[TEST] Reduce crypto_sign_signature_internal stack usage
#791
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This was referenced Dec 15, 2025
- Add mld_polymat_get_row() to retrieve matrix row pointer - Update mld_polyvec_matrix_pointwise_montgomery() to use helper Addresses #738 (steps 2-3 of #736) Signed-off-by: Hanno Becker <[email protected]>
Signed-off-by: Hanno Becker <[email protected]>
Signed-off-by: Hanno Becker <[email protected]>
crypto_sign_verify_internal stack: before: 26928/37232/49776 after: 22784/31040/41536 Signed-off-by: Matthias J. Kannwischer <[email protected]>
crypto_sign_verify_internal stack: 21743/30016/40515 Signed-off-by: Matthias J. Kannwischer <[email protected]>
crypto_sign_verify_internal stack: 17664/24864/33312 Signed-off-by: Matthias J. Kannwischer <[email protected]>
crypto_sign_verify_internal stack: 14592/19744/26144 Signed-off-by: Matthias J. Kannwischer <[email protected]>
crypto_sign_verify_internal stack: 13568/18720/25120 Signed-off-by: Matthias J. Kannwischer <[email protected]>
71168 -> 64000 Signed-off-by: Matthias J. Kannwischer <[email protected]>
64000 -> 55808 Signed-off-by: Matthias J. Kannwischer <[email protected]>
55808 -> 40448 Signed-off-by: Matthias J. Kannwischer <[email protected]>
40448 -> 34304 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
force-pushed
the
sign-internal-ram
branch
from
1588071 to
1fa9c2c
Compare
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 30, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Dec 31, 2025
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 3, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 5, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 5, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 7, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
hanno-becker
pushed a commit
that referenced
this pull request
Jan 8, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 8, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
mkannwischer
added a commit
that referenced
this pull request
Jan 8, 2026
This commit reduces the stack usage of signing by computing z = y + s1*cp incrementally (one polynomial at a time) allowing to eliminate the polyvecl z (at to cost of a single poly z). The computation of z is moved into a separate function (compute_pack_z) to vastly speed up the CBMC proofs. De-facto this saves L-1 KB irrespective of MLD_CONFIG_REDUCE_RAM. Practically, the same buffer was used early in the function too. Here we instead introduce a new polyvecl buffer tmp, but that can be placed in a union together with w1. Unfortuantely, with the current struct workaround for diffblue/cbmc#8813, this results in an increase in stack space by L KB. This gets eliminated when MLD_CONFIG_REDUCE_RAM is set. Hoisted out from #791 Signed-off-by: Matthias J. Kannwischer <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.