chore(ci): add a workflows documentation

Bumps [actions/cache](https://github.com/actions/cache) from 4.3.0 to 5.0.1.
- [Release notes](https://github.com/actions/cache/releases)
- [Changelog](https://github.com/actions/cache/blob/main/RELEASES.md)
- [Commits](0057852bfa...9255dc7a25)

---
updated-dependencies:
- dependency-name: actions/cache
  dependency-version: 5.0.1
  dependency-type: direct:production
  update-type: version-update:semver-major
...

Signed-off-by: dependabot[bot] <support@github.com>

.github/workflows/aws_tfhe_backward_compat_tests.yml

@@ -80,7 +80,7 @@ jobs:
 
       - name: Retrieve data from cache
         id: retrieve-data-cache
-        uses: actions/cache/restore@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/restore@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         with:
           path: |
             utils/tfhe-backward-compat-data/**/*.cbor
@@ -109,7 +109,7 @@ jobs:
       - name: Store data in cache
         if: steps.retrieve-data-cache.outputs.cache-hit != 'true'
         continue-on-error: true
-        uses: actions/cache/save@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/save@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         with:
           path: |
             utils/tfhe-backward-compat-data/**/*.cbor

.github/workflows/aws_tfhe_fast_tests.yml

@@ -71,7 +71,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             dependencies:
@@ -219,7 +219,7 @@ jobs:
 
       - name: Node cache restoration
         id: node-cache
-        uses: actions/cache/restore@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/restore@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         with:
           path: |
             ~/.nvm
@@ -232,7 +232,7 @@ jobs:
           make install_node
 
       - name: Node cache save
-        uses: actions/cache/save@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/save@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         if: steps.node-cache.outputs.cache-hit != 'true'
         with:
           path: |

.github/workflows/aws_tfhe_integer_tests.yml

@@ -58,7 +58,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             integer:

.github/workflows/aws_tfhe_signed_integer_tests.yml

@@ -59,7 +59,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             integer:

.github/workflows/aws_tfhe_tests.yml

@@ -80,7 +80,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             dependencies:

.github/workflows/aws_tfhe_wasm_tests.yml

@@ -80,7 +80,7 @@ jobs:
 
       - name: Node cache restoration
         id: node-cache
-        uses: actions/cache/restore@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/restore@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         with:
           path: |
             ~/.nvm
@@ -93,7 +93,7 @@ jobs:
           make install_node
 
       - name: Node cache save
-        uses: actions/cache/save@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/save@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         if: steps.node-cache.outputs.cache-hit != 'true'
         with:
           path: |

.github/workflows/benchmark_cpu_common.yml

@@ -223,7 +223,7 @@ jobs:
           results_type: ${{ inputs.additional_results_type }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ matrix.bench_type }}_${{ matrix.params_type }}
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_ct_key_sizes.yml

@@ -99,7 +99,7 @@ jobs:
           --append-results
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_ct_key_sizes
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_documentation.yml

@@ -185,7 +185,7 @@ jobs:
           persist-credentials: 'false'
 
       - name: Download SVG tables
-        uses: actions/download-artifact@018cc2cf5baa6db3ef3c5f8a56943fffe632ef53 # v6.0.0
+        uses: actions/download-artifact@37930b1c2abaa49bbe596cd826c3c89aef350131 # v7.0.0
         with:
           path: svg_tables
           merge-multiple: 'true'
@@ -203,7 +203,7 @@ jobs:
           echo "date=$(date '+%g_%m_%d_%Hh%Mm%Ss')" >> "${GITHUB_OUTPUT}"
 
       - name: Create pull-request
-        uses: peter-evans/create-pull-request@84ae59a2cdc2258d6fa0732dd66352dddae2a412 # v7.0.9
+        uses: peter-evans/create-pull-request@98357b18bf14b5342f975ff684046ec3b2a07725 # v8.0.0
         with:
           sign-commits: true # Commit will be signed by github-actions bot
           add-paths: ${{ env.PATH_TO_DOC_ASSETS }}/*.svg

.github/workflows/benchmark_gpu_4090.yml

@@ -89,7 +89,7 @@ jobs:
           REF_NAME: ${{ github.ref_name }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_integer_multi_bit_gpu_default
           path: ${{ env.RESULTS_FILENAME }}
@@ -173,7 +173,7 @@ jobs:
           REF_NAME: ${{ github.ref_name }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_core_crypto
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_gpu_common.yml

@@ -281,7 +281,7 @@ jobs:
           BENCH_TYPE: ${{ matrix.bench_type }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_${{ matrix.command }}_${{ matrix.op_flavor }}_${{ inputs.profile }}_${{ matrix.bench_type }}_${{ matrix.params_type }}
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_gpu_coprocessor.yml

@@ -192,10 +192,10 @@ jobs:
           cargo install sqlx-cli
 
       - name: Install foundry
-        uses: foundry-rs/foundry-toolchain@50d5a8956f2e319df19e6b57539d7e2acb9f8c1e
+        uses: foundry-rs/foundry-toolchain@8b0419c685ef46cb79ec93fbdc131174afceb730
 
       - name: Cache cargo
-        uses: actions/cache@0057852bfaa89a56745cba8c7296529d2fc39830 # v4.3.0
+        uses: actions/cache@9255dc7a253b0ccc959486e2bca901246202afeb # v5.0.1
         with:
           path: |
             ~/.cargo/registry
@@ -262,7 +262,7 @@ jobs:
       - name: Upload profile artifact
         env:
           REPORT_NAME: ${{ steps.nsys_profile_name.outputs.profile }}
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ env.REPORT_NAME }}
           path: fhevm/coprocessor/fhevm-engine/tfhe-worker/${{ env.REPORT_NAME }}
@@ -293,7 +293,7 @@ jobs:
         working-directory: fhevm/
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${COMMIT_SHA}_${BENCHMARKS}_${{ needs.parse-inputs.outputs.profile }}
           path: fhevm/$${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_hpu_common.yml

@@ -185,7 +185,7 @@ jobs:
           BENCH_TYPE: ${{ matrix.bench_type }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_${{ matrix.bench_type }}_integer_benchmarks
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_perf_regression.yml

@@ -280,7 +280,7 @@ jobs:
           BENCH_TYPE: ${{ env.__TFHE_RS_BENCH_TYPE }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_regression_${{ env.RESULTS_FILE_SHA }} # RESULT_FILE_SHA is needed to avoid collision between matrix.command runs
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_tfhe_fft.yml

@@ -96,7 +96,7 @@ jobs:
           REF_NAME: ${{ github.ref_name }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_fft
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_tfhe_ntt.yml

@@ -96,7 +96,7 @@ jobs:
           REF_NAME: ${{ github.ref_name }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_ntt
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/benchmark_wasm_client.yml

@@ -47,7 +47,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             wasm_bench:
@@ -119,7 +119,7 @@ jobs:
 
       - name: Node cache restoration
         id: node-cache
-        uses: actions/cache/restore@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/restore@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         with:
           path: |
             ~/.nvm
@@ -132,7 +132,7 @@ jobs:
           make install_node
 
       - name: Node cache save
-        uses: actions/cache/save@0057852bfaa89a56745cba8c7296529d2fc39830 #v4.3.0
+        uses: actions/cache/save@9255dc7a253b0ccc959486e2bca901246202afeb #v5.0.1
         if: steps.node-cache.outputs.cache-hit != 'true'
         with:
           path: |
@@ -153,6 +153,12 @@ jobs:
         env:
           BROWSER: ${{ matrix.browser }}
 
+      - name: Run benchmarks (unsafe coop)
+        run: |
+          make bench_web_js_api_unsafe_coop_"${BROWSER}"_ci
+        env:
+          BROWSER: ${{ matrix.browser }}
+
       - name: Parse results
         run: |
           make parse_wasm_benchmarks
@@ -169,7 +175,7 @@ jobs:
           REF_NAME: ${{ github.ref_name }}
 
       - name: Upload parsed results artifact
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
           name: ${{ github.sha }}_wasm_${{ matrix.browser }}
           path: ${{ env.RESULTS_FILENAME }}

.github/workflows/cargo_test_fft.yml

@@ -37,7 +37,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             fft:

.github/workflows/cargo_test_ntt.yml

@@ -39,7 +39,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             ntt:

.github/workflows/ci_lint.yml

@@ -50,7 +50,7 @@ jobs:
           version: ${{ steps.get_zizmor.outputs.version }}
 
       - name: Ensure SHA pinned actions
-        uses: zgosalvez/github-actions-ensure-sha-pinned-actions@9e9574ef04ea69da568d6249bd69539ccc704e74 # v4.0.0
+        uses: zgosalvez/github-actions-ensure-sha-pinned-actions@6124774845927d14c601359ab8138699fa5b70c3 # v4.0.1
         with:
           allowlist: |
             slsa-framework/slsa-github-generator

.github/workflows/code_coverage.yml

@@ -62,7 +62,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             tfhe:
@@ -92,7 +92,7 @@ jobs:
           make test_shortint_cov
 
       - name: Upload tfhe coverage to Codecov
-        uses: codecov/codecov-action@5a1091511ad55cbe89839c7260b706298ca349f7
+        uses: codecov/codecov-action@671740ac38dd9b0130fbe1cec585b89eea48d3de
         if: steps.changed-files.outputs.tfhe_any_changed == 'true'
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
@@ -106,7 +106,7 @@ jobs:
           make test_integer_cov
 
       - name: Upload tfhe coverage to Codecov
-        uses: codecov/codecov-action@5a1091511ad55cbe89839c7260b706298ca349f7
+        uses: codecov/codecov-action@671740ac38dd9b0130fbe1cec585b89eea48d3de
         if: steps.changed-files.outputs.tfhe_any_changed == 'true'
         with:
           token: ${{ secrets.CODECOV_TOKEN }}

.github/workflows/generate_svg_common.yml

@@ -75,6 +75,15 @@ jobs:
           DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
           DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
 
+      - name: Upload tables
+        if: inputs.backend_comparison == false
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
+        with:
+          name: ${{ github.sha }}_${{ inputs.backend }}_${{ inputs.layer }}_${{ inputs.pbs_kind }}_${{ inputs.bench_type }}_tables
+          # This will upload all the file generated
+          path: ${{ inputs.output_filename }}*.svg
+          retention-days: 60
+
       - name: Produce backends comparison table from database
         if: inputs.backend_comparison == true
         run: |
@@ -90,10 +99,11 @@ jobs:
           DATA_EXTRACTOR_DATABASE_HOST: ${{ secrets.DATA_EXTRACTOR_DATABASE_HOST }}
           DATA_EXTRACTOR_DATABASE_PASSWORD: ${{ secrets.DATA_EXTRACTOR_DATABASE_PASSWORD }}
 
-      - name: Upload tables
-        uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4
+      - name: Upload comparison tables
+        if: inputs.backend_comparison == true
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
-          name: ${{ github.sha }}_${{ inputs.backend }}_${{ inputs.layer }}_${{ inputs.pbs_kind }}_${{ inputs.bench_type }}_tables
+          name: ${{ github.sha }}_backends_comparison_tables
           # This will upload all the file generated
           path: ${{ inputs.output_filename }}*.svg
           retention-days: 60

.github/workflows/gpu_fast_h100_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_fast_tests.yml

@@ -47,7 +47,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_full_multi_gpu_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_signed_integer_classic_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_signed_integer_h100_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_signed_integer_tests.yml

@@ -49,7 +49,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_unsigned_integer_classic_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_unsigned_integer_h100_tests.yml

@@ -48,7 +48,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/gpu_unsigned_integer_tests.yml

@@ -49,7 +49,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             gpu:

.github/workflows/hpu_hlapi_tests.yml

@@ -40,7 +40,7 @@ jobs:
 
       - name: Check for file changes
         id: changed-files
-        uses: tj-actions/changed-files@24d32ffd492484c1d75e0c0b894501ddb9d30d62 # v47.0.0
+        uses: tj-actions/changed-files@e0021407031f5be11a464abee9a0776171c79891 # v47.0.1
         with:
           files_yaml: |
             hpu:

.github/workflows/make_release_common.yml

@@ -62,7 +62,7 @@ jobs:
           PACKAGE: ${{ inputs.package-name }}
         run: |
           cargo package -p "${PACKAGE}"
-      - uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # v5.0.0
+      - uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f # v6.0.0
         with:
           name: crate-${{ inputs.package-name }}
           path: target/package/*.crate
@@ -100,7 +100,7 @@ jobs:
           token: ${{ secrets.REPO_CHECKOUT_TOKEN }}
 
       - name: Download artifact
-        uses: actions/download-artifact@018cc2cf5baa6db3ef3c5f8a56943fffe632ef53 # v6.0.0
+        uses: actions/download-artifact@37930b1c2abaa49bbe596cd826c3c89aef350131 # v7.0.0
         with:
           name: crate-${{ inputs.package-name }}
           path: target/package

.github/workflows/make_release_cuda.yml

@@ -104,7 +104,7 @@ jobs:
         run: |
           cargo package -p tfhe-cuda-backend
 
-      - uses: actions/upload-artifact@330a01c490aca151604b8cf639adc76d48f6c5d4 # v5.0.0
+      - uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f # v6.0.0
         with:
           name: crate-tfhe-cuda-backend
           path: target/package/*.crate
@@ -174,7 +174,7 @@ jobs:
           GCC_VERSION: ${{ matrix.gcc }}
 
       - name: Download artifact
-        uses: actions/download-artifact@018cc2cf5baa6db3ef3c5f8a56943fffe632ef53 # v6.0.0
+        uses: actions/download-artifact@37930b1c2abaa49bbe596cd826c3c89aef350131 # v7.0.0
         with:
           name: crate-tfhe-cuda-backend
           path: target/package

Cargo.toml

@@ -36,6 +36,7 @@ rayon = "1.11"
 serde = { version = "1.0", default-features = false }
 wasm-bindgen = "0.2.101"
 getrandom = "0.2.8"
+bincode = "=1.3.3"
 
 [profile.bench]
 lto = "fat"

Makefile

@@ -1300,13 +1300,14 @@ run_web_js_api_parallel: build_web_js_api_parallel setup_venv
 	--browser-path $(browser_path) \
 	--driver-path $(driver_path) \
 	--browser-kind  $(browser_kind) \
-	--server-cmd "npm run server" \
+	--server-cmd $(server_cmd) \
 	--server-workdir "$(WEB_SERVER_DIR)" \
 	--id-pattern $(filter)
 
 test_web_js_api_parallel_chrome: browser_path = "$(WEB_RUNNER_DIR)/chrome/chrome-linux64/chrome"
 test_web_js_api_parallel_chrome: driver_path = "$(WEB_RUNNER_DIR)/chrome/chromedriver-linux64/chromedriver"
 test_web_js_api_parallel_chrome: browser_kind = chrome
+test_web_js_api_parallel_chrome: server_cmd = "npm run server:multithreaded"
 test_web_js_api_parallel_chrome: filter = Test
 
 .PHONY: test_web_js_api_parallel_chrome # Run tests for the web wasm api on Chrome
@@ -1322,6 +1323,7 @@ test_web_js_api_parallel_chrome_ci: setup_venv
 test_web_js_api_parallel_firefox: browser_path = "$(WEB_RUNNER_DIR)/firefox/firefox/firefox"
 test_web_js_api_parallel_firefox: driver_path = "$(WEB_RUNNER_DIR)/firefox/geckodriver"
 test_web_js_api_parallel_firefox: browser_kind = firefox
+test_web_js_api_parallel_firefox: server_cmd = "npm run server:multithreaded"
 test_web_js_api_parallel_firefox: filter = Test
 
 .PHONY: test_web_js_api_parallel_firefox # Run tests for the web wasm api on Firefox
@@ -1571,6 +1573,7 @@ bench_pbs128_gpu: install_rs_check_toolchain
 bench_web_js_api_parallel_chrome: browser_path = "$(WEB_RUNNER_DIR)/chrome/chrome-linux64/chrome"
 bench_web_js_api_parallel_chrome: driver_path = "$(WEB_RUNNER_DIR)/chrome/chromedriver-linux64/chromedriver"
 bench_web_js_api_parallel_chrome: browser_kind = chrome
+bench_web_js_api_parallel_chrome: server_cmd = "npm run server:multithreaded"
 bench_web_js_api_parallel_chrome: filter = Bench
 
 .PHONY: bench_web_js_api_parallel_chrome # Run benchmarks for the web wasm api
@@ -1586,6 +1589,7 @@ bench_web_js_api_parallel_chrome_ci: setup_venv
 bench_web_js_api_parallel_firefox: browser_path = "$(WEB_RUNNER_DIR)/firefox/firefox/firefox"
 bench_web_js_api_parallel_firefox: driver_path = "$(WEB_RUNNER_DIR)/firefox/geckodriver"
 bench_web_js_api_parallel_firefox: browser_kind = firefox
+bench_web_js_api_parallel_firefox: server_cmd = "npm run server:multithreaded"
 bench_web_js_api_parallel_firefox: filter = Bench
 
 .PHONY: bench_web_js_api_parallel_firefox # Run benchmarks for the web wasm api
@@ -1598,6 +1602,38 @@ bench_web_js_api_parallel_firefox_ci: setup_venv
 	nvm use $(NODE_VERSION) && \
 	$(MAKE) bench_web_js_api_parallel_firefox
 
+bench_web_js_api_unsafe_coop_chrome: browser_path = "$(WEB_RUNNER_DIR)/chrome/chrome-linux64/chrome"
+bench_web_js_api_unsafe_coop_chrome: driver_path = "$(WEB_RUNNER_DIR)/chrome/chromedriver-linux64/chromedriver"
+bench_web_js_api_unsafe_coop_chrome: browser_kind = chrome
+bench_web_js_api_unsafe_coop_chrome: server_cmd = "npm run server:unsafe-coop"
+bench_web_js_api_unsafe_coop_chrome: filter = ZeroKnowledgeBench # Only bench zk with unsafe coop
+
+.PHONY: bench_web_js_api_unsafe_coop_chrome # Run benchmarks for the web wasm api without cross-origin isolation
+bench_web_js_api_unsafe_coop_chrome: run_web_js_api_parallel
+
+.PHONY: bench_web_js_api_unsafe_coop_chrome_ci # Run benchmarks for the web wasm api without cross-origin isolation
+bench_web_js_api_unsafe_coop_chrome_ci: setup_venv
+	source ~/.nvm/nvm.sh && \
+	nvm install $(NODE_VERSION) && \
+	nvm use $(NODE_VERSION) && \
+	$(MAKE) bench_web_js_api_unsafe_coop_chrome
+
+bench_web_js_api_unsafe_coop_firefox: browser_path = "$(WEB_RUNNER_DIR)/firefox/firefox/firefox"
+bench_web_js_api_unsafe_coop_firefox: driver_path = "$(WEB_RUNNER_DIR)/firefox/geckodriver"
+bench_web_js_api_unsafe_coop_firefox: browser_kind = firefox
+bench_web_js_api_unsafe_coop_firefox: server_cmd = "npm run server:unsafe-coop"
+bench_web_js_api_unsafe_coop_firefox: filter = ZeroKnowledgeBench # Only bench zk with unsafe coop
+
+.PHONY: bench_web_js_api_unsafe_coop_firefox # Run benchmarks for the web wasm api without cross-origin isolation
+bench_web_js_api_unsafe_coop_firefox: run_web_js_api_parallel
+
+.PHONY: bench_web_js_api_unsafe_coop_firefox_ci # Run benchmarks for the web wasm api without cross-origin isolation
+bench_web_js_api_unsafe_coop_firefox_ci: setup_venv
+	source ~/.nvm/nvm.sh && \
+	nvm install $(NODE_VERSION) && \
+	nvm use $(NODE_VERSION) && \
+	$(MAKE) bench_web_js_api_unsafe_coop_firefox
+
 .PHONY: bench_hlapi # Run benchmarks for integer operations
 bench_hlapi: install_rs_check_toolchain
 	RUSTFLAGS="$(RUSTFLAGS)" __TFHE_RS_BENCH_BIT_SIZES_SET=$(BIT_SIZES_SET) \

WORKFLOWS.md

@@ -0,0 +1,704 @@
+# 🔄 TFHE-rs GitHub Workflows Documentation
+
+This document provides a comprehensive overview of all GitHub Actions workflows in the TFHE-rs project, organized by category with visual diagrams showing their triggers and purposes.
+
+## 📊 Workflow Overview
+
+The project contains **71 workflows** organized into the following categories:
+
+- **Testing & Validation** (31 workflows) - AWS CPU (7), GPU (16), HPU (1), M1 (1), special tests (4), cargo tests (2)
+- **Benchmarking** (17 workflows) - CPU, GPU, HPU, WASM, specialized benchmarks
+- **Building & Compilation** (4 workflows) - Cargo builds
+- **Release Management** (9 workflows) - Publishing to crates.io and npm
+- **CI/CD & Maintenance** (10 workflows) - Linting, PR management, security
+
+---
+
+## 🔍 Workflow Trigger Types
+
+```mermaid
+graph LR
+    A[Workflow Triggers] --> B[Pull Request]
+    A --> C[Push to main]
+    A --> D[Schedule/Cron]
+    A --> E[Workflow Dispatch]
+    A --> F[Label Events]
+    
+    B --> B1[On PR Open]
+    B --> B2[On PR Approval]
+    F --> F1[approved label]
+    F --> F2[m1_test label]
+    D --> D1[Daily]
+    D --> D2[Weekly]
+    D --> D3[Nightly]
+```
+
+---
+
+## 🧪 Testing & Validation Workflows
+
+### CPU Testing Workflows
+
+```mermaid
+flowchart TB
+    subgraph "CPU Test Workflows"
+        AWS[aws_tfhe_tests]
+        FAST[aws_tfhe_fast_tests]
+        INT[aws_tfhe_integer_tests]
+        SIGN[aws_tfhe_signed_integer_tests]
+        BACK[aws_tfhe_backward_compat_tests]
+        WASM[aws_tfhe_wasm_tests]
+        NOISE[aws_tfhe_noise_checks]
+        M1[m1_tests]
+    end
+    
+    subgraph "Triggers"
+        PR[Pull Request<br/>+ approved label]
+        SCHED[Schedule<br/>Nightly Mon-Fri]
+        DISP[Workflow Dispatch]
+        M1LABEL[m1_test label]
+    end
+    
+    PR --> AWS
+    SCHED --> AWS
+    DISP --> AWS
+    
+    PR --> FAST
+    PR --> INT
+    PR --> SIGN
+    PR --> BACK
+    PR --> WASM
+    
+    DISP --> M1
+    M1LABEL --> M1
+    SCHED --> M1
+```
+
+| Workflow | Trigger | Purpose | Runner |
+|----------|---------|---------|--------|
+| **aws_tfhe_tests** | PR (approved) / Nightly / Manual | Comprehensive CPU tests (csprng, zk-pok, core_crypto, boolean, shortint, strings, high-level API, C API, examples, apps) | AWS cpu-big |
+| **aws_tfhe_fast_tests** | PR (approved) / Manual | Fast subset of tests for quick validation | AWS cpu-small |
+| **aws_tfhe_integer_tests** | PR (approved) / Manual | Integer operations testing | AWS cpu-big |
+| **aws_tfhe_signed_integer_tests** | PR (approved) / Manual | Signed integer operations testing | AWS cpu-big |
+| **aws_tfhe_backward_compat_tests** | PR (approved) / Manual | Backward compatibility validation | AWS cpu-small |
+| **aws_tfhe_wasm_tests** | PR (approved) / Manual | WebAssembly tests | AWS cpu-small |
+| **aws_tfhe_noise_checks** | PR (approved) / Manual | Cryptographic noise validation | AWS cpu-small |
+| **m1_tests** | Manual / Schedule (10pm daily) / m1_test label | Tests on Apple M1 architecture | Self-hosted M1 Mac |
+
+---
+
+### GPU Testing Workflows
+
+```mermaid
+flowchart TB
+    subgraph "GPU Test Workflows"
+        GFAST[gpu_fast_tests]
+        G4090[gpu_4090_tests]
+        GH100F[gpu_fast_h100_tests]
+        GH100[gpu_full_h100_tests]
+        GMULTI[gpu_full_multi_gpu_tests]
+        GVAL[gpu_code_validation_tests]
+        GMEM[gpu_memory_sanitizer]
+        GMEMH[gpu_memory_sanitizer_h100]
+        GUINT[gpu_unsigned_integer_tests]
+        GSINT[gpu_signed_integer_tests]
+        GUINTC[gpu_unsigned_integer_classic_tests]
+        GSINTC[gpu_signed_integer_classic_tests]
+        GUINTH[gpu_unsigned_integer_h100_tests]
+        GSINTH[gpu_signed_integer_h100_tests]
+        GLONG[gpu_integer_long_run_tests]
+        GPCC[gpu_pcc]
+    end
+    
+    subgraph "Triggers"
+        PR[Pull Request]
+        DISP[Workflow Dispatch]
+        APPR[PR approved label]
+    end
+    
+    PR --> GFAST
+    DISP --> GFAST
+    
+    DISP --> G4090
+    DISP --> GH100F
+    DISP --> GH100
+    DISP --> GMULTI
+    DISP --> GVAL
+    
+    APPR --> GMEM
+    APPR --> GMEMH
+    APPR --> GUINT
+    APPR --> GSINT
+    APPR --> GPCC
+```
+
+| Workflow | Trigger | Purpose | GPU |
+|----------|---------|---------|-----|
+| **gpu_fast_tests** | PR / Manual | Quick GPU validation tests | Hyperstack GPU |
+| **gpu_4090_tests** | Manual | Tests on RTX 4090 hardware | RTX 4090 |
+| **gpu_fast_h100_tests** | Manual | Fast tests on H100 GPU | H100 |
+| **gpu_full_h100_tests** | Manual | Comprehensive H100 tests | H100 |
+| **gpu_full_multi_gpu_tests** | Manual | Multi-GPU testing | Multiple GPUs |
+| **gpu_code_validation_tests** | Manual | GPU code validation | GPU |
+| **gpu_memory_sanitizer** | PR (approved) / Manual | Memory leak detection | GPU |
+| **gpu_memory_sanitizer_h100** | PR (approved) / Manual | Memory sanitizer on H100 | H100 |
+| **gpu_unsigned_integer_tests** | PR (approved) / Manual | Unsigned integer GPU tests | GPU |
+| **gpu_signed_integer_tests** | PR (approved) / Manual | Signed integer GPU tests | GPU |
+| **gpu_unsigned_integer_classic_tests** | Manual | Classic unsigned integer tests | GPU |
+| **gpu_signed_integer_classic_tests** | Manual | Classic signed integer tests | GPU |
+| **gpu_unsigned_integer_h100_tests** | Manual | Unsigned integer tests on H100 | H100 |
+| **gpu_signed_integer_h100_tests** | Manual | Signed integer tests on H100 | H100 |
+| **gpu_integer_long_run_tests** | Manual | Long-running integer tests | GPU |
+| **gpu_pcc** | PR (approved) / Manual | GPU PCC checks | GPU |
+
+---
+
+### HPU Testing Workflows
+
+```mermaid
+flowchart LR
+    HPU[hpu_hlapi_tests]
+    DISP[Workflow Dispatch] --> HPU
+    HPU --> |Tests on|INTEL[Intel HPU Hardware]
+```
+
+| Workflow | Trigger | Purpose | Hardware |
+|----------|---------|---------|----------|
+| **hpu_hlapi_tests** | Manual | High-level API tests on Intel HPU | Intel HPU |
+
+---
+
+### Special Testing Workflows
+
+```mermaid
+flowchart TB
+    subgraph "Special Tests"
+        COV[code_coverage]
+        CSPRNG[csprng_randomness_tests]
+        LONG[integer_long_run_tests]
+        PARAMS[parameters_check]
+    end
+    
+    subgraph "Cargo Tests"
+        TESTFFT[cargo_test_fft]
+        TESTNTT[cargo_test_ntt]
+    end
+    
+    DISP[Workflow Dispatch] --> COV
+    DISP --> CSPRNG
+    DISP --> LONG
+    
+    APPR[PR approved label] --> CSPRNG
+    
+    PUSH[Push to main] --> PARAMS
+    PR[PR on specific paths] --> PARAMS
+    DISP --> PARAMS
+    
+    PR --> TESTFFT
+    PR --> TESTNTT
+```
+
+| Workflow | Trigger | Purpose |
+|----------|---------|---------|
+| **code_coverage** | Manual | Generate code coverage reports and upload to Codecov |
+| **csprng_randomness_tests** | Manual / PR (approved) | Dieharder randomness test suite for CSPRNG |
+| **integer_long_run_tests** | Manual | Extended integer testing |
+| **parameters_check** | Push to main / PR (specific paths) / Manual | Security check on cryptographic parameters using lattice estimator |
+| **cargo_test_fft** | PR | Run tfhe-fft tests |
+| **cargo_test_ntt** | PR | Run tfhe-ntt tests |
+
+---
+
+## 🏗️ Building & Compilation Workflows (4 workflows)
+
+```mermaid
+flowchart TB
+    subgraph "Build Workflows"
+        BUILD[cargo_build]
+        COMMON[cargo_build_common]
+        FFT[cargo_build_tfhe_fft]
+        NTT[cargo_build_tfhe_ntt]
+    end
+    
+    subgraph "Build Jobs"
+        PCC[Parallel PCC CPU]
+        PCCHPU[PCC HPU]
+        FULL[Build TFHE Full]
+        LAYERS[Build Layers]
+        CAPI[Build C API]
+    end
+    
+    PR[Pull Request] --> BUILD
+    BUILD --> PCC
+    BUILD --> PCCHPU
+    BUILD --> FULL
+    BUILD --> LAYERS
+    BUILD --> CAPI
+    
+    PR --> FFT
+    PR --> NTT
+```
+
+| Workflow | Trigger | Purpose |
+|----------|---------|---------|
+| **cargo_build** | PR | Main build workflow - coordinates all build jobs |
+| **cargo_build_common** | Reusable | Shared build logic for different targets |
+| **cargo_build_tfhe_fft** | PR | Build and validate tfhe-fft crate |
+| **cargo_build_tfhe_ntt** | PR | Build and validate tfhe-ntt crate |
+
+**Build Targets:**
+- ✅ Parallel PCC (Program Counter Checks) for CPU
+- ✅ PCC for HPU
+- ✅ Full TFHE build (Linux, macOS M1, Windows)
+- ✅ Layer-by-layer builds
+- ✅ C API builds
+
+---
+
+## 📊 Benchmarking Workflows (17 workflows)
+
+All benchmark workflows are **triggered manually** via workflow_dispatch.
+
+```mermaid
+flowchart TB
+    subgraph "CPU Benchmarks - 3 workflows"
+        BCPU[benchmark_cpu<br/>Main CPU benchmarks]
+        BCPUW[benchmark_cpu_weekly<br/>Weekly CPU benchmarks]
+        BCPUC[benchmark_cpu_common<br/>Reusable workflow]
+    end
+    
+    subgraph "GPU Benchmarks - 5 workflows"
+        BGPU[benchmark_gpu<br/>Main GPU benchmarks]
+        BGPUW[benchmark_gpu_weekly<br/>Weekly GPU benchmarks]
+        BGPUC[benchmark_gpu_common<br/>Reusable workflow]
+        BGPU4090[benchmark_gpu_4090<br/>RTX 4090 specific]
+        BGPUCOP[benchmark_gpu_coprocessor<br/>Coprocessor mode]
+    end
+    
+    subgraph "HPU Benchmarks - 2 workflows"
+        BHPU[benchmark_hpu<br/>Intel HPU benchmarks]
+        BHPUC[benchmark_hpu_common<br/>Reusable workflow]
+    end
+    
+    subgraph "Specialized Benchmarks - 7 workflows"
+        BWASM[benchmark_wasm_client<br/>WebAssembly client]
+        BCT[benchmark_ct_key_sizes<br/>Ciphertext & key sizes]
+        BFFT[benchmark_tfhe_fft<br/>FFT performance]
+        BNTT[benchmark_tfhe_ntt<br/>NTT performance]
+        BWHITE[benchmark_whitepaper<br/>Whitepaper params]
+        BREG[benchmark_perf_regression<br/>Regression detection]
+        BDOC[benchmark_documentation<br/>Generate docs]
+    end
+    
+    DISP[Workflow Dispatch<br/>Manual Trigger] --> BCPU
+    DISP --> BCPUW
+    DISP --> BGPU
+    DISP --> BGPUW
+    DISP --> BHPU
+    DISP --> BWASM
+    DISP --> BCT
+    DISP --> BFFT
+    DISP --> BNTT
+    DISP --> BWHITE
+    DISP --> BREG
+    DISP --> BDOC
+    DISP --> BGPU4090
+    DISP --> BGPUCOP
+```
+
+### CPU Benchmarks (3 workflows)
+
+| Workflow | Purpose | Operations Tested |
+|----------|---------|-------------------|
+| **benchmark_cpu** | Main CPU performance benchmarks | integer, signed_integer, integer_compression, integer_zk, shortint, shortint_oprf, hlapi, hlapi_erc20, hlapi_dex, hlapi_noise_squash, tfhe_zk_pok, boolean, pbs, pbs128, ks, ks_pbs |
+| **benchmark_cpu_weekly** | Weekly scheduled CPU benchmarks | Similar to benchmark_cpu |
+| **benchmark_cpu_common** | Reusable workflow for CPU benchmarks | Shared logic |
+
+### GPU Benchmarks (5 workflows)
+
+| Workflow | Purpose | Hardware |
+|----------|---------|----------|
+| **benchmark_gpu** | Main GPU performance benchmarks | Standard GPU |
+| **benchmark_gpu_weekly** | Weekly scheduled GPU benchmarks | Standard GPU |
+| **benchmark_gpu_4090** | Benchmarks on RTX 4090 | RTX 4090 |
+| **benchmark_gpu_coprocessor** | GPU coprocessor mode benchmarks | GPU |
+| **benchmark_gpu_common** | Reusable workflow for GPU benchmarks | Shared logic |
+
+### HPU Benchmarks (2 workflows)
+
+| Workflow | Purpose | Hardware |
+|----------|---------|----------|
+| **benchmark_hpu** | Intel HPU performance benchmarks | Intel HPU |
+| **benchmark_hpu_common** | Reusable workflow for HPU benchmarks | Shared logic |
+
+### Specialized Benchmarks (7 workflows)
+
+| Workflow | Purpose | Focus |
+|----------|---------|-------|
+| **benchmark_wasm_client** | WebAssembly client performance | WASM execution |
+| **benchmark_ct_key_sizes** | Measure ciphertext and key sizes | Memory footprint |
+| **benchmark_tfhe_fft** | FFT library performance | tfhe-fft crate |
+| **benchmark_tfhe_ntt** | NTT library performance | tfhe-ntt crate |
+| **benchmark_whitepaper** | Whitepaper parameter validation | Research params |
+| **benchmark_perf_regression** | Detect performance regressions | Regression testing |
+| **benchmark_documentation** | Generate benchmark documentation | Documentation |
+
+### Benchmark Configuration Options
+
+**📏 Operation Flavors:**
+- `default` - Standard operations
+- `fast_default` - Fast variant operations
+- `smart` - Smart operations (with automatic PBS)
+- `unchecked` - Unchecked operations (no PBS)
+- `misc` - Miscellaneous operations
+
+**🎯 Precision Sets:**
+- `fast` - Quick validation subset
+- `all` - All supported bit precisions
+- `documentation` - Precisions for documentation
+
+**⏱️ Benchmark Types:**
+- `latency` - Single operation timing
+- `throughput` - Operations per second
+- `both` - Both latency and throughput
+
+**🔧 Parameter Types:**
+- `classical` - Classical parameters
+- `multi_bit` - Multi-bit parameters
+- `classical + multi_bit` - Both parameter sets
+- `classical_documentation` - Classical for docs
+- `multi_bit_documentation` - Multi-bit for docs
+- `classical_documentation + multi_bit_documentation` - Both for docs
+
+---
+
+## 📦 Release Management Workflows (9 workflows)
+
+```mermaid
+flowchart TB
+    subgraph "Release Workflows"
+        RTFHE[make_release_tfhe]
+        RCUDA[make_release_cuda]
+        RHPU[make_release_hpu]
+        RFFT[make_release_tfhe_fft]
+        RNTT[make_release_tfhe_ntt]
+        RCSPRNG[make_release_tfhe_csprng]
+        RZK[make_release_zk_pok]
+        RVER[make_release_tfhe_versionable]
+        RCOMMON[make_release_common]
+    end
+    
+    DISP[Workflow Dispatch<br/>Manual Only] --> RTFHE
+    DISP --> RCUDA
+    DISP --> RHPU
+    DISP --> RFFT
+    DISP --> RNTT
+    DISP --> RCSPRNG
+    DISP --> RZK
+    DISP --> RVER
+    
+    RTFHE --> |Publishes to|CRATES[crates.io]
+    RTFHE --> |Publishes to|NPM[npm registry]
+    
+    style RTFHE fill:#ff6b6b
+    style DISP fill:#ffd93d
+```
+
+| Workflow | Purpose | Platforms |
+|----------|---------|-----------|
+| **make_release_tfhe** | Release main TFHE library | crates.io, npm (web & node packages) |
+| **make_release_cuda** | Release CUDA backend | crates.io |
+| **make_release_hpu** | Release HPU backend | crates.io |
+| **make_release_tfhe_fft** | Release FFT library | crates.io |
+| **make_release_tfhe_ntt** | Release NTT library | crates.io |
+| **make_release_tfhe_csprng** | Release CSPRNG library | crates.io |
+| **make_release_zk_pok** | Release Zero-Knowledge Proof of Knowledge library | crates.io |
+| **make_release_tfhe_versionable** | Release versionable trait library | crates.io |
+| **make_release_common** | Shared release logic | Reusable workflow |
+
+**Release Options:**
+- 🧪 Dry-run mode
+- 📦 Push to crates.io
+- 🌐 Push web JS package
+- 📱 Push Node.js package
+- 🏷️ Set NPM latest tag
+
+---
+
+## 🛠️ CI/CD & Maintenance Workflows (10 workflows)
+
+```mermaid
+flowchart TB
+    subgraph "Code Quality"
+        LINT[ci_lint]
+        COMMIT[check_commit]
+        AUDIT[cargo_audit]
+    end
+    
+    subgraph "PR Management"
+        APPROVE[approve_label]
+        UNVER[unverified_prs]
+        VERIFY[verify_triggering_actor]
+    end
+    
+    subgraph "Repository Sync"
+        SYNC[sync_on_push]
+    end
+    
+    subgraph "SVG Generation"
+        SVG[generate_svgs]
+        SVGC[generate_svg_common]
+    end
+    
+    PR[Pull Request] --> LINT
+    PR --> COMMIT
+    PR --> APPROVE
+    
+    SCHED1[Daily 4am UTC] --> AUDIT
+    SCHED2[Daily 1:30am UTC] --> UNVER
+    
+    PUSH[Push to main] --> SYNC
+    
+    DISP[Workflow Dispatch] --> SVG
+    DISP --> AUDIT
+    DISP --> SYNC
+```
+
+| Workflow | Trigger | Purpose |
+|----------|---------|---------|
+| **ci_lint** | PR | Lint workflows with actionlint & check security with zizmor |
+| **check_commit** | PR | Validate commit message format, line length, and signatures |
+| **approve_label** | PR / PR Review | Auto-manage "approved" label on PRs |
+| **cargo_audit** | Daily 4am UTC / Manual | Check dependencies for security vulnerabilities |
+| **unverified_prs** | Daily 1:30am UTC | Close PRs without CLA signature after 2 days |
+| **verify_triggering_actor** | Various | Verify actor permissions for sensitive workflows |
+| **sync_on_push** | Push to main / Manual | Sync repository to internal mirror |
+| **generate_svgs** | Manual | Generate parameter curve SVG visualizations |
+| **generate_svg_common** | Reusable | Common SVG generation logic |
+| **placeholder_workflow** | N/A | Template workflow |
+
+---
+
+## 🔐 Security & Quality Workflows
+
+```mermaid
+flowchart LR
+    subgraph "Security Checks"
+        A[commit signatures]
+        B[dependency audit]
+        C[zizmor security scan]
+        D[parameters security]
+    end
+    
+    subgraph "Quality Checks"
+        E[commit format]
+        F[actionlint]
+        G[code coverage]
+        H[randomness tests]
+    end
+    
+    COMMIT[check_commit] --> A
+    COMMIT --> E
+    
+    AUDIT[cargo_audit] --> B
+    
+    LINT[ci_lint] --> C
+    LINT --> F
+    
+    PARAMS[parameters_check] --> D
+    
+    COV[code_coverage] --> G
+    
+    CSPRNG[csprng_randomness_tests] --> H
+```
+
+---
+
+## 📈 Workflow Statistics
+
+### By Trigger Type
+
+| Trigger | Count | Examples |
+|---------|-------|----------|
+| **Workflow Dispatch** (Manual) | 65 | All benchmarks, releases, most tests |
+| **Pull Request** | 18 | Build, lint, fast tests, GPU tests |
+| **Pull Request (approved label)** | 12 | AWS tests, GPU memory tests |
+| **Schedule/Cron** | 5 | Nightly tests, audit, unverified PRs |
+| **Push to main** | 2 | Sync, parameters check |
+| **Label Events** | 3 | M1 tests, approve workflow |
+
+### By Runner Type
+
+| Runner | Count | Purpose |
+|--------|-------|---------|
+| **AWS CPU** | 15 | Main testing infrastructure |
+| **Hyperstack GPU** | 13 | GPU testing and benchmarks |
+| **Self-hosted M1 Mac** | 1 | Apple Silicon testing |
+| **Intel HPU** | 2 | HPU testing and benchmarks |
+| **Ubuntu Latest** | 25 | CI/CD, builds, coordination |
+| **Windows** | 1 | Windows builds |
+
+---
+
+## 🎯 Key Workflow Patterns
+
+### 1. Instance Management Pattern
+
+Many workflows follow this pattern for cost optimization:
+
+```mermaid
+sequenceDiagram
+    participant GitHub
+    participant Setup
+    participant Runner
+    participant Tests
+    participant Teardown
+    
+    GitHub->>Setup: Trigger workflow
+    Setup->>Runner: Start AWS/GPU instance
+    Runner->>Tests: Execute tests
+    Tests->>Teardown: Complete (success/fail)
+    Teardown->>Runner: Stop instance
+    Teardown->>GitHub: Send Slack notification
+```
+
+**Workflows using this pattern:**
+- All `aws_tfhe_*` workflows
+- All `gpu_*` workflows  
+- `hpu_hlapi_tests`
+- `code_coverage`
+- `parameters_check`
+- `csprng_randomness_tests`
+
+### 2. Branch Protection Rules (BPR)
+
+Workflows marked with `(bpr)` are required for PRs to be merged:
+
+- ✅ `cargo_build/cargo-builds (bpr)`
+- ✅ `ci_lint/lint-check (bpr)`
+- ✅ `check_commit/check-commit-pr (bpr)`
+
+### 3. File Change Detection
+
+Many workflows use `tj-actions/changed-files` to conditionally run tests based on changed files, optimizing CI time and resources.
+
+---
+
+## 🔄 Workflow Dependencies
+
+```mermaid
+graph TD
+    subgraph "Reusable Workflows"
+        COMMON[cargo_build_common]
+        BENCH_CPU_C[benchmark_cpu_common]
+        BENCH_GPU_C[benchmark_gpu_common]
+        BENCH_HPU_C[benchmark_hpu_common]
+        REL_COMMON[make_release_common]
+        SVG_COMMON[generate_svg_common]
+    end
+    
+    subgraph "Parent Workflows"
+        BUILD[cargo_build]
+        BENCH_CPU[benchmark_cpu]
+        BENCH_GPU[benchmark_gpu]
+        BENCH_HPU[benchmark_hpu]
+        RELEASES[make_release_*]
+        SVG[generate_svgs]
+    end
+    
+    BUILD --> COMMON
+    BENCH_CPU --> BENCH_CPU_C
+    BENCH_GPU --> BENCH_GPU_C
+    BENCH_HPU --> BENCH_HPU_C
+    RELEASES --> REL_COMMON
+    SVG --> SVG_COMMON
+```
+
+---
+
+## 📝 Workflow Naming Convention
+
+```
+<category>_<component>_<type>
+```
+
+Examples:
+- `aws_tfhe_tests` - AWS infrastructure, TFHE component, tests type
+- `gpu_fast_tests` - GPU infrastructure, fast variant, tests type
+- `benchmark_cpu_weekly` - Benchmark category, CPU target, weekly schedule
+- `make_release_tfhe` - Make/release action, TFHE component
+
+---
+
+## 🚀 Quick Reference
+
+### Running Tests on PR
+
+1. **Quick validation**: Automatic on PR creation
+   - `cargo_build` - Build checks
+   - `ci_lint` - Linting
+   - `check_commit` - Commit format
+   - `gpu_fast_tests` - Basic GPU tests
+
+2. **Full test suite**: After PR approval (add "approved" label)
+   - `aws_tfhe_tests` - Comprehensive CPU tests
+   - `gpu_memory_sanitizer` - Memory checks
+   - GPU integer tests
+
+3. **Special hardware**: Manual label addition
+   - Add `m1_test` label for M1 Mac tests
+
+### Running Benchmarks
+
+All benchmarks are **manual only** via workflow dispatch. Choose:
+- Target: CPU, GPU, HPU, or WASM
+- Operation flavor: default, smart, unchecked
+- Precision set: fast, all, documentation
+- Benchmark type: latency, throughput, both
+
+### Creating a Release
+
+1. Run appropriate `make_release_*` workflow
+2. Configure options (dry-run, push to crates, npm packages)
+3. Workflow handles versioning, building, and publishing
+4. Includes provenance and SLSA attestation
+
+---
+
+## 🔔 Notification System
+
+All critical workflows send Slack notifications on:
+- ❌ Failure
+- 🚫 Cancellation (non-PR events)
+- ⚠️ Instance teardown failures
+
+Notifications include:
+- Job status
+- Pull request link (if applicable)
+- Action run URL
+
+---
+
+## 📚 Additional Resources
+
+- **Workflow Files**: `.github/workflows/`
+- **Reusable Actions**: `.github/actions/`
+- **Configuration**: `ci/slab.toml`
+- **Scripts**: `scripts/` directory
+
+---
+
+## ✅ Verification Summary
+
+**Total Workflows: 71**
+
+Count by category:
+- Testing & Validation: **31 workflows** (7 AWS CPU + 16 GPU + 1 HPU + 1 M1 + 4 special + 2 cargo tests)
+- Benchmarking: **17 workflows** (3 CPU + 5 GPU + 2 HPU + 7 specialized)
+- Building & Compilation: **4 workflows**
+- Release Management: **9 workflows**
+- CI/CD & Maintenance: **10 workflows**
+
+**Verification:** 31 + 17 + 4 + 9 + 10 = **71** ✅
+
+*Last Updated: 2026-01-08*

apps/test-vectors/checksums.sha256

@@ -1,24 +1,32 @@
+08f31a47c29cc4d72ad32c0b5411fa20b3deef5b84558dd2fb892d3cdf90528a  data/toy_params/glwe_after_id_br_karatsuba.cbor
 29b6e3e7d27700004b70dca24d225816500490e2d6ee49b9af05837fd421896b  data/valid_params_128/lwe_after_spec_pbs.cbor
 2c70d1d78cc3760733850a353ace2b9c4705e840141b75841739e90e51247e18  data/valid_params_128/small_lwe_secret_key.cbor
+2fb4bb45c259b8383da10fc8f9459c40a6972c49b1696eb107f0a75640724be5  data/toy_params/lwe_after_id_pbs_karatsuba.cbor
 36c9080b636475fcacca503ce041bbfeee800fd3e1890dee559ea18defff9fe8  data/toy_params/glwe_after_id_br.cbor
 377761beeb4216cf5aa2624a8b64b8259f5a75c32d28e850be8bced3a0cdd6f5  data/toy_params/ksk.cbor
 59dba26d457f96478eda130cab5301fce86f23c6a8807de42f2a1e78c4985ca7  data/valid_params_128/lwe_ks.cbor
+5d80dd93fefae4f4f89484dfcd65bbe99cc32e7e3b0a90c33dd0d77516c0a023  data/valid_params_128/glwe_after_id_br_karatsuba.cbor
 656f0009c7834c5bcb61621e222047516054b9bc5d0593d474ab8f1c086b67a6  data/valid_params_128/lwe_after_id_pbs.cbor
 699580ca92b9c2f9e1f57fb1e312c9e8cb29714f7acdef9d2ba05f798546751f  data/toy_params/lwe_sum.cbor
 6e54ab41056984595b077baff70236d934308cf5c0c33b4482fbfb129b3756c6  data/valid_params_128/glwe_after_id_br.cbor
 70f5e5728822de05b49071efb5ec28551b0f5cc87aa709a455d8e7f04b9c96ee  data/toy_params/lwe_after_id_pbs.cbor
+76a5c52cab7fec1dc167da676c6cd39479cda6b2bb9f4e0573cb7d99c2692faa  data/valid_params_128/lwe_after_id_pbs_karatsuba.cbor
 7cc6803f5fbc3d5a1bf597f2b979ce17eecd3d6baca12183dea21022a7b65c52  data/toy_params/bsk.cbor
 7f3c40a134623b44779a556212477fea26eaed22450f3b6faeb8721d63699972  data/valid_params_128/lwe_sum.cbor
 837b3bd3245d4d0534ed255fdef896fb4fa6998a258a14543dfdadd0bfc9b6dd  data/toy_params/lwe_prod.cbor
+9ece8ca9c1436258b94e8c5e629b8722f9b18fdd415dd5209b6167a9dde8491c  data/toy_params/glwe_after_spec_br_karatsuba.cbor
 aa44aea29efd6d9e4d35a21a625d9cba155672e3f7ed3eddee1e211e62ad146b  data/valid_params_128/lwe_ms.cbor
 b7a037b9eaa88d6385167579b93e26a0cb6976d9b8967416fd1173e113bda199  data/valid_params_128/large_lwe_secret_key.cbor
+b7b8e3586128887bd682120f3e3a43156139bce5e3fe0b03284f8753a864d647  data/toy_params/lwe_after_spec_pbs_karatsuba.cbor
 bd00a8ae7494e400de5753029552ee1647efe7e17409b863a26a13b081099b8c  data/toy_params/lwe_after_spec_pbs.cbor
 c6df98676de04fe54b5ffc2eb30a82ebb706c9d7d5a4e0ed509700fec88761f7  data/toy_params/lwe_ms.cbor
 c7d5a864d5616a7d8ad50bbf40416e41e6c9b60c546dc14d4aa8fc40a418baa7  data/toy_params/large_lwe_secret_key.cbor
 c806533b325b1009db38be2f9bef5f3b2fad6b77b4c71f2855ccc9d3b4162e98  data/valid_params_128/lwe_b.cbor
 c9eb75bd2993639348a679cf48c06e3c38d1a513f48e5b0ce0047cea8cff6bbc  data/toy_params/lwe_a.cbor
+d3391969acf26dc69de0927ba279139d8d79999944069addc8ff469ad6c5ae2d  data/valid_params_128/lwe_after_spec_pbs_karatsuba.cbor
 d6da5baef0e787f6be56e218d8354e26904652602db964844156fdff08350ce6  data/toy_params/lwe_ks.cbor
 e591ab9af1b6a0aede273f9a3abb65a4c387feb5fa06a6959e9314058ca0f7e5  data/valid_params_128/ksk.cbor
+e59b002df3a9b01ad321ec51cf076fa35131ab9dbef141d1c54b717d61426c92  data/valid_params_128/glwe_after_spec_br_karatsuba.cbor
 e628354c81508a2d888016e8282df363dd12f1e19190b6475d4eb9d7ab8ae007  data/valid_params_128/glwe_after_spec_br.cbor
 e69d2d2c064fc8c0460b39191ca65338146990349954f5ec5ebd01d93610e7eb  data/valid_params_128/lwe_a.cbor
 e76c24b2a0c9a842ad13dda35473c2514f9e7d20983b5ea0759c4521a91626d9  data/valid_params_128/lwe_prod.cbor

apps/test-vectors/data/README.md

@@ -39,6 +39,9 @@ The following values are generated:
 | `glwe_after_spec_br` | The glwe returned by the application of the spec blind rotation on the mod switched ciphertexts.             | `GlweCiphertext<Vec<u64>>` | rot spec LUT |
 | `lwe_after_spec_pbs` | The lwe returned by the application of the sample extract operation on the output of the spec blind rotation | `LweCiphertext<Vec<u64>>`  | `spec(A)`    |
 
+Ciphertexts with the `_karatsuba` suffix are generated using the Karatsuba polynomial multiplication algorithm in the blind rotation, while default ciphertexts are generated using an FFT multiplication.
+This makes it easier to reproduce bit exact results.
+
 ### Encodings
 #### Non native encoding
 Warning: TFHE-rs uses a specific encoding for non native (ie: u32, u64) power of two ciphertext modulus. This encoding puts the encoded value in the high bits of the native integer.

apps/test-vectors/data/toy_params/glwe_after_id_br_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08f31a47c29cc4d72ad32c0b5411fa20b3deef5b84558dd2fb892d3cdf90528a
+size 4679

apps/test-vectors/data/toy_params/glwe_after_spec_br_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ece8ca9c1436258b94e8c5e629b8722f9b18fdd415dd5209b6167a9dde8491c
+size 4679

apps/test-vectors/data/toy_params/lwe_after_id_pbs_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fb4bb45c259b8383da10fc8f9459c40a6972c49b1696eb107f0a75640724be5
+size 2365

apps/test-vectors/data/toy_params/lwe_after_spec_pbs_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7b8e3586128887bd682120f3e3a43156139bce5e3fe0b03284f8753a864d647
+size 2365

apps/test-vectors/data/valid_params_128/glwe_after_id_br_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d80dd93fefae4f4f89484dfcd65bbe99cc32e7e3b0a90c33dd0d77516c0a023
+size 36935

apps/test-vectors/data/valid_params_128/glwe_after_spec_br_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e59b002df3a9b01ad321ec51cf076fa35131ab9dbef141d1c54b717d61426c92
+size 36935

apps/test-vectors/data/valid_params_128/lwe_after_id_pbs_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76a5c52cab7fec1dc167da676c6cd39479cda6b2bb9f4e0573cb7d99c2692faa
+size 18493

apps/test-vectors/data/valid_params_128/lwe_after_spec_pbs_karatsuba.cbor

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3391969acf26dc69de0927ba279139d8d79999944069addc8ff469ad6c5ae2d
+size 18493

apps/test-vectors/src/main.rs

@@ -265,6 +265,7 @@ fn generate_test_vectors<P: AsRef<Path>>(
 
     let mut id_lut = encoding.encode_lut(glwe_dimension, polynomial_size, ID_LUT);
     assert_data_not_zero(&id_lut);
+    let mut id_lut_karatsuba = id_lut.clone();
 
     blind_rotate_assign(&modswitched, &mut id_lut, &fourier_bsk);
     assert_data_not_zero(&id_lut);
@@ -287,8 +288,32 @@ fn generate_test_vectors<P: AsRef<Path>>(
     assert_data_not_zero(&lwe_pbs_id);
     store_data(path, &lwe_pbs_id, "lwe_after_id_pbs");
 
+    blind_rotate_karatsuba_assign(&modswitched, &mut id_lut_karatsuba, &bsk);
+    store_data(path, &id_lut_karatsuba, "glwe_after_id_br_karatsuba");
+
+    let mut lwe_pbs_karatsuba_id = LweCiphertext::new(
+        0u64,
+        glwe_dimension
+            .to_equivalent_lwe_dimension(polynomial_size)
+            .to_lwe_size(),
+        encoding.ciphertext_modulus,
+    );
+
+    extract_lwe_sample_from_glwe_ciphertext(
+        &id_lut_karatsuba,
+        &mut lwe_pbs_karatsuba_id,
+        MonomialDegree(0),
+    );
+
+    let decrypted_pbs_id = decrypt_lwe_ciphertext(&large_lwe_secret_key, &lwe_pbs_karatsuba_id);
+    let res = encoding.decode(decrypted_pbs_id);
+
+    assert_eq!(res, MSG_A);
+    store_data(path, &lwe_pbs_karatsuba_id, "lwe_after_id_pbs_karatsuba");
+
     let mut spec_lut = encoding.encode_lut(glwe_dimension, polynomial_size, SPEC_LUT);
     assert_data_not_zero(&spec_lut);
+    let mut spec_lut_karatsuba = spec_lut.clone();
 
     blind_rotate_assign(&modswitched, &mut spec_lut, &fourier_bsk);
     assert_data_not_zero(&spec_lut);
@@ -310,6 +335,33 @@ fn generate_test_vectors<P: AsRef<Path>>(
     assert_eq!(res, SPEC_LUT(MSG_A));
     assert_data_not_zero(&lwe_pbs_spec);
     store_data(path, &lwe_pbs_spec, "lwe_after_spec_pbs");
+
+    blind_rotate_karatsuba_assign(&modswitched, &mut spec_lut_karatsuba, &bsk);
+    store_data(path, &spec_lut_karatsuba, "glwe_after_spec_br_karatsuba");
+
+    let mut lwe_pbs_karatsuba_spec = LweCiphertext::new(
+        0u64,
+        glwe_dimension
+            .to_equivalent_lwe_dimension(polynomial_size)
+            .to_lwe_size(),
+        encoding.ciphertext_modulus,
+    );
+
+    extract_lwe_sample_from_glwe_ciphertext(
+        &spec_lut_karatsuba,
+        &mut lwe_pbs_karatsuba_spec,
+        MonomialDegree(0),
+    );
+
+    let decrypted_pbs_spec = decrypt_lwe_ciphertext(&large_lwe_secret_key, &lwe_pbs_karatsuba_spec);
+    let res = encoding.decode(decrypted_pbs_spec);
+
+    assert_eq!(res, SPEC_LUT(MSG_A));
+    store_data(
+        path,
+        &lwe_pbs_karatsuba_spec,
+        "lwe_after_spec_pbs_karatsuba",
+    );
 }
 
 fn rm_dir_except_readme<P: AsRef<Path>>(dir: P) {

backends/tfhe-hpu-backend/Cargo.toml

@@ -61,8 +61,8 @@ bitvec = { version = "1.0", optional = true }
 serde_json = { version = "1.0", optional = true }
 
 # Dependencies used for v80 pdi handling
-bincode ={ version = "1.3", optional = true}
-serde_derive ={ version = "1.0", optional = true}
+bincode = { workspace = true, optional = true }
+serde_derive = { version = "1.0", optional = true }
 
 # Binary for manual debugging
 # Enable to access Hpu register and drive some custom sequence by hand

backends/tfhe-hpu-backend/src/fw/fw_impl/llt/mod.rs

@@ -239,7 +239,12 @@ pub fn iop_erc_20(prog: &mut Program) {
 pub fn iop_erc_20_simd(prog: &mut Program) {
     // Add Comment header
     prog.push_comment("ERC_20_SIMD (new_from, new_to) <- (from, to, amount)".to_string());
-    simd(prog, crate::asm::iop::SIMD_N, fw_impl::llt::iop_erc_20_rtl, None);
+    simd(
+        prog,
+        crate::asm::iop::SIMD_N,
+        fw_impl::llt::iop_erc_20_rtl,
+        None,
+    );
 }
 
 #[instrument(level = "trace", skip(prog))]
@@ -430,7 +435,8 @@ pub fn iop_erc_20_rtl(prog: &mut Program, batch_index: u8, kogge_blk_w: Option<u
         if let Some(blk_w) = kogge_blk_w {
             kogge::add(prog, dst_to, src_to, src_amount.clone(), None, blk_w)
                 + kogge::sub(prog, dst_from, src_from, src_amount, blk_w)
-        } else { // Default to ripple carry
+        } else {
+            // Default to ripple carry
             kogge::ripple_add(dst_to, src_to, src_amount.clone(), None)
                 + kogge::ripple_sub(prog, dst_from, src_from, src_amount)
         }

ci/webdriver.py

@@ -367,6 +367,8 @@ def dump_benchmark_results(results, browser_kind):
     """
     Dump as JSON benchmark results into a file.
     If `results` is an empty dict then this function is a no-op.
+    If the file already exists, new results are merged with existing ones,
+    overwriting keys that already exist.
 
     :param results: benchmark results as :class:`dict`
     :param browser_kind: browser as :class:`BrowserKind`
@@ -376,7 +378,15 @@ def dump_benchmark_results(results, browser_kind):
             key.replace("mean", "_".join((browser_kind.name, "mean"))): val
             for key, val in results.items()
         }
-        pathlib.Path("tfhe-benchmark/wasm_benchmark_results.json").write_text(json.dumps(results))
+        results_path = pathlib.Path("tfhe-benchmark/wasm_benchmark_results.json")
+        existing_results = {}
+        if results_path.exists():
+            try:
+                existing_results = json.loads(results_path.read_text())
+            except json.JSONDecodeError:
+                pass
+        existing_results.update(results)
+        results_path.write_text(json.dumps(existing_results))
 
 
 def start_web_server(

scripts/check_memory_errors.sh

@@ -33,7 +33,11 @@ RUSTFLAGS="$RUSTFLAGS" cargo nextest list --cargo-profile "${CARGO_PROFILE}" \
           --features=integer,internal-keycache,gpu-debug,zk-pok -p tfhe &> /tmp/test_list.txt
 
 if [[ "${RUN_VALGRIND}" == "1" ]]; then
-  TESTS_TO_RUN=$(sed -e $'s/\x1b\[[0-9;]*m//g' < /tmp/test_list.txt | grep -E 'high_level_api::.*gpu.*' | grep -v 'array' | grep -v 'flip')
+  # The tests are filtered using grep (to keep only HL) GPU tests.
+  # Since, when output is directed to a file, nextest outputs a list of `<executable name> <test name>` the `grep -o '[^ ]\+$'` filter
+  # will keep only the test name and the `tfhe` executable is assumed. To sanitize tests from another
+  # executable changes might be needed
+  TESTS_TO_RUN=$(sed -e $'s/\x1b\[[0-9;]*m//g' < /tmp/test_list.txt | grep -E 'high_level_api::.*gpu.*' | grep -v 'array' | grep -v 'flip' | grep -o '[^ ]\+$')
 
   # Build the tests but don't run them
   RUSTFLAGS="$RUSTFLAGS" cargo test --no-run --profile "${CARGO_PROFILE}" \
@@ -56,7 +60,11 @@ if [[ "${RUN_VALGRIND}" == "1" ]]; then
 fi
 
 if [[ "${RUN_COMPUTE_SANITIZER}" == "1" ]]; then
-  TESTS_TO_RUN=$(sed -e $'s/\x1b\[[0-9;]*m//g' < /tmp/test_list.txt | grep -E 'high_level_api::.*gpu.*|core_crypto::.*gpu.*' | grep -v 'array' | grep -v 'modulus_switch' | grep -v '3_3' | grep -v 'noise_distribution' | grep -v 'flip')
+  # The tests are filtered using grep (to keep only HL / corecrypto) GPU tests.
+  # Since, when output is directed to a file, nextest outputs a list of `<executable name> <test name>` the `grep -o '[^ ]\+$'` filter
+  # will keep only the test name and the `tfhe` executable is assumed. To sanitize tests from another
+  # executable changes might be needed
+  TESTS_TO_RUN=$(sed -e $'s/\x1b\[[0-9;]*m//g' < /tmp/test_list.txt | grep -E 'high_level_api::.*gpu.*|core_crypto::.*gpu.*' | grep -v 'array' | grep -v 'modulus_switch' | grep -v '3_3' | grep -v 'noise_distribution' | grep -v 'flip' | grep -o '[^ ]\+$')
   # Build the tests but don't run them
   RUSTFLAGS="$RUSTFLAGS" cargo test --no-run --profile "${CARGO_PROFILE}" \
     --features=integer,internal-keycache,gpu,zk-pok -p tfhe

tfhe-benchmark/Cargo.toml

@@ -15,7 +15,7 @@ name = "benchmark"
 path = "src/lib.rs"
 
 [dependencies]
-bincode = "1.3.3"
+bincode = { workspace = true }
 # clap has to be pinned as its minimum supported rust version
 # changes often between minor releases, which breaks our CI
 clap = { version = "=4.5.30", features = ["derive"] }

tfhe-benchmark/benches/high_level_api/erc20.rs

@@ -28,12 +28,22 @@ pub fn transfer_whitepaper<FheType>(
     amount: &FheType,
 ) -> (FheType, FheType)
 where
-    FheType: Add<Output = FheType> + for<'a> FheOrd<&'a FheType>,
-    FheBool: IfThenZero<FheType>,
+    FheType: Add<Output = FheType> + for<'a> FheOrd<&'a FheType> + FheTrivialEncrypt<u64>,
+    FheBool: IfThenZero<FheType> + IfThenElse<FheType>,
     for<'a> &'a FheType: Add<Output = FheType> + Sub<Output = FheType>,
 {
     let has_enough_funds = (from_amount).ge(amount);
-    let amount_to_transfer = has_enough_funds.if_then_zero(amount);
+    let amount_to_transfer = {
+        #[cfg(not(feature = "hpu"))]
+        {
+            let zero_amount = FheType::encrypt_trivial(0u64);
+            has_enough_funds.select(amount, &zero_amount)
+        }
+        #[cfg(feature = "hpu")]
+        {
+            has_enough_funds.if_then_zero(amount)
+        }
+    };
 
     let new_to_amount = to_amount + &amount_to_transfer;
     let new_from_amount = from_amount - &amount_to_transfer;
@@ -50,13 +60,21 @@ pub fn par_transfer_whitepaper<FheType>(
 where
     FheType:
         Add<Output = FheType> + for<'a> FheOrd<&'a FheType> + Send + Sync + FheTrivialEncrypt<u64>,
-    FheBool: IfThenZero<FheType>,
+    FheBool: IfThenZero<FheType> + IfThenElse<FheType>,
     for<'a> &'a FheType: Add<Output = FheType> + Sub<Output = FheType>,
 {
     let has_enough_funds = (from_amount).ge(amount);
-    //let zero_amount = FheType::encrypt_trivial(0u64);
-    //let amount_to_transfer = has_enough_funds.select(amount, &zero_amount);
-    let amount_to_transfer = has_enough_funds.if_then_zero(amount);
+    let amount_to_transfer = {
+        #[cfg(feature = "gpu")]
+        {
+            let zero_amount = FheType::encrypt_trivial(0u64);
+            has_enough_funds.select(amount, &zero_amount)
+        }
+        #[cfg(not(feature = "gpu"))]
+        {
+            has_enough_funds.if_then_zero(amount)
+        }
+    };
 
     let (new_to_amount, new_from_amount) = rayon::join(
         || to_amount + &amount_to_transfer,

tfhe-benchmark/src/params.rs

@@ -630,7 +630,7 @@ mod integer_params {
                 #[cfg(feature = "hpu")]
                 let params = vec![BENCH_HPU_PARAM_MESSAGE_2_CARRY_2_KS32_PBS_TUNIFORM_2M128.into()];
                 #[cfg(not(feature = "hpu"))]
-                let params = vec![BENCH_PARAM_MESSAGE_2_CARRY_2_KS_PBS_TUNIFORM_2M128.into()];
+                let params = vec![BENCH_PARAM_MESSAGE_2_CARRY_2_KS32_PBS.into()];
 
                 let params_and_bit_sizes = iproduct!(params, env_config.bit_sizes());
                 Self {

tfhe-benchmark/src/params_aliases.rs

@@ -1,12 +1,10 @@
 #[cfg(any(feature = "shortint", feature = "integer"))]
 pub mod shortint_params_aliases {
     use tfhe::shortint::parameters::current_params::*;
-    #[cfg(feature = "hpu")]
-    use tfhe::shortint::parameters::KeySwitch32PBSParameters;
     use tfhe::shortint::parameters::{
         ClassicPBSParameters, CompactPublicKeyEncryptionParameters, CompressionParameters,
-        MultiBitPBSParameters, NoiseSquashingCompressionParameters, NoiseSquashingParameters,
-        ShortintKeySwitchingParameters,
+        KeySwitch32PBSParameters, MultiBitPBSParameters, NoiseSquashingCompressionParameters,
+        NoiseSquashingParameters, ShortintKeySwitchingParameters,
     };
 
     // KS PBS Gaussian
@@ -42,6 +40,8 @@ pub mod shortint_params_aliases {
         V1_5_PARAM_MESSAGE_4_CARRY_4_KS_PBS_TUNIFORM_2M128;
     pub const BENCH_PARAM_MESSAGE_2_CARRY_2_KS_PBS: ClassicPBSParameters =
         V1_5_PARAM_MESSAGE_2_CARRY_2_KS_PBS_TUNIFORM_2M128;
+    pub const BENCH_PARAM_MESSAGE_2_CARRY_2_KS32_PBS: KeySwitch32PBSParameters =
+        V1_5_PARAM_MESSAGE_2_CARRY_2_KS32_PBS_TUNIFORM_2M128;
 
     pub const BENCH_ALL_CLASSIC_PBS_PARAMETERS: [(&ClassicPBSParameters, &str); 141] =
         VEC_ALL_CLASSIC_PBS_PARAMETERS;

tfhe-fft/Cargo.toml

@@ -30,7 +30,7 @@ serde = ["dep:serde", "num-complex/serde"]
 [dev-dependencies]
 rustfft = "6.0"
 rand = { workspace = true }
-bincode = "1.3"
+bincode = { workspace = true }
 more-asserts = "0.3.1"
 serde_json = "1.0.96"
 dyn-stack = { workspace = true, features = ["alloc"] }

tfhe-zk-pok/Cargo.toml

@@ -31,7 +31,7 @@ experimental = []
 [dev-dependencies]
 serde_json = "~1.0"
 itertools = { workspace = true }
-bincode = "1.3.3"
+bincode = { workspace = true }
 criterion = "0.5.1"
 
 [[bench]]

tfhe/Cargo.toml

@@ -58,7 +58,7 @@ tfhe-csprng = { version = "0.8.0", path = "../tfhe-csprng", features = [
 ] }
 serde = { workspace = true, features = ["default", "derive"] }
 rayon = { workspace = true }
-bincode = "1.3.3"
+bincode = { workspace = true }
 tfhe-fft = { version = "0.10.0", path = "../tfhe-fft", features = [
     "serde",
     "fft128",

tfhe/src/core_crypto/algorithms/lwe_programmable_bootstrapping/karatsuba_pbs.rs

@@ -0,0 +1,415 @@
+use aligned_vec::CACHELINE_ALIGN;
+use dyn_stack::{PodStack, StackReq};
+
+use crate::core_crypto::commons::traits::*;
+use crate::core_crypto::commons::utils::izip_eq;
+use crate::core_crypto::entities::*;
+use crate::core_crypto::fft_impl::fft64::crypto::ggsw::collect_next_term;
+use crate::core_crypto::fft_impl::fft64::math::decomposition::TensorSignedDecompositionLendingIter;
+use crate::core_crypto::prelude::polynomial_algorithms::*;
+use crate::core_crypto::prelude::{
+    extract_lwe_sample_from_glwe_ciphertext, lwe_ciphertext_modulus_switch, ComputationBuffers,
+    DecompositionBaseLog, DecompositionLevelCount, GlweSize, ModulusSwitchedLweCiphertext,
+    MonomialDegree, PolynomialSize, SignedDecomposer,
+};
+
+pub fn programmable_bootstrap_karatsuba_lwe_ciphertext_mem_optimized_requirement<Scalar>(
+    glwe_size: GlweSize,
+    polynomial_size: PolynomialSize,
+) -> StackReq {
+    StackReq::all_of(&[
+        // local accumulator
+        StackReq::new_aligned::<Scalar>(glwe_size.0 * polynomial_size.0, CACHELINE_ALIGN),
+        // blind rotation
+        blind_rotate_karatsuba_assign_scratch::<Scalar>(glwe_size, polynomial_size),
+    ])
+}
+
+/// Return the required memory for [`blind_rotate_karatsuba_assign`].
+pub fn blind_rotate_karatsuba_assign_scratch<Scalar>(
+    glwe_size: GlweSize,
+    polynomial_size: PolynomialSize,
+) -> StackReq {
+    StackReq::any_of(&[
+        // tmp_poly allocation
+        StackReq::new_aligned::<Scalar>(polynomial_size.0, CACHELINE_ALIGN),
+        StackReq::all_of(&[
+            // ct1 allocation
+            StackReq::new_aligned::<Scalar>(glwe_size.0 * polynomial_size.0, CACHELINE_ALIGN),
+            // external product
+            karatsuba_add_external_product_assign_scratch::<Scalar>(glwe_size, polynomial_size),
+        ]),
+    ])
+}
+
+/// Return the required memory for [`karatsuba_add_external_product_assign`].
+pub fn karatsuba_add_external_product_assign_scratch<Scalar>(
+    glwe_size: GlweSize,
+    polynomial_size: PolynomialSize,
+) -> StackReq {
+    StackReq::all_of(&[
+        // Output buffer
+        StackReq::new_aligned::<Scalar>(glwe_size.0 * polynomial_size.0, CACHELINE_ALIGN),
+        // decomposition
+        StackReq::new_aligned::<Scalar>(glwe_size.0 * polynomial_size.0, CACHELINE_ALIGN),
+        // decomposition term
+        StackReq::new_aligned::<Scalar>(glwe_size.0 * polynomial_size.0, CACHELINE_ALIGN),
+    ])
+}
+
+/// Perform a programmable bootstrap given an input [`LWE ciphertext`](`LweCiphertext`), a
+/// look-up table passed as a [`GLWE ciphertext`](`GlweCiphertext`) and an [`LWE bootstrap
+/// key`](`LweBootstrapKey`) using the karatsuba polynomial multiplication. The result is written in
+/// the provided output [`LWE ciphertext`](`LweCiphertext`).
+///
+/// If you want to manage the computation memory manually you can use
+/// [`programmable_bootstrap_karatsuba_lwe_ciphertext_mem_optimized`].
+///
+/// # Warning
+/// For a more efficient implementation of the programmable bootstrap, see
+/// [`programmable_bootstrap_lwe_ciphertext`](super::programmable_bootstrap_lwe_ciphertext)
+pub fn programmable_bootstrap_karatsuba_lwe_ciphertext<InputCont, OutputCont, AccCont, KeyCont>(
+    input: &LweCiphertext<InputCont>,
+    output: &mut LweCiphertext<OutputCont>,
+    accumulator: &GlweCiphertext<AccCont>,
+    bsk: &LweBootstrapKey<KeyCont>,
+) where
+    InputCont: Container<Element = u64>,
+    OutputCont: ContainerMut<Element = u64>,
+    AccCont: Container<Element = u64>,
+    KeyCont: Container<Element = u64>,
+{
+    assert!(
+        input.ciphertext_modulus().is_power_of_two(),
+        "This operation requires the input to have a power of two modulus."
+    );
+    assert_eq!(
+        output.ciphertext_modulus(),
+        accumulator.ciphertext_modulus()
+    );
+
+    let mut buffers = ComputationBuffers::new();
+
+    buffers.resize(
+        programmable_bootstrap_karatsuba_lwe_ciphertext_mem_optimized_requirement::<u64>(
+            bsk.glwe_size(),
+            bsk.polynomial_size(),
+        )
+        .unaligned_bytes_required(),
+    );
+
+    programmable_bootstrap_karatsuba_lwe_ciphertext_mem_optimized(
+        input,
+        output,
+        accumulator,
+        bsk,
+        buffers.stack(),
+    );
+}
+
+/// Perform a programmable bootstrap given an input [`LWE ciphertext`](`LweCiphertext`), a
+/// look-up table passed as a [`GLWE ciphertext`](`GlweCiphertext`) and an [`LWE bootstrap
+/// key`](`LweBootstrapKey`) using the karatsuba polynomial multiplication. The result is written in
+/// the provided output [`LWE ciphertext`](`LweCiphertext`).
+///
+/// # Warning
+/// For a more efficient implementation of the programmable bootstrap, see
+/// [`programmable_bootstrap_lwe_ciphertext_mem_optimized`](super::programmable_bootstrap_lwe_ciphertext_mem_optimized)
+pub fn programmable_bootstrap_karatsuba_lwe_ciphertext_mem_optimized<
+    InputCont,
+    OutputCont,
+    AccCont,
+    KeyCont,
+>(
+    input: &LweCiphertext<InputCont>,
+    output: &mut LweCiphertext<OutputCont>,
+    accumulator: &GlweCiphertext<AccCont>,
+    bsk: &LweBootstrapKey<KeyCont>,
+    stack: &mut PodStack,
+) where
+    InputCont: Container<Element = u64>,
+    OutputCont: ContainerMut<Element = u64>,
+    AccCont: Container<Element = u64>,
+    KeyCont: Container<Element = u64>,
+{
+    assert_eq!(
+        output.ciphertext_modulus(),
+        accumulator.ciphertext_modulus()
+    );
+    assert_eq!(accumulator.ciphertext_modulus(), bsk.ciphertext_modulus());
+
+    let (local_accumulator_data, stack) =
+        stack.collect_aligned(CACHELINE_ALIGN, accumulator.as_ref().iter().copied());
+    let mut local_accumulator = GlweCiphertextMutView::from_container(
+        &mut *local_accumulator_data,
+        accumulator.polynomial_size(),
+        accumulator.ciphertext_modulus(),
+    );
+
+    let log_modulus = accumulator
+        .polynomial_size()
+        .to_blind_rotation_input_modulus_log();
+
+    let msed = lwe_ciphertext_modulus_switch(input.as_view(), log_modulus);
+
+    blind_rotate_karatsuba_assign_mem_optimized(&msed, &mut local_accumulator, bsk, stack);
+
+    extract_lwe_sample_from_glwe_ciphertext(&local_accumulator, output, MonomialDegree(0));
+}
+
+/// Perform a blind rotation given an input [`modulus switched LWE
+/// ciphertext`](`ModulusSwitchedLweCiphertext`), modifying a look-up table passed as a [`GLWE
+/// ciphertext`](`GlweCiphertext`) and an [`LWE bootstrap key`](`LweBootstrapKey`) using the
+/// karatsuba polynomial multiplication.
+///
+/// If you want to manage the computation memory manually you can use
+/// [`blind_rotate_karatsuba_assign_mem_optimized`].
+///
+/// # Warning
+/// For a more efficient implementation of the blind rotation, see
+/// [`blind_rotate_assign`](super::blind_rotate_assign)
+pub fn blind_rotate_karatsuba_assign<OutputScalar, OutputCont, KeyCont>(
+    msed_input: &impl ModulusSwitchedLweCiphertext<usize>,
+    lut: &mut GlweCiphertext<OutputCont>,
+    bsk: &LweBootstrapKey<KeyCont>,
+) where
+    OutputScalar: UnsignedTorus + CastInto<usize>,
+    OutputCont: ContainerMut<Element = OutputScalar>,
+    KeyCont: Container<Element = OutputScalar>,
+    GlweCiphertext<OutputCont>: PartialEq<GlweCiphertext<OutputCont>>,
+{
+    let mut buffers = ComputationBuffers::new();
+
+    buffers.resize(
+        blind_rotate_karatsuba_assign_scratch::<u64>(bsk.glwe_size(), bsk.polynomial_size())
+            .unaligned_bytes_required(),
+    );
+
+    blind_rotate_karatsuba_assign_mem_optimized(msed_input, lut, bsk, buffers.stack())
+}
+
+/// Perform a blind rotation given an input [`modulus switched LWE
+/// ciphertext`](`ModulusSwitchedLweCiphertext`), modifying a look-up table passed as a [`GLWE
+/// ciphertext`](`GlweCiphertext`) and an [`LWE bootstrap key`](`LweBootstrapKey`) using the
+/// karatsuba polynomial multiplication.
+///
+/// # Warning
+/// For a more efficient implementation of the blind rotation, see
+/// [`blind_rotate_assign`](super::blind_rotate_assign)
+pub fn blind_rotate_karatsuba_assign_mem_optimized<OutputScalar, OutputCont, KeyCont>(
+    msed_input: &impl ModulusSwitchedLweCiphertext<usize>,
+    lut: &mut GlweCiphertext<OutputCont>,
+    bsk: &LweBootstrapKey<KeyCont>,
+    stack: &mut PodStack,
+) where
+    OutputScalar: UnsignedTorus + CastInto<usize>,
+    OutputCont: ContainerMut<Element = OutputScalar>,
+    KeyCont: Container<Element = OutputScalar>,
+    GlweCiphertext<OutputCont>: PartialEq<GlweCiphertext<OutputCont>>,
+{
+    assert!(lut.ciphertext_modulus().is_power_of_two());
+
+    assert_eq!(
+        bsk.input_lwe_dimension(),
+        msed_input.lwe_dimension(),
+        "Mismatched input LweDimension. \
+        LweBootstrapKey input LweDimension: {:?}, input LweCiphertext LweDimension {:?}.",
+        bsk.input_lwe_dimension(),
+        msed_input.lwe_dimension(),
+    );
+    assert_eq!(
+        bsk.glwe_size(),
+        lut.glwe_size(),
+        "Mismatched GlweSize. \
+        LweBootstrapKey GlweSize: {:?}, lut GlweSize {:?}.",
+        bsk.glwe_size(),
+        lut.glwe_size(),
+    );
+    assert_eq!(
+        lut.polynomial_size(),
+        bsk.polynomial_size(),
+        "Mismatched PolynomialSize. \
+        LweBootstrapKey PolynomialSize: {:?}, lut PolynomialSize {:?}.",
+        bsk.polynomial_size(),
+        lut.polynomial_size(),
+    );
+
+    let msed_lwe_mask = msed_input.mask();
+
+    let msed_lwe_body = msed_input.body();
+
+    let monomial_degree = MonomialDegree(msed_lwe_body.cast_into());
+
+    let lut_poly_size = lut.polynomial_size();
+    let ciphertext_modulus = lut.ciphertext_modulus();
+    assert!(ciphertext_modulus.is_compatible_with_native_modulus());
+
+    lut.as_mut_polynomial_list()
+        .iter_mut()
+        .for_each(|mut poly| {
+            let (tmp_poly, _) = stack.make_aligned_raw(poly.as_ref().len(), CACHELINE_ALIGN);
+
+            let mut tmp_poly = Polynomial::from_container(&mut *tmp_poly);
+            tmp_poly.as_mut().copy_from_slice(poly.as_ref());
+            polynomial_wrapping_monic_monomial_div(&mut poly, &tmp_poly, monomial_degree);
+        });
+
+    // We initialize the ct_0 used for the successive cmuxes
+    let ct0 = lut;
+    let (ct1, stack) = stack.make_aligned_raw(ct0.as_ref().len(), CACHELINE_ALIGN);
+    let mut ct1 =
+        GlweCiphertextMutView::from_container(&mut *ct1, lut_poly_size, ciphertext_modulus);
+
+    for (lwe_mask_element, bootstrap_key_ggsw) in izip_eq!(msed_lwe_mask, bsk.iter()) {
+        if lwe_mask_element != 0 {
+            let monomial_degree = MonomialDegree(lwe_mask_element);
+
+            // we effectively inline the body of cmux here, merging the initial subtraction
+            // operation with the monic polynomial multiplication, then performing the
+            // external product manually
+
+            // We rotate ct_1 and subtract ct_0 (first step of cmux) by performing
+            // ct_1 <- (ct_0 * X^a_i) - ct_0
+            for (mut ct1_poly, ct0_poly) in izip_eq!(
+                ct1.as_mut_polynomial_list().iter_mut(),
+                ct0.as_polynomial_list().iter(),
+            ) {
+                polynomial_wrapping_monic_monomial_mul_and_subtract(
+                    &mut ct1_poly,
+                    &ct0_poly,
+                    monomial_degree,
+                );
+            }
+
+            // second step of cmux:
+            // ct_0 <- ct_0 + ct1 * s_i
+            // with ct_0 + ct1s_i = ct_0 + ((ct_0 * X^a_i) - ct_0)s_i
+            //                    = ct_0          if s_i= 0
+            //                      ct_0 * X^a_i  otherwise
+            //                    = ct_0 * X^(a_i * s_i)
+            //
+            // as_mut_view is required to keep borrow rules consistent
+            karatsuba_add_external_product_assign(
+                ct0.as_mut_view(),
+                bootstrap_key_ggsw,
+                ct1.as_view(),
+                stack,
+            );
+        }
+    }
+
+    if !ciphertext_modulus.is_native_modulus() {
+        let signed_decomposer = SignedDecomposer::new(
+            DecompositionBaseLog(ciphertext_modulus.get_custom_modulus().ilog2() as usize),
+            DecompositionLevelCount(1),
+        );
+        ct0.as_mut()
+            .iter_mut()
+            .for_each(|x| *x = signed_decomposer.closest_representable(*x));
+    }
+}
+
+/// Perform the external product of `ggsw` and `glwe`, and adds the result to `out`.
+#[cfg_attr(feature = "__profiling", inline(never))]
+pub fn karatsuba_add_external_product_assign<Scalar>(
+    mut out: GlweCiphertextMutView<'_, Scalar>,
+    ggsw: GgswCiphertextView<Scalar>,
+    glwe: GlweCiphertextView<Scalar>,
+    stack: &mut PodStack,
+) where
+    Scalar: UnsignedTorus,
+{
+    // we check that the polynomial sizes match
+    debug_assert_eq!(ggsw.polynomial_size(), glwe.polynomial_size());
+    debug_assert_eq!(ggsw.polynomial_size(), out.polynomial_size());
+    // we check that the glwe sizes match
+    debug_assert_eq!(ggsw.glwe_size(), glwe.glwe_size());
+    debug_assert_eq!(ggsw.glwe_size(), out.glwe_size());
+
+    let align = CACHELINE_ALIGN;
+    let poly_size = ggsw.polynomial_size().0;
+
+    // we round the input mask and body
+    let decomposer = SignedDecomposer::<Scalar>::new(
+        ggsw.decomposition_base_log(),
+        ggsw.decomposition_level_count(),
+    );
+
+    let (output_buffer, substack0) =
+        stack.make_aligned_raw::<Scalar>(poly_size * ggsw.glwe_size().0, align);
+    // output_fft_buffer is initially uninitialized, considered to be implicitly zero, to avoid
+    // the cost of filling it up with zeros. `is_output_uninit` is set to `false` once
+    // it has been fully initialized for the first time.
+    let output_buffer = &mut *output_buffer;
+    let mut is_output_uninit = true;
+
+    let (mut decomposition, substack1) = TensorSignedDecompositionLendingIter::new(
+        glwe.as_ref()
+            .iter()
+            .map(|s| decomposer.init_decomposer_state(*s)),
+        DecompositionBaseLog(decomposer.base_log),
+        DecompositionLevelCount(decomposer.level_count),
+        substack0,
+    );
+
+    // We loop through the levels
+    for ggsw_decomp_matrix in ggsw.iter() {
+        // We retrieve the decomposition of this level.
+        let (_glwe_level, glwe_decomp_term, _substack2) =
+            collect_next_term(&mut decomposition, substack1, align);
+        let glwe_decomp_term = GlweCiphertextView::from_container(
+            &*glwe_decomp_term,
+            ggsw.polynomial_size(),
+            out.ciphertext_modulus(),
+        );
+
+        // For each level we have to add the result of the vector-matrix product between the
+        // decomposition of the glwe, and the ggsw level matrix to the output. To do so, we
+        // iteratively add to the output, the product between every line of the matrix, and
+        // the corresponding (scalar) polynomial in the glwe decomposition:
+        //
+        //                ggsw_mat                        ggsw_mat
+        //   glwe_dec   | - - - - | <        glwe_dec   | - - - - |
+        //  | - - - | x | - - - - |         | - - - | x | - - - - | <
+        //    ^         | - - - - |             ^       | - - - - |
+        //
+        //        t = 1                           t = 2                     ...
+
+        for (ggsw_row, glwe_poly) in izip_eq!(
+            ggsw_decomp_matrix.as_glwe_list().iter(),
+            glwe_decomp_term.as_polynomial_list().iter()
+        ) {
+            let row_as_poly_list = ggsw_row.as_polynomial_list();
+            if is_output_uninit {
+                for (mut output_poly, row_poly) in output_buffer
+                    .chunks_exact_mut(poly_size)
+                    .map(Polynomial::from_container)
+                    .zip(row_as_poly_list.iter())
+                {
+                    polynomial_wrapping_mul(&mut output_poly, &row_poly, &glwe_poly);
+                }
+            } else {
+                for (mut output_poly, row_poly) in output_buffer
+                    .chunks_exact_mut(poly_size)
+                    .map(Polynomial::from_container)
+                    .zip(row_as_poly_list.iter())
+                {
+                    polynomial_wrapping_add_mul_assign(&mut output_poly, &row_poly, &glwe_poly);
+                }
+            }
+
+            is_output_uninit = false;
+        }
+    }
+
+    // We iterate over the polynomials in the output.
+    if !is_output_uninit {
+        izip_eq!(
+            out.as_mut_polynomial_list().iter_mut(),
+            output_buffer
+                .into_chunks(poly_size)
+                .map(Polynomial::from_container),
+        )
+        .for_each(|(mut out, res)| polynomial_wrapping_add_assign(&mut out, &res));
+    }
+}

tfhe/src/core_crypto/algorithms/lwe_programmable_bootstrapping/mod.rs

@@ -1,10 +1,12 @@
 pub mod fft128_pbs;
 pub mod fft64_pbs;
+pub mod karatsuba_pbs;
 pub mod ntt64_bnf_pbs;
 pub mod ntt64_pbs;
 
 pub use fft128_pbs::*;
 pub use fft64_pbs::*;
+pub use karatsuba_pbs::*;
 pub use ntt64_bnf_pbs::*;
 pub use ntt64_pbs::*;
 

tfhe/src/core_crypto/algorithms/test/lwe_programmable_bootstrapping.rs

@@ -1161,3 +1161,91 @@ fn lwe_encrypt_pbs_ntt64_bnf_decrypt(params: ClassicTestParams<u64>) {
 create_parameterized_test!(lwe_encrypt_pbs_ntt64_bnf_decrypt {
     TEST_PARAMS_3_BITS_SOLINAS_U64
 });
+
+fn lwe_encrypt_pbs_karatsuba_decrypt_custom_mod(params: ClassicTestParams<u64>) {
+    let lwe_noise_distribution = params.lwe_noise_distribution;
+    let ciphertext_modulus = params.ciphertext_modulus;
+    let message_modulus_log = params.message_modulus_log;
+    let msg_modulus = 1 << (message_modulus_log.0);
+    let encoding_with_padding = get_encoding_with_padding(ciphertext_modulus);
+    let glwe_dimension = params.glwe_dimension;
+    let polynomial_size = params.polynomial_size;
+
+    let mut rsc = TestResources::new();
+
+    let f = |x: u64| x;
+
+    let delta: u64 = encoding_with_padding / msg_modulus;
+    let mut msg = msg_modulus;
+
+    let accumulator = generate_programmable_bootstrap_glwe_lut(
+        polynomial_size,
+        glwe_dimension.to_glwe_size(),
+        msg_modulus.cast_into(),
+        ciphertext_modulus,
+        delta,
+        f,
+    );
+
+    assert!(check_encrypted_content_respects_mod(
+        &accumulator,
+        ciphertext_modulus
+    ));
+
+    while msg != 0 {
+        msg = msg.wrapping_sub(1);
+
+        let mut keys_gen = |params| generate_keys(params, &mut rsc);
+        let keys = gen_keys_or_get_from_cache_if_enabled(params, &mut keys_gen);
+        let (input_lwe_secret_key, output_lwe_secret_key, bsk) =
+            (keys.small_lwe_sk, keys.big_lwe_sk, keys.bsk);
+
+        for _ in 0..NB_TESTS {
+            let plaintext = Plaintext(msg * delta);
+
+            let lwe_ciphertext_in = allocate_and_encrypt_new_lwe_ciphertext(
+                &input_lwe_secret_key,
+                plaintext,
+                lwe_noise_distribution,
+                ciphertext_modulus,
+                &mut rsc.encryption_random_generator,
+            );
+
+            assert!(check_encrypted_content_respects_mod(
+                &lwe_ciphertext_in,
+                ciphertext_modulus
+            ));
+
+            let mut out_pbs_ct = LweCiphertext::new(
+                0,
+                output_lwe_secret_key.lwe_dimension().to_lwe_size(),
+                ciphertext_modulus,
+            );
+
+            programmable_bootstrap_karatsuba_lwe_ciphertext(
+                &lwe_ciphertext_in,
+                &mut out_pbs_ct,
+                &accumulator,
+                &bsk,
+            );
+
+            assert!(check_encrypted_content_respects_mod(
+                &out_pbs_ct,
+                ciphertext_modulus
+            ));
+
+            let decrypted = decrypt_lwe_ciphertext(&output_lwe_secret_key, &out_pbs_ct);
+
+            let decoded = round_decode(decrypted.0, delta) % msg_modulus;
+
+            assert_eq!(decoded, f(msg));
+        }
+
+        // In coverage, we break after one while loop iteration, changing message values does not
+        // yield higher coverage
+        #[cfg(tarpaulin)]
+        break;
+    }
+}
+
+create_parameterized_test!(lwe_encrypt_pbs_karatsuba_decrypt_custom_mod);

tfhe/src/high_level_api/integers/unsigned/tests/cpu.rs

@@ -386,6 +386,12 @@ fn test_if_then_else() {
     super::test_case_if_then_else(&client_key);
 }
 
+#[test]
+fn test_if_then_zero() {
+    let client_key = setup_default_cpu();
+    super::test_case_if_then_zero(&client_key);
+}
+
 #[test]
 fn test_flip() {
     let client_key = setup_default_cpu();

tfhe/src/high_level_api/integers/unsigned/tests/hpu.rs

@@ -89,6 +89,12 @@ fn test_case_if_then_else_hpu() {
     super::test_case_if_then_else(&client_key);
 }
 
+#[test]
+fn test_case_if_then_zero_hpu() {
+    let client_key = setup_default_hpu();
+    super::test_case_if_then_zero(&client_key);
+}
+
 #[test]
 fn test_case_flip_hpu() {
     let client_key = setup_default_hpu();

tfhe/src/high_level_api/integers/unsigned/tests/mod.rs

@@ -568,6 +568,28 @@ fn test_case_if_then_else(client_key: &ClientKey) {
     );
 }
 
+fn test_case_if_then_zero(client_key: &ClientKey) {
+    let clear_a = 42u8;
+    let clear_b = 128u8;
+
+    let a = FheUint8::encrypt(clear_a, client_key);
+    let b = FheUint8::encrypt(clear_b, client_key);
+
+    let result = a.le(&b).if_then_zero(&a);
+    let decrypted_result: u8 = result.decrypt(client_key);
+    assert_eq!(
+        decrypted_result,
+        if clear_a <= clear_b { clear_a } else { 0 }
+    );
+
+    let result = a.ge(&b).if_then_zero(&a);
+    let decrypted_result: u8 = result.decrypt(client_key);
+    assert_eq!(
+        decrypted_result,
+        if clear_a >= clear_b { clear_a } else { 0 }
+    );
+}
+
 fn test_case_flip(client_key: &ClientKey) {
     let clear_a = rand::random::<u32>();
     let clear_b = rand::random::<u32>();

tfhe/src/high_level_api/prelude.rs

@@ -9,9 +9,9 @@
 pub use crate::high_level_api::traits::{
     BitSlice, CiphertextList, DivRem, FheDecrypt, FheEncrypt, FheEq, FheKeyswitch, FheMax, FheMin,
     FheOrd, FheTrivialEncrypt, FheTryEncrypt, FheTryTrivialEncrypt, FheWait, Flip, IfThenElse,
-    OverflowingAdd, OverflowingMul, OverflowingNeg, OverflowingSub, ReRandomize, RotateLeft,
-    RotateLeftAssign, RotateRight, RotateRightAssign, ScalarIfThenElse, SquashNoise, Tagged,
-    IfThenZero,
+    IfThenZero, OverflowingAdd, OverflowingMul, OverflowingNeg, OverflowingSub, ReRandomize,
+    RotateLeft, RotateLeftAssign, RotateRight, RotateRightAssign, ScalarIfThenElse, SquashNoise,
+    Tagged,
 };
 #[cfg(feature = "hpu")]
 pub use crate::high_level_api::traits::{FheHpu, HpuHandle};

tfhe/web_wasm_parallel_tests/index.js

@@ -12,14 +12,22 @@ function setButtonsDisabledState(buttonIds, state) {
 
 async function setup() {
   let supportsThreads = await threads();
-  if (!supportsThreads) {
-    console.error("This browser does not support threads");
-    return;
+  // This variable is set to true if we are using the `serve.multithreaded.json` config
+  if (crossOriginIsolated) {
+    if (supportsThreads) {
+      console.info("Running in multithreaded mode");
+    } else {
+      console.error("This browser does not support threads");
+      return;
+    }
+  } else {
+    console.warn("Running in unsafe coop mode");
   }
 
   const worker = new Worker(new URL("worker.js", import.meta.url), {
     type: "module",
   });
+
   const demos = await Comlink.wrap(worker).demos;
 
   const demoNames = [

tfhe/web_wasm_parallel_tests/package.json

@@ -5,7 +5,9 @@
   "main": "index.js",
   "scripts": {
     "build": "cp -r ../../tfhe/pkg ./ && webpack build ./index.js --mode production -o dist --output-filename index.js && cp index.html dist/ && cp favicon.ico dist/",
-    "server": "serve --config ../serve.json dist/",
+    "server": "npm run server:multithreaded",
+    "server:multithreaded": "serve --config ../serve.multithreaded.json dist/",
+    "server:unsafe-coop": "serve --config ../serve.unsafe-coop.json dist/",
     "format": "prettier . --write",
     "check-format": "prettier . --check"
   },

tfhe/web_wasm_parallel_tests/serve.unsafe-coop.json

@@ -0,0 +1,11 @@
+{
+  "headers": [
+    {
+      "source": "**/*.@(js|html)",
+      "headers": [
+        { "key": "Cross-Origin-Embedder-Policy", "value": "unsafe-none" },
+        { "key": "Cross-Origin-Opener-Policy", "value": "unsafe-none" }
+      ]
+    }
+  ]
+}

tfhe/web_wasm_parallel_tests/worker.js

@@ -1,4 +1,5 @@
 import * as Comlink from "comlink";
+import { threads } from "wasm-feature-detect";
 import init, {
   initThreadPool,
   init_panic_hook,
@@ -753,7 +754,10 @@ async function compactPublicKeyZeroKnowledgeBench() {
 
 async function main() {
   await init();
-  await initThreadPool(navigator.hardwareConcurrency);
+  let supportsThreads = await threads();
+  if (supportsThreads) {
+    await initThreadPool(navigator.hardwareConcurrency);
+  }
   await init_panic_hook();
 
   return Comlink.proxy({

utils/tfhe-backward-compat-data/Cargo.toml

@@ -8,7 +8,7 @@ license = "BSD-3-Clause-Clear"
 serde = { version = "1.0", features = ["derive"] }
 ron = { version = "0.8", features = ["integer128"] }
 ciborium = "0.2"
-bincode = "1.3"
+bincode = { workspace = true }
 strum = { version = "0.26", features = ["derive"] }
 semver = { version = "1.0" }
 clap = { version = "4.5", features = ["derive"] }

utils/tfhe-versionable/Cargo.toml

@@ -17,7 +17,7 @@ static_assertions = "1.1"
 trybuild = { version = "1", features = ["diff"] }
 
 # used to test various serialization formats
-bincode = "1.3"
+bincode = { workspace = true }
 serde_json = "1.0"
 ciborium = "0.2"
 rmp-serde = "1.3"