forgot

This reverts commit 3db41fbce6.

src/CMakeLists.txt

@@ -121,7 +121,7 @@ else()
         -Werror=unused
 
         -Wno-attributes
-        -Wno-invalid-offsetof
+        $<$<COMPILE_LANGUAGE:CXX>:-Wno-invalid-offsetof>
         -Wno-unused-parameter
         -Wno-missing-field-initializers
     )

src/common/host_memory.cpp

@@ -730,7 +730,9 @@ void HostMemory::Map(size_t virtual_offset, size_t host_offset, size_t length,
     ASSERT(virtual_offset % PageAlignment == 0);
     ASSERT(host_offset % PageAlignment == 0);
     ASSERT(length % PageAlignment == 0);
-    ASSERT(virtual_offset + length <= virtual_size);
+    if (impl && virtual_base) {
+        ASSERT(virtual_offset + length <= virtual_size);
+    }
     ASSERT(host_offset + length <= backing_size);
     if (length == 0 || !virtual_base || !impl) {
         return;
@@ -741,7 +743,9 @@ void HostMemory::Map(size_t virtual_offset, size_t host_offset, size_t length,
 void HostMemory::Unmap(size_t virtual_offset, size_t length, bool separate_heap) {
     ASSERT(virtual_offset % PageAlignment == 0);
     ASSERT(length % PageAlignment == 0);
-    ASSERT(virtual_offset + length <= virtual_size);
+    if (impl && virtual_base) {
+        ASSERT(virtual_offset + length <= virtual_size);
+    }
     if (length == 0 || !virtual_base || !impl) {
         return;
     }
@@ -751,7 +755,9 @@ void HostMemory::Unmap(size_t virtual_offset, size_t length, bool separate_heap)
 void HostMemory::Protect(size_t virtual_offset, size_t length, MemoryPermission perm) {
     ASSERT(virtual_offset % PageAlignment == 0);
     ASSERT(length % PageAlignment == 0);
-    ASSERT(virtual_offset + length <= virtual_size);
+    if (impl && virtual_base) {
+        ASSERT(virtual_offset + length <= virtual_size);
+    }
     if (length == 0 || !virtual_base || !impl) {
         return;
     }

src/core/hle/service/hle_ipc.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -393,6 +396,24 @@ std::size_t HLERequestContext::WriteBuffer(const void* buffer, std::size_t size,
     const bool is_buffer_b{BufferDescriptorB().size() > buffer_index &&
                            BufferDescriptorB()[buffer_index].Size()};
     const std::size_t buffer_size{GetWriteBufferSize(buffer_index)};
+    
+    // Defensive check: if client didn't provide output buffer, log detailed error but don't crash
+    if (buffer_size == 0) {
+        LOG_ERROR(Core,
+                  "WriteBuffer called but client provided NO output buffer! "
+                  "Requested size: 0x{:X}, buffer_index: {}, is_buffer_b: {}, "
+                  "BufferB count: {}, BufferC count: {}",
+                  size, buffer_index, is_buffer_b, BufferDescriptorB().size(),
+                  BufferDescriptorC().size());
+        
+        // Log command context for debugging
+        LOG_ERROR(Core, "IPC Command: 0x{:X}, Type: {}", GetCommand(), 
+                  static_cast<u32>(GetCommandType()));
+        
+        // Return 0 instead of crashing - let service handle error
+        return 0;
+    }
+    
     if (size > buffer_size) {
         LOG_CRITICAL(Core, "size ({:016X}) is greater than buffer_size ({:016X})", size,
                      buffer_size);

src/shader_recompiler/backend/glsl/glsl_emit_context.cpp

@@ -380,13 +380,14 @@ void EmitContext::SetupExtensions() {
     if (info.uses_int64 && profile.support_int64) {
         header += "#extension GL_ARB_gpu_shader_int64 : enable\n";
     }
-    if (info.uses_int64_bit_atomics) {
+    if (info.uses_int64_bit_atomics && profile.support_gl_shader_atomic_int64) {
         header += "#extension GL_NV_shader_atomic_int64 : enable\n";
     }
-    if (info.uses_atomic_f32_add) {
+    if (info.uses_atomic_f32_add && profile.support_gl_shader_atomic_float) {
         header += "#extension GL_NV_shader_atomic_float : enable\n";
     }
-    if (info.uses_atomic_f16x2_add || info.uses_atomic_f16x2_min || info.uses_atomic_f16x2_max) {
+    if ((info.uses_atomic_f16x2_add || info.uses_atomic_f16x2_min || info.uses_atomic_f16x2_max) &&
+        profile.support_gl_shader_atomic_fp16_vector) {
         header += "#extension GL_NV_shader_atomic_fp16_vector : enable\n";
     }
     if (info.uses_fp16) {

src/shader_recompiler/backend/spirv/emit_spirv_memory.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -92,7 +95,7 @@ void EmitLoadGlobalS16(EmitContext&) {
 }
 
 Id EmitLoadGlobal32(EmitContext& ctx, Id address) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         return ctx.OpFunctionCall(ctx.U32[1], ctx.load_global_func_u32, address);
     }
     LOG_WARNING(Shader_SPIRV, "Int64 not supported, ignoring memory operation");
@@ -100,7 +103,7 @@ Id EmitLoadGlobal32(EmitContext& ctx, Id address) {
 }
 
 Id EmitLoadGlobal64(EmitContext& ctx, Id address) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         return ctx.OpFunctionCall(ctx.U32[2], ctx.load_global_func_u32x2, address);
     }
     LOG_WARNING(Shader_SPIRV, "Int64 not supported, ignoring memory operation");
@@ -108,7 +111,7 @@ Id EmitLoadGlobal64(EmitContext& ctx, Id address) {
 }
 
 Id EmitLoadGlobal128(EmitContext& ctx, Id address) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         return ctx.OpFunctionCall(ctx.U32[4], ctx.load_global_func_u32x4, address);
     }
     LOG_WARNING(Shader_SPIRV, "Int64 not supported, ignoring memory operation");
@@ -132,7 +135,7 @@ void EmitWriteGlobalS16(EmitContext&) {
 }
 
 void EmitWriteGlobal32(EmitContext& ctx, Id address, Id value) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         ctx.OpFunctionCall(ctx.void_id, ctx.write_global_func_u32, address, value);
         return;
     }
@@ -140,7 +143,7 @@ void EmitWriteGlobal32(EmitContext& ctx, Id address, Id value) {
 }
 
 void EmitWriteGlobal64(EmitContext& ctx, Id address, Id value) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         ctx.OpFunctionCall(ctx.void_id, ctx.write_global_func_u32x2, address, value);
         return;
     }
@@ -148,7 +151,7 @@ void EmitWriteGlobal64(EmitContext& ctx, Id address, Id value) {
 }
 
 void EmitWriteGlobal128(EmitContext& ctx, Id address, Id value) {
-    if (ctx.profile.support_int64) {
+    if (ctx.SupportsNativeInt64() || ctx.UsesInt64Emulation()) {
         ctx.OpFunctionCall(ctx.void_id, ctx.write_global_func_u32x4, address, value);
         return;
     }

src/shader_recompiler/backend/spirv/spirv_emit_context.cpp

@@ -460,9 +460,14 @@ void VectorTypes::Define(Sirit::Module& sirit_ctx, Id base_type, std::string_vie
 
 EmitContext::EmitContext(const Profile& profile_, const RuntimeInfo& runtime_info_,
                          IR::Program& program, Bindings& bindings)
-    : Sirit::Module(profile_.supported_spirv), profile{profile_}, runtime_info{runtime_info_},
+        : Sirit::Module(profile_.supported_spirv), profile{profile_}, runtime_info{runtime_info_},
-      stage{program.stage}, texture_rescaling_index{bindings.texture_scaling_index},
+            stage{program.stage},
-      image_rescaling_index{bindings.image_scaling_index} {
+            // Enable int64 emulation if host lacks int64 but we either use int64 ops
+            // or we need 64-bit addressing for global memory operations.
+            emulate_int64{!profile.support_int64 &&
+                          (program.info.uses_int64 || program.info.uses_global_memory)},
+            texture_rescaling_index{bindings.texture_scaling_index},
+            image_rescaling_index{bindings.image_scaling_index} {
     const bool is_unified{profile.unified_descriptor_binding};
     u32& uniform_binding{is_unified ? bindings.unified : bindings.uniform_buffer};
     u32& storage_binding{is_unified ? bindings.unified : bindings.storage_buffer};
@@ -932,11 +937,163 @@ void EmitContext::DefineWriteStorageCasLoopFunction(const Info& info) {
 }
 
 void EmitContext::DefineGlobalMemoryFunctions(const Info& info) {
-    if (!info.uses_global_memory || !profile.support_int64) {
+    if (!info.uses_global_memory) {
         return;
     }
     using DefPtr = Id StorageDefinitions::*;
     const Id zero{u32_zero_value};
+
+    if (SupportsNativeInt64()) {
+        const auto define_body{[&](DefPtr ssbo_member, Id addr, Id element_pointer, u32 shift,
+                                   auto&& callback) {
+            AddLabel();
+            const size_t num_buffers{info.storage_buffers_descriptors.size()};
+            for (size_t index = 0; index < num_buffers; ++index) {
+                if (!info.nvn_buffer_used[index]) {
+                    continue;
+                }
+                const auto& ssbo{info.storage_buffers_descriptors[index]};
+                const Id ssbo_addr_cbuf_offset{Const(ssbo.cbuf_offset / 8)};
+                const Id ssbo_size_cbuf_offset{Const(ssbo.cbuf_offset / 4 + 2)};
+                const Id ssbo_addr_pointer{OpAccessChain(
+                    uniform_types.U32x2, cbufs[ssbo.cbuf_index].U32x2, zero,
+                    ssbo_addr_cbuf_offset)};
+                const Id ssbo_size_pointer{OpAccessChain(
+                    uniform_types.U32, cbufs[ssbo.cbuf_index].U32, zero, ssbo_size_cbuf_offset)};
+
+                const u64 ssbo_align_mask{~(profile.min_ssbo_alignment - 1U)};
+                const Id unaligned_addr{OpBitcast(U64, OpLoad(U32[2], ssbo_addr_pointer))};
+                const Id ssbo_addr{OpBitwiseAnd(U64, unaligned_addr, Constant(U64, ssbo_align_mask))};
+                const Id ssbo_size{OpUConvert(U64, OpLoad(U32[1], ssbo_size_pointer))};
+                const Id ssbo_end{OpIAdd(U64, ssbo_addr, ssbo_size)};
+                const Id cond{OpLogicalAnd(U1, OpUGreaterThanEqual(U1, addr, ssbo_addr),
+                                           OpULessThan(U1, addr, ssbo_end))};
+                const Id then_label{OpLabel()};
+                const Id else_label{OpLabel()};
+                OpSelectionMerge(else_label, spv::SelectionControlMask::MaskNone);
+                OpBranchConditional(cond, then_label, else_label);
+                AddLabel(then_label);
+                const Id ssbo_id{ssbos[index].*ssbo_member};
+                const Id ssbo_offset{OpUConvert(U32[1], OpISub(U64, addr, ssbo_addr))};
+                const Id ssbo_index{OpShiftRightLogical(U32[1], ssbo_offset, Const(shift))};
+                const Id ssbo_pointer{OpAccessChain(element_pointer, ssbo_id, zero, ssbo_index)};
+                callback(ssbo_pointer);
+                AddLabel(else_label);
+            }
+        }};
+        const auto define_load{[&](DefPtr ssbo_member, Id element_pointer, Id type, u32 shift) {
+            const Id function_type{TypeFunction(type, U64)};
+            const Id func_id{OpFunction(type, spv::FunctionControlMask::MaskNone, function_type)};
+            const Id addr{OpFunctionParameter(U64)};
+            define_body(ssbo_member, addr, element_pointer, shift,
+                        [&](Id ssbo_pointer) { OpReturnValue(OpLoad(type, ssbo_pointer)); });
+            OpReturnValue(ConstantNull(type));
+            OpFunctionEnd();
+            return func_id;
+        }};
+        const auto define_write{[&](DefPtr ssbo_member, Id element_pointer, Id type, u32 shift) {
+            const Id function_type{TypeFunction(void_id, U64, type)};
+            const Id func_id{
+                OpFunction(void_id, spv::FunctionControlMask::MaskNone, function_type)};
+            const Id addr{OpFunctionParameter(U64)};
+            const Id data{OpFunctionParameter(type)};
+            define_body(ssbo_member, addr, element_pointer, shift, [&](Id ssbo_pointer) {
+                OpStore(ssbo_pointer, data);
+                OpReturn();
+            });
+            OpReturn();
+            OpFunctionEnd();
+            return func_id;
+        }};
+        const auto define{
+            [&](DefPtr ssbo_member, const StorageTypeDefinition& type_def, Id type, size_t size) {
+                const Id element_type{type_def.element};
+                const u32 shift{static_cast<u32>(std::countr_zero(size))};
+                const Id load_func{define_load(ssbo_member, element_type, type, shift)};
+                const Id write_func{define_write(ssbo_member, element_type, type, shift)};
+                return std::make_pair(load_func, write_func);
+            }};
+        std::tie(load_global_func_u32, write_global_func_u32) =
+            define(&StorageDefinitions::U32, storage_types.U32, U32[1], sizeof(u32));
+        std::tie(load_global_func_u32x2, write_global_func_u32x2) =
+            define(&StorageDefinitions::U32x2, storage_types.U32x2, U32[2], sizeof(u32[2]));
+        std::tie(load_global_func_u32x4, write_global_func_u32x4) =
+            define(&StorageDefinitions::U32x4, storage_types.U32x4, U32[4], sizeof(u32[4]));
+        return;
+    }
+
+    if (!UsesInt64Emulation()) {
+        return;
+    }
+
+    const auto make_pair = [&](Id lo, Id hi) {
+        return OpCompositeConstruct(U32[2], lo, hi);
+    };
+    const auto split_pair = [&](Id value) {
+        return std::array<Id, 2>{OpCompositeExtract(U32[1], value, 0U),
+                                 OpCompositeExtract(U32[1], value, 1U)};
+    };
+    const auto bool_to_u32 = [&](Id predicate) {
+        return OpSelect(U32[1], predicate, Const(1u), zero);
+    };
+    const auto and_pair = [&](Id value, Id mask) {
+        const auto value_parts{split_pair(value)};
+        const auto mask_parts{split_pair(mask)};
+        return make_pair(OpBitwiseAnd(U32[1], value_parts[0], mask_parts[0]),
+                         OpBitwiseAnd(U32[1], value_parts[1], mask_parts[1]));
+    };
+    const auto add_pair = [&](Id lhs, Id rhs) {
+        const auto lhs_parts{split_pair(lhs)};
+        const auto rhs_parts{split_pair(rhs)};
+        const Id sum_lo{OpIAdd(U32[1], lhs_parts[0], rhs_parts[0])};
+        const Id carry{OpULessThan(U1, sum_lo, lhs_parts[0])};
+        Id sum_hi{OpIAdd(U32[1], lhs_parts[1], rhs_parts[1])};
+        sum_hi = OpIAdd(U32[1], sum_hi, bool_to_u32(carry));
+        return make_pair(sum_lo, sum_hi);
+    };
+    const auto sub_pair = [&](Id lhs, Id rhs) {
+        const auto lhs_parts{split_pair(lhs)};
+        const auto rhs_parts{split_pair(rhs)};
+        const Id borrow{OpULessThan(U1, lhs_parts[0], rhs_parts[0])};
+        const Id diff_lo{OpISub(U32[1], lhs_parts[0], rhs_parts[0])};
+        Id diff_hi{OpISub(U32[1], lhs_parts[1], rhs_parts[1])};
+        diff_hi = OpISub(U32[1], diff_hi, bool_to_u32(borrow));
+        return make_pair(diff_lo, diff_hi);
+    };
+    const auto shift_right_pair = [&](Id value, u32 shift) {
+        if (shift == 0) {
+            return value;
+        }
+        const auto parts{split_pair(value)};
+        const Id shift_id{Const(shift)};
+        const Id high_shifted{OpShiftRightLogical(U32[1], parts[1], shift_id)};
+        Id low_shifted{OpShiftRightLogical(U32[1], parts[0], shift_id)};
+        const Id carry_bits{OpShiftLeftLogical(U32[1], parts[1], Const(32u - shift))};
+        low_shifted = OpBitwiseOr(U32[1], low_shifted, carry_bits);
+        return make_pair(low_shifted, high_shifted);
+    };
+    const auto greater_equal_pair = [&](Id lhs, Id rhs) {
+        const auto lhs_parts{split_pair(lhs)};
+        const auto rhs_parts{split_pair(rhs)};
+        const Id hi_gt{OpUGreaterThan(U1, lhs_parts[1], rhs_parts[1])};
+        const Id hi_eq{OpIEqual(U1, lhs_parts[1], rhs_parts[1])};
+        const Id lo_ge{OpUGreaterThanEqual(U1, lhs_parts[0], rhs_parts[0])};
+        return OpLogicalOr(U1, hi_gt, OpLogicalAnd(U1, hi_eq, lo_ge));
+    };
+    const auto less_than_pair = [&](Id lhs, Id rhs) {
+        const auto lhs_parts{split_pair(lhs)};
+        const auto rhs_parts{split_pair(rhs)};
+        const Id hi_lt{OpULessThan(U1, lhs_parts[1], rhs_parts[1])};
+        const Id hi_eq{OpIEqual(U1, lhs_parts[1], rhs_parts[1])};
+        const Id lo_lt{OpULessThan(U1, lhs_parts[0], rhs_parts[0])};
+        return OpLogicalOr(U1, hi_lt, OpLogicalAnd(U1, hi_eq, lo_lt));
+    };
+
+    const u64 ssbo_align_mask_value{~(profile.min_ssbo_alignment - 1U)};
+    const Id ssbo_align_mask{
+        Const(static_cast<u32>(ssbo_align_mask_value & 0xFFFFFFFFu),
+              static_cast<u32>(ssbo_align_mask_value >> 32))};
+
     const auto define_body{[&](DefPtr ssbo_member, Id addr, Id element_pointer, u32 shift,
                                auto&& callback) {
         AddLabel();
@@ -953,40 +1110,44 @@ void EmitContext::DefineGlobalMemoryFunctions(const Info& info) {
             const Id ssbo_size_pointer{OpAccessChain(uniform_types.U32, cbufs[ssbo.cbuf_index].U32,
                                                      zero, ssbo_size_cbuf_offset)};
 
-            const u64 ssbo_align_mask{~(profile.min_ssbo_alignment - 1U)};
+            const Id unaligned_addr_pair{OpLoad(U32[2], ssbo_addr_pointer)};
-            const Id unaligned_addr{OpBitcast(U64, OpLoad(U32[2], ssbo_addr_pointer))};
+            const Id ssbo_addr_pair{and_pair(unaligned_addr_pair, ssbo_align_mask)};
-            const Id ssbo_addr{OpBitwiseAnd(U64, unaligned_addr, Constant(U64, ssbo_align_mask))};
+            const Id ssbo_size_value{OpLoad(U32[1], ssbo_size_pointer)};
-            const Id ssbo_size{OpUConvert(U64, OpLoad(U32[1], ssbo_size_pointer))};
+            const Id ssbo_size_pair{make_pair(ssbo_size_value, zero)};
-            const Id ssbo_end{OpIAdd(U64, ssbo_addr, ssbo_size)};
+            const Id ssbo_end_pair{add_pair(ssbo_addr_pair, ssbo_size_pair)};
-            const Id cond{OpLogicalAnd(U1, OpUGreaterThanEqual(U1, addr, ssbo_addr),
+            const Id cond{OpLogicalAnd(U1, greater_equal_pair(addr, ssbo_addr_pair),
-                                       OpULessThan(U1, addr, ssbo_end))};
+                                       less_than_pair(addr, ssbo_end_pair))};
             const Id then_label{OpLabel()};
             const Id else_label{OpLabel()};
             OpSelectionMerge(else_label, spv::SelectionControlMask::MaskNone);
             OpBranchConditional(cond, then_label, else_label);
             AddLabel(then_label);
             const Id ssbo_id{ssbos[index].*ssbo_member};
-            const Id ssbo_offset{OpUConvert(U32[1], OpISub(U64, addr, ssbo_addr))};
+            const Id ssbo_offset_pair{sub_pair(addr, ssbo_addr_pair)};
-            const Id ssbo_index{OpShiftRightLogical(U32[1], ssbo_offset, Const(shift))};
+            const Id ssbo_index_pair{shift_right_pair(ssbo_offset_pair, shift)};
+            const Id ssbo_index{OpCompositeExtract(U32[1], ssbo_index_pair, 0U)};
             const Id ssbo_pointer{OpAccessChain(element_pointer, ssbo_id, zero, ssbo_index)};
             callback(ssbo_pointer);
             AddLabel(else_label);
         }
     }};
+
     const auto define_load{[&](DefPtr ssbo_member, Id element_pointer, Id type, u32 shift) {
-        const Id function_type{TypeFunction(type, U64)};
+        const Id function_type{TypeFunction(type, U32[2])};
         const Id func_id{OpFunction(type, spv::FunctionControlMask::MaskNone, function_type)};
-        const Id addr{OpFunctionParameter(U64)};
+        const Id addr{OpFunctionParameter(U32[2])};
         define_body(ssbo_member, addr, element_pointer, shift,
                     [&](Id ssbo_pointer) { OpReturnValue(OpLoad(type, ssbo_pointer)); });
         OpReturnValue(ConstantNull(type));
         OpFunctionEnd();
         return func_id;
     }};
+
     const auto define_write{[&](DefPtr ssbo_member, Id element_pointer, Id type, u32 shift) {
-        const Id function_type{TypeFunction(void_id, U64, type)};
+        const Id function_type{TypeFunction(void_id, U32[2], type)};
-        const Id func_id{OpFunction(void_id, spv::FunctionControlMask::MaskNone, function_type)};
+        const Id func_id{
-        const Id addr{OpFunctionParameter(U64)};
+            OpFunction(void_id, spv::FunctionControlMask::MaskNone, function_type)};
+        const Id addr{OpFunctionParameter(U32[2])};
         const Id data{OpFunctionParameter(type)};
         define_body(ssbo_member, addr, element_pointer, shift, [&](Id ssbo_pointer) {
             OpStore(ssbo_pointer, data);
@@ -996,6 +1157,7 @@ void EmitContext::DefineGlobalMemoryFunctions(const Info& info) {
         OpFunctionEnd();
         return func_id;
     }};
+
     const auto define{
         [&](DefPtr ssbo_member, const StorageTypeDefinition& type_def, Id type, size_t size) {
             const Id element_type{type_def.element};
@@ -1004,6 +1166,7 @@ void EmitContext::DefineGlobalMemoryFunctions(const Info& info) {
             const Id write_func{define_write(ssbo_member, element_type, type, shift)};
             return std::make_pair(load_func, write_func);
         }};
+
     std::tie(load_global_func_u32, write_global_func_u32) =
         define(&StorageDefinitions::U32, storage_types.U32, U32[1], sizeof(u32));
     std::tie(load_global_func_u32x2, write_global_func_u32x2) =

src/shader_recompiler/backend/spirv/spirv_emit_context.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -207,6 +210,15 @@ public:
     const Profile& profile;
     const RuntimeInfo& runtime_info;
     Stage stage{};
+    const bool emulate_int64{};
+
+    bool SupportsNativeInt64() const {
+        return profile.support_int64;
+    }
+
+    bool UsesInt64Emulation() const {
+        return emulate_int64;
+    }
 
     Id void_id{};
     Id U1{};

src/shader_recompiler/ir_opt/global_memory_to_storage_buffer_pass.cpp

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -293,6 +296,14 @@ std::optional<LowAddrInfo> TrackLowAddress(IR::Inst* inst) {
     }
     // This address is expected to either be a PackUint2x32, a IAdd64, or a CompositeConstructU32x2
     IR::Inst* addr_inst{addr.InstRecursive()};
+    // Unwrap Identity ops introduced by lowerings (e.g., PackUint2x32 -> Identity)
+    while (addr_inst->GetOpcode() == IR::Opcode::Identity) {
+        const IR::Value id_arg{addr_inst->Arg(0)};
+        if (id_arg.IsImmediate()) {
+            return std::nullopt;
+        }
+        addr_inst = id_arg.InstRecursive();
+    }
     s32 imm_offset{0};
     if (addr_inst->GetOpcode() == IR::Opcode::IAdd64) {
         // If it's an IAdd64, get the immediate offset it is applying and grab the address
@@ -308,6 +319,14 @@ std::optional<LowAddrInfo> TrackLowAddress(IR::Inst* inst) {
             return std::nullopt;
         }
         addr_inst = iadd_addr.InstRecursive();
+        // Unwrap Identity again if present after folding IAdd64
+        while (addr_inst->GetOpcode() == IR::Opcode::Identity) {
+            const IR::Value id_arg{addr_inst->Arg(0)};
+            if (id_arg.IsImmediate()) {
+                return std::nullopt;
+            }
+            addr_inst = id_arg.InstRecursive();
+        }
     }
     // With IAdd64 handled, now PackUint2x32 is expected
     if (addr_inst->GetOpcode() == IR::Opcode::PackUint2x32) {
@@ -317,6 +336,14 @@ std::optional<LowAddrInfo> TrackLowAddress(IR::Inst* inst) {
             return std::nullopt;
         }
         addr_inst = vector.InstRecursive();
+        // Unwrap Identity that may replace PackUint2x32
+        while (addr_inst->GetOpcode() == IR::Opcode::Identity) {
+            const IR::Value id_arg{addr_inst->Arg(0)};
+            if (id_arg.IsImmediate()) {
+                return std::nullopt;
+            }
+            addr_inst = id_arg.InstRecursive();
+        }
     }
     // The vector is expected to be a CompositeConstructU32x2
     if (addr_inst->GetOpcode() != IR::Opcode::CompositeConstructU32x2) {

src/shader_recompiler/profile.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -38,6 +41,9 @@ struct Profile {
     bool support_gl_nv_gpu_shader_5{};
     bool support_gl_amd_gpu_shader_half_float{};
     bool support_gl_texture_shadow_lod{};
+    bool support_gl_shader_atomic_float{};
+    bool support_gl_shader_atomic_fp16_vector{};
+    bool support_gl_shader_atomic_int64{};
     bool support_gl_warp_intrinsics{};
     bool support_gl_variable_aoffi{};
     bool support_gl_sparse_textures{};

src/video_core/host_shaders/CMakeLists.txt

@@ -76,6 +76,7 @@ set(SHADER_FILES
     vulkan_quad_indexed.comp
     vulkan_turbo_mode.comp
     vulkan_uint8.comp
+    vulkan_qcom_msaa_resolve.frag
     convert_rgba8_to_bgra8.frag
     convert_yuv420_to_rgb.comp
     convert_rgb_to_yuv420.comp

src/video_core/host_shaders/vulkan_qcom_msaa_resolve.frag

@@ -0,0 +1,39 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
+#version 450
+
+// VK_QCOM_render_pass_shader_resolve fragment shader
+// Resolves MSAA attachment to single-sample within render pass
+// Requires VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM in subpass flags
+
+// Use combined image sampler for MSAA texture instead of input attachment
+// This allows us to sample MSAA textures from previous rendering
+layout(set = 0, binding = 0) uniform sampler2DMS msaa_texture;
+
+layout(location = 0) out vec4 color_output;
+
+layout(push_constant) uniform PushConstants {
+    vec2 tex_scale;
+    vec2 tex_offset;
+} push_constants;
+
+// Custom MSAA resolve using box filter (simple average)
+// Assumes 4x MSAA (can be extended with push constant for dynamic sample count)
+void main() {
+    ivec2 coord = ivec2(gl_FragCoord.xy);
+    ivec2 tex_size = textureSize(msaa_texture);
+    
+    // Clamp coordinates to texture bounds
+    coord = clamp(coord, ivec2(0), tex_size - ivec2(1));
+    
+    vec4 accumulated_color = vec4(0.0);
+    int sample_count = 4; // Adreno typically uses 4x MSAA max
+    
+    // Box filter: simple average of all MSAA samples
+    for (int i = 0; i < sample_count; i++) {
+        accumulated_color += texelFetch(msaa_texture, coord, i);
+    }
+    
+    color_output = accumulated_color / float(sample_count);
+}

src/video_core/renderer_opengl/gl_device.cpp

@@ -225,6 +225,9 @@ Device::Device(Core::Frontend::EmuWindow& emu_window) {
     has_amd_shader_half_float = GLAD_GL_AMD_gpu_shader_half_float;
     has_sparse_texture_2 = GLAD_GL_ARB_sparse_texture2;
     has_draw_texture = GLAD_GL_NV_draw_texture;
+    has_shader_atomic_float = GLAD_GL_NV_shader_atomic_float;
+    has_shader_atomic_fp16_vector = GLAD_GL_NV_shader_atomic_fp16_vector;
+    has_shader_atomic_int64 = GLAD_GL_NV_shader_atomic_int64;
     warp_size_potentially_larger_than_guest = !is_nvidia && !is_intel;
     need_fastmath_off = is_nvidia;
     can_report_memory = GLAD_GL_NVX_gpu_memory_info;

src/video_core/renderer_opengl/gl_device.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -152,6 +155,18 @@ public:
         return has_draw_texture;
     }
 
+    bool HasShaderAtomicFloat() const {
+        return has_shader_atomic_float;
+    }
+
+    bool HasShaderAtomicFp16Vector() const {
+        return has_shader_atomic_fp16_vector;
+    }
+
+    bool HasShaderAtomicInt64() const {
+        return has_shader_atomic_int64;
+    }
+
     bool IsWarpSizePotentiallyLargerThanGuest() const {
         return warp_size_potentially_larger_than_guest;
     }
@@ -235,6 +250,9 @@ private:
     bool has_amd_shader_half_float{};
     bool has_sparse_texture_2{};
     bool has_draw_texture{};
+    bool has_shader_atomic_float{};
+    bool has_shader_atomic_fp16_vector{};
+    bool has_shader_atomic_int64{};
     bool warp_size_potentially_larger_than_guest{};
     bool need_fastmath_off{};
     bool has_cbuf_ftou_bug{};

src/video_core/renderer_opengl/gl_shader_cache.cpp

@@ -215,6 +215,9 @@ ShaderCache::ShaderCache(Tegra::MaxwellDeviceMemoryManager& device_memory_,
           .support_gl_nv_gpu_shader_5 = device.HasNvGpuShader5(),
           .support_gl_amd_gpu_shader_half_float = device.HasAmdShaderHalfFloat(),
           .support_gl_texture_shadow_lod = device.HasTextureShadowLod(),
+          .support_gl_shader_atomic_float = device.HasShaderAtomicFloat(),
+          .support_gl_shader_atomic_fp16_vector = device.HasShaderAtomicFp16Vector(),
+          .support_gl_shader_atomic_int64 = device.HasShaderAtomicInt64(),
           .support_gl_warp_intrinsics = false,
           .support_gl_variable_aoffi = device.HasVariableAoffi(),
           .support_gl_sparse_textures = device.HasSparseTexture2(),

src/video_core/renderer_opengl/gl_texture_cache.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -97,6 +100,10 @@ public:
         return true;
     }
 
+    bool CanDownloadMSAA() const noexcept {
+        return true;
+    }
+
     void CopyImage(Image& dst, Image& src, std::span<const VideoCommon::ImageCopy> copies);
 
     void CopyImageMSAA(Image& dst, Image& src, std::span<const VideoCommon::ImageCopy> copies);

src/video_core/renderer_vulkan/blit_image.cpp

@@ -40,6 +40,7 @@
 #include "video_core/host_shaders/convert_rgba16f_to_rgba8_frag_spv.h"
 #include "video_core/host_shaders/dither_temporal_frag_spv.h"
 #include "video_core/host_shaders/dynamic_resolution_scale_comp_spv.h"
+#include "video_core/host_shaders/vulkan_qcom_msaa_resolve_frag_spv.h"
 
 namespace Vulkan {
 
@@ -545,6 +546,7 @@ BlitImageHelper::BlitImageHelper(const Device& device_, Scheduler& scheduler_,
       convert_rgba16f_to_rgba8_frag(BuildShader(device, CONVERT_RGBA16F_TO_RGBA8_FRAG_SPV)),
       dither_temporal_frag(BuildShader(device, DITHER_TEMPORAL_FRAG_SPV)),
       dynamic_resolution_scale_comp(BuildShader(device, DYNAMIC_RESOLUTION_SCALE_COMP_SPV)),
+      qcom_msaa_resolve_frag(BuildShader(device, VULKAN_QCOM_MSAA_RESOLVE_FRAG_SPV)),
       linear_sampler(device.GetLogical().CreateSampler(SAMPLER_CREATE_INFO<VK_FILTER_LINEAR>)),
       nearest_sampler(device.GetLogical().CreateSampler(SAMPLER_CREATE_INFO<VK_FILTER_NEAREST>)) {}
 
@@ -1240,4 +1242,30 @@ void BlitImageHelper::ApplyDynamicResolutionScale(const Framebuffer* dst_framebu
     Convert(*dynamic_resolution_scale_pipeline, dst_framebuffer, src_image_view);
 }
 
+void BlitImageHelper::ResolveMSAAQcom(const Framebuffer* dst_framebuffer,
+                                      const ImageView& src_image_view) {
+    // VK_QCOM_render_pass_shader_resolve implementation
+    // This must be used within a render pass with VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM
+    ConvertPipeline(qcom_msaa_resolve_pipeline,
+                    dst_framebuffer->RenderPass(),
+                    false);
+    
+    RecordShaderReadBarrier(scheduler, src_image_view);
+    scheduler.RequestRenderpass(dst_framebuffer);
+    
+    const VkImageView src_view = src_image_view.Handle(Shader::TextureType::Color2D);
+    const VkPipelineLayout layout = *one_texture_pipeline_layout;
+    const VkPipeline pipeline = *qcom_msaa_resolve_pipeline;
+    
+    scheduler.Record([this, src_view, layout, pipeline](vk::CommandBuffer cmdbuf) {
+        const VkDescriptorSet descriptor_set = one_texture_descriptor_allocator.Commit();
+        UpdateOneTextureDescriptorSet(device, descriptor_set, *nearest_sampler, src_view);
+        cmdbuf.BindPipeline(VK_PIPELINE_BIND_POINT_GRAPHICS, pipeline);
+        cmdbuf.BindDescriptorSets(VK_PIPELINE_BIND_POINT_GRAPHICS, layout, 0, descriptor_set, nullptr);
+        cmdbuf.Draw(3, 1, 0, 0);
+    });
+    
+    scheduler.InvalidateState();
+}
+
 } // namespace Vulkan

src/video_core/renderer_vulkan/blit_image.h

@@ -95,6 +95,8 @@ public:
     void ConvertRGBA16FtoRGBA8(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
     void ApplyDitherTemporal(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
     void ApplyDynamicResolutionScale(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
+    
+    void ResolveMSAAQcom(const Framebuffer* dst_framebuffer, const ImageView& src_image_view);
 
 private:
     void Convert(VkPipeline pipeline, const Framebuffer* dst_framebuffer,
@@ -159,6 +161,7 @@ private:
     vk::ShaderModule convert_rgba16f_to_rgba8_frag;
     vk::ShaderModule dither_temporal_frag;
     vk::ShaderModule dynamic_resolution_scale_comp;
+    vk::ShaderModule qcom_msaa_resolve_frag;
     vk::Sampler linear_sampler;
     vk::Sampler nearest_sampler;
 
@@ -188,6 +191,7 @@ private:
     vk::Pipeline convert_rgba16f_to_rgba8_pipeline;
     vk::Pipeline dither_temporal_pipeline;
     vk::Pipeline dynamic_resolution_scale_pipeline;
+    vk::Pipeline qcom_msaa_resolve_pipeline;
 };
 
 } // namespace Vulkan

src/video_core/renderer_vulkan/maxwell_to_vk.cpp

@@ -214,7 +214,7 @@ struct FormatTuple {
     {VK_FORMAT_ASTC_8x6_SRGB_BLOCK},                           // ASTC_2D_8X6_SRGB
     {VK_FORMAT_ASTC_6x5_UNORM_BLOCK},                          // ASTC_2D_6X5_UNORM
     {VK_FORMAT_ASTC_6x5_SRGB_BLOCK},                           // ASTC_2D_6X5_SRGB
-    {VK_FORMAT_E5B9G9R9_UFLOAT_PACK32},                        // E5B9G9R9_FLOAT
+    {VK_FORMAT_E5B9G9R9_UFLOAT_PACK32, Attachable | Storage},  // E5B9G9R9_FLOAT
 
     // Depth formats
     {VK_FORMAT_D32_SFLOAT, Attachable},          // D32_FLOAT

src/video_core/renderer_vulkan/pipeline_helper.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 

src/video_core/renderer_vulkan/present/util.cpp

@@ -5,7 +5,6 @@
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #include "common/assert.h"
-#include <ranges>
 #include "video_core/renderer_vulkan/present/util.h"
 
 namespace Vulkan {

src/video_core/renderer_vulkan/vk_render_pass_cache.cpp

@@ -101,8 +101,15 @@ VkRenderPass RenderPassCache::Get(const RenderPassKey& key) {
         };
         descriptions.push_back(AttachmentDescription(*device, key.depth_format, key.samples));
     }
+    VkSubpassDescriptionFlags subpass_flags = 0;
+    if (key.qcom_shader_resolve) {
+        // VK_QCOM_render_pass_shader_resolve: enables custom shader resolve in fragment shader
+        // This must be the last subpass in the dependency chain
+        subpass_flags |= 0x00000004; // VK_SUBPASS_DESCRIPTION_SHADER_RESOLVE_BIT_QCOM
+    }
+    
     const VkSubpassDescription subpass{
-        .flags = 0,
+        .flags = subpass_flags,
         .pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS,
         .inputAttachmentCount = 0,
         .pInputAttachments = nullptr,

src/video_core/renderer_vulkan/vk_render_pass_cache.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2021 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -17,6 +20,14 @@ struct RenderPassKey {
     std::array<VideoCore::Surface::PixelFormat, 8> color_formats;
     VideoCore::Surface::PixelFormat depth_format;
     VkSampleCountFlagBits samples;
+    
+    // TBDR optimization hints - only affect tile-based GPUs (Qualcomm, ARM, Imagination)
+    // These flags indicate the expected usage pattern to optimize load/store operations
+    bool tbdr_will_clear{false};  // Attachment will be cleared with vkCmdClearAttachments
+    bool tbdr_discard_after{false};  // Attachment won't be read after render pass
+    
+    // VK_QCOM_render_pass_shader_resolve support
+    bool qcom_shader_resolve{false};  // Use shader resolve instead of fixed-function (last subpass)
 };
 
 } // namespace Vulkan

src/video_core/renderer_vulkan/vk_swapchain.cpp

@@ -171,6 +171,10 @@ void Swapchain::Create(
 
     resource_ticks.clear();
     resource_ticks.resize(image_count);
+
+    // Initialize incremental-present probe flags for this swapchain.
+    incremental_present_usable = device.IsKhrIncrementalPresentSupported();
+    incremental_present_probed = false;
 }
 
 bool Swapchain::AcquireNextImage() {
@@ -202,7 +206,13 @@ bool Swapchain::AcquireNextImage() {
 
 void Swapchain::Present(VkSemaphore render_semaphore) {
     const auto present_queue{device.GetPresentQueue()};
-    const VkPresentInfoKHR present_info{
+    // If the device advertises VK_KHR_incremental_present, we attempt a one-time probe
+    // on the first present to validate the driver/compositor accepts present-region info.
+    VkPresentRegionsKHR present_regions{};
+    VkPresentRegionKHR region{};
+    VkRectLayerKHR layer{};
+
+    VkPresentInfoKHR present_info{
         .sType = VK_STRUCTURE_TYPE_PRESENT_INFO_KHR,
         .pNext = nullptr,
         .waitSemaphoreCount = render_semaphore ? 1U : 0U,
@@ -212,6 +222,20 @@ void Swapchain::Present(VkSemaphore render_semaphore) {
         .pImageIndices = &image_index,
         .pResults = nullptr,
     };
+
+    if (incremental_present_usable && !incremental_present_probed) {
+        // Build a minimal present-region describing a single 1x1 dirty rect at (0,0).
+        layer.offset = {0, 0};
+        layer.extent = {1, 1};
+        region.rectangleCount = 1;
+        region.pRectangles = &layer;
+        present_regions.sType = VK_STRUCTURE_TYPE_PRESENT_REGIONS_KHR;
+        present_regions.pNext = nullptr;
+        present_regions.swapchainCount = 1;
+        present_regions.pRegions = &region;
+
+        present_info.pNext = &present_regions;
+    }
     std::scoped_lock lock{scheduler.submit_mutex};
     switch (const VkResult result = present_queue.Present(present_info)) {
     case VK_SUCCESS:
@@ -227,8 +251,18 @@ void Swapchain::Present(VkSemaphore render_semaphore) {
         break;
     default:
         LOG_CRITICAL(Render_Vulkan, "Failed to present with error {}", string_VkResult(result));
+        // If the first present with incremental-present pNext failed, disable future use.
+        if (incremental_present_usable && !incremental_present_probed) {
+            incremental_present_usable = false;
+            LOG_WARNING(Render_Vulkan, "Disabling VK_KHR_incremental_present for this swapchain due to present failure: {}", string_VkResult(result));
+        }
         break;
     }
+    if (incremental_present_usable && !incremental_present_probed) {
+        // Mark probe as completed if we reached here (success or handled failure above).
+        incremental_present_probed = true;
+        LOG_INFO(Render_Vulkan, "VK_KHR_incremental_present probe completed: usable={}", incremental_present_usable);
+    }
     ++frame_index;
     if (frame_index >= image_count) {
         frame_index = 0;

src/video_core/renderer_vulkan/vk_swapchain.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -158,6 +161,8 @@ private:
 
     bool is_outdated{};
     bool is_suboptimal{};
+    bool incremental_present_usable{};
+    bool incremental_present_probed{};
 };
 
 } // namespace Vulkan

src/video_core/renderer_vulkan/vk_texture_cache.cpp

@@ -857,6 +857,9 @@ TextureCacheRuntime::TextureCacheRuntime(const Device& device_, Scheduler& sched
         astc_decoder_pass.emplace(device, scheduler, descriptor_pool, staging_buffer_pool,
                                   compute_pass_descriptor_queue, memory_allocator);
     }
+    
+    // MSAA copy support via compute shader (only for non-Qualcomm with shaderStorageImageMultisample)
+    // Qualcomm uses VK_QCOM_render_pass_shader_resolve (fragment shader in render pass)
     if (device.IsStorageImageMultisampleSupported()) {
         msaa_copy_pass = std::make_unique<MSAACopyPass>(
             device, scheduler, descriptor_pool, staging_buffer_pool, compute_pass_descriptor_queue);
@@ -1549,6 +1552,24 @@ Image::Image(TextureCacheRuntime& runtime_, const ImageInfo& info_, GPUVAddr gpu
                 MakeStorageView(device, level, *original_image, VK_FORMAT_A8B8G8R8_UNORM_PACK32);
         }
     }
+
+    // Proactive warning for problematic HDR format + MSAA combinations on Android
+    // These combinations commonly cause texture flickering/black screens across multiple game engines
+    // Note: MSAA is native Switch rendering technique, cannot be disabled by emulator
+    if (info.num_samples > 1) {
+        const auto vk_format = MaxwellToVK::SurfaceFormat(runtime->device, FormatType::Optimal,
+                                                          false, info.format).format;
+        const bool is_hdr_format = vk_format == VK_FORMAT_B10G11R11_UFLOAT_PACK32 ||
+                                   vk_format == VK_FORMAT_E5B9G9R9_UFLOAT_PACK32;
+
+        if (is_hdr_format) {
+            LOG_WARNING(Render_Vulkan,
+                      "Creating MSAA image ({}x samples) with HDR format {} (Maxwell: {}). "
+                      "Driver support may be limited on Android (Qualcomm < 800, Mali pre-maintenance5). "
+                      "Format fallback to RGBA16F should prevent issues.",
+                      info.num_samples, vk_format, info.format);
+        }
+    }
 }
 
 Image::Image(const VideoCommon::NullImageParams& params) : VideoCommon::ImageBase{params} {}

src/video_core/renderer_vulkan/vk_texture_cache.h

@@ -1,3 +1,6 @@
+// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
+// SPDX-License-Identifier: GPL-3.0-or-later
+
 // SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
 
@@ -85,6 +88,10 @@ public:
         return msaa_copy_pass.operator bool();
     }
 
+    bool CanDownloadMSAA() const noexcept {
+        return msaa_copy_pass.operator bool();
+    }
+
     void AccelerateImageUpload(Image&, const StagingBufferRef&,
                                std::span<const VideoCommon::SwizzleParameters>);
 

src/video_core/texture_cache/image_base.cpp

@@ -131,10 +131,6 @@ bool ImageBase::IsSafeDownload() const noexcept {
     if (True(flags & ImageFlagBits::CpuModified)) {
         return false;
     }
-    if (info.num_samples > 1) {
-        LOG_WARNING(HW_GPU, "MSAA image downloads are not implemented");
-        return false;
-    }
     return true;
 }
 

src/video_core/texture_cache/texture_cache.h

@@ -101,8 +101,12 @@ void TextureCache<P>::RunGarbageCollector() {
         if (!aggressive_mode && True(image.flags & ImageFlagBits::CostlyLoad)) {
             return false;
         }
-        const bool must_download =
+        const bool supports_msaa_download = HasMsaaDownloadSupport(image.info);
-            image.IsSafeDownload() && False(image.flags & ImageFlagBits::BadOverlap);
+        if (!supports_msaa_download && image.info.num_samples > 1) {
+            LOG_WARNING(HW_GPU, "MSAA image downloads are not implemented");
+        }
+        const bool must_download = supports_msaa_download && image.IsSafeDownload() &&
+                                    False(image.flags & ImageFlagBits::BadOverlap);
         if (!high_priority_mode && must_download) {
             return false;
         }
@@ -548,10 +552,14 @@ void TextureCache<P>::WriteMemory(DAddr cpu_addr, size_t size) {
 template <class P>
 void TextureCache<P>::DownloadMemory(DAddr cpu_addr, size_t size) {
     boost::container::small_vector<ImageId, 16> images;
-    ForEachImageInRegion(cpu_addr, size, [&images](ImageId image_id, ImageBase& image) {
+    ForEachImageInRegion(cpu_addr, size, [this, &images](ImageId image_id, ImageBase& image) {
         if (!image.IsSafeDownload()) {
             return;
         }
+        if (!HasMsaaDownloadSupport(image.info)) {
+            LOG_WARNING(HW_GPU, "MSAA image downloads are not implemented");
+            return;
+        }
         image.flags &= ~ImageFlagBits::GpuModified;
         images.push_back(image_id);
     });
@@ -930,6 +938,17 @@ ImageId TextureCache<P>::DmaImageId(const Tegra::DMA::ImageOperand& operand, boo
         return NULL_IMAGE_ID;
     }
     auto& image = slot_images[dst_id];
+    if (image.info.num_samples > 1) {
+        if (is_upload) {
+            if (!HasMsaaUploadSupport(image.info)) {
+                return NULL_IMAGE_ID;
+            }
+        } else {
+            if (!HasMsaaDownloadSupport(image.info)) {
+                return NULL_IMAGE_ID;
+            }
+        }
+    }
     if (False(image.flags & ImageFlagBits::GpuModified)) {
         // No need to waste time on an image that's synced with guest
         return NULL_IMAGE_ID;
@@ -1056,7 +1075,7 @@ void TextureCache<P>::RefreshContents(Image& image, ImageId image_id) {
     image.flags &= ~ImageFlagBits::CpuModified;
     TrackImage(image, image_id);
 
-    if (image.info.num_samples > 1 && !runtime.CanUploadMSAA()) {
+    if (!HasMsaaUploadSupport(image.info)) {
         LOG_WARNING(HW_GPU, "MSAA image uploads are not implemented");
         runtime.TransitionImageLayout(image);
         return;
@@ -1274,6 +1293,16 @@ u64 TextureCache<P>::GetScaledImageSizeBytes(const ImageBase& image) {
     return fitted_size;
 }
 
+template <class P>
+bool TextureCache<P>::HasMsaaUploadSupport(const ImageInfo& info) const noexcept {
+    return info.num_samples <= 1 || runtime.CanUploadMSAA();
+}
+
+template <class P>
+bool TextureCache<P>::HasMsaaDownloadSupport(const ImageInfo& info) const noexcept {
+    return info.num_samples <= 1 || runtime.CanDownloadMSAA();
+}
+
 template <class P>
 void TextureCache<P>::QueueAsyncDecode(Image& image, ImageId image_id) {
     UNIMPLEMENTED_IF(False(image.flags & ImageFlagBits::Converted));
@@ -1491,7 +1520,31 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, DA
     for (const ImageId overlap_id : join_ignore_textures) {
         Image& overlap = slot_images[overlap_id];
         if (True(overlap.flags & ImageFlagBits::GpuModified)) {
-            UNIMPLEMENTED();
+            // Merge GPU-modified contents from the overlapping image into the newly
+            // created image to preserve guest-visible data. Compute shrink/scale
+            // copies and dispatch a GPU-side copy. This mirrors the behavior used
+            // for overlaps handled in join_copies_to_do above.
+            new_image.flags |= ImageFlagBits::GpuModified;
+            const auto& resolution = Settings::values.resolution_info;
+            const auto base_opt = new_image.TryFindBase(overlap.gpu_addr);
+            if (base_opt) {
+                const SubresourceBase base = base_opt.value();
+                const u32 up_scale = can_rescale ? resolution.up_scale : 1;
+                const u32 down_shift = can_rescale ? resolution.down_shift : 0;
+                auto copies = MakeShrinkImageCopies(new_info, overlap.info, base, up_scale, down_shift);
+                if (overlap.info.num_samples != new_image.info.num_samples) {
+                    runtime.CopyImageMSAA(new_image, overlap, FixSmallVectorADL(copies));
+                } else {
+                    runtime.CopyImage(new_image, overlap, FixSmallVectorADL(copies));
+                }
+                new_image.modification_tick = overlap.modification_tick;
+            } else {
+                // If we cannot determine a base mapping, fallback to preserving the
+                // overlap (avoid deleting GPU-modified data) and log the event so
+                // it can be investigated, we're trying to pinpoint the issue of texture flickering.
+                LOG_WARNING(HW_GPU, "Could not map overlap gpu_addr {:#x} into new image; preserving overlap", u64(overlap.gpu_addr));
+                continue;
+            }
         }
         if (True(overlap.flags & ImageFlagBits::Tracked)) {
             UntrackImage(overlap, overlap_id);
@@ -1551,6 +1604,10 @@ ImageId TextureCache<P>::JoinImages(const ImageInfo& info, GPUVAddr gpu_addr, DA
     for (const auto& copy_object : join_copies_to_do) {
         Image& overlap = slot_images[copy_object.id];
         if (copy_object.is_alias) {
+            if (!HasMsaaDownloadSupport(overlap.info)) {
+                LOG_WARNING(HW_GPU, "MSAA image downloads are not implemented");
+                continue;
+            }
             if (!overlap.IsSafeDownload()) {
                 continue;
             }
@@ -2467,8 +2524,13 @@ void TextureCache<P>::BindRenderTarget(ImageViewId* old_id, ImageViewId new_id)
     if (new_id) {
         const ImageViewBase& old_view = slot_image_views[new_id];
         if (True(old_view.flags & ImageViewFlagBits::PreemtiveDownload)) {
-            const PendingDownload new_download{true, 0, old_view.image_id};
+            const ImageBase& image = slot_images[old_view.image_id];
-            uncommitted_downloads.emplace_back(new_download);
+            if (!HasMsaaDownloadSupport(image.info)) {
+                LOG_WARNING(HW_GPU, "MSAA image downloads are not implemented");
+            } else {
+                const PendingDownload new_download{true, 0, old_view.image_id};
+                uncommitted_downloads.emplace_back(new_download);
+            }
         }
     }
     *old_id = new_id;

src/video_core/texture_cache/texture_cache_base.h

@@ -426,6 +426,8 @@ private:
     bool ScaleUp(Image& image);
     bool ScaleDown(Image& image);
     u64 GetScaledImageSizeBytes(const ImageBase& image);
+    [[nodiscard]] bool HasMsaaUploadSupport(const ImageInfo& info) const noexcept;
+    [[nodiscard]] bool HasMsaaDownloadSupport(const ImageInfo& info) const noexcept;
 
     void QueueAsyncDecode(Image& image, ImageId image_id);
     void TickAsyncDecode();

src/video_core/vulkan_common/vulkan.h

@@ -22,6 +22,34 @@
 
 #include <vulkan/vulkan.h>
 
+#ifndef VK_KHR_MAINTENANCE_1_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_1_EXTENSION_NAME "VK_KHR_maintenance1"
+#endif
+#ifndef VK_KHR_MAINTENANCE_2_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_2_EXTENSION_NAME "VK_KHR_maintenance2"
+#endif
+#ifndef VK_KHR_MAINTENANCE_3_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_3_EXTENSION_NAME "VK_KHR_maintenance3"
+#endif
+#ifndef VK_KHR_MAINTENANCE_4_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_4_EXTENSION_NAME "VK_KHR_maintenance4"
+#endif
+#ifndef VK_KHR_MAINTENANCE_5_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_5_EXTENSION_NAME "VK_KHR_maintenance5"
+#endif
+#ifndef VK_KHR_MAINTENANCE_6_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_6_EXTENSION_NAME "VK_KHR_maintenance6"
+#endif
+#ifndef VK_KHR_MAINTENANCE_7_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_7_EXTENSION_NAME "VK_KHR_maintenance7"
+#endif
+#ifndef VK_KHR_MAINTENANCE_8_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_8_EXTENSION_NAME "VK_KHR_maintenance8"
+#endif
+#ifndef VK_KHR_MAINTENANCE_9_EXTENSION_NAME
+#   define VK_KHR_MAINTENANCE_9_EXTENSION_NAME "VK_KHR_maintenance9"
+#endif
+
 // Sanitize macros
 #undef CreateEvent
 #undef CreateSemaphore

src/video_core/vulkan_common/vulkan_device.cpp

@@ -1,6 +1,5 @@
 // SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
 // SPDX-License-Identifier: GPL-3.0-or-later
-
 // SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
 // SPDX-License-Identifier: GPL-2.0-or-later
 
@@ -90,6 +89,23 @@ constexpr std::array VK_FORMAT_A4B4G4R4_UNORM_PACK16{
     VK_FORMAT_UNDEFINED,
 };
 
+// B10G11R11_UFLOAT (R11G11B10 float) is used by Unreal Engine 5 for HDR textures
+// Some Android drivers (Qualcomm pre-800, Mali pre-maintenance5) have issues with this format
+// when used with MSAA or certain tiling modes, causing texture flickering/black screens
+constexpr std::array B10G11R11_UFLOAT_PACK32{
+    VK_FORMAT_R16G16B16A16_SFLOAT, // Fallback: RGBA16F (more memory, but widely supported)
+    VK_FORMAT_E5B9G9R9_UFLOAT_PACK32, // Alternative: E5B9G9R9 shared exponent format
+    VK_FORMAT_UNDEFINED,
+};
+
+// E5B9G9R9_UFLOAT (shared exponent RGB9E5) used by various engines (Unity, custom engines)
+// Also problematic on some Android drivers, especially with MSAA and as render target
+constexpr std::array E5B9G9R9_UFLOAT_PACK32{
+    VK_FORMAT_R16G16B16A16_SFLOAT, // Fallback: RGBA16F (safest option)
+    VK_FORMAT_B10G11R11_UFLOAT_PACK32, // Alternative: might work if E5B9G9R9 fails
+    VK_FORMAT_UNDEFINED,
+};
+
 } // namespace Alternatives
 
 template <typename T>
@@ -122,6 +138,10 @@ constexpr const VkFormat* GetFormatAlternatives(VkFormat format) {
         return Alternatives::VK_FORMAT_R32G32B32_SFLOAT.data();
     case VK_FORMAT_A4B4G4R4_UNORM_PACK16_EXT:
         return Alternatives::VK_FORMAT_A4B4G4R4_UNORM_PACK16.data();
+    case VK_FORMAT_B10G11R11_UFLOAT_PACK32:
+        return Alternatives::B10G11R11_UFLOAT_PACK32.data();
+    case VK_FORMAT_E5B9G9R9_UFLOAT_PACK32:
+        return Alternatives::E5B9G9R9_UFLOAT_PACK32.data();
     default:
         return nullptr;
     }
@@ -416,7 +436,9 @@ Device::Device(VkInstance instance_, vk::PhysicalDevice physical_, VkSurfaceKHR
     const bool is_suitable = GetSuitability(surface != nullptr);
 
     const VkDriverId driver_id = properties.driver.driverID;
-    const auto device_id = properties.properties.deviceID;
+    // uncomment this if you want per-device overrides :P
+    // const u32 device_id = properties.properties.deviceID;
+
     const bool is_radv = driver_id == VK_DRIVER_ID_MESA_RADV;
     const bool is_amd_driver =
         driver_id == VK_DRIVER_ID_AMD_PROPRIETARY || driver_id == VK_DRIVER_ID_AMD_OPEN_SOURCE;
@@ -427,7 +449,6 @@ Device::Device(VkInstance instance_, vk::PhysicalDevice physical_, VkSurfaceKHR
     const bool is_mvk = driver_id == VK_DRIVER_ID_MOLTENVK;
     const bool is_qualcomm = driver_id == VK_DRIVER_ID_QUALCOMM_PROPRIETARY;
     const bool is_turnip = driver_id == VK_DRIVER_ID_MESA_TURNIP;
-    const bool is_s8gen2 = device_id == 0x43050a01;
     const bool is_arm = driver_id == VK_DRIVER_ID_ARM_PROPRIETARY;
 
     if ((is_mvk || is_qualcomm || is_turnip || is_arm) && !is_suitable) {
@@ -494,11 +515,23 @@ Device::Device(VkInstance instance_, vk::PhysicalDevice physical_, VkSurfaceKHR
     CollectPhysicalMemoryInfo();
     CollectToolingInfo();
 
-    if (is_qualcomm || is_turnip) {
+    // Driver-specific handling for VK_EXT_custom_border_color
-        LOG_WARNING(Render_Vulkan,
+    // On some Qualcomm/Turnip/ARM drivers the extension may be partially implemented.
-                    "Qualcomm and Turnip drivers have broken VK_EXT_custom_border_color");
+    // Enable it if ANY useful feature bit is reported; otherwise, let the removal pass drop it.
-        //RemoveExtensionFeature(extensions.custom_border_color, features.custom_border_color,
+    if (is_qualcomm || is_turnip || is_arm) {
-        //VK_EXT_CUSTOM_BORDER_COLOR_EXTENSION_NAME);
+        const bool has_any_custom_border_color =
+            features.custom_border_color.customBorderColors ||
+            features.custom_border_color.customBorderColorWithoutFormat;
+        if (!has_any_custom_border_color) {
+            LOG_WARNING(Render_Vulkan,
+                        "Disabling VK_EXT_custom_border_color on '{}' — no usable custom border color features reported",
+                        properties.driver.driverName);
+            // Do not clear here; final removal happens in RemoveUnsuitableExtensions based on bits.
+        } else {
+            LOG_INFO(Render_Vulkan,
+                     "Partial VK_EXT_custom_border_color support detected on '{}' — enabling available features",
+                     properties.driver.driverName);
+        }
     }
 
     if (is_qualcomm) {
@@ -681,9 +714,40 @@ Device::Device(VkInstance instance_, vk::PhysicalDevice physical_, VkSurfaceKHR
     has_broken_compute =
         CheckBrokenCompute(properties.driver.driverID, properties.properties.driverVersion) &&
         !Settings::values.enable_compute_pipelines.GetValue();
-    if (is_intel_anv || (is_qualcomm && !is_s8gen2)) {
+    must_emulate_bgr565 = false; // Default: assume emulation isn't required
-        LOG_WARNING(Render_Vulkan, "Driver does not support native BGR format");
+
+    if (is_intel_anv) {
+        LOG_WARNING(Render_Vulkan, "Intel ANV driver does not support native BGR format");
         must_emulate_bgr565 = true;
+    } else if (is_qualcomm) {
+        // Qualcomm driver version where VK_KHR_maintenance5 and A1B5G5R5 become reliable
+        constexpr uint32_t QUALCOMM_FIXED_DRIVER_VERSION = VK_MAKE_VERSION(512, 800, 1);
+        // Check if VK_KHR_maintenance5 is supported
+        if (extensions.maintenance5 && properties.properties.driverVersion >= QUALCOMM_FIXED_DRIVER_VERSION) {
+            LOG_INFO(Render_Vulkan, "Qualcomm driver supports VK_KHR_maintenance5, disabling BGR emulation");
+            must_emulate_bgr565 = false;
+        } else {
+            LOG_WARNING(Render_Vulkan, "Qualcomm driver doesn't support native BGR, emulating formats");
+            must_emulate_bgr565 = true;
+        }
+    } else if (is_turnip) {
+        // Mesa Turnip added support for maintenance5 in Mesa 25.0
+        if (extensions.maintenance5) {
+            LOG_INFO(Render_Vulkan, "Turnip driver supports VK_KHR_maintenance5, disabling BGR emulation");
+            must_emulate_bgr565 = false;
+        } else {
+            LOG_WARNING(Render_Vulkan, "Turnip driver doesn't support native BGR, emulating formats");
+            must_emulate_bgr565 = true;
+        }
+    } else if (is_arm) {
+        // ARM Mali: stop emulating BGR5 formats when VK_KHR_maintenance5 is available
+        if (extensions.maintenance5) {
+            LOG_INFO(Render_Vulkan, "ARM driver supports VK_KHR_maintenance5, disabling BGR emulation");
+            must_emulate_bgr565 = false;
+        } else {
+            LOG_WARNING(Render_Vulkan, "ARM driver doesn't support native BGR, emulating formats");
+            must_emulate_bgr565 = true;
+        }
     }
     if (extensions.push_descriptor && is_intel_anv) {
         const u32 version = (properties.properties.driverVersion << 3) >> 3;
@@ -800,15 +864,33 @@ Device::~Device() {
 VkFormat Device::GetSupportedFormat(VkFormat wanted_format, VkFormatFeatureFlags wanted_usage,
                                     FormatType format_type) const {
     if (IsFormatSupported(wanted_format, wanted_usage, format_type)) {
-        return wanted_format;
+        // CRITICAL FIX: Even if format is "supported", check for STORAGE + HDR + no MSAA support
+        // Driver may report STORAGE_IMAGE_BIT but shaderStorageImageMultisample=false means
+        // it will fail at runtime when used with MSAA (CopyImageMSAA silently fails)
+        const bool requests_storage = (wanted_usage & VK_FORMAT_FEATURE_STORAGE_IMAGE_BIT) != 0;
+        const bool is_hdr_format = wanted_format == VK_FORMAT_B10G11R11_UFLOAT_PACK32 ||
+                                   wanted_format == VK_FORMAT_E5B9G9R9_UFLOAT_PACK32;
+        
+        // If driver doesn't support shader storage image with MSAA, and we're requesting storage
+        // for an HDR format (which will likely be used with MSAA), force fallback
+        if (requests_storage && is_hdr_format && !features.features.shaderStorageImageMultisample) {
+            LOG_WARNING(Render_Vulkan,
+                      "Format {} reports STORAGE_IMAGE_BIT but driver doesn't support "
+                      "shaderStorageImageMultisample. Forcing fallback for MSAA compatibility.",
+                      wanted_format);
+            // Continue to alternatives search below
+        } else {
+            return wanted_format;
+        }
     }
     // The wanted format is not supported by hardware, search for alternatives
     const VkFormat* alternatives = GetFormatAlternatives(wanted_format);
     if (alternatives == nullptr) {
         LOG_ERROR(Render_Vulkan,
-                  "Format={} with usage={} and type={} has no defined alternatives and host "
+                  "Format={} (0x{:X}) with usage={} and type={} has no defined alternatives and host "
-                  "hardware does not support it",
+                  "hardware does not support it. Driver: {} Device: {}",
-                  wanted_format, wanted_usage, format_type);
+                  wanted_format, static_cast<u32>(wanted_format), wanted_usage, format_type,
+                  GetDriverName(), properties.properties.deviceName);
         return wanted_format;
     }
 
@@ -817,9 +899,22 @@ VkFormat Device::GetSupportedFormat(VkFormat wanted_format, VkFormatFeatureFlags
         if (!IsFormatSupported(alternative, wanted_usage, format_type)) {
             continue;
         }
-        LOG_DEBUG(Render_Vulkan,
+        // Special logging for HDR formats (common across multiple engines) on problematic drivers
+        if (wanted_format == VK_FORMAT_B10G11R11_UFLOAT_PACK32) {
+            LOG_WARNING(Render_Vulkan,
+                  "Emulating B10G11R11_UFLOAT (HDR format: UE5, custom engines) with {} on {}. "
+                  "Native format not supported by driver, using fallback.",
+                  alternative, properties.properties.deviceName);
+        } else if (wanted_format == VK_FORMAT_E5B9G9R9_UFLOAT_PACK32) {
+            LOG_WARNING(Render_Vulkan,
+                  "Emulating E5B9G9R9_UFLOAT (HDR format: Unity, RE Engine) with {} on {}. "
+                  "Native format not supported by driver, using fallback.",
+                  alternative, properties.properties.deviceName);
+        } else {
+            LOG_DEBUG(Render_Vulkan,
                   "Emulating format={} with alternative format={} with usage={} and type={}",
                   wanted_format, alternative, wanted_usage, format_type);
+        }
         return alternative;
     }
 
@@ -1267,6 +1362,43 @@ void Device::RemoveUnsuitableExtensions() {
                                VK_EXT_PROVOKING_VERTEX_EXTENSION_NAME);
     }
 
+    // VK_KHR_shader_float16_int8
+    const bool float16_int8_requested = extensions.shader_float16_int8;
+    const bool float16_int8_usable =
+        features.shader_float16_int8.shaderFloat16 || features.shader_float16_int8.shaderInt8;
+    if (float16_int8_requested && !float16_int8_usable) {
+        LOG_WARNING(Render_Vulkan,
+                    "Disabling VK_KHR_shader_float16_int8 — no shaderFloat16/shaderInt8 features reported");
+    }
+    extensions.shader_float16_int8 = float16_int8_requested && float16_int8_usable;
+    RemoveExtensionFeatureIfUnsuitable(float16_int8_usable, features.shader_float16_int8,
+                                       VK_KHR_SHADER_FLOAT16_INT8_EXTENSION_NAME);
+
+    // VK_EXT_shader_atomic_float
+    const bool atomic_float_requested = extensions.shader_atomic_float;
+    const auto& atomic_float_features = features.shader_atomic_float;
+    const bool supports_buffer_f32 = atomic_float_features.shaderBufferFloat32Atomics ||
+                                     atomic_float_features.shaderBufferFloat32AtomicAdd;
+    const bool supports_shared_f32 = atomic_float_features.shaderSharedFloat32Atomics ||
+                                     atomic_float_features.shaderSharedFloat32AtomicAdd;
+    const bool supports_image_f32 = atomic_float_features.shaderImageFloat32Atomics ||
+                                    atomic_float_features.shaderImageFloat32AtomicAdd;
+    const bool supports_sparse_f32 = atomic_float_features.sparseImageFloat32Atomics ||
+                                     atomic_float_features.sparseImageFloat32AtomicAdd;
+    const bool supports_buffer_f64 = atomic_float_features.shaderBufferFloat64Atomics ||
+                                     atomic_float_features.shaderBufferFloat64AtomicAdd;
+    const bool supports_shared_f64 = atomic_float_features.shaderSharedFloat64Atomics ||
+                                     atomic_float_features.shaderSharedFloat64AtomicAdd;
+    const bool atomic_float_usable = supports_buffer_f32 || supports_shared_f32 || supports_image_f32 ||
+                                     supports_sparse_f32 || supports_buffer_f64 || supports_shared_f64;
+    if (atomic_float_requested && !atomic_float_usable) {
+        LOG_WARNING(Render_Vulkan,
+                    "Disabling VK_EXT_shader_atomic_float — no usable atomic float feature bits reported");
+    }
+    extensions.shader_atomic_float = atomic_float_requested && atomic_float_usable;
+    RemoveExtensionFeatureIfUnsuitable(atomic_float_usable, features.shader_atomic_float,
+                                       VK_EXT_SHADER_ATOMIC_FLOAT_EXTENSION_NAME);
+
     // VK_KHR_shader_atomic_int64
     extensions.shader_atomic_int64 = features.shader_atomic_int64.shaderBufferInt64Atomics &&
                                      features.shader_atomic_int64.shaderSharedInt64Atomics;
@@ -1300,6 +1432,21 @@ void Device::RemoveUnsuitableExtensions() {
     RemoveExtensionFeatureIfUnsuitable(extensions.transform_feedback, features.transform_feedback,
                                        VK_EXT_TRANSFORM_FEEDBACK_EXTENSION_NAME);
 
+    // VK_EXT_robustness2
+    extensions.robustness_2 =
+        features.robustness2.robustBufferAccess2 && features.robustness2.robustImageAccess2;
+    RemoveExtensionFeatureIfUnsuitable(extensions.robustness_2, features.robustness2,
+                                       VK_EXT_ROBUSTNESS_2_EXTENSION_NAME);
+
+    // VK_EXT_image_robustness
+    extensions.image_robustness = features.image_robustness.robustImageAccess;
+    RemoveExtensionFeatureIfUnsuitable(extensions.image_robustness, features.image_robustness,
+                                       VK_EXT_IMAGE_ROBUSTNESS_EXTENSION_NAME);
+
+    // VK_EXT_swapchain_maintenance1
+    extensions.swapchain_maintenance1 = loaded_extensions.contains(VK_EXT_SWAPCHAIN_MAINTENANCE_1_EXTENSION_NAME);
+    RemoveExtensionIfUnsuitable(extensions.swapchain_maintenance1, VK_EXT_SWAPCHAIN_MAINTENANCE_1_EXTENSION_NAME);
+
     // VK_EXT_vertex_input_dynamic_state
     extensions.vertex_input_dynamic_state =
         features.vertex_input_dynamic_state.vertexInputDynamicState;

src/video_core/vulkan_common/vulkan_device.h

@@ -49,9 +49,11 @@ VK_DEFINE_HANDLE(VmaAllocator)
     FEATURE(EXT, ExtendedDynamicState, EXTENDED_DYNAMIC_STATE, extended_dynamic_state)             \
     FEATURE(EXT, ExtendedDynamicState2, EXTENDED_DYNAMIC_STATE_2, extended_dynamic_state2)         \
     FEATURE(EXT, ExtendedDynamicState3, EXTENDED_DYNAMIC_STATE_3, extended_dynamic_state3)         \
+    FEATURE(EXT, ShaderAtomicFloat, SHADER_ATOMIC_FLOAT, shader_atomic_float)                      \
     FEATURE(EXT, 4444Formats, 4444_FORMATS, format_a4b4g4r4)                                       \
     FEATURE(EXT, IndexTypeUint8, INDEX_TYPE_UINT8, index_type_uint8)                               \
     FEATURE(EXT, LineRasterization, LINE_RASTERIZATION, line_rasterization)                        \
+    FEATURE(EXT, ImageRobustness, IMAGE_ROBUSTNESS, image_robustness)                              \
     FEATURE(EXT, PrimitiveTopologyListRestart, PRIMITIVE_TOPOLOGY_LIST_RESTART,                    \
             primitive_topology_list_restart)                                                       \
     FEATURE(EXT, ProvokingVertex, PROVOKING_VERTEX, provoking_vertex)                              \
@@ -82,7 +84,9 @@ VK_DEFINE_HANDLE(VmaAllocator)
     EXTENSION(KHR, SHADER_FLOAT_CONTROLS, shader_float_controls)                                   \
     EXTENSION(KHR, SPIRV_1_4, spirv_1_4)                                                           \
     EXTENSION(KHR, SWAPCHAIN, swapchain)                                                           \
+    EXTENSION(KHR, INCREMENTAL_PRESENT, incremental_present)                                       \
     EXTENSION(KHR, SWAPCHAIN_MUTABLE_FORMAT, swapchain_mutable_format)                             \
+    EXTENSION(EXT, SWAPCHAIN_MAINTENANCE_1, swapchain_maintenance1)                                \
     EXTENSION(KHR, IMAGE_FORMAT_LIST, image_format_list)                                           \
     EXTENSION(NV, DEVICE_DIAGNOSTICS_CONFIG, device_diagnostics_config)                            \
     EXTENSION(NV, GEOMETRY_SHADER_PASSTHROUGH, geometry_shader_passthrough)                        \
@@ -90,7 +94,17 @@ VK_DEFINE_HANDLE(VmaAllocator)
     EXTENSION(NV, VIEWPORT_SWIZZLE, viewport_swizzle)                                              \
     EXTENSION(EXT, DESCRIPTOR_INDEXING, descriptor_indexing)                                       \
     EXTENSION(EXT, FILTER_CUBIC, filter_cubic)                                                     \
-    EXTENSION(QCOM, FILTER_CUBIC_WEIGHTS, filter_cubic_weights)
+    EXTENSION(QCOM, FILTER_CUBIC_WEIGHTS, filter_cubic_weights)                                    \
+    EXTENSION(QCOM, RENDER_PASS_SHADER_RESOLVE, render_pass_shader_resolve)                        \
+    EXTENSION(KHR, MAINTENANCE_1, maintenance1)                                                    \
+    EXTENSION(KHR, MAINTENANCE_2, maintenance2)                                                    \
+    EXTENSION(KHR, MAINTENANCE_3, maintenance3)                                                    \
+    EXTENSION(KHR, MAINTENANCE_4, maintenance4)                                                    \
+    EXTENSION(KHR, MAINTENANCE_5, maintenance5)                                                    \
+    EXTENSION(KHR, MAINTENANCE_6, maintenance6)                                                    \
+    EXTENSION(KHR, MAINTENANCE_7, maintenance7)                                                    \
+    EXTENSION(KHR, MAINTENANCE_8, maintenance8)                                                    \
+    EXTENSION(KHR, MAINTENANCE_9, maintenance9)
 
 // Define extensions which must be supported.
 #define FOR_EACH_VK_MANDATORY_EXTENSION(EXTENSION_NAME)                                            \
@@ -455,6 +469,11 @@ public:
         return extensions.image_format_list || instance_version >= VK_API_VERSION_1_2;
     }
 
+    /// Returns true if the device supports VK_KHR_incremental_present.
+    bool IsKhrIncrementalPresentSupported() const {
+        return extensions.incremental_present;
+    }
+
     /// Returns true if the device supports VK_EXT_primitive_topology_list_restart.
     bool IsTopologyListPrimitiveRestartSupported() const {
         return features.primitive_topology_list_restart.primitiveTopologyListRestart;
@@ -564,6 +583,21 @@ public:
         return extensions.filter_cubic_weights;
     }
 
+    /// Returns true if the device supports VK_QCOM_render_pass_shader_resolve
+    bool IsQcomRenderPassShaderResolveSupported() const {
+        return extensions.render_pass_shader_resolve;
+    }
+
+    /// Returns true if MSAA copy operations are supported via compute shader (upload/download)
+    /// Qualcomm uses render pass shader resolve instead, so this returns false for Qualcomm
+    bool CanUploadMSAA() const {
+        return IsStorageImageMultisampleSupported();
+    }
+
+    bool CanDownloadMSAA() const {
+        return CanUploadMSAA();
+    }
+
     /// Returns true if the device supports VK_EXT_line_rasterization.
     bool IsExtLineRasterizationSupported() const {
         return extensions.line_rasterization;
@@ -594,6 +628,11 @@ public:
         return extensions.shader_atomic_int64;
     }
 
+    /// Returns true if the device supports VK_EXT_shader_atomic_float.
+    bool IsExtShaderAtomicFloatSupported() const {
+        return extensions.shader_atomic_float;
+    }
+
     bool IsExtConditionalRendering() const {
         return extensions.conditional_rendering;
     }
@@ -817,7 +856,6 @@ private:
         VkPhysicalDevicePushDescriptorPropertiesKHR push_descriptor{};
         VkPhysicalDeviceSubgroupSizeControlProperties subgroup_size_control{};
         VkPhysicalDeviceTransformFeedbackPropertiesEXT transform_feedback{};
-
         VkPhysicalDeviceProperties properties{};
     };