Optimizations for Shamir-based Protocol #880
Open
+5,799
−105
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Pull Request ## What problem does this PR solve? Issue Number: Fixed # ## Possible side effects? - Performance: - Backward compatibility:
|
All contributors have signed the CLA ✍️ ✅ |
|
I have read the CLA Document and I hereby sign the CLA |
|
recheck |
w-gc
reviewed
Nov 5, 2024
| @@ -623,12 +625,10 @@ Value xor_pp(SPUContext* ctx, const Value& x, const Value& y) { | |||
| Value lshift_s(SPUContext* ctx, const Value& x, const Sizes& bits) { | |||
| SPU_TRACE_MPC_DISP(ctx, x, bits); | |||
| TRY_DISPATCH(ctx, x, bits); | |||
| if (IsA(x)) { | |||
| if (IsA(x) && ctx->hasKernel("lshift_a")) { | |||
There was a problem hiding this comment.
若是ArithShare，但有没有lshift_a，也走lshift_b吗？
There was a problem hiding this comment.
若是ArithShare，但有没有
lshift_a，也走lshift_b吗？
是的
| @@ -25,6 +25,7 @@ namespace spu::mpc::test { | |||
| namespace { | |||
|
|
|||
| Shape kShape = {20, 30}; | |||
| // Shape kShape = {3}; | |||
tpppppub
reviewed
Nov 5, 2024
| @@ -623,12 +625,10 @@ Value xor_pp(SPUContext* ctx, const Value& x, const Value& y) { | |||
| Value lshift_s(SPUContext* ctx, const Value& x, const Sizes& bits) { | |||
| SPU_TRACE_MPC_DISP(ctx, x, bits); | |||
| TRY_DISPATCH(ctx, x, bits); | |||
| if (IsA(x)) { | |||
| if (IsA(x) && ctx->hasKernel("lshift_a")) { | |||
There was a problem hiding this comment.
若是ArithShare，但有没有
lshift_a，也走lshift_b吗？
是的
| public: | ||
| static constexpr const char* kBindName() {return "rand_a"; } | ||
|
|
||
| Kind kind() const override { return Kind::Dynamic; } |
There was a problem hiding this comment.
please add the communication complexity
w-gc
reviewed
Nov 5, 2024
| @@ -0,0 +1,16 @@ | |||
| { | |||
|
|
||
| ce::CExpr comm() const override { return ce::Const(0); } | ||
|
|
||
| NdArrayRef proc(KernelEvalContext* ctx, const NdArrayRef& in) const override; |
There was a problem hiding this comment.
Please remove class NotA if not need.
w-gc
reviewed
Nov 5, 2024
| _prefix_prod[i] = mul_mod(_prefix_prod[i - 1], _x[i]); | ||
| } | ||
|
|
||
| ring2k_t pprod_inverse = mul_inv(_prefix_prod[numel - 1]); |
|
|
||
| ce::CExpr comm() const override { return ce::Const(0); } | ||
|
|
||
| NdArrayRef proc(KernelEvalContext* ctx, const NdArrayRef& in) const override; |
There was a problem hiding this comment.
Please remove class NotA if not need.
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.
Pull Request
What problem does this PR solve?
主要修改&优化：
libspu/mpc/utils/gfmp_ops.cc
libspu/mpc/shamir/arithmetic.cc
libspu/mpc/shamir/boolean.cc
libspu/mpc/shamir/conversion.cc
Issue Number: Fixed #
Possible side effects?
Performance:
Backward compatibility: