use workspace size from get_solution()

src/targets/gpu/convolution.cpp

@@ -119,7 +119,6 @@ shape miopen_convolution::find(context& ctx, const shape& output_shape, std::vec
                                                         false);
     if(status != miopenStatusSuccess)
         MIGRAPHX_THROW("MIOpen Convolution: find convolution failed");
-    algo = perf.fwd_algo;
 
     size_t solution_count;
 
@@ -146,8 +145,10 @@ shape miopen_convolution::find(context& ctx, const shape& output_shape, std::vec
         MIGRAPHX_THROW("MIOpen Convolution: get solution failed");
 
     solution_id = solutions.front().solution_id;
+    algo = solutions.front().algorithm;
+    workspace_size = solutions.front().workspace_size;
 
-    return shape{shape::int8_type, {perf.memory}};
+    return shape{shape::int8_type, {workspace_size}};
 }
 
 void miopen_convolution::finalize(context& ctx,

src/targets/gpu/deconvolution.cpp

@@ -119,7 +119,6 @@ shape miopen_deconvolution::find(context& ctx, const shape& output_shape, std::v
                                                         false);
     if(status != miopenStatusSuccess)
         MIGRAPHX_THROW("MIOpen Deconvolution: find convolution failed");
-    algo = perf.fwd_algo;
 
     size_t solution_count;
 
@@ -146,8 +145,10 @@ shape miopen_deconvolution::find(context& ctx, const shape& output_shape, std::v
         MIGRAPHX_THROW("MIOpen Deconvolution: get solution failed");
 
     solution_id = solutions.front().solution_id;
+    algo = solutions.front().algorithm;
+    workspace_size = solutions.front().workspace_size;
 
-    return shape{shape::int8_type, {perf.memory}};
+    return shape{shape::int8_type, {workspace_size}};
 }
 
 void miopen_deconvolution::finalize(context& ctx,

src/targets/gpu/fuse_ops.cpp

@@ -210,7 +210,7 @@ MIGRAPHX_PRED_MATCHER(fusable_conv, instruction_ref ins)
     auto conv = any_cast<miopen_convolution>(ins->get_operator());
     if(conv.op.group > 1)
         return false;
-    if(wei.lens()[1] > 512 and conv.algo != miopenConvolutionFwdAlgoWinograd)
+    if(wei.lens()[1] > 512 and conv.algo != miopenConvolutionAlgoWinograd)
         return false;
 
     // Do not fuse non-symmetric input
@@ -220,7 +220,7 @@ MIGRAPHX_PRED_MATCHER(fusable_conv, instruction_ref ins)
 
     auto op = conv.op;
     // Dont fuse winograd for non-3x3s since there is no fused windograd for those configs
-    if(conv.algo == miopenConvolutionFwdAlgoWinograd and wei.lens()[2] != 3 and
+    if(conv.algo == miopenConvolutionAlgoWinograd and wei.lens()[2] != 3 and
        wei.lens()[3] != 3 and contains({{1, 1}}, op.stride))
         return false;
     return contains({{0, 0, 0, 0}, {1, 1, 1, 1}, {2, 2, 2, 2}}, op.padding) and

src/targets/gpu/include/migraphx/gpu/convolution.hpp

@@ -38,7 +38,7 @@ struct miopen_convolution
 {
     op::convolution op;
     shared<convolution_descriptor> cd = nullptr;
-    miopenConvFwdAlgorithm_t algo{};
+    miopenConvAlgorithm_t algo{};
     uint64_t solution_id = 0;
 
     template <class Self, class F>

src/targets/gpu/include/migraphx/gpu/deconvolution.hpp

@@ -38,7 +38,7 @@ struct miopen_deconvolution
 {
     op::deconvolution op;
     shared<convolution_descriptor> cd;
-    miopenConvFwdAlgorithm_t algo{};
+    miopenConvAlgorithm_t algo{};
     uint64_t solution_id = 0;
 
     template <class Self, class F>

src/targets/gpu/include/migraphx/gpu/quant_convolution.hpp

@@ -40,7 +40,7 @@ struct miopen_quant_convolution
     op::quant_convolution op;
     bool int8_x4_format = false;
     shared<convolution_descriptor> cd;
-    miopenConvFwdAlgorithm_t algo{};
+    miopenConvAlgorithm_t algo{};
     uint64_t solution_id = 0;
 
     template <class Self, class F>

src/targets/gpu/quant_convolution.cpp

@@ -43,26 +43,23 @@ argument miopen_quant_convolution::compute(context& ctx,
     auto w_desc = make_tensor(args[1].get_shape(), int8_x4_format);
     auto y_desc = make_tensor(output_shape);
 
-    float alpha = 1;
-    float beta  = 0;
+    if(solution_id == 0)
+        MIGRAPHX_THROW("MIOpen Convolution: invalid solution ID");
 
-    auto status = miopenConvolutionForward(ctx.get_stream().get_miopen(),
-                                           &alpha,
-                                           x_desc.get(),
-                                           args[0].implicit(),
+    auto status = miopenConvolutionForwardImmediate(ctx.get_stream().get_miopen(),
                                                     w_desc.get(),
                                                     args[1].implicit(),
+                                                    x_desc.get(),
+                                                    args[0].implicit(),
                                                     cd.get(),
-                                           algo,
-                                           &beta,
                                                     y_desc.get(),
                                                     args[3].implicit(),
                                                     args[2].implicit(),
-                                           args[2].get_shape().bytes());
+                                                    args[2].get_shape().bytes(),
+                                                    solution_id);
+
     if(status != miopenStatusSuccess)
-    {
         MIGRAPHX_THROW("QUANT_CONVOLUTION: run convolution forward failed");
-    }
 
     return args[3];
 }
@@ -115,7 +112,6 @@ shape miopen_quant_convolution::find(context& ctx,
                                                         false);
     if(status != miopenStatusSuccess)
         MIGRAPHX_THROW("MIOpen Quant Convolution: find convolution failed");
-    algo = perf.fwd_algo;
 
     size_t solution_count;
 
@@ -143,7 +139,10 @@ shape miopen_quant_convolution::find(context& ctx,
 
     solution_id = solutions.front().solution_id;
 
-    return shape{shape::int8_type, {perf.memory}};
+    algo = solutions.front().algorithm;
+    workspace_size = solutions.front().workspace_size;
+
+    return shape{shape::int8_type, {workspace_size}};
 }
 
 void miopen_quant_convolution::finalize(context& ctx,