +01:00
David Testé
fd6323b311
chore(ci): add throughput and hpu support to data extractor
...
Now throughput results can be fetched.
HPU backend is supported for integer formatting
2025-10-27 14:39:46 +01:00
Arthur Meyre
b02a3b16ff
test: add rerand atomic pattern for noise checks
...
- make sure it works with KS32 parameters
2025-10-27 13:21:50 +01:00
Arthur Meyre
a95ee140f5
refactor: remove noise check function with PBS for sanity check
...
- it's a lot of code to "just" compute an additional PBS to make shortint
sanity checks, so run the function which gives the ms result, and complete
the AP by running the PBS as shortint would, gets rid of a big function
that was doing the same thing
2025-10-27 13:21:50 +01:00
Guillermo Oyarzun
62780ac500
fix(gpu): fix decompression mem leak
2025-10-24 13:02:41 +02:00
Thomas Montaigu
c10f1def70
fix: Tag propagation in XofKeySet
2025-10-24 10:40:22 +02:00
Mayeul@Zama
31a0136655
test(all): test multi bit decompression
2025-10-24 09:28:17 +02:00
Mayeul@Zama
859d5e4e1f
chore: add backward multi bit decompression keys
2025-10-24 09:28:17 +02:00
Mayeul@Zama
92dcd38e30
chore: add decompression_grouping_factor to TestCompressionParameterSet
2025-10-24 09:28:17 +02:00
Mayeul@Zama
777bbe437a
feat(shortint): add multi bit decompression
2025-10-24 09:28:17 +02:00
Mayeul@Zama
3842032f08
chore(shortint): fix unused function
2025-10-24 09:28:17 +02:00
Arthur Meyre
23246f63f7
chore: update fast_dedup opset to match the latency benchmarks in the docs
...
- signed bench update
2025-10-23 10:42:19 +02:00
Arthur Meyre
11c79b5237
chore: update fast_dedup opset to match the latency benchmarks in the docs
2025-10-23 10:42:19 +02:00
Nicolas Sarlin
a694e08ddc
fix(core): par_encrypt_and_prove was using sequential encryption
2025-10-23 10:08:06 +02:00
Guillermo Oyarzun
e12638dabe
feat(gpu): extend specialized version to classical pbs
2025-10-22 09:20:40 +02:00
pgardratzama
79f1d22573
fix(hpu): scalar rot & shift were not doing anything and not tested in test/hpu.rs
2025-10-21 13:29:59 +02:00
pgardratzama
f9c89212ea
fix(hpu): display name on shift looked wrong
2025-10-21 13:29:59 +02:00
pgardratzama
b918f77859
chore(hpu): add force_reload option in v80 config, remove added line in sim config
2025-10-21 13:29:59 +02:00
Helder Campos
054c5028a1
feat(hpu): Added the option to forcefully reload the HPU
2025-10-21 13:29:59 +02:00
Helder Campos
7b621e57b0
feat(hpu): LLT ROT/SHIFT IOPs
2025-10-21 13:29:59 +02:00
Agnes Leroy
b4b6275ca5
chore(gpu): remove device synchronize in drop for cudavec
2025-10-21 11:33:46 +02:00
Agnes Leroy
42644349ef
chore(gpu): remove remaining async functions from the integer gpu api
2025-10-20 16:19:19 +02:00
Thomas Montaigu
20b7b06ffb
chore: add check_fmt_js to pcc_batch
2025-10-20 14:37:36 +02:00
Thomas Montaigu
39fbc20360
fix(js): catch undefined variant using Option<>
...
In the JS ShortintParametersName, users could
make typo in the variant used e.g:
`ShortintParametersName.PARAM_MESSAGE_2_CARRY128`
In JS this returns `undefined` which is then later casted to an
int and it becomes 0, leading to match the first variant
We modify the input to receive an `Option<ShortintParametersName>`
as it seems to allow us to catch the `undefined` and return a proper
error
2025-10-20 14:37:36 +02:00
