fix(compiler): use dyn sized tensors in CAPI func definitions

zama-ai · Jul 21, 2023 · 545bda9 · 545bda9
1 parent 7e138bf
commit 545bda9
Showing 1 changed file with 34 additions and 11 deletions.
diff --git a/compilers/concrete-compiler/compiler/lib/Conversion/SimulateTFHE/SimulateTFHE.cpp b/compilers/concrete-compiler/compiler/lib/Conversion/SimulateTFHE/SimulateTFHE.cpp
@@ -54,6 +54,13 @@ class SimulateTFHETypeConverter : public mlir::TypeConverter {
 
 namespace {
 
+mlir::RankedTensorType toDynamicTensorType(mlir::TensorType staticSizedTensor) {
+  std::vector<int64_t> dynSizedShape(staticSizedTensor.getShape().size(),
+                                     mlir::ShapedType::kDynamic);
+  return mlir::RankedTensorType::get(dynSizedShape,
+                                     staticSizedTensor.getElementType());
+}
+
 struct NegOpPattern : public mlir::OpConversionPattern<TFHE::NegGLWEOp> {
 
   NegOpPattern(mlir::MLIRContext *context, mlir::TypeConverter &typeConverter)
@@ -131,27 +138,37 @@ struct EncodeExpandLutForBootstrapOpPattern
             eeOp.getResult().getType().cast<mlir::RankedTensorType>(),
             mlir::ValueRange{});
 
+    auto dynamicResultType = toDynamicTensorType(eeOp.getResult().getType());
+    auto dynamicLutType =
+        toDynamicTensorType(eeOp.getInputLookupTable().getType());
+
+    mlir::Value castedOutputBuffer = rewriter.create<mlir::tensor::CastOp>(
+        eeOp.getLoc(), dynamicResultType, outputBuffer);
+
+    mlir::Value castedLUT = rewriter.create<mlir::tensor::CastOp>(
+        eeOp.getLoc(),
+        toDynamicTensorType(eeOp.getInputLookupTable().getType()),
+        adaptor.getInputLookupTable());
+
     // sim_encode_expand_lut_for_boostrap(uint64_t *out_allocated, uint64_t
     // *out_aligned, uint64_t out_offset, uint64_t out_size, uint64_t
     // out_stride, uint64_t *in_allocated, uint64_t *in_aligned, uint64_t
     // in_offset, uint64_t in_size, uint64_t in_stride, uint32_t poly_size,
     // uint32_t output_bits, bool is_signed)
     if (insertForwardDeclaration(
             eeOp, rewriter, funcName,
-            rewriter.getFunctionType({eeOp.getResult().getType(),
-                                      eeOp.getInputLookupTable().getType(),
-                                      rewriter.getIntegerType(32),
-                                      rewriter.getIntegerType(32),
-                                      rewriter.getIntegerType(1)},
-                                     {}))
+            rewriter.getFunctionType(
+                {dynamicResultType, dynamicLutType, rewriter.getIntegerType(32),
+                 rewriter.getIntegerType(32), rewriter.getIntegerType(1)},
+                {}))
             .failed()) {
       return mlir::failure();
     }
 
     rewriter.create<mlir::func::CallOp>(
         eeOp.getLoc(), funcName, mlir::TypeRange{},
-        mlir::ValueRange({outputBuffer, adaptor.getInputLookupTable(),
-                          polySizeCst, outputBitsCst, isSignedCst}));
+        mlir::ValueRange({castedOutputBuffer, castedLUT, polySizeCst,
+                          outputBitsCst, isSignedCst}));
 
     rewriter.replaceOp(eeOp, outputBuffer);
 
@@ -244,14 +261,19 @@ struct BootstrapGLWEOpPattern
     auto inputLweDimensionCst = rewriter.create<mlir::arith::ConstantIntOp>(
         bsOp.getLoc(), inputLweDimension, 32);
 
+    auto dynamicLutType = toDynamicTensorType(bsOp.getLookupTable().getType());
+
+    mlir::Value castedLUT = rewriter.create<mlir::tensor::CastOp>(
+        bsOp.getLoc(), dynamicLutType, adaptor.getLookupTable());
+
     // uint64_t sim_bootstrap_lwe_u64(uint64_t plaintext, uint64_t
     // *tlu_allocated, uint64_t *tlu_aligned, uint64_t tlu_offset, uint64_t
     // tlu_size, uint64_t tlu_stride, uint32_t input_lwe_dim, uint32_t
     // poly_size, uint32_t level, uint32_t base_log, uint32_t glwe_dim)
     if (insertForwardDeclaration(
             bsOp, rewriter, funcName,
             rewriter.getFunctionType(
-                {rewriter.getIntegerType(64), bsOp.getLookupTable().getType(),
+                {rewriter.getIntegerType(64), dynamicLutType,
                  rewriter.getIntegerType(32), rewriter.getIntegerType(32),
                  rewriter.getIntegerType(32), rewriter.getIntegerType(32),
                  rewriter.getIntegerType(32)},
@@ -262,7 +284,7 @@ struct BootstrapGLWEOpPattern
 
     rewriter.replaceOpWithNewOp<mlir::func::CallOp>(
         bsOp, funcName, this->getTypeConverter()->convertType(resultType),
-        mlir::ValueRange({adaptor.getCiphertext(), adaptor.getLookupTable(),
+        mlir::ValueRange({adaptor.getCiphertext(), castedLUT,
                           inputLweDimensionCst, polySizeCst, levelsCst,
                           baseLogCst, glweDimensionCst}));
 
@@ -376,7 +398,8 @@ void SimulateTFHEPass::runOnOperation() {
   SimulateTFHETypeConverter converter;
 
   target.addLegalDialect<mlir::arith::ArithDialect>();
-  target.addLegalOp<mlir::func::CallOp, mlir::bufferization::AllocTensorOp>();
+  target.addLegalOp<mlir::func::CallOp, mlir::bufferization::AllocTensorOp,
+                    mlir::tensor::CastOp>();
   // Make sure that no ops from `TFHE` remain after the lowering
   target.addIllegalDialect<TFHE::TFHEDialect>();