[dynarmic] Add A64:SM4/SM3 tests

Signed-off-by: lizzie <lizzie@eden-emu.dev>

src/dynarmic/src/dynarmic/backend/x64/constant_pool.cpp

@@ -7,27 +7,34 @@
  */
 
 #include <cstring>
-#include <cassert>
+
+#include "common/assert.h"
+
 #include "dynarmic/backend/x64/block_of_code.h"
 #include "dynarmic/backend/x64/constant_pool.h"
 
 namespace Dynarmic::Backend::X64 {
 
-ConstantPool::ConstantPool(BlockOfCode& code, size_t size) noexcept {
-    auto const align_size = 64; //always align to cacheline
-    code.EnsureMemoryCommitted(2 * align_size + size);
+ConstantPool::ConstantPool(BlockOfCode& code, size_t size)
+    : insertion_point(0)
+{
+    code.EnsureMemoryCommitted(align_size + size);
+    code.int3();
     code.align(align_size);
-    pool = std::span<u128>(static_cast<u128*>(code.AllocateFromCodeSpace(size)), size / sizeof(u128));
+    pool = std::span<ConstantT>(reinterpret_cast<ConstantT*>(code.AllocateFromCodeSpace(size)), size / align_size);
 }
 
-Xbyak::Address ConstantPool::GetConstant(BlockOfCode& code, const Xbyak::AddressFrame& frame, u64 lo, u64 hi) noexcept {
-    // offset grows iff new constant is added
-    auto const key = std::make_pair(lo, hi);
-    auto const val = u32(constant_info.size());
-    auto const [it, _] = constant_info.insert({key, val});
-    // if it != constant_info.end(): assume(pool[it->second] == c);
-    pool[it->second] = u128{lo, hi};
-    return frame[code.rip + std::addressof(pool[it->second])];
+Xbyak::Address ConstantPool::GetConstant(BlockOfCode& code, const Xbyak::AddressFrame& frame, u64 lower, u64 upper) {
+    const auto constant = ConstantT(lower, upper);
+    auto it = constant_info.find(constant);
+    if (it == constant_info.end()) {
+        ASSERT(insertion_point < pool.size());
+        ConstantT& target_constant = pool[insertion_point];
+        target_constant = constant;
+        it = constant_info.insert({constant, &target_constant}).first;
+        ++insertion_point;
+    }
+    return frame[code.rip + it->second];
 }
 
 }  // namespace Dynarmic::Backend::X64

src/dynarmic/src/dynarmic/backend/x64/constant_pool.h

@@ -12,7 +12,6 @@
 #include <cstddef>
 #include <span>
 #include <utility>
-#include <boost/functional/hash.hpp>
 
 #include "common/common_types.h"
 #include "common/container/unordered_map.h"
@@ -22,17 +21,31 @@ namespace Dynarmic::Backend::X64 {
 
 class BlockOfCode;
 
-/// @brief ConstantPool allocates a block of memory from BlockOfCode.
+/// ConstantPool allocates a block of memory from BlockOfCode.
 /// It places constants into this block of memory, returning the address
 /// of the memory location where the constant is placed. If the constant
 /// already exists, its memory location is reused.
 class ConstantPool final {
 public:
-    ConstantPool(BlockOfCode& code, size_t size) noexcept;
-    Xbyak::Address GetConstant(BlockOfCode& code, const Xbyak::AddressFrame& frame, u64 lower, u64 upper = 0) noexcept;
-    // key = identity hash, value = offset from pool.data()
-    ::Common::unordered_map<std::pair<u64, u64>, u32, boost::hash<std::pair<u64, u64>>> constant_info;
-    std::span<u128> pool;
+    ConstantPool(BlockOfCode& code, size_t size);
+
+    Xbyak::Address GetConstant(BlockOfCode& code, const Xbyak::AddressFrame& frame, u64 lower, u64 upper = 0);
+
+private:
+    static constexpr size_t align_size = 16;  // bytes
+
+    using ConstantT = std::pair<u64, u64>;
+    static_assert(sizeof(ConstantT) == align_size);
+
+    struct ConstantHash {
+        std::size_t operator()(const ConstantT& constant) const noexcept {
+            return constant.first ^ std::rotl<u64>(constant.second, 1);
+        }
+    };
+
+    ::Common::unordered_map<ConstantT, void*, ConstantHash> constant_info;
+    std::span<ConstantT> pool;
+    std::size_t insertion_point;
 };
 
 }  // namespace Dynarmic::Backend::X64

src/dynarmic/tests/A64/a64.cpp

@@ -2632,3 +2632,66 @@ TEST_CASE("A64: Rounding", "[a64]") {
     CHECK(jit.GetVector(5) == Vector{0x4000000040000000, 0x4000000040000000});
     CHECK(jit.GetVector(6) == Vector{0x4000000040000000, 0x4000000040000000});
 }
+
+TEST_CASE("A64: SM4", "[a64]") {
+    A64TestEnv env;
+    A64::UserConfig jit_user_config{};
+    jit_user_config.callbacks = &env;
+    A64::Jit jit{jit_user_config};
+
+    oaknut::VectorCodeGenerator code{env.code_mem, nullptr};
+
+    code.SM4EKEY(V4.S4(), V0.S4(), V1.S4());
+    code.SM4E(V5.S4(), V0.S4(), V1.S4());
+
+    jit.SetPC(0);
+    jit.SetVector(0, {0xefcdab8967452301, 0x1032547698badcfe});
+    jit.SetVector(1, {0xefcdab8967452301, 0x1032547698badcfe});
+    env.ticks_left = env.code_mem.size();
+    CheckedRun([&]() { jit.Run(); });
+
+    CHECK(jit.GetVector(4) == Vector{0x486a0c2e00224466, 0x5c7e183a62402604});
+    CHECK(jit.GetVector(5) == Vector{0xc2e086a43c1e785a, 0xd7f593b104264062});
+}
+
+TEST_CASE("A64: SM3", "[a64]") {
+    A64TestEnv env;
+    A64::UserConfig jit_user_config{};
+    jit_user_config.callbacks = &env;
+    SECTION("SM3SS1") {
+        A64::Jit jit{jit_user_config};
+        oaknut::VectorCodeGenerator code{env.code_mem, nullptr};
+        code.SM3SS1(V0.S4(), V0.S4(), V1.S4(), V2.S4());
+        jit.SetPC(0);
+        jit.SetVector(0, {0xec4aff517369c667, 0x46e3fbf2abbacd29});
+        jit.SetVector(1, {0x8de7e81bf854c27c, 0x9ac99f33632e5a76});
+        jit.SetVector(2, {0x5aa35831b70d3266, 0xd45ee95817055d25});
+        env.ticks_left = env.code_mem.size();
+        CheckedRun([&]() { jit.Run(); });
+        CHECK(jit.GetVector(0) == Vector{0x0000000000000000, 0x73d67cd700000000});
+    }
+    SECTION("SM3PARTW1") {
+        A64::Jit jit{jit_user_config};
+        oaknut::VectorCodeGenerator code{env.code_mem, nullptr};
+        code.SM3PARTW1(V0.S4(), V0.S4(), V1.S4(), V2.S4());
+        jit.SetPC(0);
+        jit.SetVector(0, {0xec4aff517369c667, 0x46e3fbf2abbacd29});
+        jit.SetVector(1, {0x8de7e81bf854c27c, 0x9ac99f33632e5a76});
+        jit.SetVector(2, {0x5aa35831b70d3266, 0xd45ee95817055d25});
+        env.ticks_left = env.code_mem.size();
+        CheckedRun([&]() { jit.Run(); });
+        CHECK(jit.GetVector(0) == Vector{0x653c64995693280c, 0xabcc7db97adebf12});
+    }
+    SECTION("SM3PARTW2") {
+        A64::Jit jit{jit_user_config};
+        oaknut::VectorCodeGenerator code{env.code_mem, nullptr};
+        code.SM3PARTW2(V0.S4(), V0.S4(), V1.S4(), V2.S4());
+        jit.SetPC(0);
+        jit.SetVector(0, {0xec4aff517369c667, 0x46e3fbf2abbacd29});
+        jit.SetVector(1, {0x8de7e81bf854c27c, 0x9ac99f33632e5a76});
+        jit.SetVector(2, {0x5aa35831b70d3266, 0xd45ee95817055d25});
+        env.ticks_left = env.code_mem.size();
+        CheckedRun([&]() { jit.Run(); });
+        CHECK(jit.GetVector(0) == Vector{0x30010fe70da43740, 0x6702425b4a3a05d4});
+    }
+}