bump version to v131

The -mconsole and -mwindows flags don't seem to do much of anything when
Clang isn't targeting MinGW. On the other hand, directly overriding the
subsystem with a linker flag results in a bunch of unresolved symbols.
However, if the right entry point exists (either WinMain or main),
some apparent magic goes on under the hood to ensure that everything
links as expected, so we can just take advantage of that.

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,27 @@
+---
+name: Bug report
+about: Report a bug *not* relating to game compatibility
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+**Describe the bug**
+A clear and concise description of what the bug is.
+
+**To Reproduce**
+Steps to reproduce the behavior:
+1. Go to '...'
+2. Click on '....'
+3. Scroll down to '....'
+4. See error
+
+**Expected behavior**
+A clear and concise description of what you expected to happen.
+
+**Screenshots**
+If applicable, add screenshots to help explain your problem.
+
+**Additional context**
+Add any other context about the problem here.

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,20 @@
+---
+name: Feature request
+about: Suggest an idea for this project
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+**Is your feature request related to a problem? Please describe.**
+A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
+
+**Describe the solution you'd like**
+A clear and concise description of what you want to happen.
+
+**Describe alternatives you've considered**
+A clear and concise description of any alternative solutions or features you've considered.
+
+**Additional context**
+Add any other context or screenshots about the feature request here.

.github/ISSUE_TEMPLATE/game-issue-report.md

@@ -0,0 +1,21 @@
+---
+name: Game Issue Report
+about: Report an issue running a specific game in ares
+title: "[system] Game"
+labels: bug
+assignees: ''
+
+---
+
+Before submitting an issue, please use the search to make sure an issue doesn't already exist, you should also make sure your rom is not a bad dump, by comparing against the hashes in the No-Intro (Cartridges) or Redump (CD-ROM) databases.
+
+**Describe the issue**
+A clear and concise description of what the bug is.
+
+**Rom Information**
+Specify the impacted rom filename and any associated hashes (sha256 preferred)
+
+```
+file: Some Game (REGION).ext
+sha256: <hash>
+```

.github/workflows/build.yml

@@ -11,27 +11,131 @@ jobs:
       matrix:
         program:
         - ares
-        os:
+        platform:
         - name: windows
-          version: latest
+          os: windows-latest
+          compiler: clang++
+          windres: windres
+          shell: 'msys2 {0}'
+          msystem: clang64
+          msys-env: clang-x86_64
+        - name: windows-arm64
+          os: windows-latest
+          compiler: clang++ --target=aarch64-w64-windows-gnu --sysroot=/clangarm64 -resource-dir=/clangarm64/lib/clang/$resourcever
+          windres: windres --target=aarch64-w64-windows-gnu
+          shell: 'msys2 {0}'
+          msystem: clang64
+          msys-env: clang-x86_64
+        - name: windows-msvc-x64
+          os: windows-latest
+          compiler: clang++
+          windres: llvm-rc
+          shell: bash
+          msvc-arch: x64
+        - name: windows-msvc-arm64
+          os: windows-latest
+          compiler: clang++ --target=aarch64-w64-windows-msvc
+          windres: llvm-rc
+          shell: bash
+          msvc-arch: x64_arm64
         - name: macos
-          version: latest
+          os: macos-latest
+          compiler: clang++
+          shell: sh
         - name: ubuntu
-          version: latest
-    runs-on: ${{ matrix.os.name }}-${{ matrix.os.version }}
+          os: ubuntu-latest
+          compiler: g++
+          shell: sh
+    name: ${{ matrix.program }}-${{ matrix.platform.name }}
+    runs-on: ${{ matrix.platform.os }}
+    defaults:
+      run:
+        shell: ${{ matrix.platform.shell }}
     steps:
-    - uses: actions/checkout@v2
-    - name: Install Dependencies
-      if: matrix.os.name == 'ubuntu'
+    - name: Install MSYS2 Dependencies
+      if: matrix.platform.shell == 'msys2 {0}'
+      uses: msys2/setup-msys2@v2
+      with:
+        msystem: ${{ matrix.platform.msystem }}
+        update: true
+        install: make mingw-w64-${{ matrix.platform.msys-env }}-toolchain
+    - name: Install MSYS2 ARM64 Dependencies
+      if: matrix.platform.name == 'windows-arm64'
+      run: | # enable clang arm64 package source, refresh package database, and install clang package
+        echo "resourcever=$(basename "$(clang++ -print-resource-dir)")" >> $GITHUB_ENV
+        grep -qFx '[clangarm64]' /etc/pacman.conf || sed -i '/^# \[clangarm64\]/,/^$/ s|^# ||g' /etc/pacman.conf
+        pacman --noconfirm -Sy mingw-w64-clang-aarch64-clang
+    - name: Install Linux Dependencies
+      if: runner.os == 'Linux'
       run: |
         sudo apt-get update -y -qq
-        sudo apt-get install libsdl2-dev libgtksourceview2.0-dev libgtk2.0-dev libao-dev libopenal-dev   
+        sudo apt-get install libsdl2-dev libgtk2.0-dev libao-dev libopenal-dev   
+    - name: "macOS: Import Certificate"
+      if: runner.os == 'macOS' && github.event_name != 'pull_request'
+      uses: devbotsxyz/import-signing-certificate@2ac4f44d28045073d23153256efbb4c4b2d8aa22    # Don't use rolling branch for security reasons
+      with:
+        certificate-data: ${{ secrets.MACOS_CERTIFICATE_DATA }}
+        certificate-passphrase: ${{ secrets.MACOS_CERTIFICATE_PASSPHRASE }}
+        keychain-name: ares-macos-keychain
+        keychain-password: ${{ secrets.MACOS_KEYCHAIN_PASSWORD }}
+    - name: Checkout source code
+      uses: actions/checkout@v2
+    - name: "macOS: recover cache"
+      if: runner.os == 'macOS'
+      uses: actions/cache@v3
+      env:
+          cache-name: moltenvk
+      with:
+          path: thirdparty/MoltenVK/MoltenVK
+          key: ${{ runner.os }}-build-${{ env.cache-name }}-${{ hashFiles('thirdparty/MoltenVK/HEAD') }}
+    - name: Install macOS Dependencies
+      if: runner.os == 'macOS'
+      run: |
+        brew install make ninja
+        echo "MAKE=gmake" >> $GITHUB_ENV
+        pushd thirdparty/MoltenVK
+        ./build-moltenvk.sh
+        popd
+    - name: Set up MSVC environment
+      if: matrix.msvc-arch != ''
+      uses: ilammy/msvc-dev-cmd@v1
+      with:
+        arch: ${{ matrix.msvc-arch }}
     - name: Make
-      run: make -j4 -C desktop-ui build=optimized local=false lto=true
+      if: runner.os != 'macOS'
+      run: ${MAKE:-make} -j4 -C desktop-ui build=optimized local=false compiler="${{ matrix.platform.compiler }}" windres="${{ matrix.platform.windres }}"
+    - name: "macOS: Make universal app"
+      if: runner.os == 'macOS'
+      run: scripts/macos-make-universal.sh
+      env:
+        MAKEFLAGS: -j3
+        MACOS_CERTIFICATE_NAME: ${{ secrets.MACOS_CERTIFICATE_NAME }}
+        MACOS_KEYCHAIN_NAME: ares-macos-keychain
+        MACOS_KEYCHAIN_PASSWORD: ${{ secrets.MACOS_KEYCHAIN_PASSWORD }}
+    - name: "Windows: sign"
+      if: runner.os == 'Windows' && github.event_name != 'pull_request'
+      uses: DanaBear/code-sign-action@98c79121b376beab8d6a9484f445089db4461bca
+      with:
+        certificate: '${{ secrets.WIN_CERTIFICATE }}'
+        password: '${{ secrets.WIN_PASSWORD }}'
+        certificatesha1: '${{ secrets.WIN_CERTHASH }}'
+        certificatename: '${{ secrets.WIN_CERTNAME }}'
+        folder: 'desktop-ui/out'
+        timestampUrl: 'http://timestamp.digicert.com' 
+    - name: "macOS: notarize"
+      if: runner.os == 'macOS' && github.event_name != 'pull_request'
+      run: |
+        ditto -c -k --keepParent desktop-ui/out/ares.app /tmp/ares.zip
+        xcrun notarytool submit /tmp/ares.zip --apple-id "$MACOS_NOTARIZATION_USERNAME" --password "$MACOS_NOTARIZATION_PASSWORD" --team-id "$MACOS_NOTARIZATION_TEAMID" --wait
+        xcrun stapler staple desktop-ui/out/ares.app
+      env:
+        MACOS_NOTARIZATION_USERNAME: ${{ secrets.MACOS_NOTARIZATION_USERNAME }}
+        MACOS_NOTARIZATION_PASSWORD: ${{ secrets.MACOS_NOTARIZATION_PASSWORD }}
+        MACOS_NOTARIZATION_TEAMID: ${{ secrets.MACOS_NOTARIZATION_TEAMID }}
     - name: Upload
       uses: actions/upload-artifact@v2
       with:
-        name: ${{ matrix.program }}-${{ matrix.os.name }}
+        name: ${{ matrix.program }}-${{ matrix.platform.name }}
         path: desktop-ui/out/*
 
      
@@ -63,16 +167,16 @@ jobs:
         # Hack: Workaround for GitHub artifacts losing attributes.
         chmod +x ${bindir}/ares-macos/ares.app/Contents/MacOS/ares
 
-        for os in windows macos
+        for package in windows-msvc-x64 windows-msvc-arm64 macos
         do
-          mkdir "${os}"
-          cd "${os}"
+          mkdir "${package}"
+          cd "${package}"
 
           # Package ares.
           outdir=ares${suffix}
           mkdir ${outdir}
-          cp -ar ${bindir}/ares-${os}/* ${outdir}
-          zip -r ../ares-${os}.zip ${outdir}
+          cp -ar ${bindir}/ares-${package}/* ${outdir}
+          zip -r ../ares-${package}.zip ${outdir}
 
           cd -
         done
@@ -155,10 +259,14 @@ jobs:
             echo "UPLOAD_URL=$(make_nightly_release | jq -r .upload_url)" >> $GITHUB_ENV
             ;;
         esac
-    - name: Upload ares-windows
+    - name: Upload ares-windows-msvc-x64
       uses: actions/upload-release-asset@v1
       env: { GITHUB_TOKEN: '${{ secrets.GITHUB_TOKEN }}' }
-      with: { upload_url: '${{ env.UPLOAD_URL }}', asset_path: 'ares-windows.zip', asset_name: 'ares-windows.zip', asset_content_type: 'application/zip' }
+      with: { upload_url: '${{ env.UPLOAD_URL }}', asset_path: 'ares-windows-msvc-x64.zip', asset_name: 'ares-windows-msvc-x64.zip', asset_content_type: 'application/zip' }
+    - name: Upload ares-windows-msvc-arm64
+      uses: actions/upload-release-asset@v1
+      env: { GITHUB_TOKEN: '${{ secrets.GITHUB_TOKEN }}' }
+      with: { upload_url: '${{ env.UPLOAD_URL }}', asset_path: 'ares-windows-msvc-arm64.zip', asset_name: 'ares-windows-msvc-arm64.zip', asset_content_type: 'application/zip' }      
     - name: Upload ares-macos
       uses: actions/upload-release-asset@v1
       env: { GITHUB_TOKEN: '${{ secrets.GITHUB_TOKEN }}' }

.gitignore

@@ -0,0 +1,6 @@
+obj/
+out/
+obj-amd64/
+obj-arm64/
+out-amd64/
+out-arm64/

GNUmakefile

@@ -0,0 +1,6 @@
+targets = all clean install uninstall verbose
+
+$(targets):
+	@$(MAKE) -C desktop-ui $@
+
+.PHONY: $(targets)

LICENSE

@@ -1,3 +1,4 @@
+----------------------------------------------------------------------
 ares
 
 Copyright (c) 2004-2021 ares team, Near et al
@@ -13,3 +14,430 @@ ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
 WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
 ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+Stack-less Just-In-Time compiler
+
+Copyright Zoltan Herczeg ([email protected]). All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+1. Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright
+   notice, this list of conditions and the following disclaimer in the
+   documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER(S) AND CONTRIBUTORS
+``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+HOLDER(S) OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
+USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
+ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+MAME
+Copyright (c) 1997-2021 MAMEdev and contributors
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+1. Redistributions of source code must retain the above copyright
+   notice, this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright
+   notice, this list of conditions and the following disclaimer in the
+   documentation and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+   contributors may be used to endorse or promote products derived from
+   this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
+IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
+TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
+PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT
+HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
+TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+libchdr
+
+Copyright Romain Tisserand
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.
+    * Redistributions in binary form must reproduce the above copyright
+      notice, this list of conditions and the following disclaimer in the
+      documentation and/or other materials provided with the distribution.
+    * Neither the name of the <organization> nor the
+      names of its contributors may be used to endorse or promote products
+      derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
+DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+LZMA SDK is placed in the public domain.
+
+Anyone is free to copy, modify, publish, use, compile, sell, or distribute
+the original LZMA SDK code, either in source code form or as a compiled binary,
+for any purpose, commercial or non-commercial, and by any means.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+zlib
+
+ (C) 1995-2017 Jean-loup Gailly and Mark Adler
+
+  This software is provided 'as-is', without any express or implied
+  warranty.  In no event will the authors be held liable for any damages
+  arising from the use of this software.
+
+  Permission is granted to anyone to use this software for any purpose,
+  including commercial applications, and to alter it and redistribute it
+  freely, subject to the following restrictions:
+
+  1. The origin of this software must not be misrepresented; you must not
+     claim that you wrote the original software. If you use this software
+     in a product, an acknowledgment in the product documentation would be
+     appreciated but is not required.
+  2. Altered source versions must be plainly marked as such, and must not be
+     misrepresented as being the original software.
+  3. This notice may not be removed or altered from any source distribution.
+
+  Jean-loup Gailly        Mark Adler
+  [email protected]          [email protected]
+
+If you use the zlib library in a product, we would appreciate *not* receiving
+lengthy legal documents to sign.  The sources are provided for free but without
+warranty of any kind.  The library has been entirely written by Jean-loup
+Gailly and Mark Adler; it does not include third-party code.
+
+If you redistribute modified sources, we would appreciate that you include in
+the file ChangeLog history information documenting your changes.  Please read
+the FAQ for more information on the distribution of modified source versions.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+
+xxHash - Extremely Fast Hash algorithm
+Header File
+Copyright (C) 2012-2021 Yann Collet
+
+BSD 2-Clause License (https://www.opensource.org/licenses/bsd-license.php)
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+* Redistributions of source code must retain the above copyright
+  notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above
+  copyright notice, this list of conditions and the following disclaimer
+  in the documentation and/or other materials provided with the
+  distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+You can contact the author at:
+- xxHash homepage: https://www.xxhash.com
+- xxHash source repository: https://github.com/Cyan4973/xxHash
+
+----------------------------------------------------------------------
+
+MoltenVK
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.
+   
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+OpenGL API and Extension Registry
+https://github.com/KhronosGroup/OpenGL-Registry
+
+Copyright 2013-2020 The Khronos Group Inc.
+
+Permission is hereby granted, free of charge, to any person obtaining a
+copy of this software and/or associated documentation files (the
+"Materials"), to deal in the Materials without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Materials, and to
+permit persons to whom the Materials are furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be included
+in all copies or substantial portions of the Materials.
+
+THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
+----------------------------------------------------------------------
+
+----------------------------------------------------------------------
+EGL API and Extension Registry
+https://github.com/KhronosGroup/EGL-Registry
+
+Copyright (c) 2008-2018 The Khronos Group Inc.
+
+Permission is hereby granted, free of charge, to any person obtaining a
+copy of this software and/or associated documentation files (the
+"Materials"), to deal in the Materials without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Materials, and to
+permit persons to whom the Materials are furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be included
+in all copies or substantial portions of the Materials.
+
+THE MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+MATERIALS OR THE USE OR OTHER DEALINGS IN THE MATERIALS.
+----------------------------------------------------------------------

README.md

@@ -1,4 +1,4 @@
-# ares
+![ares logo](https://github.com/ares-emulator/ares/blob/master/ares/ares/resource/logo.png)
 
 [![License: ISC](https://img.shields.io/badge/License-ISC-blue.svg)](https://github.com/higan-emu/ares/blob/master/LICENSE)
 
@@ -9,10 +9,90 @@ Official Releases
 -----------------
 
 Official releases are available from
-[the ares website](https://ares-emulator.github.io).
+[the ares website](https://ares-emu.net).
 
 Nightly Builds
 --------------
 
 Automated, untested builds of ares are available for Windows and macOS as a [pre-release](https://github.com/higan-emu/ares/releases/tag/nightly). 
 Only the latest nightly build is kept.
+
+Prerequisites
+-------------
+
+#### *nix building
+
+```
+g++ make pkg-config libgtk2.0-dev libcanberra-gtk-module libgl-dev libasound2-dev
+```
+
+By default, GTK2 is used, but support for GTK3 is available. You will need to install the additional package `libgtk-3-dev` as well
+as specifying the command line option `hiro=gtk3` at compile time.
+
+#### Windows building
+
+To build on Windows, using MSYS2 is recommended which can be download [here](https://www.msys2.org/). Follow the instructions
+on this page to install and setup an appropriate MINGW64 environment. Running the command  `pacman -S --needed base-devel mingw-w64-x86_64-toolchain` from the MSYS2 MSYS terminal should setup everything you need to compile Ares. Note that in order 
+to compile, you will want to be in a MINGW64 terminal window after install and setup is complete. 
+
+Compilation
+-----------
+
+Check out the source code by running this command:
+
+```
+git clone https://github.com/ares-emulator/ares.git
+```
+
+From the root of the project directory run:
+
+```
+make -j4 build=release
+```
+
+that builds with build type of type 'release'. 
+`-j4` indicates number of parallel build processes, and shouldn't be set higher than N-1 cores on your processor. Specifying this option can significantly decrease the time to build this project. There are multiple build types available (debug, etc.). Most additional options can be 
+found in nall's make file (nall/GNUmakefile).
+
+To start compilation from the beginning, run the following prior to compiling:
+
+```
+make clean
+```
+
+Build Output
+------------
+
+There is a single binary produced at the end of compilation which can be found in `desktop-ui/out`. On OS's besides Linux, the `Database` & `Shader` directories are copied over here as well. On Linux, running `make install` after compilation will copy these directories and binary into suitable locations (see desktop-ui/GNUmakefile for details). Alternatively, these directories can be copied from `ares/Shaders/*` and `mia/Database/*`.
+
+
+Command-line options
+--------------------
+
+When started from the command-line, ares accepts a few options.
+
+```
+Usage: ./ares [options] game
+
+  --help                 Displays available options and exit
+  --fullscreen           Start in full screen mode
+  --system system        Specify the system name
+```
+
+The --system option is useful when the system type cannot be auto-detected.
+--fullscreen will only have an effect if a game is also passed in argument.
+
+Example:
+`ares --system MSX examples.rom --fullscreen`
+
+
+High-level Components
+---------------------
+
+* __ares__:       emulator cores and component implementations
+* __desktop-ui__: main GUI implementation for this project
+* __hiro__:       cross-platform GUI toolkit that utilizes native APIs on supported platforms
+* __nall__:       Near's alternative to the C++ standard library
+* __ruby__:       interface between a hiro application and platform-specific APIs for emulator video, audio, and input
+* __mia__:        internal ROM database and ROM/image loader
+* __libco__:      cooperative multithreading library

ares/GNUmakefile

@@ -1,7 +1,5 @@
 ifeq ($(platform),windows)
   options += -luuid -lkernel32 -luser32 -lgdi32 -lcomctl32 -lcomdlg32 -lshell32
-  options += -Wl,-enable-auto-import
-  options += -Wl,-enable-runtime-pseudo-reloc
 else ifeq ($(platform),macos)
 else ifneq ($(filter $(platform),linux bsd),)
   options += -Wl,-export-dynamic
@@ -15,6 +13,10 @@ ares.objects := ares ares-fixed-allocator
 $(object.path)/ares.o: $(ares.path)/ares/ares.cpp
 $(object.path)/ares-fixed-allocator.o: $(ares.path)/ares/memory/fixed-allocator.cpp
 
+ifeq ($(platform),macos)
+  ares.dylibs :=
+endif
+
 ifeq ($(vulkan),true)
   flags += -DVULKAN
 endif
@@ -27,6 +29,10 @@ ifeq ($(profile),performance)
   flags += -DPROFILE_PERFORMANCE
 endif
 
+ifneq ($(filter $(cores),a26),)
+  include $(ares.path)/a26/GNUmakefile
+endif
+
 ifneq ($(filter $(cores),fc),)
   include $(ares.path)/fc/GNUmakefile
 endif

ares/Shaders/CRT-Lottes2.shader/lottes.vs

@@ -1,16 +1,16 @@
-#version 150
-
-in vec4 position;
-in vec2 texCoord;
-
-out Vertex {
-   vec2 vTexCoord;
-};
-
-uniform vec4 targetSize;
-uniform vec4 sourceSize[];
-
-void main() {
-   gl_Position = position;
-   vTexCoord = texCoord;
+#version 150
+
+in vec4 position;
+in vec2 texCoord;
+
+out Vertex {
+   vec2 vTexCoord;
+};
+
+uniform vec4 targetSize;
+uniform vec4 sourceSize[];
+
+void main() {
+   gl_Position = position;
+   vTexCoord = texCoord;
 }

ares/Shaders/CRT-Lottes2.shader/manifest.bml

@@ -1,12 +1,12 @@
-input
-  filter: nearest
-
-program
-  filter: nearest
-  vertex: lottes.vs
-  fragment: lottes.fs
-  
-output
-  height: 0
-  width: 0
+input
+  filter: nearest
+
+program
+  filter: nearest
+  vertex: lottes.vs
+  fragment: lottes.fs
+  
+output
+  height: 0
+  width: 0
   filter: linear

ares/Shaders/GritsScanlines.shader/GritsScanlines.fs

@@ -32,18 +32,18 @@ out vec4 fragColor;
 #ifdef LuminanceLUT
 #define LUT_SizeLum 16.0
 
-	// Code taken from RetroArch's LUT shader
+  // Code taken from RetroArch's LUT shader
 float luminancelut(vec4 org)
 {
-	vec4 imgColorLum = org;
-	float redLum = ( imgColorLum.r * (LUT_SizeLum - 1.0) + 0.4999 ) / (LUT_SizeLum * LUT_SizeLum);
-	float greenLum = ( imgColorLum.g * (LUT_SizeLum - 1.0) + 0.4999 ) / LUT_SizeLum;
-	float blue1Lum = (floor( imgColorLum.b  * (LUT_SizeLum - 1.0) ) / LUT_SizeLum) + redLum;
-	float blue2Lum = (ceil( imgColorLum.b  * (LUT_SizeLum - 1.0) ) / LUT_SizeLum) + redLum;
-	float mixerLum = clamp(max((imgColorLum.b - blue1Lum) / (blue2Lum - blue1Lum), 0.0), 0.0, 32.0);
-	float color1Lum = texture(pixmap[1], vec2( blue1Lum, greenLum )).x;
-	float color2Lum = texture(pixmap[1], vec2( blue2Lum, greenLum )).x;
-	return mix(color1Lum, color2Lum, mixerLum);
+  vec4 imgColorLum = org;
+  float redLum = ( imgColorLum.r * (LUT_SizeLum - 1.0) + 0.4999 ) / (LUT_SizeLum * LUT_SizeLum);
+  float greenLum = ( imgColorLum.g * (LUT_SizeLum - 1.0) + 0.4999 ) / LUT_SizeLum;
+  float blue1Lum = (floor( imgColorLum.b  * (LUT_SizeLum - 1.0) ) / LUT_SizeLum) + redLum;
+  float blue2Lum = (ceil( imgColorLum.b  * (LUT_SizeLum - 1.0) ) / LUT_SizeLum) + redLum;
+  float mixerLum = clamp(max((imgColorLum.b - blue1Lum) / (blue2Lum - blue1Lum), 0.0), 0.0, 32.0);
+  float color1Lum = texture(pixmap[1], vec2( blue1Lum, greenLum )).x;
+  float color2Lum = texture(pixmap[1], vec2( blue2Lum, greenLum )).x;
+  return mix(color1Lum, color2Lum, mixerLum);
 }
 #endif
 
@@ -52,48 +52,48 @@ float luminancelut(vec4 org)
 
 vec4 TrinitronD50(vec4 org)
 {
-	vec4 imgColorTrinitron = org;
-	float redTrinitron = ( imgColorTrinitron.r * (LUT_SizeTrinitron - 1.0) + 0.4999 ) / (LUT_SizeTrinitron * LUT_SizeTrinitron);
-	float greenTrinitron = ( imgColorTrinitron.g * (LUT_SizeTrinitron - 1.0) + 0.4999 ) / LUT_SizeTrinitron;
-	float blue1Trinitron = (floor( imgColorTrinitron.b  * (LUT_SizeTrinitron - 1.0) ) / LUT_SizeTrinitron) + redTrinitron;
-	float blue2Trinitron = (ceil( imgColorTrinitron.b  * (LUT_SizeTrinitron - 1.0) ) / LUT_SizeTrinitron) + redTrinitron;
-	float mixerTrinitron = clamp(max((imgColorTrinitron.b - blue1Trinitron) / (blue2Trinitron - blue1Trinitron), 0.0), 0.0, 32.0);
-	vec4 color1Trinitron = texture(pixmap[2], vec2( blue1Trinitron, greenTrinitron ));
-	vec4 color2Trinitron = texture(pixmap[2], vec2( blue2Trinitron, greenTrinitron ));
-	vec4 fragColorTrinitron = mix(color1Trinitron, color2Trinitron, mixerTrinitron);
-	return vec4(pow(fragColorTrinitron.rgb,vec3(GammaCorrection,GammaCorrection,GammaCorrection)),1.0);
+  vec4 imgColorTrinitron = org;
+  float redTrinitron = ( imgColorTrinitron.r * (LUT_SizeTrinitron - 1.0) + 0.4999 ) / (LUT_SizeTrinitron * LUT_SizeTrinitron);
+  float greenTrinitron = ( imgColorTrinitron.g * (LUT_SizeTrinitron - 1.0) + 0.4999 ) / LUT_SizeTrinitron;
+  float blue1Trinitron = (floor( imgColorTrinitron.b  * (LUT_SizeTrinitron - 1.0) ) / LUT_SizeTrinitron) + redTrinitron;
+  float blue2Trinitron = (ceil( imgColorTrinitron.b  * (LUT_SizeTrinitron - 1.0) ) / LUT_SizeTrinitron) + redTrinitron;
+  float mixerTrinitron = clamp(max((imgColorTrinitron.b - blue1Trinitron) / (blue2Trinitron - blue1Trinitron), 0.0), 0.0, 32.0);
+  vec4 color1Trinitron = texture(pixmap[2], vec2( blue1Trinitron, greenTrinitron ));
+  vec4 color2Trinitron = texture(pixmap[2], vec2( blue2Trinitron, greenTrinitron ));
+  vec4 fragColorTrinitron = mix(color1Trinitron, color2Trinitron, mixerTrinitron);
+  return vec4(pow(fragColorTrinitron.rgb,vec3(GammaCorrection,GammaCorrection,GammaCorrection)),1.0);
 } 
 #endif
 
 void main() {
 
 //Source Image
-	vec4 org = texture(source[0], texCoord);
+  vec4 org = texture(source[0], texCoord);
    
 #ifdef LuminanceLUT
 // Use a 3DLUT instead of an equation so that it can use any arbitrary mess you can come up with.
-	float luminance = luminancelut(org);
+  float luminance = luminancelut(org);
 #elif defined LuminanceDawnbringer
 // Dawnbringer's brightness equation from Dawnbringer's Toolbox scripts for Grafx2
-	float luminance = sqrt(org.r*org.r*0.0676 + org.g*org.g*0.3025 + org.b*org.b*0.0361) * 1.5690256395005606;
+  float luminance = sqrt(org.r*org.r*0.0676 + org.g*org.g*0.3025 + org.b*org.b*0.0361) * 1.5690256395005606;
 #else
 // Plain, standard, fine; slightly faster
-	float luminance = ((0.299*org.r) + (0.587*org.g) + (0.114*org.b));
+  float luminance = ((0.299*org.r) + (0.587*org.g) + (0.114*org.b));
 #endif
 
 // Don't let it exceed 1.0
-	luminance = clamp(luminance, 0.0, 1.0);
+  luminance = clamp(luminance, 0.0, 1.0);
 
 // Scanline Mapping, based on the Phosphor LUT shader's method of tiling a texture over the screen
-	vec2 LUTeffectiveCoord = vec2(luminance,fract((texCoord.y*targetSize.y)/4.0));
+  vec2 LUTeffectiveCoord = vec2(luminance,fract((texCoord.y*targetSize.y)/4.0));
 
 // Scanline Layer
-	vec4 screen = texture(pixmap[0], LUTeffectiveCoord);
+  vec4 screen = texture(pixmap[0], LUTeffectiveCoord);
 
 // Output multiplying the scanlines into the original image, with control over opacity
 #ifdef TrinitronColors
-	org = TrinitronD50(org);
+  org = TrinitronD50(org);
 #endif
-	fragColor = ((screen*ScanlinesOpacity)+(1.0 - ScanlinesOpacity)) * (org);
+  fragColor = ((screen*ScanlinesOpacity)+(1.0 - ScanlinesOpacity)) * (org);
 }
 

ares/Shaders/PAL-r57shell.shader/manifest.bml

@@ -1,11 +1,11 @@
-input
-  filter: nearest
-
-program
-  filter: nearest
-  height: 100%
-  fragment: pal-r57shell.fs
-  
-output
-  height: 0
+input
+  filter: nearest
+
+program
+  filter: nearest
+  height: 100%
+  fragment: pal-r57shell.fs
+  
+output
+  height: 0
   filter: linear

ares/Shaders/PAL-r57shell.shader/pal-r57shell.fs

@@ -1,368 +1,368 @@
-#version 150
-
-// NES PAL composite signal simulation for RetroArch
-// shader by r57shell
-// thanks to feos & HardWareMan & NewRisingSun
-
-// also TV subpixels and scanlines
-
-// LICENSE: PUBLIC DOMAIN
-
-// NOTE: for nice TV subpixels and scanlines I recommend to
-// disable this features here and apply CRT-specialized shader.
-
-// Quality considerations
-
-// there are three main options:
-// USE_RAW (R), USE_DELAY_LINE (D), USE_COLORIMETRY (C)
-// here is table of quality in decreasing order:
-// RDC, RD, RC, DC, D, C
-
-// compatibility macros
-#define float2 vec2
-#define float3 vec3
-#define float4 vec4
-#define frac(c) fract(c)
-#define saturate(c) clamp(c, 0.0, 1.0)
-#define fmod(x,y) mod(x,y)
-#define mul(x,y) (y*x)
-#define float2x2 mat2
-#define float3x3 mat3
-#define float4x4 mat4
-#define bool2 bvec2
-#define bool3 bvec3
-#define bool4 bvec4
-#define static 
-
-uniform sampler2D source[];
-uniform vec4 sourceSize[];
-uniform vec4 outputSize;
-uniform int phase;
-
-in Vertex {
-   vec2 texCoord;
-};
-
-out vec4 fragColor;
-
-// TWEAKS start
-
-// use delay line technique
-// without delay line technique, color would interleave
-// to avoid this, set HueRotation to zero.
-#define USE_DELAY_LINE
-
-// use this if you need to swap even/odd V sign.
-// sign of V changes each scanline
-// so if some scanline is positive, then next is negative
-// and if you want to match picture
-// to actual running PAL NES on TV
-// you may want to have this option, to change signs
-// if they don't match
-//#define SWAP_VSIGN
-
-// rough simulation of scanlines
-// better if you use additional shader instead
-// if you still use it, make sure that SizeY
-// is at least twice lower than output height
-#define USE_SCANLINES // FIXME: scanlines are broken and I'm too lazy to fix it right now
-
-// to change gamma of virtual TV from 2.2 to something else
-#define USE_GAMMA
-
-// use sampled version. it's much more slower version of shader.
-// because it is computing x4 more values. NOT RECOMMENDED.
-//#define USE_SAMPLED
-
-#ifndef PARAMETER_UNIFORM
-// NTSC standard gamma = 2.2
-// PAL standard gamma = 2.8
-// according to many sources, very unlikely gamma of TV is 2.8
-// most likely gamma of PAL TV is in range 2.4-2.5
-const float Gamma_static = 2.8; // gamma of virtual TV
-
-const float Brightness_static = 0.25;
-const float Contrast_static = 1.1;
-const float Saturation_static = 1.1;
-
-const int
-	Ywidth_static = 12,
-	Uwidth_static = 23,
-	Vwidth_static = 23;
-   
-#define Brightness Brightness_static
-#define Gamma Gamma_static
-
-#define Ywidth Ywidth_static
-#define Uwidth Uwidth_static
-#define Vwidth Vwidth_static
-
-#define SCANLINE_MUL (sw.x*dark_scanline+sw.y)
-
-int Mwidth = int(max(float(Ywidth), max(float(Uwidth), float(Vwidth))));
-
-// correct one is -2.5
-// works only with USE_RAW
-const float HueShift = -2.5;
-
-// rotation of hue due to luma level.
-const float HueRotation = 2.;
-
-// touch this only if you know what you doing
-const float Phase_Y = 2.; // fmod(341*10,12)
-const float Phase_One = 0.; // alternating phases.
-const float Phase_Two = 8.;
-
-// screen size, scanlines = y*2; y one field, and y other field.
-const int SizeX = 256;
-const int SizeY = 240;
-
-// count of pixels of virtual TV.
-// value close to 1000 produce small artifacts
-const int TV_Pixels = 400;
-
-const float dark_scanline = 0.5; // half
-#endif
-
-// this is using following matrixes.
-// it provides more scientific approach
-// by conversion into linear XYZ space
-// and back to sRGB.
-// it's using Gamma setting too.
-// define USE_GAMMA is not required.
-#define USE_COLORIMETRY
-
-const float3x3 RGB_to_XYZ =
-mat3(
-	0.4306190, 0.3415419, 0.1783091,
-	0.2220379, 0.7066384, 0.0713236,
-	0.0201853, 0.1295504, 0.9390944
-);
-
-const float3x3 XYZ_to_sRGB =
-mat3(
-	 3.2406, -1.5372, -0.4986,
-	-0.9689,  1.8758,  0.0415,
-	 0.0557, -0.2040,  1.0570
-);
-
-// TWEAKS end
-
-const float YUV_u = 0.492;
-const float YUV_v = 0.877;
-
-const mat3 RGB_to_YUV =
-mat3(
-	float3( 0.299, 0.587, 0.114), //Y
-	float3(-0.299,-0.587, 0.886)*YUV_u, //B-Y
-	float3( 0.701,-0.587,-0.114)*YUV_v //R-Y
-);
-
-#ifdef USE_DELAY_LINE
-const float comb_line = 1.;
-#else
-const float comb_line = 2.;
-#endif
-
-static const float Voltage_0 = 0.518;
-static const float Voltage_1 = 1.962;
-static const float DeltaV = (Voltage_1-Voltage_0);
-
-float RGB_y = Contrast_static/Ywidth_static/DeltaV;
-float RGB_u = comb_line*Contrast_static*Saturation_static/YUV_u/Uwidth_static/DeltaV;
-float RGB_v = comb_line*Contrast_static*Saturation_static/YUV_v/Vwidth_static/DeltaV;
-
-mat3 YUV_to_RGB =
-mat3(
-	float3(1., 1., 1.)*RGB_y,
-	float3(0., -0.114/0.587, 1.)*RGB_u,
-	float3(1., -0.299/0.587, 0.)*RGB_v
-);
-
-const float pi = 3.1415926535897932384626433832795;
-
-float sinn(float x)
-{
-	return sin(/*fmod(x,24)*/x*(pi*2./24.));
-}
-
-float coss(float x)
-{
-	return cos(/*fmod(x,24)*/x*(pi*2./24.));
-}
-
-float3 monitor(sampler2D tex, float2 p)
-{
-   float2 size = sourceSize[0].xy;//float2(SizeX,SizeY);
-	// align vertical coord to center of texel
-	float2 uv = float2(
-
-		p.x,
-
-		(floor(p.y*sourceSize[0].y)+0.5)/sourceSize[0].y);
-#ifdef USE_DELAY_LINE
-	float2 sh = (sourceSize[0].xy/sourceSize[0].xy/size)*float2(14./10.,-1.0);
-#endif
-	float2 pc = uv*sourceSize[0].xy/sourceSize[0].xy*size*float2(10.,1.);
-	float alpha = dot(floor(float2(pc.x,pc.y)),float2(2.,Phase_Y*2.));
-	alpha += Phase_One*2.;
-
-	// 1/size.x of screen in uv coords = sourceSize[0].x/sourceSize[0].x/size.x;
-	// then 1/10*size.x of screen:
-	float ustep = sourceSize[0].x/sourceSize[0].x/size.x/10.;
-
-	float border = sourceSize[0].x/sourceSize[0].x;
-	float ss = 2.0;
-#ifdef SWAP_VSIGN
-#define PAL_SWITCH(A) A < 1.
-#else
-#define PAL_SWITCH(A) A > 1.
-#endif
-	if (PAL_SWITCH(fmod(uv.y*sourceSize[0].y/sourceSize[0].y*size.y,2.0)))
-	{
-		// cos(pi-alpha) = -cos(alpha)
-		// sin(pi-alpha) = sin(alpha)
-		// pi - alpha
-		alpha = -alpha+12012.0;
-		ss = -2.0;
-	}
-   
-	float ysum = 0., usum = 0., vsum = 0.;
-	for (int i=0; i<Mwidth; ++i)
-	{
-		float4 res = texture(tex, uv);
-      float3 yuv = mul(RGB_to_YUV, res.xyz);
-		float a1 = alpha+(HueShift+2.5)*2.-yuv.x*ss*HueRotation;
-		float sig = yuv.x+dot(yuv.yz,sign(float2(sinn(a1),coss(a1))));
-#ifdef USE_DELAY_LINE
-		float4 res1 = texture(tex, uv+sh);
-		float3 yuv1 = mul(RGB_to_YUV, res1.xyz);
-		float a2 = (HueShift+2.5)*2.+12012.0-alpha+yuv.x*ss*HueRotation;
-		float sig1 = yuv1.x+dot(yuv1.yz,sign(float2(sinn(a2),coss(a2))));
-#endif
-		if (i < Ywidth)
-			ysum += sig;
-
-#ifdef USE_DELAY_LINE
-		if (i < Uwidth)
-			usum += (sig+sig1)*sinn(alpha);
-		if (i < Vwidth)
-			vsum += (sig-sig1)*coss(alpha);
-#else
-		if (i < Uwidth)
-			usum += sig*sinn(alpha);
-		if (i < Vwidth)
-			vsum += sig*coss(alpha);
-#endif
-		alpha -= ss;
-		uv.x -= ustep;
-	}
-   
-	float3 rgb = mul(float3(ysum+Brightness*Ywidth_static,usum,vsum), YUV_to_RGB);
-#if defined(USE_GAMMA) && !defined(USE_COLORIMETRY)
-	float3 rgb1 = saturate(rgb);
-	rgb = pow(rgb1, Gamma/2.2);
-#endif
-
-#ifdef USE_COLORIMETRY
-	float3 rgb1 = saturate(rgb);
-	rgb = pow(rgb1, float3(Gamma, Gamma, Gamma));
-#endif
-
-#if (defined(USE_SCANLINES))
-	float2 q = (p*sourceSize[0].xy/sourceSize[0].xy)*float2(TV_Pixels*3.,size.y*2.);
-#endif
-
-#ifdef USE_SCANLINES
-	float scanlines = size.y * outputSize.z;
-	float top = fmod(q.y-0.5*scanlines*2.,2.);
-	float bottom = top+frac(scanlines)*2.;
-	float2 sw = saturate(min(float2(1.,2.),float2(bottom, bottom))
-		-max(float2(0.,1.),float2(top)))
-		+saturate(min(float2(3.,4.),float2(bottom, bottom))
-		-max(float2(2.,3.),float2(top)))
-		+floor(scanlines);
-
-	rgb = rgb*SCANLINE_MUL/(sw.x+sw.y);
-   
-   /*
-	//old stupid method
-	float z =
-#ifdef ANIMATE_SCANLINE
-	fmod(phase,2.0)+
-#endif
-		0.5;
-
-	if (abs(fmod(q.y+0.5,2)-z)<0.5)
-		rgb *= dark_scanline;
-*/
-#endif
-
-	// size of pixel screen in texture coords:
-	//float output_pixel_size = sourceSize[0].x/(outputSize.x*sourceSize[0].x);
-
-	// correctness check
-	//if (fmod(p.x*output_pixel_size,2.0) < 1.0)
-	//	rgb = float3(0.,0.,0.);
-
-#ifdef USE_COLORIMETRY
-	float3 xyz1 = mul(RGB_to_XYZ,rgb);
-	float3 srgb = saturate(mul(XYZ_to_sRGB,xyz1));
-	float3 a1 = 12.92*srgb;
-	float3 a2 = 1.055*pow(srgb,float3(1./2.4))-0.055;
-	float3 ssrgb;
-   ssrgb.x = (srgb.x<0.0031308?a1.x:a2.x);
-   ssrgb.y = (srgb.y<0.0031308?a1.y:a2.y);
-   ssrgb.z = (srgb.z<0.0031308?a1.z:a2.z);
-	return ssrgb;
-#else
-	return rgb;
-#endif
-}
-
-// pos (left corner, sample size)
-float4 monitor_sample(sampler2D tex, float2 p, float2 sample_)
-{
-	// linear interpolation was...
-	// now other thing.
-	// http://imgur.com/m8Z8trV
-	// AT LAST IT WORKS!!!!
-	// going to check in retroarch...
-	float2 size = sourceSize[0].xy;
-	float2 next = float2(.25,1.)/size;
-	float2 f = frac(float2(4.,1.)*size*p);
-	sample_ *= float2(4.,1.)*size;
-	float2 l;
-	float2 r;
-	if (f.x+sample_.x < 1.)
-	{
-		l.x = f.x+sample_.x;
-		r.x = 0.;
-	}
-	else
-	{
-		l.x = 1.-f.x;
-		r.x = min(1.,f.x+sample_.x-1.);
-	}
-	if (f.y+sample_.y < 1.)
-	{
-		l.y = f.y+sample_.y;
-		r.y = 0.;
-	}
-	else
-	{
-		l.y = 1.-f.y;
-		r.y = min(1.,f.y+sample_.y-1.);
-	}
-	float3 top = mix(monitor(tex, p), monitor(tex, p+float2(next.x,0.)), r.x/(l.x+r.x));
-	float3 bottom = mix(monitor(tex, p+float2(0.,next.y)), monitor(tex, p+next), r.x/(l.x+r.x));
-	return float4(mix(top,bottom, r.y/(l.y+r.y)),1.0);
-}
-
-void main() {
-#ifdef USE_SAMPLED
-	fragColor = vec4(monitor_sample(source[0], texCoord, outputSize.zw).rgb, 1.0);
-#else
-	fragColor = float4(monitor(source[0], texCoord), 1.);
-#endif
+#version 150
+
+// NES PAL composite signal simulation for RetroArch
+// shader by r57shell
+// thanks to feos & HardWareMan & NewRisingSun
+
+// also TV subpixels and scanlines
+
+// LICENSE: PUBLIC DOMAIN
+
+// NOTE: for nice TV subpixels and scanlines I recommend to
+// disable this features here and apply CRT-specialized shader.
+
+// Quality considerations
+
+// there are three main options:
+// USE_RAW (R), USE_DELAY_LINE (D), USE_COLORIMETRY (C)
+// here is table of quality in decreasing order:
+// RDC, RD, RC, DC, D, C
+
+// compatibility macros
+#define float2 vec2
+#define float3 vec3
+#define float4 vec4
+#define frac(c) fract(c)
+#define saturate(c) clamp(c, 0.0, 1.0)
+#define fmod(x,y) mod(x,y)
+#define mul(x,y) (y*x)
+#define float2x2 mat2
+#define float3x3 mat3
+#define float4x4 mat4
+#define bool2 bvec2
+#define bool3 bvec3
+#define bool4 bvec4
+#define static 
+
+uniform sampler2D source[];
+uniform vec4 sourceSize[];
+uniform vec4 outputSize;
+uniform int phase;
+
+in Vertex {
+   vec2 texCoord;
+};
+
+out vec4 fragColor;
+
+// TWEAKS start
+
+// use delay line technique
+// without delay line technique, color would interleave
+// to avoid this, set HueRotation to zero.
+#define USE_DELAY_LINE
+
+// use this if you need to swap even/odd V sign.
+// sign of V changes each scanline
+// so if some scanline is positive, then next is negative
+// and if you want to match picture
+// to actual running PAL NES on TV
+// you may want to have this option, to change signs
+// if they don't match
+//#define SWAP_VSIGN
+
+// rough simulation of scanlines
+// better if you use additional shader instead
+// if you still use it, make sure that SizeY
+// is at least twice lower than output height
+#define USE_SCANLINES // FIXME: scanlines are broken and I'm too lazy to fix it right now
+
+// to change gamma of virtual TV from 2.2 to something else
+#define USE_GAMMA
+
+// use sampled version. it's much more slower version of shader.
+// because it is computing x4 more values. NOT RECOMMENDED.
+//#define USE_SAMPLED
+
+#ifndef PARAMETER_UNIFORM
+// NTSC standard gamma = 2.2
+// PAL standard gamma = 2.8
+// according to many sources, very unlikely gamma of TV is 2.8
+// most likely gamma of PAL TV is in range 2.4-2.5
+const float Gamma_static = 2.8; // gamma of virtual TV
+
+const float Brightness_static = 0.25;
+const float Contrast_static = 1.1;
+const float Saturation_static = 1.1;
+
+const int
+   Ywidth_static = 12,
+   Uwidth_static = 23,
+   Vwidth_static = 23;
+   
+#define Brightness Brightness_static
+#define Gamma Gamma_static
+
+#define Ywidth Ywidth_static
+#define Uwidth Uwidth_static
+#define Vwidth Vwidth_static
+
+#define SCANLINE_MUL (sw.x*dark_scanline+sw.y)
+
+int Mwidth = int(max(float(Ywidth), max(float(Uwidth), float(Vwidth))));
+
+// correct one is -2.5
+// works only with USE_RAW
+const float HueShift = -2.5;
+
+// rotation of hue due to luma level.
+const float HueRotation = 2.;
+
+// touch this only if you know what you doing
+const float Phase_Y = 2.; // fmod(341*10,12)
+const float Phase_One = 0.; // alternating phases.
+const float Phase_Two = 8.;
+
+// screen size, scanlines = y*2; y one field, and y other field.
+const int SizeX = 256;
+const int SizeY = 240;
+
+// count of pixels of virtual TV.
+// value close to 1000 produce small artifacts
+const int TV_Pixels = 400;
+
+const float dark_scanline = 0.5; // half
+#endif
+
+// this is using following matrixes.
+// it provides more scientific approach
+// by conversion into linear XYZ space
+// and back to sRGB.
+// it's using Gamma setting too.
+// define USE_GAMMA is not required.
+#define USE_COLORIMETRY
+
+const float3x3 RGB_to_XYZ =
+mat3(
+   0.4306190, 0.3415419, 0.1783091,
+   0.2220379, 0.7066384, 0.0713236,
+   0.0201853, 0.1295504, 0.9390944
+);
+
+const float3x3 XYZ_to_sRGB =
+mat3(
+    3.2406, -1.5372, -0.4986,
+   -0.9689,  1.8758,  0.0415,
+    0.0557, -0.2040,  1.0570
+);
+
+// TWEAKS end
+
+const float YUV_u = 0.492;
+const float YUV_v = 0.877;
+
+const mat3 RGB_to_YUV =
+mat3(
+   float3( 0.299, 0.587, 0.114), //Y
+   float3(-0.299,-0.587, 0.886)*YUV_u, //B-Y
+   float3( 0.701,-0.587,-0.114)*YUV_v //R-Y
+);
+
+#ifdef USE_DELAY_LINE
+const float comb_line = 1.;
+#else
+const float comb_line = 2.;
+#endif
+
+static const float Voltage_0 = 0.518;
+static const float Voltage_1 = 1.962;
+static const float DeltaV = (Voltage_1-Voltage_0);
+
+float RGB_y = Contrast_static/Ywidth_static/DeltaV;
+float RGB_u = comb_line*Contrast_static*Saturation_static/YUV_u/Uwidth_static/DeltaV;
+float RGB_v = comb_line*Contrast_static*Saturation_static/YUV_v/Vwidth_static/DeltaV;
+
+mat3 YUV_to_RGB =
+mat3(
+   float3(1., 1., 1.)*RGB_y,
+   float3(0., -0.114/0.587, 1.)*RGB_u,
+   float3(1., -0.299/0.587, 0.)*RGB_v
+);
+
+const float pi = 3.1415926535897932384626433832795;
+
+float sinn(float x)
+{
+   return sin(/*fmod(x,24)*/x*(pi*2./24.));
+}
+
+float coss(float x)
+{
+   return cos(/*fmod(x,24)*/x*(pi*2./24.));
+}
+
+float3 monitor(sampler2D tex, float2 p)
+{
+   float2 size = sourceSize[0].xy;//float2(SizeX,SizeY);
+   // align vertical coord to center of texel
+   float2 uv = float2(
+
+      p.x,
+
+      (floor(p.y*sourceSize[0].y)+0.5)/sourceSize[0].y);
+#ifdef USE_DELAY_LINE
+   float2 sh = (sourceSize[0].xy/sourceSize[0].xy/size)*float2(14./10.,-1.0);
+#endif
+   float2 pc = uv*sourceSize[0].xy/sourceSize[0].xy*size*float2(10.,1.);
+   float alpha = dot(floor(float2(pc.x,pc.y)),float2(2.,Phase_Y*2.));
+   alpha += Phase_One*2.;
+
+   // 1/size.x of screen in uv coords = sourceSize[0].x/sourceSize[0].x/size.x;
+   // then 1/10*size.x of screen:
+   float ustep = sourceSize[0].x/sourceSize[0].x/size.x/10.;
+
+   float border = sourceSize[0].x/sourceSize[0].x;
+   float ss = 2.0;
+#ifdef SWAP_VSIGN
+#define PAL_SWITCH(A) A < 1.
+#else
+#define PAL_SWITCH(A) A > 1.
+#endif
+   if (PAL_SWITCH(fmod(uv.y*sourceSize[0].y/sourceSize[0].y*size.y,2.0)))
+   {
+      // cos(pi-alpha) = -cos(alpha)
+      // sin(pi-alpha) = sin(alpha)
+      // pi - alpha
+      alpha = -alpha+12012.0;
+      ss = -2.0;
+   }
+   
+   float ysum = 0., usum = 0., vsum = 0.;
+   for (int i=0; i<Mwidth; ++i)
+   {
+      float4 res = texture(tex, uv);
+      float3 yuv = mul(RGB_to_YUV, res.xyz);
+      float a1 = alpha+(HueShift+2.5)*2.-yuv.x*ss*HueRotation;
+      float sig = yuv.x+dot(yuv.yz,sign(float2(sinn(a1),coss(a1))));
+#ifdef USE_DELAY_LINE
+      float4 res1 = texture(tex, uv+sh);
+      float3 yuv1 = mul(RGB_to_YUV, res1.xyz);
+      float a2 = (HueShift+2.5)*2.+12012.0-alpha+yuv.x*ss*HueRotation;
+      float sig1 = yuv1.x+dot(yuv1.yz,sign(float2(sinn(a2),coss(a2))));
+#endif
+      if (i < Ywidth)
+         ysum += sig;
+
+#ifdef USE_DELAY_LINE
+      if (i < Uwidth)
+         usum += (sig+sig1)*sinn(alpha);
+      if (i < Vwidth)
+         vsum += (sig-sig1)*coss(alpha);
+#else
+      if (i < Uwidth)
+         usum += sig*sinn(alpha);
+      if (i < Vwidth)
+         vsum += sig*coss(alpha);
+#endif
+      alpha -= ss;
+      uv.x -= ustep;
+   }
+   
+   float3 rgb = mul(float3(ysum+Brightness*Ywidth_static,usum,vsum), YUV_to_RGB);
+#if defined(USE_GAMMA) && !defined(USE_COLORIMETRY)
+   float3 rgb1 = saturate(rgb);
+   rgb = pow(rgb1, Gamma/2.2);
+#endif
+
+#ifdef USE_COLORIMETRY
+   float3 rgb1 = saturate(rgb);
+   rgb = pow(rgb1, float3(Gamma, Gamma, Gamma));
+#endif
+
+#if (defined(USE_SCANLINES))
+   float2 q = (p*sourceSize[0].xy/sourceSize[0].xy)*float2(TV_Pixels*3.,size.y*2.);
+#endif
+
+#ifdef USE_SCANLINES
+   float scanlines = size.y * outputSize.z;
+   float top = fmod(q.y-0.5*scanlines*2.,2.);
+   float bottom = top+frac(scanlines)*2.;
+   float2 sw = saturate(min(float2(1.,2.),float2(bottom, bottom))
+      -max(float2(0.,1.),float2(top)))
+      +saturate(min(float2(3.,4.),float2(bottom, bottom))
+      -max(float2(2.,3.),float2(top)))
+      +floor(scanlines);
+
+   rgb = rgb*SCANLINE_MUL/(sw.x+sw.y);
+   
+   /*
+   //old stupid method
+   float z =
+#ifdef ANIMATE_SCANLINE
+   fmod(phase,2.0)+
+#endif
+      0.5;
+
+   if (abs(fmod(q.y+0.5,2)-z)<0.5)
+      rgb *= dark_scanline;
+*/
+#endif
+
+   // size of pixel screen in texture coords:
+   //float output_pixel_size = sourceSize[0].x/(outputSize.x*sourceSize[0].x);
+
+   // correctness check
+   //if (fmod(p.x*output_pixel_size,2.0) < 1.0)
+   //   rgb = float3(0.,0.,0.);
+
+#ifdef USE_COLORIMETRY
+   float3 xyz1 = mul(RGB_to_XYZ,rgb);
+   float3 srgb = saturate(mul(XYZ_to_sRGB,xyz1));
+   float3 a1 = 12.92*srgb;
+   float3 a2 = 1.055*pow(srgb,float3(1./2.4))-0.055;
+   float3 ssrgb;
+   ssrgb.x = (srgb.x<0.0031308?a1.x:a2.x);
+   ssrgb.y = (srgb.y<0.0031308?a1.y:a2.y);
+   ssrgb.z = (srgb.z<0.0031308?a1.z:a2.z);
+   return ssrgb;
+#else
+   return rgb;
+#endif
+}
+
+// pos (left corner, sample size)
+float4 monitor_sample(sampler2D tex, float2 p, float2 sample_)
+{
+   // linear interpolation was...
+   // now other thing.
+   // http://imgur.com/m8Z8trV
+   // AT LAST IT WORKS!!!!
+   // going to check in retroarch...
+   float2 size = sourceSize[0].xy;
+   float2 next = float2(.25,1.)/size;
+   float2 f = frac(float2(4.,1.)*size*p);
+   sample_ *= float2(4.,1.)*size;
+   float2 l;
+   float2 r;
+   if (f.x+sample_.x < 1.)
+   {
+      l.x = f.x+sample_.x;
+      r.x = 0.;
+   }
+   else
+   {
+      l.x = 1.-f.x;
+      r.x = min(1.,f.x+sample_.x-1.);
+   }
+   if (f.y+sample_.y < 1.)
+   {
+      l.y = f.y+sample_.y;
+      r.y = 0.;
+   }
+   else
+   {
+      l.y = 1.-f.y;
+      r.y = min(1.,f.y+sample_.y-1.);
+   }
+   float3 top = mix(monitor(tex, p), monitor(tex, p+float2(next.x,0.)), r.x/(l.x+r.x));
+   float3 bottom = mix(monitor(tex, p+float2(0.,next.y)), monitor(tex, p+next), r.x/(l.x+r.x));
+   return float4(mix(top,bottom, r.y/(l.y+r.y)),1.0);
+}
+
+void main() {
+#ifdef USE_SAMPLED
+   fragColor = vec4(monitor_sample(source[0], texCoord, outputSize.zw).rgb, 1.0);
+#else
+   fragColor = float4(monitor(source[0], texCoord), 1.);
+#endif
 }

ares/a26/GNUmakefile

@@ -0,0 +1,15 @@
+ares.components += mos6502
+
+ares.objects += ares-a26-system
+ares.objects += ares-a26-controller
+ares.objects += ares-a26-cartridge
+ares.objects += ares-a26-cpu
+ares.objects += ares-a26-tia
+ares.objects += ares-a26-riot
+
+$(object.path)/ares-a26-system.o:     	$(ares.path)/a26/system/system.cpp
+$(object.path)/ares-a26-controller.o: 	$(ares.path)/a26/controller/controller.cpp
+$(object.path)/ares-a26-cartridge.o:  	$(ares.path)/a26/cartridge/cartridge.cpp
+$(object.path)/ares-a26-cpu.o:     		$(ares.path)/a26/cpu/cpu.cpp
+$(object.path)/ares-a26-tia.o:     		$(ares.path)/a26/tia/tia.cpp
+$(object.path)/ares-a26-riot.o:     	$(ares.path)/a26/riot/riot.cpp

ares/a26/a26.hpp

@@ -0,0 +1,25 @@
+#pragma once
+
+#include <ares/ares.hpp>
+
+#include <component/processor/mos6502/mos6502.hpp>
+
+namespace ares::Atari2600 {
+  #include <ares/inline.hpp>
+  auto enumerate() -> vector<string>;
+  auto load(Node::System& node, string name) -> bool;
+
+  struct Region {
+    static inline auto NTSC() -> bool;
+    static inline auto PAL() -> bool;
+    static inline auto SECAM() -> bool;
+  };
+
+  #include <a26/controller/controller.hpp>
+  #include <a26/system/system.hpp>
+  #include <a26/cartridge/cartridge.hpp>
+  #include <a26/cpu/cpu.hpp>
+  #include <a26/tia/tia.hpp>
+  #include <a26/riot/riot.hpp>
+}
+

ares/a26/cartridge/board/atari16k.cpp

@@ -0,0 +1,58 @@
+struct Atari16k : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  Memory::Writable<n8> ram;
+  n2 bank;
+  bool hasRam;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address == 0x1ff6) bank = 0;
+    if(address == 0x1ff7) bank = 1;
+    if(address == 0x1ff8) bank = 2;
+    if(address == 0x1ff9) bank = 3;
+
+    if(address.bit(12)) {
+      if(hasRam && address >= 0x1080 && address <= 0x10ff) return ram.read(address & 0x7f);
+      return rom.read((bank * 0x1000) + (address & 0xfff));
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address == 0x1ff6) bank = 0;
+    if(address == 0x1ff7) bank = 1;
+    if(address == 0x1ff8) bank = 2;
+    if(address == 0x1ff9) bank = 3;
+
+    if(address >= 0x1000 && address <= 0x107f) {
+      hasRam = true;
+      ram.write(address & 0x7f, data);
+      return true;
+    }
+
+    return false;
+  }
+
+  auto power() -> void override {
+    bank = 0;
+    hasRam = 0;
+    ram.allocate(128);
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(bank);
+    s(hasRam);
+    s(ram);
+  }
+};

ares/a26/cartridge/board/atari32k.cpp

@@ -0,0 +1,66 @@
+struct Atari32k : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  Memory::Writable<n8> ram;
+  n3 bank;
+  bool hasRam;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address == 0x1ff4) bank = 0;
+    if(address == 0x1ff5) bank = 1;
+    if(address == 0x1ff6) bank = 2;
+    if(address == 0x1ff7) bank = 3;
+    if(address == 0x1ff8) bank = 4;
+    if(address == 0x1ff9) bank = 5;
+    if(address == 0x1ffa) bank = 6;
+    if(address == 0x1ffb) bank = 7;
+
+    if(address.bit(12)) {
+      if(hasRam && address >= 0x1080 && address <= 0x10ff) return ram.read(address & 0x7f);
+      return rom.read((bank * 0x1000) + (address & 0xfff));
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address == 0x1ff4) bank = 0;
+    if(address == 0x1ff5) bank = 1;
+    if(address == 0x1ff6) bank = 2;
+    if(address == 0x1ff7) bank = 3;
+    if(address == 0x1ff8) bank = 4;
+    if(address == 0x1ff9) bank = 5;
+    if(address == 0x1ffa) bank = 6;
+    if(address == 0x1ffb) bank = 7;
+
+    if(address >= 0x1000 && address <= 0x107f) {
+      hasRam = true;
+      ram.write(address & 0x7f, data);
+      return true;
+    }
+
+    return false;
+  }
+
+  auto power() -> void override {
+    bank = 0;
+    hasRam = 0;
+    ram.allocate(128);
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(bank);
+    s(hasRam);
+    s(ram);
+  }
+};

ares/a26/cartridge/board/atari8k.cpp

@@ -0,0 +1,54 @@
+struct Atari8k : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  Memory::Writable<n8> ram;
+  n1 bank;
+  bool hasRam;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address == 0x1ff8) bank = 0;
+    if(address == 0x1ff9) bank = 1;
+
+    if(address.bit(12)) {
+      if(hasRam && address >= 0x1080 && address <= 0x10ff) return ram.read(address & 0x7f);
+      return rom.read((bank * 0x1000) + (address & 0xfff));
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address == 0x1ff8) bank = 0;
+    if(address == 0x1ff9) bank = 1;
+
+    if(address >= 0x1000 && address <= 0x107f) {
+      hasRam = true;
+      ram.write(address & 0x7f, data);
+      return true;
+    }
+
+    return false;
+  }
+
+  auto power() -> void override {
+    bank = 0;
+    hasRam = 0;
+    ram.allocate(128);
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(bank);
+    s(hasRam);
+    s(ram);
+  }
+};

ares/a26/cartridge/board/board.cpp

@@ -0,0 +1,37 @@
+namespace Board {
+
+#include "linear.cpp"
+#include "atari8k.cpp"
+#include "atari16k.cpp"
+#include "atari32k.cpp"
+#include "commavid.cpp"
+#include "parker-bros.cpp"
+#include "tigervision.cpp"
+
+auto Interface::load(Memory::Readable<n8>& memory, string name) -> bool {
+  if(auto fp = pak->read(name)) {
+    memory.allocate(fp->size());
+    memory.load(fp);
+    return true;
+  }
+  return false;
+}
+
+auto Interface::load(Memory::Writable<n8>& memory, string name) -> bool {
+  if(auto fp = pak->read(name)) {
+    memory.allocate(fp->size());
+    memory.load(fp);
+    return true;
+  }
+  return false;
+}
+
+auto Interface::save(Memory::Writable<n8>& memory, string name) -> bool {
+  if(auto fp = pak->write(name)) {
+    memory.save(fp);
+    return true;
+  }
+  return false;
+}
+
+}

ares/a26/cartridge/board/board.hpp

@@ -0,0 +1,23 @@
+namespace Board {
+
+struct Interface {
+  VFS::Pak pak;
+
+  Interface(Cartridge& cartridge) : cartridge(cartridge) {}
+  virtual ~Interface() = default;
+  virtual auto load() -> void {}
+  virtual auto save() -> void {}
+  virtual auto unload() -> void {}
+  virtual auto read(n16 address) -> n8 { return 0xff; }
+  virtual auto write(n16 address, n8 data) -> bool { return false; }
+  virtual auto power() -> void {}
+  virtual auto serialize(serializer&) -> void {}
+
+  auto load(Memory::Readable<n8>&, string name) -> bool;
+  auto load(Memory::Writable<n8>&, string name) -> bool;
+  auto save(Memory::Writable<n8>&, string name) -> bool;
+
+  Cartridge& cartridge;
+};
+
+}

ares/a26/cartridge/board/commavid.cpp

@@ -0,0 +1,41 @@
+struct Commavid : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  Memory::Writable<n8> ram;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address.bit(12)) {
+      if(address >= 0x1000 && address <= 0x13ff) return ram.read(address & 0x3ff);
+      return rom.read(address & 0xfff);
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address >= 0x1400 && address <= 0x17ff) {
+      ram.write(address & 0x3ff, data);
+      return true;
+    }
+
+    return false;
+  }
+
+  auto power() -> void override {
+    ram.allocate(1_KiB);
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(ram);
+  }
+};

ares/a26/cartridge/board/linear.cpp

@@ -0,0 +1,45 @@
+struct Linear : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  Memory::Writable<n8> ram;
+  bool hasRam;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address.bit(12)) {
+      if(hasRam && address >= 0x1080 && address <= 0x10ff) return ram.read(address & 0x7f);
+      return rom.read(address & 0xfff);
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address >= 0x1000 && address <= 0x107f) {
+      hasRam = true;
+      ram.write(address & 0x7f, data);
+      return true;
+    }
+
+    return false;
+  }
+
+  auto power() -> void override {
+    hasRam = 0;
+    ram.allocate(128);
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(hasRam);
+    s(ram);
+  }
+};

ares/a26/cartridge/board/parker-bros.cpp

@@ -0,0 +1,51 @@
+struct ParkerBros : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  n3 banks[3];
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    bankswitch(address);
+
+    if(address.bit(12)) {
+      if(address >= 0x1000 && address <= 0x13ff) return rom.read((banks[0] * 0x400) + (address & 0x3ff));
+      if(address >= 0x1400 && address <= 0x17ff) return rom.read((banks[1] * 0x400) + (address & 0x3ff));
+      if(address >= 0x1800 && address <= 0x1bff) return rom.read((banks[2] * 0x400) + (address & 0x3ff));
+      if(address >= 0x1c00 && address <= 0x1fff) return rom.read((      7  * 0x400) + (address & 0x3ff));
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    bankswitch(address);
+    return false;
+  }
+
+  auto bankswitch(n16 address) -> void {
+    if(address >= 0x1fe0 && address <= 0x1fe7) banks[0] = address - 0x1fe0;
+    if(address >= 0x1fe8 && address <= 0x1fef) banks[1] = address - 0x1fe8;
+    if(address >= 0x1ff0 && address <= 0x1ff7) banks[2] = address - 0x1ff0;
+  }
+
+  auto power() -> void override {
+    banks[0] = 0;
+    banks[1] = 1;
+    banks[2] = 2;
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(banks[0]);
+    s(banks[1]);
+    s(banks[2]);
+  }
+};

ares/a26/cartridge/board/tigervision.cpp

@@ -0,0 +1,42 @@
+struct Tigervision : Interface {
+  using Interface::Interface;
+  Memory::Readable<n8> rom;
+  n8 bank;
+  n8 nextBank;
+
+  auto load() -> void override {
+    Interface::load(rom, "program.rom");
+  }
+
+  auto save() -> void override {
+  }
+
+  auto unload() -> void override {
+  }
+
+  auto read(n16 address) -> n8 override {
+    if(address.bit(12)) {
+      if(bank != nextBank) bank = nextBank;
+      if(address <= 0x17ff) return rom.read((bank * 0x800) + (address & 0x7ff));
+      return rom.read((rom.size() - 0x800) + (address & 0x7ff));
+    }
+
+    return 0xff;
+  }
+   
+  auto write(n16 address, n8 data) -> bool override {
+    if(address.bit(12) && bank != nextBank) bank = nextBank;
+    if(address <= 0x3f) nextBank = data;
+    return false;
+  }
+
+  auto power() -> void override {
+    bank = 0;
+    nextBank = 0;
+  }
+
+  auto serialize(serializer& s) -> void override {
+    s(bank);
+    s(nextBank);
+  }
+};

ares/a26/cartridge/cartridge.cpp

@@ -0,0 +1,66 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+Cartridge& cartridge = cartridgeSlot.cartridge;
+#include "board/board.cpp"
+#include "slot.cpp"
+#include "serialization.cpp"
+
+auto Cartridge::allocate(Node::Port parent) -> Node::Peripheral {
+  return node = parent->append<Node::Peripheral>(string{system.name(), " Cartridge"});
+}
+
+auto Cartridge::connect() -> void {
+  if(!node->setPak(pak = platform->pak(node))) return;
+
+  information = {};
+  information.title  = pak->attribute("title");
+  information.region = pak->attribute("region");
+  information.board  = pak->attribute("board");
+
+  if(information.board == "Linear") board = new Board::Linear{*this};
+  if(information.board == "Atari8k") board = new Board::Atari8k{*this};
+  if(information.board == "Atari16k") board = new Board::Atari16k{*this};
+  if(information.board == "Atari32k") board = new Board::Atari32k{*this};
+  if(information.board == "Commavid") board = new Board::Commavid{*this};
+  if(information.board == "ParkerBros8k") board = new Board::ParkerBros{*this};
+  if(information.board == "Tigervision") board = new Board::Tigervision{*this};
+
+  if(!board) board = new Board::Interface{*this};
+  board->pak = pak;
+  board->load();
+  power();
+}
+
+auto Cartridge::disconnect() -> void {
+  if(!node || !board) return;
+  board->unload();
+  board->pak.reset();
+  board.reset();
+  node.reset();
+}
+
+auto Cartridge::save() -> void {
+  if(!node) return;
+  if(board) board->save();
+}
+
+auto Cartridge::power() -> void {
+  if(board) board->power();
+}
+
+auto Cartridge::read(n16 address) -> n8 {
+  if(!node) return 0xff;
+  if(board) return board->read(address);
+
+  return 0xff;
+}
+
+auto Cartridge::write(n16 address, n8 data) -> bool {
+  if(!node) return false;
+  if(board) return board->write(address, data);
+  return false;
+}
+
+}

ares/a26/cartridge/cartridge.hpp

@@ -0,0 +1,36 @@
+struct Cartridge;
+#include "board/board.hpp"
+
+struct Cartridge {
+  Node::Peripheral node;
+  VFS::Pak pak;
+
+  auto title() const -> string { return information.title; }
+  auto region() const -> string { return information.region; }
+
+  //cartridge.cpp
+  auto allocate(Node::Port) -> Node::Peripheral;
+  auto connect() -> void;
+  auto disconnect() -> void;
+
+  auto save() -> void;
+  auto power() -> void;
+
+  auto read(n16 address) -> n8;
+  auto write(n16 address, n8 data) -> bool;
+
+  //serialization.cpp
+  auto serialize(serializer&) -> void;
+
+  unique_pointer<Board::Interface> board;
+
+//private:
+  struct Information {
+    string title;
+    string region;
+    string board;
+  } information;
+};
+
+#include "slot.hpp"
+extern Cartridge& cartridge;

ares/a26/cartridge/serialization.cpp

@@ -0,0 +1,3 @@
+auto Cartridge::serialize(serializer& s) -> void {
+  if(board) s(*board);
+}

ares/a26/cartridge/slot.cpp

@@ -0,0 +1,18 @@
+CartridgeSlot cartridgeSlot{"Cartridge Slot"};
+
+CartridgeSlot::CartridgeSlot(string name) : name(name) {
+}
+
+auto CartridgeSlot::load(Node::Object parent) -> void {
+  port = parent->append<Node::Port>(name);
+  port->setFamily(system.name());
+  port->setType("Cartridge");
+  port->setAllocate([&](auto name) { return cartridge.allocate(port); });
+  port->setConnect([&] { return cartridge.connect(); });
+  port->setDisconnect([&] { return cartridge.disconnect(); });
+}
+
+auto CartridgeSlot::unload() -> void {
+  cartridge.disconnect();
+  port = {};
+}

ares/a26/cartridge/slot.hpp

@@ -0,0 +1,13 @@
+struct CartridgeSlot {
+  Node::Port port;
+  Cartridge cartridge;
+
+  //slot.cpp
+  CartridgeSlot(string name);
+  auto load(Node::Object) -> void;
+  auto unload() -> void;
+
+  const string name;
+};
+
+extern CartridgeSlot cartridgeSlot;

ares/a26/controller/controller.cpp

@@ -0,0 +1,8 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+#include "port.cpp"
+#include "gamepad/gamepad.cpp"
+
+}

ares/a26/controller/controller.hpp

@@ -0,0 +1,11 @@
+struct Controller {
+  Node::Peripheral node;
+
+  virtual ~Controller() = default;
+
+  virtual auto read() -> n8 { return 0xff; }
+  virtual auto write(n8 data) -> void {}
+};
+
+#include "port.hpp"
+#include "gamepad/gamepad.hpp"

ares/a26/controller/gamepad/gamepad.cpp

@@ -0,0 +1,37 @@
+Gamepad::Gamepad(Node::Port parent) {
+  node = parent->append<Node::Peripheral>("Gamepad");
+
+  up    = node->append<Node::Input::Button>("Up");
+  down  = node->append<Node::Input::Button>("Down");
+  left  = node->append<Node::Input::Button>("Left");
+  right = node->append<Node::Input::Button>("Right");
+  fire  = node->append<Node::Input::Button>("Fire");
+}
+
+auto Gamepad::read() -> n8 {
+  platform->input(up);
+  platform->input(down);
+  platform->input(left);
+  platform->input(right);
+  platform->input(fire);
+
+  if(!(up->value() & down->value())) {
+    yHold = 0, upLatch = up->value(), downLatch = down->value();
+  } else if(!yHold) {
+    yHold = 1, swap(upLatch, downLatch);
+  }
+
+  if(!(left->value() & right->value())) {
+    xHold = 0, leftLatch = left->value(), rightLatch = right->value();
+  } else if(!xHold) {
+    xHold = 1, swap(leftLatch, rightLatch);
+  }
+
+  n8 data = 0xff;
+  data.bit(0) = !upLatch;
+  data.bit(1) = !downLatch;
+  data.bit(2) = !leftLatch;
+  data.bit(3) = !rightLatch;
+  data.bit(4) = !fire->value();
+  return data;
+}

ares/a26/controller/gamepad/gamepad.hpp

@@ -0,0 +1,19 @@
+struct Gamepad : Controller {
+  Node::Input::Button up;
+  Node::Input::Button down;
+  Node::Input::Button left;
+  Node::Input::Button right;
+  Node::Input::Button fire;
+
+  Gamepad(Node::Port);
+
+  auto read() -> n8 override;
+
+private:
+  b1 yHold;
+  b1 upLatch;
+  b1 downLatch;
+  b1 xHold;
+  b1 leftLatch;
+  b1 rightLatch;
+};

ares/a26/controller/port.cpp

@@ -0,0 +1,28 @@
+ControllerPort controllerPort1{"Controller Port 1"};
+ControllerPort controllerPort2{"Controller Port 2"};
+
+ControllerPort::ControllerPort(string name) : name(name) {
+}
+
+auto ControllerPort::load(Node::Object parent) -> void {
+  port = parent->append<Node::Port>(name);
+  port->setFamily(system.name());
+  port->setType("Controller");
+  port->setHotSwappable(true);
+  port->setAllocate([&](auto name) { return allocate(name); });
+  port->setSupported({"Gamepad"});
+}
+
+auto ControllerPort::unload() -> void {
+  device = {};
+  port = {};
+}
+
+auto ControllerPort::allocate(string name) -> Node::Peripheral {
+  if(name == "Gamepad") device = new Gamepad(port);
+  if(device) return device->node;
+  return {};
+}
+
+auto ControllerPort::serialize(serializer& s) -> void {
+}

ares/a26/controller/port.hpp

@@ -0,0 +1,20 @@
+struct ControllerPort {
+  Node::Port port;
+  unique_pointer<Controller> device;
+
+  //port.cpp
+  ControllerPort(string name);
+  auto load(Node::Object) -> void;
+  auto unload() -> void;
+  auto allocate(string name) -> Node::Peripheral;
+
+  auto read() -> n8 { if(device) return device->read(); return 0xff; }
+  auto write(n8 data) -> void { if(device) return device->write(data); }
+
+  auto serialize(serializer&) -> void;
+
+  const string name;
+};
+
+extern ControllerPort controllerPort1;
+extern ControllerPort controllerPort2;

ares/a26/cpu/cpu.cpp

@@ -0,0 +1,42 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+CPU cpu;
+#include "memory.cpp"
+#include "timing.cpp"
+#include "debugger.cpp"
+#include "serialization.cpp"
+
+auto CPU::load(Node::Object parent) -> void {
+  node = parent->append<Node::Object>("CPU");
+  debugger.load(node);
+}
+
+auto CPU::unload() -> void {
+  debugger = {};
+  node = {};
+}
+
+auto CPU::main() -> void {
+  debugger.instruction();
+  instruction();
+}
+
+auto CPU::step(u32 clocks) -> void {
+  Thread::step(clocks);
+  Thread::synchronize();
+}
+
+auto CPU::power(bool reset) -> void {
+  MOS6502::BCD = 1;
+  MOS6502::power();
+  Thread::create(system.frequency() / 3, {&CPU::main, this});
+
+  PC.byte(0) = readBus(0xfffc);
+  PC.byte(1) = readBus(0xfffd);
+
+  io = {};
+}
+
+}

ares/a26/cpu/cpu.hpp

@@ -0,0 +1,48 @@
+struct CPU : MOS6502, Thread {
+  Node::Object node;
+
+  struct Debugger {
+    //debugger.cpp
+    auto load(Node::Object) -> void;
+    auto instruction() -> void;
+
+    struct Tracer {
+      Node::Debugger::Tracer::Instruction instruction;
+    } tracer;
+  } debugger;
+
+  //cpu.cpp
+  auto load(Node::Object) -> void;
+  auto unload() -> void;
+
+  auto main() -> void;
+  auto step(u32 clocks) -> void;
+
+  auto power(bool reset) -> void;
+
+  //memory.cpp
+  auto readBus(n16 address) -> n8;
+  auto writeBus(n16 address, n8 data) -> void;
+
+  auto readIO(n16 address) -> n8;
+  auto writeIO(n16 address, n8 data) -> void;
+
+  auto readDebugger(n16 address) -> n8 override;
+
+  auto serialize(serializer&) -> void;
+
+  //timing.cpp
+  auto read(n16 address) -> n8 override;
+  auto write(n16 address, n8 data) -> void override;
+  auto lastCycle() -> void override;
+  auto nmi(n16& vector) -> void override;
+
+  auto rdyLine(bool) -> void;
+
+//protected:
+  struct IO {
+    n1  rdyLine = 1;
+  } io;
+};
+
+extern CPU cpu;

ares/a26/cpu/debugger.cpp

@@ -0,0 +1,12 @@
+auto CPU::Debugger::load(Node::Object parent) -> void {
+  tracer.instruction = parent->append<Node::Debugger::Tracer::Instruction>("Instruction", "CPU");
+  tracer.instruction->setAddressBits(16);
+}
+
+auto CPU::Debugger::instruction() -> void {
+  if(tracer.instruction->enabled() && tracer.instruction->address(cpu.PC)) {
+    tracer.instruction->notify(cpu.disassembleInstruction(), cpu.disassembleContext());
+  }
+}
+
+

ares/a26/cpu/memory.cpp

@@ -0,0 +1,40 @@
+inline auto CPU::readBus(n16 address) -> n8 {
+  address &= 0x1fff;
+  n8 data = cartridge.read(address);
+
+  if(address.bit(12) == 0 && address.bit(7) == 0) {
+    return tia.read(address & 0xf);
+  }
+
+  if(address.bit(12) == 0 && address.bit(9) == 0 && address.bit(7) == 1) {
+    return riot.readRam(address & 0x7f);
+  }
+
+  if(address.bit(12) == 0 && address.bit(9) == 1 && address.bit(7) == 1) {
+    return riot.readIo((address - 0x80) & 0x1f);
+  }
+
+  return data;
+}
+
+inline auto CPU::writeBus(n16 address, n8 data) -> void {
+  address &= 0x1fff;
+  cartridge.write(address, data);
+
+  if(address.bit(12) == 0 && address.bit(7) == 0) {
+    return tia.write(address & 0x3f, data);
+  }
+
+ if(address.bit(12) == 0 && address.bit(9) == 0 && address.bit(7) == 1) {
+   return riot.writeRam(address & 0x7f, data);
+ }
+
+ if(address.bit(12) == 0 && address.bit(9) == 1 && address.bit(7) == 1) {
+   return riot.writeIo((address - 0x80) & 0x1f, data);
+ }
+
+}
+
+auto CPU::readDebugger(n16 address) -> n8 {
+  return readBus(address);
+}

ares/a26/cpu/serialization.cpp

@@ -0,0 +1,5 @@
+auto CPU::serialize(serializer& s) -> void {
+  MOS6502::serialize(s);
+  Thread::serialize(s);
+  s(io.rdyLine);
+}

ares/a26/cpu/timing.cpp

@@ -0,0 +1,30 @@
+auto CPU::read(n16 address) -> n8 {
+  if(io.rdyLine == 0) {
+    while (io.rdyLine == 0) {
+      step(1);
+    }
+
+    MDR = readBus(address);
+    return MDR;
+  }
+
+  step(1);
+  MDR = readBus(address);
+  return MDR;
+}
+
+auto CPU::write(n16 address, n8 data) -> void {
+  step(1);
+  writeBus(address, MDR = data);
+}
+
+auto CPU::lastCycle() -> void {
+}
+
+auto CPU::nmi(n16& vector) -> void {
+
+}
+
+auto CPU::rdyLine(bool line) -> void {
+  io.rdyLine = line;
+}

ares/a26/riot/io.cpp

@@ -0,0 +1,87 @@
+auto RIOT::readRam(n8 address) -> n8 {
+  return ram[address];
+}
+
+auto RIOT::writeRam(n8 address, n8 data) -> void {
+  ram[address] = data;
+}
+
+auto RIOT::readIo(n8 address) -> n8 {
+  switch(address) {
+  case 0x00: return readPortA();
+  case 0x01: return port[0].direction;
+  case 0x02: return readPortB();
+  case 0x03: return port[1].direction;
+  case 0x04: return timer.counter;
+  }
+
+  debug(unimplemented, "[RIOT] IO read ", hex(address));
+  return 0xff;
+}
+
+auto RIOT::writeIo(n8 address, n8 data) -> void {
+  switch(address) {
+  case 0x00: writePortA(data);         return;
+  case 0x01: port[0].direction = data; return;
+  case 0x02: writePortB(data);         return;
+  case 0x03: port[1].direction = data; return;
+  case 0x14: case 0x15: case 0x16: case 0x17: {
+    const n16 intervals[] = {1, 8, 64, 1024};
+    timer.counter = data;
+    timer.interval = intervals[address - 0x14];
+    timer.reload = timer.interval;
+    return; }
+  }
+
+  debug(unimplemented, "[RIOT] IO write ",hex(address), " = ", hex(data));
+}
+
+auto RIOT::readPortA() -> n8 {
+  n8 data;
+  // TODO: Handle when configured as outputs
+  data.bit(0, 3) = controllerPort2.read().bit(0, 4);
+  data.bit(4, 7) = controllerPort1.read().bit(0, 4);
+  return data;
+}
+
+auto RIOT::writePortA(n8 data) -> void {
+  port[0].data = data;
+}
+
+auto RIOT::readPortB() -> n8 {
+    system.controls.poll();
+
+    n8 data;
+
+    for(auto n : range(8)) if(port[1].direction.bit(n)) { data.bit(n) = port[1].data.bit(n); }
+
+    // Atari hard-wires these bits as input
+    data.bit(0) = !system.controls.reset->value();
+    data.bit(1) = !system.controls.select->value();
+
+    // These are toggle switches; flip the values on the rising edge
+    if(system.controls.tvType->value() && !tvTypeLatch) {
+      tvType = !tvType;
+    }
+    tvTypeLatch = system.controls.tvType->value();
+
+    if(system.controls.leftDifficulty->value()  && !leftDifficultyLatch) {
+      leftDifficulty = !leftDifficulty;
+    }
+    leftDifficultyLatch = system.controls.leftDifficulty->value();
+
+    if(system.controls.rightDifficulty->value() && !rightDifficultyLatch) {
+      rightDifficulty = !rightDifficulty;
+    }
+    rightDifficultyLatch = system.controls.rightDifficulty->value();
+
+    data.bit(3) = tvType;
+    data.bit(6) = leftDifficulty;
+    data.bit(7) = rightDifficulty;
+
+    return data;
+}
+
+auto RIOT::writePortB(n8 data) -> void {
+    port[1].data = data;
+}

ares/a26/riot/riot.cpp

@@ -0,0 +1,51 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+RIOT riot;
+
+#include "io.cpp"
+#include "serialization.cpp"
+
+auto RIOT::load(Node::Object parent) -> void {
+  node = parent->append<Node::Object>("RIOT");
+
+  ram.allocate(128);
+}
+
+auto RIOT::unload() -> void {
+  ram.reset();
+  node.reset();
+}
+
+auto RIOT::main() -> void {
+  decrementTimer();
+  step(1);
+}
+
+auto RIOT::decrementTimer() -> void {
+  if(--timer.interval == 0) {
+    if(--timer.counter == 0xff) {
+      timer.reload = 1;
+    }
+
+    timer.interval = timer.reload;
+  }
+}
+
+auto RIOT::step(u32 clocks) -> void {
+  Thread::step(clocks);
+  Thread::synchronize();
+}
+
+auto RIOT::power(bool reset) -> void {
+  Thread::create(system.frequency() / 3, {&RIOT::main, this});
+  timer = {};
+  port[0] = {.data = 0xff};
+  port[1] = {.data = 0xff};
+  leftDifficulty = 1;
+  rightDifficulty = 1;
+  tvType = 1;
+}
+
+}

ares/a26/riot/riot.hpp

@@ -0,0 +1,48 @@
+struct RIOT : Thread {
+  Node::Object node;
+  Memory::Writable<n8> ram;
+
+  //riot.cpp
+  auto load(Node::Object) -> void;
+  auto unload() -> void;
+
+  auto main() -> void;
+  auto step(u32 clocks) -> void;
+  auto power(bool reset) -> void;
+
+  auto decrementTimer() -> void;
+
+  //io.cpp
+  auto readRam(n8 address) -> n8;
+  auto writeRam(n8 address, n8 data) -> void;
+  auto readIo(n8 address) -> n8;
+  auto writeIo(n8 address, n8 data) -> void;
+
+  auto readPortA() -> n8;
+  auto readPortB() -> n8;
+  auto writePortA(n8 data) -> void;
+  auto writePortB(n8 data) -> void;
+
+  //serialization.cpp
+  auto serialize(serializer&) -> void;
+
+  struct {
+    u8 counter;
+    n16 interval;
+    n16 reload;
+  } timer;
+
+  struct Port {
+    n8 data;
+    n8 direction;
+  } port[2];
+
+  n1 leftDifficulty;
+  n1 leftDifficultyLatch;
+  n1 rightDifficulty;
+  n1 rightDifficultyLatch;
+  n1 tvType;
+  n1 tvTypeLatch;
+};
+
+extern RIOT riot;

ares/a26/riot/serialization.cpp

@@ -0,0 +1,21 @@
+//serialization.cpp
+auto RIOT::serialize(serializer& s) -> void {
+  Thread::serialize(s);
+
+  s(ram);
+  s(timer.counter);
+  s(timer.interval);
+  s(timer.reload);
+
+  for(auto n : range(2)) {
+    s(port[n].data);
+    s(port[n].direction);
+  }
+
+  s(leftDifficulty);
+  s(leftDifficultyLatch);
+  s(rightDifficulty);
+  s(rightDifficultyLatch);
+  s(tvType);
+  s(tvTypeLatch);
+}

ares/a26/system/controls.cpp

@@ -0,0 +1,17 @@
+auto System::Controls::load(Node::Object parent) -> void {
+  node = parent->append<Node::Object>("Controls");
+
+  reset             = node->append<Node::Input::Button>("Reset");
+  select            = node->append<Node::Input::Button>("Select");
+  leftDifficulty    = node->append<Node::Input::Button>("Left Difficulty");
+  rightDifficulty   = node->append<Node::Input::Button>("Right Difficulty");
+  tvType            = node->append<Node::Input::Button>("TV Type");
+}
+
+auto System::Controls::poll() -> void {
+  platform->input(reset);
+  platform->input(select);
+  platform->input(leftDifficulty);
+  platform->input(rightDifficulty);
+  platform->input(tvType);
+}

ares/a26/system/serialization.cpp

@@ -0,0 +1,46 @@
+auto System::serialize(bool synchronize) -> serializer {
+  if(synchronize) scheduler.enter(Scheduler::Mode::Synchronize);
+  serializer s;
+
+  u32  signature = SerializerSignature;
+  char version[16] = {};
+  char description[512] = {};
+  memory::copy(&version, (const char*)SerializerVersion, SerializerVersion.size());
+
+  s(signature);
+  s(synchronize);
+  s(version);
+  s(description);
+
+  serialize(s, synchronize);
+  return s;
+}
+
+auto System::unserialize(serializer& s) -> bool {
+  u32  signature = 0;
+  bool synchronize = true;
+  char version[16] = {};
+  char description[512] = {};
+
+  s(signature);
+  s(synchronize);
+  s(version);
+  s(description);
+
+  if(signature != SerializerSignature) return false;
+  if(string{version} != SerializerVersion) return false;
+
+  if(synchronize) power(/* reset = */ false);
+  serialize(s, synchronize);
+  return true;
+}
+
+auto System::serialize(serializer& s, bool synchronize) -> void {
+  scheduler.setSynchronize(synchronize);
+  s(cartridge);
+  s(cpu);
+  s(tia);
+  s(riot);
+  s(controllerPort1);
+  s(controllerPort2);
+}

ares/a26/system/system.cpp

@@ -0,0 +1,105 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+auto enumerate() -> vector<string> {
+  return {
+    "[Atari] Atari 2600 (NTSC)",
+    "[Atari] Atari 2600 (PAL)",
+    "[Atari] Atari 2600 (SECAM)",
+  };
+}
+
+auto load(Node::System& node, string name) -> bool {
+  if(!enumerate().find(name)) return false;
+  return system.load(node, name);
+}
+
+Random random;
+Scheduler scheduler;
+System system;
+#include "controls.cpp"
+#include "serialization.cpp"
+
+auto System::game() -> string {
+  if(cartridge.node) {
+    return cartridge.title();
+  }
+
+  return "(no cartridge connected)";
+}
+
+auto System::run() -> void {
+  scheduler.enter();
+}
+
+auto System::load(Node::System& root, string name) -> bool {
+  if(node) unload();
+
+  information = {};
+  if(name.find("NTSC")) {
+    information.name = "Atari 2600";
+    information.region = Region::NTSC;
+    information.frequency = 3579546;
+  }
+  if(name.find("PAL")) {
+    information.name = "Atari 2600";
+    information.region = Region::PAL;
+    information.frequency = 3546894;
+  }
+  if(name.find("SECAM")) {
+    information.name = "Atari 2600";
+    information.region = Region::SECAM;
+    information.frequency = 3546894;
+  }
+
+  node = Node::System::create(information.name);
+  node->setGame({&System::game, this});
+  node->setRun({&System::run, this});
+  node->setPower({&System::power, this});
+  node->setSave({&System::save, this});
+  node->setUnload({&System::unload, this});
+  node->setSerialize({&System::serialize, this});
+  node->setUnserialize({&System::unserialize, this});
+  root = node;
+
+  scheduler.reset();
+  controls.load(node);
+  riot.load(node);
+  cpu.load(node);
+  tia.load(node);
+  cartridgeSlot.load(node);
+  controllerPort1.load(node);
+  controllerPort2.load(node);
+  return true;
+}
+
+auto System::save() -> void {
+  if(!node) return;
+  cartridge.save();
+}
+
+auto System::unload() -> void {
+  if(!node) return;
+  save();
+  riot.unload();
+  cpu.unload();
+  tia.unload();
+  cartridgeSlot.unload();
+  controllerPort1.unload();
+  controllerPort2.unload();
+  node = {};
+}
+
+auto System::power(bool reset) -> void {
+  for(auto& setting : node->find<Node::Setting::Setting>()) setting->setLatch();
+
+  random.entropy(Random::Entropy::Low);
+  cartridge.power();
+  riot.power(reset);
+  cpu.power(reset);
+  tia.power(reset);
+  scheduler.power(cpu);
+}
+
+}

ares/a26/system/system.hpp

@@ -0,0 +1,53 @@
+extern Random random;
+
+struct System {
+  Node::System node;
+
+  struct Controls {
+    Node::Object node;
+    Node::Input::Button reset;
+    Node::Input::Button select;
+    Node::Input::Button leftDifficulty;
+    Node::Input::Button rightDifficulty;
+    Node::Input::Button tvType;
+
+    //controls.cpp
+    auto load(Node::Object) -> void;
+    auto poll() -> void;
+  } controls;
+
+  enum class Region : u32 { NTSC, PAL, SECAM };
+
+  auto name() const -> string { return information.name; }
+  auto region() const -> Region { return information.region; }
+  auto frequency() const -> u32 { return information.frequency; }
+
+  //system.cpp
+  auto game() -> string;
+  auto run() -> void;
+
+  auto load(Node::System& node, string name) -> bool;
+  auto save() -> void;
+  auto unload() -> void;
+  auto power(bool reset) -> void;
+
+  //serialization.cpp
+  auto serialize(bool synchronize) -> serializer;
+  auto unserialize(serializer&) -> bool;
+
+private:
+  struct Information {
+    string name = "Atari 2600";
+    Region region = Region::NTSC;
+    u32 frequency = Constants::Colorburst::NTSC;
+  } information;
+
+  //serialization.cpp
+  auto serialize(serializer&, bool synchronize) -> void;
+};
+
+extern System system;
+
+auto Region::NTSC()  -> bool { return system.region() == System::Region::NTSC; }
+auto Region::PAL()   -> bool { return system.region() == System::Region::PAL; }
+auto Region::SECAM() -> bool { return system.region() == System::Region::SECAM; }

ares/a26/tia/audio.cpp

@@ -0,0 +1,64 @@
+auto TIA::runAudio() -> void {
+  if(io.hcounter == 9 || io.hcounter == 81) {
+    audio[0].phase0();
+    audio[1].phase0();
+  }
+
+  if(io.hcounter == 37 || io.hcounter == 149) {
+    auto a = audio[0].phase1();
+    auto b = audio[1].phase1();
+
+    double output = 0;
+    output += volume[a];
+    output += volume[b];
+
+    stream->frame(output / 2.0);
+  }
+}
+
+auto TIA::AudioChannel::phase0() -> void {
+  if (enable) {
+    switch (control.bit(0, 1)) {
+    case 0x00: case 0x01: pulseCounterPaused = 0;                break;
+    case 0x02: pulseCounterPaused = noiseCounter.bit(1, 4) != 1; break;
+    case 0x03: pulseCounterPaused = !noiseCounter.bit(0);        break;
+    }
+
+    noiseFeedback = (noiseCounter.bit(2) ^ noiseCounter.bit(0)) || noiseCounter == 0;
+    if(control.bit(0, 2) == 0) {
+      noiseFeedback = !(control.bit(2, 3)) ||
+                      !(noiseCounter || (pulseCounter != 0x0a)) ||
+                      (pulseCounter.bit(0) ^ noiseCounter.bit(0));
+    }
+  }
+
+  enable = divCounter == frequency;
+  if (divCounter == frequency || divCounter == 0x1f) {
+    divCounter = 0;
+    return;
+  }
+
+  divCounter++;
+}
+
+auto TIA::AudioChannel::phase1() -> u8 {
+  if(!enable) return pulseCounter.bit(0) * volume;
+
+  pulseFeedback = 0;
+  switch(control.bit(2, 3)) {
+    case 0x00: pulseFeedback = pulseCounter != 0x0a && control.bit(0, 2) && (pulseCounter.bit(1) ^ pulseCounter.bit(0)); break;
+    case 0x01: pulseFeedback = !pulseCounter.bit(3);                                                                   ; break;
+    case 0x02: pulseFeedback = !noiseCounter.bit(0);                                                                   ; break;
+    case 0x03: pulseFeedback = !(pulseCounter.bit(1) || !pulseCounter.bit(1, 3));                                      ; break;
+  }
+
+  noiseCounter >>= 1;
+  if (noiseFeedback) noiseCounter.bit(4) = 1;
+
+  if(!pulseCounterPaused) {
+    pulseCounter = ~(pulseCounter >> 1) & 7;
+    if (pulseFeedback) pulseCounter.bit(3) = 1;
+  }
+
+  return pulseCounter.bit(0) * volume;
+}

ares/a26/tia/color.cpp

@@ -0,0 +1,42 @@
+auto TIA::color(n32 color) -> n64 {
+  u64 ntscColors[128] = {
+    0x0000'0000'0000ull, 0x4040'4040'4040ull, 0x6c6c'6c6c'6c6cull, 0x9090'9090'9090ull, 0xb0b0'b0b0'b0b0ull, 0xc8c8'c8c8'c8c8ull, 0xdcdc'dcdc'dcdcull, 0xecec'ecec'ececull,
+    0x4444'4444'0000ull, 0x6464'6464'1010ull, 0x8484'8484'2424ull, 0xa0a0'a0a0'3434ull, 0xb8b8'b8b8'4040ull, 0xd0d0'd0d0'5050ull, 0xe8e8'e8e8'5c5cull, 0xfcfc'fcfc'6868ull,
+    0x7070'2828'0000ull, 0x8484'4444'1414ull, 0x9898'5c5c'2828ull, 0xacac'7878'3c4cull, 0xbcbc'8c8c'4c4cull, 0xcccc'a0a0'5c5cull, 0xdcdc'b4b4'6868ull, 0xecec'c8c8'7878ull,
+    0x8484'1818'0000ull, 0x9898'3434'1818ull, 0xacac'5050'3030ull, 0xc0c0'6868'4848ull, 0xd0d0'8080'5c5cull, 0xe0e0'9494'7070ull, 0xecec'a8a8'8080ull, 0xfcfc'bcbc'9494ull,
+    0x8888'0000'0000ull, 0x9c9c'2020'2020ull, 0xb0b0'3c3c'3c3cull, 0xc0c0'5858'5858ull, 0xd0d0'7070'7070ull, 0xe0e0'8888'8888ull, 0xecec'a0a0'a0a0ull, 0xfcfc'b4b4'b4b4ull,
+    0x7878'0000'5c5cull, 0x8c8c'2020'7474ull, 0xa0a0'3c3c'8888ull, 0xb0b0'5858'9c9cull, 0xc0c0'7070'b0b0ull, 0xd0d0'8484'c0c0ull, 0xdcdc'9c9c'd0d0ull, 0xecec'b0b0'e0e0ull,
+    0x4848'0000'7878ull, 0x6060'2020'9090ull, 0x7878'3c3c'a4a4ull, 0x8c8c'5858'b8b8ull, 0xa0a0'7070'ccccull, 0xb4b4'8484'dcdcull, 0xc4c4'9c9c'ececull, 0xd4d4'b0b0'fcfcull,
+    0x1414'0000'8484ull, 0x3030'2020'9898ull, 0x4c4c'3c3c'acacull, 0x6868'5858'c0c0ull, 0x7c7c'7070'd0d0ull, 0x9494'8888'e0e0ull, 0xa8a8'a0a0'ececull, 0xbcbc'b4b4'fcfcull,
+    0x0000'0000'8888ull, 0x1c1c'2020'9c9cull, 0x3838'4040'b0b0ull, 0x5050'5c5c'c0c0ull, 0x6868'7474'd0d0ull, 0x7c7c'8c8c'e0e0ull, 0x9090'a4a4'ececull, 0xa4a4'b8b8'fcfcull,
+    0x0000'1818'7c7cull, 0x1c1c'3838'9090ull, 0x3838'5454'a8a8ull, 0x5050'7070'bcbcull, 0x6868'8888'ccccull, 0x7c7c'9c9c'dcdcull, 0x9090'b4b4'ececull, 0xa4a4'c8c8'fcfcull,
+    0x0000'2c2c'5c5cull, 0x1c1c'4c4c'7878ull, 0x3838'6868'9090ull, 0x5050'8484'acacull, 0x6868'9c9c'c0c0ull, 0x7c7c'b4b4'd4d4ull, 0x9090'cccc'e8e8ull, 0xa4a4'e0e0'fcfcull,
+    0x0000'3c3c'2c2cull, 0x1c1c'5c5c'4848ull, 0x3838'7c7c'6464ull, 0x5050'9c9c'8080ull, 0x6868'b4b4'9494ull, 0x7c7c'd0d0'acacull, 0x9090'e4e4'c0c0ull, 0xa4a4'fcfc'd4d4ull,
+    0x0000'3c3c'0000ull, 0x2020'5c5c'2020ull, 0x4040'7c7c'4040ull, 0x5c5c'9c9c'5c5cull, 0x7474'b4b4'7474ull, 0x8c8c'd0d0'8c8cull, 0xa4a4'e4e4'a4a4ull, 0xb8b8'fcfc'b8b8ull,
+    0x1414'3838'0000ull, 0x3434'5c5c'1c1cull, 0x5050'7c7c'3838ull, 0x6c6c'9898'5050ull, 0x8484'b4b4'6868ull, 0x9c9c'cccc'7c7cull, 0xb4b4'e4e4'9090ull, 0xc8c8'fcfc'a4a4ull,
+    0x2c2c'3030'0000ull, 0x4c4c'5050'1c1cull, 0x6868'7070'3434ull, 0x8484'8c8c'4c4cull, 0x9c9c'a8a8'6464ull, 0xb4b4'c0c0'7878ull, 0xcccc'd4d4'8888ull, 0xe0e0'ecec'9c9cull,
+    0x4444'2828'0000ull, 0x6464'4848'1818ull, 0x8484'6868'3030ull, 0xa0a0'8484'4444ull, 0xb8b8'9c9c'5858ull, 0xd0d0'b4b4'6c6cull, 0x8e8c'cccc'7c7cull, 0xfcfc'c0c0'8c8cull,
+  };
+
+  u64 palColors[128] = {
+    0x0000'0000'0000ull, 0x2828'2828'2828ull, 0x5050'5050'5050ull, 0x7474'7474'7474ull, 0x9494'9494'9494ull, 0xb4b4'b4b4'b4b4ull, 0xd0d0'd0d0'd0d0ull, 0xecec'ecec'ececull,
+    0x0000'0000'0000ull, 0x2828'2828'2828ull, 0x5050'5050'5050ull, 0x7474'7474'7474ull, 0x9494'9494'9494ull, 0xb4b4'b4b4'b4b4ull, 0xd0d0'd0d0'd0d0ull, 0xecec'ecec'ececull,
+    0x8080'5858'0000ull, 0x9494'7070'2020ull, 0xa8a8'8484'3c3cull, 0xbcbc'9c9c'5858ull, 0xcccc'acac'7070ull, 0xdcdc'c0c0'8484ull, 0xecec'd0d0'9c9cull, 0xfcfc'e0e0'b0b0ull,
+    0x4444'5c5c'0000ull, 0x5c5c'7878'2020ull, 0x7474'9090'3c3cull, 0x8c8c'acac'5858ull, 0xa0a0'c0c0'7070ull, 0xb0b0'd4d4'8484ull, 0xc4c4'e8e8'9c9cull, 0xd4d4'fcfc'b0b0ull,
+    0x7070'3434'0000ull, 0x8888'5050'2020ull, 0xa0a0'6868'3c3cull, 0xb4b4'8484'5858ull, 0xc8c8'9898'7070ull, 0xdcdc'acac'8484ull, 0xecec'c0c0'9c9cull, 0xfcfc'd4d4'b0b0ull,
+    0x0000'6464'1414ull, 0x2020'8080'3434ull, 0x3c3c'9898'5050ull, 0x5858'b0b0'6c6cull, 0x7070'c4c4'8484ull, 0x8484'd8d8'9c9cull, 0x9c9c'e8e8'b4b4ull, 0xb0b0'fcfc'c8c8ull,
+    0x7070'0000'1414ull, 0x8888'2020'3434ull, 0xa0a0'3c3c'5050ull, 0xb4b4'5858'6c6cull, 0xc8c8'7070'8484ull, 0xdcdc'8484'9c9cull, 0xecec'9c9c'b4b4ull, 0xfcfc'b0b0'c8c8ull,
+    0x0000'5c5c'5c5cull, 0x2020'7474'7474ull, 0x3c3c'8c8c'8c8cull, 0x5858'a4a4'a4a4ull, 0x7070'b8b8'b8b8ull, 0x8484'c8c8'c8c8ull, 0x9c9c'dcdc'dcdcull, 0xb0b0'ecec'ececull,
+    0x7070'0000'5c5cull, 0x8484'2020'7474ull, 0x9494'3c3c'8888ull, 0xa8a8'5858'9c9cull, 0xb4b4'7070'b0b0ull, 0xc4c4'8484'c0c0ull, 0xd0d0'9c9c'd0d0ull, 0xe0e0'b0b0'e0e0ull,
+    0x0000'3c3c'7070ull, 0x1c1c'5858'8888ull, 0x3838'7474'a0a0ull, 0x5050'8c8c'b4b4ull, 0x6868'a4a4'c8c8ull, 0x7c7c'b8b8'dcdcull, 0x9090'cccc'ececull, 0xa4a4'e0e0'fcfcull,
+    0x5858'0000'7070ull, 0x6c6c'2020'8888ull, 0x8080'3c3c'a0a0ull, 0x9494'5858'b4b4ull, 0xa4a4'7070'c8c8ull, 0xb4b4'8484'dcdcull, 0xc4c4'9c9c'ececull, 0xd4d4'b0b0'fcfcull,
+    0x0000'2020'7070ull, 0x1c1c'3c3c'8888ull, 0x3838'5858'a0a0ull, 0x5050'7474'b4b4ull, 0x6868'8888'c8c8ull, 0x7c7c'a0a0'dcdcull, 0x9090'b4b4'ececull, 0xa4a4'c8c8'fcfcull,
+    0x3c3c'0000'8080ull, 0x5454'2020'9494ull, 0x6c6c'3c3c'a8a8ull, 0x8080'5858'bcbcull, 0x9494'7070'ccccull, 0xa8a8'8484'dcdcull, 0xb8b8'9c9c'ececull, 0xc8c8'b0b0'fcfcull,
+    0x0000'0000'8888ull, 0x2020'2020'9c9cull, 0x3c3c'3c3c'b0b0ull, 0x5858'5858'c0c0ull, 0x7070'7070'd0d0ull, 0x8484'8484'e0e0ull, 0x9c9c'9c9c'ececull, 0xb0b0'b0b0'fcfcull,
+    0x0000'0000'0000ull, 0x2828'2828'2828ull, 0x5050'5050'5050ull, 0x7474'7474'7474ull, 0x9494'9494'9494ull, 0xb4b4'b4b4'b4b4ull, 0xd0d0'd0d0'd0d0ull, 0xecec'ecec'ececull,
+    0x0000'0000'0000ull, 0x2828'2828'2828ull, 0x5050'5050'5050ull, 0x7474'7474'7474ull, 0x9494'9494'9494ull, 0xb4b4'b4b4'b4b4ull, 0xd0d0'd0d0'd0d0ull, 0xecec'ecec'ececull,
+  };
+
+  return Region::PAL() ? palColors[color & 0x7f] : ntscColors[color & 0x7f];
+}
+

ares/a26/tia/io.cpp

@@ -0,0 +1,147 @@
+auto TIA::read(n8 address) -> n8 {
+  n8 value;
+
+  switch(address) {
+  case 0x00: value.bit(7) = collision.M0P1; value.bit(6) = collision.M0P0; return value; // CXM0P
+  case 0x01: value.bit(7) = collision.M1P0; value.bit(6) = collision.M1P1; return value; // CXM1P
+  case 0x02: value.bit(7) = collision.P0PF; value.bit(6) = collision.P0BL; return value; // CX0FB
+  case 0x03: value.bit(7) = collision.P1PF; value.bit(6) = collision.P1BL; return value; // CX1FB
+  case 0x04: value.bit(7) = collision.M0PF; value.bit(6) = collision.M0BL; return value; // CXM0FB
+  case 0x05: value.bit(7) = collision.M1PF; value.bit(6) = collision.M1BL; return value; // CXM1FB
+  case 0x06: value.bit(7) = collision.BLPF;                                return value; // CXBLPF
+  case 0x07: value.bit(7) = collision.P0P1; value.bit(6) = collision.M0M1; return value; // CXPPMM
+  case 0x0c: value.bit(7) = controllerPort1.read().bit(4);                 return value; // INPT4
+  case 0x0d: value.bit(7) = controllerPort2.read().bit(4);                 return value; // INPT5
+  }
+
+  debug(unimplemented, "[TIA] read ", hex(address));
+  return 0xff;
+}
+
+auto TIA::write(n8 address, n8 data) -> void {
+  switch(address) {
+  case 0x00: vsync(data.bit(1));                                               return; // VSYNC
+  case 0x01: writeQueue.add(address, data, 1);                                 return; // VBLANK
+  case 0x02: cpu.rdyLine(0);                                                   return; // WSYNC
+  case 0x03: io.hcounter = 0;                                                  return; // RSYNC
+  case 0x04: player[0].size = data.bit(0,2); missile[0].size = data.bit(4, 5); return; // NUSIZ1
+  case 0x05: player[1].size = data.bit(0,2); missile[1].size = data.bit(4, 5); return; // NUSIZ1
+  case 0x06: io.p0Color = data.bit(1, 7);                                      return; // COLUP0
+  case 0x07: io.p1Color = data.bit(1, 7);                                      return; // COLUP1
+  case 0x08: io.fgColor = data.bit(1, 7);                                      return; // COLUPF
+  case 0x09: io.bgColor = data.bit(1, 7);                                      return; // COLUBK
+  case 0x0a: ctrlpf(data);                                                     return; // CTRLPF
+  case 0x0b: writeQueue.add(address, data, 1);                                 return; // REFP0
+  case 0x0c: writeQueue.add(address, data, 1);                                 return; // REFP1
+  case 0x0d: writeQueue.add(address, data, 6);                                 return; // PF0
+  case 0x0e: writeQueue.add(address, data, 6);                                 return; // PF1
+  case 0x0f: writeQueue.add(address, data, 6);                                 return; // PF2
+  case 0x10: resp(0);                                                          return; // RESP0
+  case 0x11: resp(1);                                                          return; // RESP1
+  case 0x12: resm(0);                                                          return; // RESM0
+  case 0x13: resm(1);                                                          return; // RESM1
+  case 0x14: resbl();                                                          return; // RESBL
+  case 0x15: audio[0].control = data;                                          return; // AUDC0
+  case 0x16: audio[1].control = data;                                          return; // AUDC1
+  case 0x17: audio[0].frequency = data;                                        return; // AUDF0
+  case 0x18: audio[1].frequency = data;                                        return; // AUDF1
+  case 0x19: audio[0].volume = data;                                           return; // AUDV0
+  case 0x1a: audio[1].volume = data;                                           return; // AUDV1
+  case 0x1b: writeQueue.add(address, data, 3);                                 return; // GRP0
+  case 0x1c: writeQueue.add(address, data, 3);                                 return; // GRP1
+  case 0x1d: writeQueue.add(address, data, 1);                                 return; // ENAM0
+  case 0x1e: writeQueue.add(address, data, 1);                                 return; // ENAM1
+  case 0x1f: writeQueue.add(address, data, 1);                                 return; // ENAB1
+  case 0x20: writeQueue.add(address, data, 2);                                 return; // HMP0
+  case 0x21: writeQueue.add(address, data, 2);                                 return; // HMP1
+  case 0x22: writeQueue.add(address, data, 2);                                 return; // HMM0
+  case 0x23: writeQueue.add(address, data, 2);                                 return; // HMM1
+  case 0x24: writeQueue.add(address, data, 2);                                 return; // HMBL
+  case 0x25: player[0].delay = data.bit(0);                                    return; // VDELP0
+  case 0x26: player[1].delay = data.bit(0);                                    return; // VDELP1
+  case 0x27: ball.delay      = data.bit(0);                                    return; // VDELBL
+  case 0x28: resmp(0, data);                                                   return; // RESMP0
+  case 0x29: resmp(1, data);                                                   return; // RESMP1
+  case 0x2a: writeQueue.add(address, data, 6);                                 return; // HMOVE
+  case 0x2b: writeQueue.add(address, data, 2);                                 return; // HMCLR
+  case 0x2c: cxclr();                                                          return; // CXCLR
+  }
+
+  debug(unimplemented, "[TIA] write ",hex(address), " = ", hex(data));
+}
+
+
+auto TIA::vsync(n1 state) -> void {
+  if(io.vsync && !state) {
+    io.vcounter = 0;
+    frame();
+  }
+
+  io.vsync = state;
+}
+
+auto TIA::ctrlpf(n8 data) -> void {
+  playfield.mirror    = data.bit(0);
+  playfield.scoreMode = data.bit(1);
+  playfield.priority  = data.bit(2);
+  ball.size           = data.bit(4, 5);
+}
+
+auto TIA::cxclr() -> void {
+  collision.M0P0 = 0;
+  collision.M0P1 = 0;
+  collision.M1P0 = 0;
+  collision.M1P1 = 0;
+  collision.P0PF = 0;
+  collision.P0BL = 0;
+  collision.P1PF = 0;
+  collision.P1BL = 0;
+  collision.M0PF = 0;
+  collision.M0BL = 0;
+  collision.M1PF = 0;
+  collision.M1BL = 0;
+  collision.BLPF = 0;
+  collision.P0P1 = 0;
+  collision.M0M1 = 0;
+}
+
+auto TIA::hmove() -> void {
+  for(auto& p : player)  p.position = (p.position - p.offset + 160) % 160;
+  for(auto& m : missile) m.position = (m.position - m.offset + 160) % 160;
+  ball.position = (ball.position - ball.offset + 160) % 160;
+  io.hmoveTriggered = io.vcounter;
+}
+
+auto TIA::hmclr() -> void {
+  for(auto& p : player)  p.offset = 0;
+  for(auto& m : missile) m.offset = 0;
+  ball.offset = 0;
+}
+
+auto TIA::grp(n1 index, n8 data) -> void {
+  player[index].graphics[0] = data;
+  player[!index].graphics[1] = player[!index].graphics[0];
+  if(index) ball.enable[1] = ball.enable[0];
+}
+
+auto TIA::resp(n1 index) -> void {
+  player[index].position = (io.hcounter < 68 ? 3 : io.hcounter - 68 + 8) % 160;
+}
+
+auto TIA::resm(n1 index) -> void {
+  missile[index].position = (io.hcounter < 68 ? 2 : io.hcounter - 68 + 7) % 160;
+}
+
+auto TIA::resmp(n1 index, n8 data) -> void {
+  missile[index].reset = data.bit(1);
+  if(missile[index].reset) {
+    auto offset = 3;
+    if(player[index].size == 5) offset = 6;
+    if(player[index].size == 7) offset = 10;
+    missile[index].position = player[index].position + offset;
+  }
+}
+
+auto TIA::resbl() -> void {
+  ball.position = (io.hcounter < 68 ? 2 : io.hcounter - 68 + 7);
+}

ares/a26/tia/serialization.cpp

@@ -0,0 +1,85 @@
+//serialization.cpp
+auto TIA::serialize(serializer& s) -> void {
+  Thread::serialize(s);
+
+  for(auto n : range(writeQueue.maxItems)) {
+    s(writeQueue.items[n].active);
+    s(writeQueue.items[n].address);
+    s(writeQueue.items[n].data);
+    s(writeQueue.items[n].delay);
+  }
+
+  s(io.vcounter);
+  s(io.hcounter);
+  s(io.hmoveTriggered);
+  s(io.vsync);
+  s(io.vblank);
+  s(io.bgColor);
+  s(io.p0Color);
+  s(io.p1Color);
+  s(io.fgColor);
+
+  s(playfield.graphics);
+  s(playfield.mirror);
+  s(playfield.scoreMode);
+  s(playfield.priority);
+
+  for(auto n : range(2)) {
+    s(player[n].graphics[0]);
+    s(player[n].graphics[1]);
+    s(player[n].reflect);
+    s(player[n].size);
+    s(player[n].offset);
+    s(player[n].position);
+    s(player[n].delay);
+  }
+
+  for(auto n : range(2)) {
+    s(missile[n].enable);
+    s(missile[n].reset);
+    s(missile[n].size);
+    s(missile[n].offset);
+    s(missile[n].position);
+  }
+
+  for(auto n : range(2)) {
+    s(ball.enable[0]);
+    s(ball.enable[1]);
+    s(ball.size);
+    s(ball.offset);
+    s(ball.position);
+  }
+
+  s(collision.M0P0);
+  s(collision.M0P1);
+  s(collision.M1P0);
+  s(collision.M1P1);
+  s(collision.P0PF);
+  s(collision.P0BL);
+  s(collision.P1PF);
+  s(collision.P1BL);
+  s(collision.M0PF);
+  s(collision.M0BL);
+  s(collision.M1PF);
+  s(collision.M1BL);
+  s(collision.BLPF);
+  s(collision.P0P1);
+  s(collision.M0M1);
+
+  for(auto n : range(26)) {
+    s(volume[n]);
+  }
+
+  for(auto n : range(2)) {
+    s(audio[n].enable);
+    s(audio[n].divCounter);
+    s(audio[n].noiseCounter);
+    s(audio[n].noiseFeedback);
+    s(audio[n].pulseCounter);
+    s(audio[n].pulseCounterPaused);
+    s(audio[n].pulseFeedback);
+    s(audio[n].volume);
+    s(audio[n].control);
+    s(audio[n].frequency);
+  }
+}

ares/a26/tia/tia.cpp

@@ -0,0 +1,173 @@
+#include <a26/a26.hpp>
+
+namespace ares::Atari2600 {
+
+TIA tia;
+
+#include "audio.cpp"
+#include "color.cpp"
+#include "io.cpp"
+#include "serialization.cpp"
+#include "write-queue.cpp"
+
+auto TIA::load(Node::Object parent) -> void {
+  node = parent->append<Node::Object>("TIA");
+
+  screen = node->append<Node::Video::Screen>("Screen", 160, vlines());
+  screen->colors(1 << 7, {&TIA::color, this});
+  screen->setSize(160, vlines());
+  screen->setScale(2.0, 2.0);
+  screen->setAspect(12.0, 7.0);
+  screen->setViewport(0, 0, 160, vlines());
+
+  stream = node->append<Node::Audio::Stream>("Audio");
+  stream->setChannels(1);
+  stream->setFrequency(31'403);
+  stream->addHighPassFilter(  20.0, 1);
+  stream->addLowPassFilter (2840.0, 1);
+}
+
+auto TIA::unload() -> void {
+  screen->quit();
+  node->remove(screen);
+  screen.reset();
+  node.reset();
+}
+
+auto TIA::main() -> void {
+    scanline();
+    io.vcounter++;
+    io.hmoveTriggered = 255;
+}
+
+auto TIA::scanline() -> void {
+  cpu.rdyLine(1);
+  for(io.hcounter = 0; io.hcounter < 228; io.hcounter++) {
+    auto x = io.hcounter - 68;
+    if(x >= 0 && io.vcounter < vlines()) pixel(x);
+    writeQueue.step();
+    runAudio();
+    step();
+  }
+}
+
+auto TIA::pixel(n8 x) -> void {
+  auto output = screen->pixels().data() + (io.vcounter * 160);
+
+  // Output only black during vblank, or for the first 8px of a scanline where hmove was triggered
+  if (io.vblank || (io.hmoveTriggered == io.vcounter && x < 8)) {
+    output[x] = 0;
+    return;
+  }
+
+  output[x] = io.bgColor;
+  auto pf = runPlayfield(x);
+  auto bl = runBall(x);
+  auto p0 = runPlayer(x, 0);
+  auto p1 = runPlayer(x, 1);
+  auto m0 = runMissile(x, 0);
+  auto m1 = runMissile(x, 1);
+
+  if(!playfield.priority && (pf || bl)) output[x] = io.fgColor;
+  if(p1 || m1)                          output[x] = io.p1Color;
+  if(p0 || m0)                          output[x] = io.p0Color;
+  if(playfield.priority && (pf || bl))  output[x] = io.fgColor;
+
+  // Update Collision
+  if(m0 && p1) collision.M0P1 = 1;
+  if(m0 && p0) collision.M0P0 = 1;
+  if(m1 && p0) collision.M1P0 = 1;
+  if(m1 && p1) collision.M1P1 = 1;
+  if(p0 && pf) collision.P0PF = 1;
+  if(p0 && bl) collision.P0BL = 1;
+  if(p1 && pf) collision.P1PF = 1;
+  if(p1 && bl) collision.P1BL = 1;
+  if(m0 && pf) collision.M0PF = 1;
+  if(m0 && bl) collision.M0BL = 1;
+  if(m1 && pf) collision.M1PF = 1;
+  if(m1 && bl) collision.M1BL = 1;
+  if(bl && pf) collision.BLPF = 1;
+  if(p0 && p1) collision.P0P1 = 1;
+  if(m0 && m1) collision.M0M1 = 1;
+}
+
+auto TIA::runPlayfield(n8 x) -> n1 {
+  auto pos = x >> 2;
+
+  if(!playfield.mirror || pos < 20) return playfield.graphics.bit(pos % 20);
+  return playfield.graphics.bit(19 - (pos % 20));
+}
+
+auto TIA::runPlayer(n8 x, n1 index) -> n1 {
+  auto& player = this->player[index];
+  auto position = player.position;
+
+  // Handle player stretch capability
+  auto width = 8;
+  if(player.size == 5) width = 16;
+  if(player.size == 7) width = 32;
+
+  // Handle repeat capability
+  auto repeat = 1;
+  if(player.size == 1 || player.size == 2 || player.size == 4) repeat = 2;
+  if(player.size == 3 || player.size == 6)                     repeat = 3;
+
+  auto spacing = 8;
+  if(player.size == 2 || player.size == 6) spacing = 24;
+  if(player.size == 4                    ) spacing = 56;
+
+  for(int i = 0; i < repeat; i++) {
+    if (x >= position && x < position + width) {
+      auto bit = player.reflect ? (x - position) / (width / 8) : 7 - ((x - position) / (width / 8));
+      return player.graphics[player.delay].bit(bit);
+    }
+
+    position = (position + (spacing + width)) % 160;
+  }
+
+  return 0;
+}
+
+auto TIA::runMissile(n8 x, n1 index) -> n1 {
+  auto& missile = this->missile[index];
+  if (!missile.enable || missile.reset) return 0;
+
+  const int missileSizes[4] = {1, 2, 4, 8};
+  return x >= missile.position && x < missile.position + missileSizes[missile.size];
+}
+
+auto TIA::runBall(n8 x) -> n1 {
+  if(!ball.enable[ball.delay]) return 0;
+
+  const int ballSizes[4] = {1, 2, 4, 8};
+  return x >= ball.position && x < ball.position + ballSizes[ball.size];
+}
+
+auto TIA::frame() -> void {
+  screen->frame();
+  scheduler.exit(Event::Frame);
+}
+
+auto TIA::step(u32 clocks) -> void {
+  Thread::step(clocks);
+  Thread::synchronize();
+}
+
+auto TIA::power(bool reset) -> void {
+  Thread::create(system.frequency(), {&TIA::main, this});
+  screen->power();
+  io = {};
+  playfield = {};
+  for(auto& p : player)  p = {};
+  for(auto& m : missile) m = {};
+  ball = {};
+  collision = {};
+  for(auto n : range(writeQueue.maxItems)) writeQueue.items[n] = {};
+
+  for(u32 level : range(15)) {
+    volume[level] = pow(2, level * -2.0 / 6.0);
+  }
+  volume[15] = 0;
+}
+
+}

ares/a26/tia/tia.hpp

@@ -0,0 +1,147 @@
+struct TIA : Thread {
+  Node::Object node;
+  Node::Video::Screen screen;
+  Node::Audio::Stream stream;
+
+  auto vlines() const -> u32 { return Region::NTSC() ? 262 : 312; }
+
+  //tia.cpp
+  auto load(Node::Object) -> void;
+  auto unload() -> void;
+
+  auto main() -> void;
+  auto step(u32 clocks = 1) -> void;
+  auto power(bool reset) -> void;
+  auto frame() -> void;
+  auto scanline() -> void;
+  auto pixel(n8 x) -> void;
+
+  auto runPlayfield(n8 x) -> n1;
+  auto runBall(n8 x) -> n1;
+  auto runPlayer(n8 x, n1 index) -> n1;
+  auto runMissile(n8 x, n1 index) -> n1;
+  auto runCollision() -> void;
+
+  //io.cpp
+  auto read(n8 address) -> n8;
+  auto write(n8 address, n8 data) -> void;
+  auto vsync(n1 state) -> void;
+  auto ctrlpf(n8 data) -> void;
+  auto cxclr() -> void;
+  auto hmove() -> void;
+  auto hmclr() -> void;
+  auto grp(n1 index, n8 data) -> void;
+  auto resp(n1 index) -> void;
+  auto resm(n1 index) -> void;
+  auto resmp(n1 index, n8 data) -> void;
+  auto resbl() -> void;
+
+  // audio.cpp
+  auto runAudio() -> void;
+
+  //color.cpp
+  auto color(n32) -> n64;
+
+  //serialization.cpp
+  auto serialize(serializer&) -> void;
+
+  // write-queue.cpp
+  struct WriteQueue {
+    inline auto add(u8 address, n8 data, i8 delay) -> void;
+    inline auto step() -> void;
+
+    struct QueueItem {
+      inline auto commit() -> void;
+      n1 active;
+      i8 address;
+      n8 data;
+      i8 delay;
+    };
+
+    QueueItem items[16];
+    const int maxItems = 16;
+  } writeQueue;
+
+  struct {
+    u9 vcounter;
+    u8 hcounter;
+    u8 hmoveTriggered;
+
+    n1 vsync;
+    n1 vblank;
+
+    n7 bgColor;
+    n7 p0Color;
+    n7 p1Color;
+    n7 fgColor;
+  } io;
+
+  struct {
+    n20 graphics;
+    n1 mirror;
+    n1 scoreMode;
+    n1 priority;
+  } playfield;
+
+  struct {
+    n8 graphics[2];
+    n1 reflect;
+    n3 size;
+    i4 offset;
+    n8 position;
+    n1 delay;
+  } player[2];
+
+  struct {
+    n1 enable;
+    n1 reset;
+    n2 size;
+    i4 offset;
+    n8 position;
+  } missile[2];
+
+  struct {
+    n1 enable[2];
+    n2 size;
+    i4 offset;
+    n1 delay;
+    n8 position;
+  } ball;
+
+  struct {
+    n1 M0P0;
+    n1 M0P1;
+    n1 M1P0;
+    n1 M1P1;
+    n1 P0PF;
+    n1 P0BL;
+    n1 P1PF;
+    n1 P1BL;
+    n1 M0PF;
+    n1 M0BL;
+    n1 M1PF;
+    n1 M1BL;
+    n1 BLPF;
+    n1 P0P1;
+    n1 M0M1;
+  } collision;
+
+  struct AudioChannel {
+    auto phase0() -> void;
+    auto phase1() -> u8;
+    n1 enable;
+    n8 divCounter;
+    n8 noiseCounter;
+    n1 noiseFeedback;
+    n8 pulseCounter;
+    n1 pulseCounterPaused;
+    n1 pulseFeedback;
+    n4 volume;
+    n4 control;
+    n5 frequency;
+  } audio[2];
+
+  f64 volume[16];
+};
+
+extern TIA tia;

ares/a26/tia/write-queue.cpp

@@ -0,0 +1,46 @@
+auto TIA::WriteQueue::add(u8 address, n8 data, i8 delay) -> void {
+  for(auto n : range(maxItems)) {
+    if(!items[n].active) {
+      items[n].address = address;
+      items[n].data = data;
+      items[n].delay = delay;
+      items[n].active = 1;
+      return;
+    }
+  }
+}
+
+auto TIA::WriteQueue::step() -> void {
+  auto& items = tia.writeQueue.items;
+
+  for(auto n : range(maxItems)) {
+    if(items[n].active && --items[n].delay == 0) {
+      items[n].commit();
+      items[n].active = 0;
+    }
+  }
+}
+
+auto TIA::WriteQueue::QueueItem::commit() -> void {
+  switch(address) {
+  case 0x01: tia.io.vblank = data.bit(1);                                                return; // VBLANK
+  case 0x0b: tia.player[0].reflect = data.bit(3);                                        return; // REFP0
+  case 0x0c: tia.player[1].reflect = data.bit(3);                                        return; // REFP1
+  case 0x0d: tia.playfield.graphics.bit(0, 3) = data.bit(4, 7);                          return; // PF0
+  case 0x0e: for(auto n : range(8)) tia.playfield.graphics.bit(4 + n) = data.bit(7 - n); return; // PF1
+  case 0x0f: tia.playfield.graphics.bit(12, 19) = data;                                  return; // PF2
+  case 0x1b: tia.grp(0, data);                                                           return; // GRP0
+  case 0x1c: tia.grp(1, data);                                                           return; // GRP1
+  case 0x1d: tia.missile[0].enable = data.bit(1);                                        return; // ENAM0
+  case 0x1e: tia.missile[1].enable = data.bit(1);                                        return; // ENAM1
+  case 0x1f: tia.ball.enable[0] = data.bit(1);                                           return; // ENABL
+  case 0x20: tia.player[0].offset = data.bit(4, 7);                                      return; // HMP0
+  case 0x21: tia.player[1].offset = data.bit(4, 7);                                      return; // HMP1
+  case 0x22: tia.missile[0].offset = data.bit(4, 7);                                     return; // HMM0
+  case 0x23: tia.missile[1].offset = data.bit(4, 7);                                     return; // HMM1
+  case 0x24: tia.ball.offset = data.bit(4, 7);                                           return; // HMBL
+  case 0x2a: tia.hmove();                                                                return; // HMOVE
+  case 0x2b: tia.hmclr();                                                                return; // HMCLR
+  }
+  debug(unimplemented, "[TIA] writeQueue: ",hex(address), " = ", hex(data));
+}

ares/ares/ares.hpp

@@ -1,6 +1,7 @@
 #pragma once
 
 #include <libco/libco.h>
+#include <sljit.h>
 
 #include <nall/platform.hpp>
 #include <nall/adaptive-array.hpp>
@@ -11,6 +12,7 @@
 #include <nall/directory.hpp>
 #include <nall/dl.hpp>
 #include <nall/endian.hpp>
+#include <nall/hashset.hpp>
 #include <nall/image.hpp>
 #include <nall/literals.hpp>
 #include <nall/priority-queue.hpp>
@@ -34,19 +36,20 @@
 #include <nall/hash/crc32.hpp>
 #include <nall/hash/sha256.hpp>
 using namespace nall;
+using namespace nall::primitives;
 
 namespace ares {
   static const string Name       = "ares";
-  static const string Version    = "124";
+  static const string Version    = "131";
   static const string Copyright  = "ares team, Near";
   static const string License    = "ISC";
   static const string LicenseURI = "https://opensource.org/licenses/ISC";
-  static const string Website    = "ares-emulator.github.io";
-  static const string WebsiteURI = "https://ares-emulator.github.io";
+  static const string Website    = "ares-emu.net";
+  static const string WebsiteURI = "https://ares-emu.net/";
 
   //incremented only when serialization format changes
   static const u32    SerializerSignature = 0x31545342;  //"BST1" (little-endian)
-  static const string SerializerVersion   = "123.1";
+  static const string SerializerVersion   = "131";
 
   namespace VFS {
     using Pak = shared_pointer<vfs::directory>;

ares/ares/debug/debug.hpp

@@ -4,19 +4,19 @@ struct Debug {
   auto reset() -> void;
 
   template<typename... P> auto unhandled(P&&... p) -> void {
-    return _unhandled({forward<P>(p)...});
+    return _unhandled({std::forward<P>(p)...});
   }
 
   template<typename... P> auto unimplemented(P&&... p) -> void {
-    return _unimplemented({forward<P>(p)...});
+    return _unimplemented({std::forward<P>(p)...});
   }
 
   template<typename... P> auto unusual(P&&... p) -> void {
-    return _unusual({forward<P>(p)...});
+    return _unusual({std::forward<P>(p)...});
   }
 
   template<typename... P> auto unverified(P&&... p) -> void {
-    return _unverified({forward<P>(p)...});
+    return _unverified({std::forward<P>(p)...});
   }
 
 private:

ares/ares/memory/fixed-allocator.cpp

@@ -1,16 +1,22 @@
-#include <ares/ares.hpp>
-
-namespace ares::Memory {
-
-alignas(4096) u8 fixedBuffer[1_GiB];
-
-FixedAllocator::FixedAllocator() {
-  _allocator.resize(sizeof(fixedBuffer), 0, fixedBuffer);
-}
-
-auto FixedAllocator::get() -> bump_allocator& {
-  static FixedAllocator allocator;
-  return allocator._allocator;
-}
-
-}
+#include <ares/ares.hpp>
+
+namespace ares::Memory {
+
+#if defined(PLATFORM_MACOS)
+//stub for unsupported platforms
+FixedAllocator::FixedAllocator() {
+}
+#else
+alignas(4096) u8 fixedBuffer[128_MiB];
+
+FixedAllocator::FixedAllocator() {
+  _allocator.resize(sizeof(fixedBuffer), 0, fixedBuffer);
+}
+#endif
+
+auto FixedAllocator::get() -> bump_allocator& {
+  static FixedAllocator allocator;
+  return allocator._allocator;
+}
+
+}

ares/ares/memory/fixed-allocator.hpp

@@ -1,14 +1,14 @@
-#pragma once
-
-namespace ares::Memory {
-
-struct FixedAllocator {
-  static auto get() -> bump_allocator&;
-
-private:
-  FixedAllocator();
-
-  bump_allocator _allocator;
-};
-
-}
+#pragma once
+
+namespace ares::Memory {
+
+struct FixedAllocator {
+  static auto get() -> bump_allocator&;
+
+private:
+  FixedAllocator();
+
+  bump_allocator _allocator;
+};
+
+}

ares/ares/node/audio/stream.hpp

@@ -25,7 +25,7 @@ struct Stream : Audio {
   template<typename... P>
   auto frame(P&&... p) -> void {
     if(runAhead()) return;
-    f64 samples[sizeof...(p)] = {forward<P>(p)...};
+    f64 samples[sizeof...(p)] = {std::forward<P>(p)...};
     write(samples);
   }
 

ares/ares/node/debugger/tracer/instruction.hpp

@@ -18,23 +18,25 @@ struct Instruction : Tracer {
 
   auto setMask(bool mask) -> void {
     _mask = mask;
+    _masks.reset();
   }
 
   auto setDepth(u32 depth) -> void {
     _depth = depth;
     _history.reset();
     _history.resize(depth);
-    for(auto& history : _history) history = ~0;
+    for(auto& history : _history) history = ~0ull;
   }
 
-  auto address(u32 address) -> bool {
-    address &= (1ull << _addressBits) - 1;  //mask upper bits of address
+  auto address(u64 address) -> bool {
+    address &= ~0ull >> (64 - _addressBits);  //mask upper bits of address
     _address = address;
     address >>= _addressMask;  //clip unneeded alignment bits (to reduce _masks size)
 
-    if(_mask && updateMasks()) {
-      if(_masks[address >> 3] & 1 << (address & 7)) return false;  //do not trace twice
-      _masks[address >> 3] |= 1 << (address & 7);
+    if(_mask) {
+      auto mask = _masks.find(address);
+      if(!mask) mask = _masks.insert(address);
+      if(mask->visit(address)) return false;  //do not trace twice
     }
 
     if(_depth) {
@@ -55,11 +57,13 @@ struct Instruction : Tracer {
 
   //mark an already-executed address as not executed yet for trace masking.
   //call when writing to executable RAM to support self-modifying code.
-  auto invalidate(u32 address) -> void {
-    if(unlikely(_mask && updateMasks())) {
-      address &= (1ull << _addressBits) - 1;
+  auto invalidate(u64 address) -> void {
+    if(unlikely(_mask)) {
+      address &= ~0ull >> (64 - _addressBits);
       address >>= _addressMask;
-      _masks[address >> 3] &= ~(1 << (address & 7));
+
+      auto mask = _masks.find(address);
+      if(mask) mask->unvisit(address);
     }
   }
 
@@ -103,14 +107,27 @@ struct Instruction : Tracer {
   }
 
 protected:
-  auto updateMasks() -> bool {
-    auto size = 1ull << _addressBits >> _addressMask >> 3;
-    if(!_mask || !size) return _masks.reset(), false;
-    if(_masks.size() == size) return true;
-    _masks.reset();
-    _masks.resize(size);
-    return true;
-  }
+  struct VisitMask {
+    VisitMask(u64 address) : upper(address >> 6), mask(0) {}
+    auto operator==(const VisitMask& source) const -> bool { return upper == source.upper; }
+    auto hash() const -> u32 { return upper; }
+
+    auto visit(u64 address) -> bool {
+      const u64 bit = 1ull << (address & 0x3f);
+      if(mask & bit) return true;
+      mask |= bit;
+      return false;
+    }
+
+    auto unvisit(u64 address) -> void {
+      const u64 bit = 1ull << (address & 0x3f);
+      mask &= ~bit;
+    }
+
+  private:
+    u64 upper;
+    u64 mask;
+  };
 
   u32  _addressBits = 32;
   u32  _addressMask = 0;
@@ -120,6 +137,6 @@ protected:
 //unserialized:
   n64 _address = 0;
   n64 _omitted = 0;
-  vector<u32> _history;
-  vector<u08> _masks;
+  vector<u64> _history;
+  hashset<VisitMask> _masks;
 };

ares/ares/node/input/input.hpp

@@ -1,4 +1,5 @@
 struct Input : Object {
   DeclareClass(Input, "input")
   using Object::Object;
+  u64 lastPoll;
 };

ares/ares/node/object.hpp

@@ -33,7 +33,7 @@ struct Object : shared_pointer_this<Object> {
   template<typename T, typename... P>
   auto prepend(P&&... p) -> Node::Object {
     using Type = typename T::type;
-    return prepend(shared_pointer<Type>::create(forward<P>(p)...));
+    return prepend(shared_pointer<Type>::create(std::forward<P>(p)...));
   }
 
   auto append(Node::Object node) -> Node::Object {
@@ -47,7 +47,7 @@ struct Object : shared_pointer_this<Object> {
   template<typename T, typename... P>
   auto append(P&&... p) -> Node::Object {
     using Type = typename T::type;
-    return append(shared_pointer<Type>::create(forward<P>(p)...));
+    return append(shared_pointer<Type>::create(std::forward<P>(p)...));
   }
 
   auto remove(Node::Object node) -> void {

ares/ares/node/video/screen.cpp

@@ -36,6 +36,7 @@ auto Screen::main(uintptr_t) -> void {
 auto Screen::quit() -> void {
   _kill = true;
   _thread.join();
+  _sprites.reset();
 }
 
 auto Screen::power() -> void {

ares/ares/node/video/screen.hpp

@@ -82,10 +82,10 @@ protected:
   u32  _rotation = 0;  //counter-clockwise (90 = left, 270 = right)
 
   function<n64 (n32)> _color;
-  unique_pointer<u32> _inputA;
-  unique_pointer<u32> _inputB;
-  unique_pointer<u32> _output;
-  unique_pointer<u32> _rotate;
+  unique_pointer<u32[]> _inputA;
+  unique_pointer<u32[]> _inputB;
+  unique_pointer<u32[]> _output;
+  unique_pointer<u32[]> _rotate;
   unique_pointer<u32[]> _palette;
   vector<Node::Video::Sprite> _sprites;
 

ares/ares/platform.hpp

@@ -16,6 +16,7 @@ struct Platform {
   virtual auto pak(Node::Object) -> shared_pointer<vfs::directory> { return {}; }
   virtual auto event(Event) -> void {}
   virtual auto log(string_view message) -> void {}
+  virtual auto status(string_view message) -> void {}
   virtual auto video(Node::Video::Screen, const u32* data, u32 pitch, u32 width, u32 height) -> void {}
   virtual auto audio(Node::Audio::Stream) -> void {}
   virtual auto input(Node::Input::Input) -> void {}

ares/ares/resource/GNUmakefile

@@ -1,6 +1,10 @@
+build := stable
+nall.path := ../../../nall
+include $(nall.path)/GNUmakefile
+
 all:
 	sourcery resource.bml resource.cpp resource.hpp
 
 clean:
-	rm resource.cpp
-	rm resource.hpp
+	$(call delete,resource.cpp)
+	$(call delete,resource.hpp)

ares/ares/scheduler/thread.cpp

@@ -84,14 +84,15 @@ inline auto Thread::synchronize() -> void {
 template<typename... P>
 inline auto Thread::synchronize(Thread& thread, P&&... p) -> void {
   //switching to another thread does not guarantee it will catch up before switching back.
-  while(thread.clock() < clock()) {
+  //make sure not to switch to threads that were destroyed during synchronization
+  while(thread.clock() < clock() && thread.handle()) {
     //disable synchronization for auxiliary threads during scheduler synchronization.
     //synchronization can begin inside of this while loop.
     if(scheduler.synchronizing()) break;
     co_switch(thread.handle());
   }
   //convenience: allow synchronizing multiple threads with one function call.
-  if constexpr(sizeof...(p) > 0) synchronize(forward<P>(p)...);
+  if constexpr(sizeof...(p) > 0) synchronize(std::forward<P>(p)...);
 }
 
 inline auto Thread::serialize(serializer& s) -> void {

ares/component/audio/ay38910/ay38910.cpp

@@ -196,9 +196,11 @@ auto AY38910::write(n8 data) -> void {
     envelope.output    = !envelope.attacking ? 15 : 0;
     break;
   case 14:
+    portA.data = data;
     writeIO(0, data);
     break;
   case 15:
+    portB.data = data;
     writeIO(1, data);
     break;
   }

ares/component/audio/sn76489/sn76489.cpp

@@ -63,6 +63,10 @@ auto SN76489::write(n8 data) -> void {
     case 0: tone0.pitch.bit(4,9) = data.bit(0,5); break;
     case 1: tone1.pitch.bit(4,9) = data.bit(0,5); break;
     case 2: tone2.pitch.bit(4,9) = data.bit(0,5); noise.pitch = tone2.pitch; break;
+    case 3: noise.rate   = data.bit(0,1);
+            noise.enable = data.bit(2);
+            noise.lfsr   = 0x8000;
+            break;
     }
   }
 }

ares/component/audio/ym2610/io.cpp

@@ -1,47 +1,84 @@
 auto YM2610::read(n2 address) -> n8 {
   switch(address) {
   case 0: return fm.readStatus();
+  case 1: return ssg.read();
+
   }
+
+  debug(unimplemented, "[YM2610] read ", hex(address));
   return 0;
 }
 
 auto YM2610::write(n2 address, n8 data) -> void {
   switch(address) {
-  case 0: register = 0x000 | data; return;
+  case 0: {
+    registerAddress = 0x000 | data;
+    if (registerAddress <= 0xf) ssg.select(registerAddress);
+    else fm.writeAddress(registerAddress);
+    return;
+  }
   case 1: return writeLower(data);
-  case 2: register = 0x100 | data; return;
+  case 2: registerAddress = 0x100 | data; fm.writeAddress(registerAddress); return;
   case 3: return writeUpper(data);
   }
 }
 
 auto YM2610::writeLower(n8 data) -> void {
-  switch(register) {
+  switch(registerAddress) {
   case 0x000 ... 0x00f:
-    ssg.select(register);
     ssg.write(data);
     return;
-  case 0x010 ... 0x01b:
-    //ADPCM-B
-    return;
+  case 0x010:
+              pcmB.repeat  = data.bit(4);
+              if(data.bit(0)) pcmB.power();
+              if(data.bit(7)) pcmB.beginPlay();
+              else            pcmB.playing = 0;
+              return;
+  case 0x011: pcmB.left    = data.bit(7);
+              pcmB.right   = data.bit(6);           return;
+  case 0x012: pcmB.startAddress.bit( 8, 15) = data; return;
+  case 0x013: pcmB.startAddress.bit(16, 23) = data; return;
+  case 0x014: pcmB.endAddress.bit  ( 8, 15) = data; return;
+  case 0x015: pcmB.endAddress.bit  (16, 23) = data; return;
+  case 0x016 ... 0x018: debug(unimplemented, "[YM2610::ADPCMB] Write ", hex(registerAddress), " = ", data);
+  case 0x019: pcmB.delta.bit       ( 0,  7) = data; return;
+  case 0x01a: pcmB.delta.bit       ( 8, 15) = data; return;
+  case 0x01b: pcmB.volume                   = data; return;
   case 0x01c:
     //EOS
+    debug(unimplemented, "[YM2610::EOS] Write ", hex(registerAddress), " = ", data);
     return;
   case 0x01d ... 0x1ff:
-    fm.writeAddress(register);
     fm.writeData(data);
     return;
   }
-  unreachable;
+
+  debug(unimplemented, "[YM2610] Write ", hex(registerAddress), " = ", data);
 }
 
 auto YM2610::writeUpper(n8 data) -> void {
-  switch(register) {
-  case 0x000 ... 0x12f:
-    //ADPCM-A
-    return;
-  case 0x130 ... 0x1ff:
-    fm.writeAddress(register);
-    fm.writeData(data);
+  switch(registerAddress) {
+  case 0x100:
+    for(auto n : range(6)) {
+      if(!data.bit(n)) continue;
+      if(data.bit(7)) {
+        pcmA.channels[n].keyOff();
+        continue;
+      }
+
+      pcmA.channels[n].keyOn();
+    }
     return;
+  case 0x101:           pcmA.volume = data.bit(0, 5);                                           return;
+  case 0x108 ... 0x10d: pcmA.channels[registerAddress - 0x108].volume = data.bit(0,4);
+                        pcmA.channels[registerAddress - 0x108].left   = data.bit(7);
+                        pcmA.channels[registerAddress - 0x108].right  = data.bit(6);            return;
+  case 0x110 ... 0x115: pcmA.channels[registerAddress - 0x110].startAddress.bit( 8, 15) = data; return;
+  case 0x118 ... 0x11d: pcmA.channels[registerAddress - 0x118].startAddress.bit(16, 23) = data; return;
+  case 0x120 ... 0x125: pcmA.channels[registerAddress - 0x120].endAddress.bit  ( 8, 15) = data; return;
+  case 0x128 ... 0x12d: pcmA.channels[registerAddress - 0x128].endAddress.bit  (16, 23) = data; return;
+  case 0x130 ... 0x1ff: fm.writeAddress(registerAddress); fm.writeData(data);                   return;
   }
+
+  debug(unimplemented, "[YM2610] Write ", hex(registerAddress), " = ", data);
 }

ares/component/audio/ym2610/pcm.cpp

@@ -0,0 +1,149 @@
+auto YM2610::PCMA::power() -> void {
+  for(auto c : range(6)) {
+    channels[c].startAddress.bit(0, 7) = 0x00;
+    channels[c].endAddress.bit(0, 7) = 0x00;
+    channels[c].decodeAccumulator = 0;
+    channels[c].decodeStep = 0;
+  }
+
+  i16 stepSize[] = {
+    16, 17, 19, 21, 23, 25, 28, 31, 34, 37,
+    41, 45, 50, 55, 60, 66, 73, 80, 88, 97,
+    107, 118, 130, 143, 157, 173, 190, 209, 230, 253,
+    279, 307, 337, 371, 408, 449, 494, 544, 598, 658,
+    724, 796, 876, 963, 1060, 1166, 1282, 1411, 1552
+  };
+
+  // Generate a decode table
+  for (int step = 0; step < 49; step++) {
+    for (int nibble = 0; nibble < 16; nibble++) {
+      int value = (2 * (nibble & 0x07) + 1) * stepSize[step] / 8;
+      decodeTable[step * 16 + nibble] = ((nibble & 0x08) != 0) ? -value : value;
+    }
+  }
+}
+
+auto YM2610::PCMA::clock() -> array<i16[2]> {
+  f32 left = 0;
+  f32 right = 0;
+
+   for(auto c : range(6)) {
+    if(!channels[c].playing) continue;
+    if(((channels[c].currentAddress ^ (channels[c].endAddress + (1 << 8))) & 0xfffff) == 0) {
+      channels[c].playing = 0;
+      continue;
+    }
+
+    auto sample = channels[c].decode(self.readPCMA(channels[c].currentAddress));
+
+    int totalVolume = (volume ^ 0x3f) + (channels[c].volume ^ 0x1f);
+    int volumeMultiplier = 15 - (totalVolume & 7);
+    int volumeShift = 5 + (totalVolume >> 3);
+
+    sample = ((sample * volumeMultiplier) >> volumeShift) & ~3;
+
+    if(channels[c].left)   left += sample / 2047.0f;
+    if(channels[c].right) right += sample / 2047.0f;
+  }
+
+  return {sclamp<16>(left * 32768.0), sclamp<16>(right * 32768.0)};
+}
+
+auto YM2610::PCMA::Channel::keyOn() -> void {
+  playing = 1;
+  currentAddress = startAddress;
+  decodeAccumulator = 0;
+  decodeStep = 0;
+  currentNibble = 0;
+}
+
+auto YM2610::PCMA::Channel::keyOff() -> void {
+  playing = 0;
+}
+
+auto YM2610::PCMA::Channel::decode(n8 value) -> i12 {
+  int stepAdjust[] = { -1, -1, -1, -1, 2, 5, 7, 9, -1, -1, -1, -1, 2, 5, 7, 9 };
+
+  value = currentNibble ? value.bit(0, 3) : value.bit(4, 7);
+  if(currentNibble) currentAddress++;
+  currentNibble = !currentNibble;
+
+  decodeAccumulator += self.decodeTable[decodeStep + value];
+
+  decodeStep += stepAdjust[value & 7] * 16;
+  decodeStep = std::clamp((int)decodeStep, 0, 48 * 16);
+
+  return decodeAccumulator;
+}
+
+auto YM2610::PCMB::power() -> void {
+  currentNibble       = 0;
+  currentAddress      = 0;
+  decodeAccumulator   = 0;
+  previousAccumulator = 0;
+  decodePosition      = 0;
+  decodeStep          = 127;
+  volume              = 255;
+}
+
+auto YM2610::PCMB::clock() -> array<i16[2]> {
+  f32 leftSample = 0;
+  f32 rightSample = 0;
+
+  decode();
+
+  i32 sample = (previousAccumulator * (i32)((decodePosition ^ 0xffff) + 1) + decodeAccumulator * (i32)(decodePosition)) >> 16;
+  sample = (sample * (i32)volume) >> 9;
+
+  if(left)  leftSample  += (sample / 65535.0);
+  if(right) rightSample += (sample / 65535.0);
+
+  return {sclamp<16>(leftSample * 32768.0), sclamp<16>(rightSample * 32768.0)};
+}
+
+auto YM2610::PCMB::decode() -> void {
+  if(!playing) {
+    return;
+  }
+
+  u8 stepSize[8] = { 57, 57, 57, 57, 77, 102, 128, 153 };
+
+  u32 position = decodePosition + delta;
+  decodePosition = (u16)position;
+  if(position < 0x10000) return;
+
+  n8 value = self.readPCMB(currentAddress);
+  value = currentNibble ? value.bit(0, 3) : value.bit(4, 7);
+  currentNibble = !currentNibble;
+
+  if(currentNibble == 0) {
+    currentAddress++;
+    if(currentAddress == endAddress) {
+      if(repeat) beginPlay();
+      else {
+        decodeAccumulator = 0;
+        previousAccumulator = 0;
+        playing = 0;
+      }
+    }
+  }
+
+  previousAccumulator = decodeAccumulator;
+
+  i32 _delta = (value.bit(0, 2) * 2 + 1) * decodeStep / 8;
+  if(value.bit(3)) _delta = -_delta;
+
+  decodeAccumulator = sclamp<16>(decodeAccumulator + _delta);
+
+  decodeStep = std::clamp(((int)decodeStep * stepSize[value.bit(0, 2)]) / 64, 127, 24576);
+}
+
+auto YM2610::PCMB::beginPlay() -> void {
+  currentAddress      = startAddress;
+  currentNibble       = 0;
+  decodeAccumulator   = 0;
+  previousAccumulator = 0;
+  decodePosition      = 0;
+  decodeStep          = 127;
+  playing             = 1;
+}

ares/component/audio/ym2610/serialization.cpp

@@ -1,5 +1,41 @@
 auto YM2610::serialize(serializer& s) -> void {
-  s(register);
+  s(registerAddress);
   s(fm);
   s(ssg);
+  s(pcmA);
+  s(pcmB);
 }
+
+auto YM2610::PCMA::serialize(serializer& s) -> void {
+  for(auto n : range(6)) {
+   s(channels[n].playing);
+   s(channels[n].ended);
+   s(channels[n].left);
+   s(channels[n].right);
+   s(channels[n].volume);
+   s(channels[n].startAddress);
+   s(channels[n].endAddress);
+   s(channels[n].currentAddress);
+   s(channels[n].decodeAccumulator);
+   s(channels[n].decodeStep);
+   s(channels[n].currentNibble);
+  }
+}
+
+auto YM2610::PCMB::serialize(serializer& s) -> void {
+  s(playing);
+  s(repeat);
+  s(left);
+  s(right);
+  s(startAddress);
+  s(endAddress);
+  s(delta);
+  s(volume);
+
+  s(currentAddress);
+  s(currentNibble);
+  s(decodeAccumulator);
+  s(decodePosition);
+  s(previousAccumulator);
+  s(decodeStep);
+}

ares/component/audio/ym2610/ym2610.cpp

@@ -4,11 +4,14 @@
 namespace ares {
 
 #include "io.cpp"
+#include "pcm.cpp"
 #include "serialization.cpp"
 
 auto YM2610::power() -> void {
   fm.power();
   ssg.power();
+  pcmA.power();
+  pcmB.power();
 }
 
 }

ares/component/audio/ym2610/ym2610.hpp

@@ -20,10 +20,74 @@ struct YM2610 {
   //serialization.cpp
   auto serialize(serializer&) -> void;
 
+  virtual auto readPCMA(u32 address) -> u8 { return 0; }
+  virtual auto readPCMB(u32 address) -> u8 { return 0; }
+
 protected:
-  n9 register;
+  n9 registerAddress;
   YM2612 fm;
   YM2149 ssg;
+
+  struct PCMA {
+    YM2610& self;
+
+    // pcm.cpp
+    auto power() -> void;
+    auto clock() -> array<i16[2]>;
+
+    //serialization.cpp
+    auto serialize(serializer&) -> void;
+
+    struct Channel {
+      PCMA& self;
+      auto keyOn() -> void;
+      auto keyOff() -> void;
+      auto decode(n8) -> i12;
+      n1 playing;
+      n1 ended;
+      n1 left;
+      n1 right;
+      n5 volume;
+      n32 startAddress;
+      n32 endAddress;
+      n32 currentAddress;
+      i12 decodeAccumulator;
+      i32 decodeStep;
+      bool currentNibble;
+    } channels[6]{{*this}, {*this}, {*this}, {*this}, {*this}, {*this}};
+
+    n6 volume;
+    int decodeTable[16 * 49];
+  } pcmA{*this};
+
+  struct PCMB {
+    YM2610& self;
+
+    // pcm.cpp
+    auto power() -> void;
+    auto clock() -> array<i16[2]>;
+    auto decode() -> void;
+    auto beginPlay() -> void;
+
+    //serialization.cpp
+    auto serialize(serializer&) -> void;
+
+    n1  playing;
+    n1  repeat;
+    n1  left;
+    n2  right;
+    n32 startAddress;
+    n32 endAddress;
+    n16 delta;
+    n8  volume;
+
+    n32 currentAddress;
+    bool currentNibble;
+    i32 decodeAccumulator;
+    u32 decodePosition;
+    i32 previousAccumulator;
+    i32 decodeStep;
+  } pcmB{*this};
 };
 
 }

ares/component/audio/ym2612/channel.cpp

@@ -1,9 +1,6 @@
-auto YM2612::Channel::Operator::trigger(bool state) -> void {
+auto YM2612::Channel::Operator::updateKeyState(bool state) -> void {
   if(keyOn == state) return;  //no change
-
   keyOn = state;
-  envelope.state = Release;
-  updateEnvelope();
 
   if(keyOn) {
     //restart phase and envelope generators
@@ -13,42 +10,46 @@ auto YM2612::Channel::Operator::trigger(bool state) -> void {
     updateEnvelope();
 
     if(envelope.rate >= 62) {
-      //skip attack and possibly decay stages
+      //skip attack phase
       envelope.value = 0;
-      envelope.state = envelope.sustainLevel ? Decay : Sustain;
-      updateEnvelope();
     }
-  } else if(ssg.enable && ssg.attack != ssg.invert) {
-    //SSG-EG key-off
-    envelope.value = 0x200 - envelope.value;
+  } else {
+    envelope.state = Release;
+    updateEnvelope();
+
+    if(ssg.enable && ssg.attack != ssg.invert) {
+      //SSG-EG key-off
+      envelope.value = 0x200 - envelope.value;
+    }
   }
 
   updateLevel();
 }
 
 auto YM2612::Channel::Operator::runEnvelope() -> void {
-  u32 sustain = envelope.sustainLevel < 15 ? envelope.sustainLevel << 5 : 0x3f0;
+  updateKeyState(keyLine);
   if(ym2612.envelope.clock & (1 << envelope.divider) - 1) return;
 
   u32 value = ym2612.envelope.clock >> envelope.divider;
   u32 step = envelope.steps >> ((~value & 7) << 2) & 0xf;
-  if(ssg.enable) step <<= 2;  //SSG results in a 4x faster envelope
+  u32 sustain = envelope.sustainLevel < 15 ? envelope.sustainLevel << 5 : 0x1f << 5;
 
   if(envelope.state == Attack) {
-    u32 next = envelope.value + (~u16(envelope.value) * step >> 4) & 0x3ff;
-    if(next <= envelope.value) {
-      envelope.value = next;
-    } else {
-      envelope.value = 0;
-      envelope.state = envelope.value < sustain ? Decay : Sustain;
+    if(envelope.value == 0) {
+      envelope.state = Decay;
       updateEnvelope();
+    } else if(envelope.rate < 62) {
+      // will stop updating if attack rate is increased to upper threshold during attack phase (confirmed behavior)
+      envelope.value += ~u16(envelope.value) * step >> 4;
     }
-  } else if(!ssg.enable || envelope.value < 0x200) {
-    envelope.value = min(envelope.value + step, 0x3ff);
+  }
+  if(envelope.state != Attack) {
     if(envelope.state == Decay && envelope.value >= sustain) {
       envelope.state = Sustain;
       updateEnvelope();
     }
+    if(ssg.enable) step = envelope.value < 0x200 ? step << 2 : 0;  //SSG results in a 4x faster envelope
+    envelope.value = min(envelope.value + step, 0x3ff);
   }
 
   updateLevel();
@@ -59,18 +60,19 @@ auto YM2612::Channel::Operator::runPhase() -> void {
   if(!(ssg.enable && envelope.value >= 0x200)) return;  //SSG loop check
 
   if(!ssg.hold && !ssg.alternate) phase.value = 0;
-  if(!ssg.hold || ssg.attack == ssg.invert) ssg.invert ^= ssg.alternate;
+  if(!(ssg.hold && ssg.invert)) ssg.invert ^= ssg.alternate;
 
   if(envelope.state == Attack) {
     //do nothing; SSG is meant to skip the attack phase
   } else if(envelope.state != Release && !ssg.hold) {
     //if still looping, reset the envelope
     envelope.state = Attack;
-    if(envelope.attackRate >= 62) {
-      envelope.value = 0;
-      envelope.state = envelope.sustainLevel ? Decay : Sustain;
-    }
     updateEnvelope();
+
+    if(envelope.rate >= 62) {
+      //skip attack phase
+      envelope.value = 0;
+    }
   } else if(envelope.state == Release || (ssg.hold && ssg.attack == ssg.invert)) {
     //clear envelope once finished
     envelope.value = 0x3ff;
@@ -115,11 +117,9 @@ auto YM2612::Channel::Operator::updatePhase() -> void {
   u32 tuning = detune & 3 ? detunes[(detune & 3) - 1][ksr & 7] >> (3 - (ksr >> 3)) : 0;
 
   u32 lfo = ym2612.lfo.clock >> 2 & 0x1f;
-  u32 pm = 4 * vibratos[channel.vibrato][lfo & 15] * (-lfo >> 4);
-  u32 msb = 10;
-  while(msb > 4 && ~pitch.value & 1 << msb) msb--;
+  s32 pm = (pitch.value * vibratos[channel.vibrato][lfo & 15] >> 9) * (lfo > 15 ? -1 : 1);
 
-  phase.delta = pitch.value + (pm >> 10 - msb) << 6 >> 7 - octave.value;
+  phase.delta = pitch.value + pm << 6 >> 7 - octave.value;
   phase.delta = (!detune.bit(2) ? phase.delta + tuning : phase.delta - tuning) & 0x1ffff;
   phase.delta = (multiple ? phase.delta * multiple : phase.delta >> 1) & 0xfffff;
 }
@@ -147,6 +147,7 @@ auto YM2612::Channel::power() -> void {
 
   for(auto& op : operators) {
     op.keyOn = 0;
+    op.keyLine = 0;
     op.lfoEnable = 0;
     op.detune = 0;
     op.multiple = 0;

ares/component/audio/ym2612/io.cpp

@@ -14,6 +14,8 @@ auto YM2612::writeData(n8 data) -> void {
   case 0x022: {
     lfo.rate = data.bit(0,2);
     lfo.enable = data.bit(3);
+    lfo.clock = 0;
+    lfo.divider = 0;
     break;
   }
 
@@ -37,11 +39,12 @@ auto YM2612::writeData(n8 data) -> void {
 
   //timer control
   case 0x027: {
-    //d6,d7 = mode (unimplemented; treated as mode 0 always)
-
     //reload period on 0->1 transition
     if(!timerA.enable && data.bit(0)) timerA.counter = timerA.period;
-    if(!timerB.enable && data.bit(1)) timerB.counter = timerB.period;
+    if(!timerB.enable && data.bit(1)) {
+      timerB.counter = timerB.period;
+      timerB.divider = 0;
+    }
 
     timerA.enable = data.bit(0);
     timerB.enable = data.bit(1);
@@ -51,6 +54,7 @@ auto YM2612::writeData(n8 data) -> void {
     if(data.bit(4)) timerA.line = 0;
     if(data.bit(5)) timerB.line = 0;
 
+    //d7 is enable bit for CSM mode (not yet implemented)
     channels[2].mode = data.bit(6,7);
     for(auto& op : channels[2].operators) op.updatePitch();
 
@@ -64,10 +68,10 @@ auto YM2612::writeData(n8 data) -> void {
     if(index == 3 || index == 7) break;
     if(index >= 4) index--;
 
-    channels[index][0].trigger(data.bit(4));
-    channels[index][1].trigger(data.bit(5));
-    channels[index][2].trigger(data.bit(6));
-    channels[index][3].trigger(data.bit(7));
+    channels[index][0].keyLine = data.bit(4);
+    channels[index][1].keyLine = data.bit(5);
+    channels[index][2].keyLine = data.bit(6);
+    channels[index][3].keyLine = data.bit(7);
 
     break;
   }
@@ -145,10 +149,11 @@ auto YM2612::writeData(n8 data) -> void {
 
   //SSG-EG
   case 0x090: {
-    op.ssg.hold = data.bit(0);
-    op.ssg.alternate = data.bit(1);
-    op.ssg.attack = data.bit(2);
     op.ssg.enable = data.bit(3);
+    op.ssg.hold      = op.ssg.enable ? data.bit(0) : 0;
+    op.ssg.alternate = op.ssg.enable ? data.bit(1) : 0;
+    op.ssg.attack    = op.ssg.enable ? data.bit(2) : 0;
+    channel[index].updateLevel();
     break;
   }
 

ares/component/audio/ym2612/serialization.cpp

@@ -42,6 +42,7 @@ auto YM2612::Channel::serialize(serializer& s) -> void {
 
 auto YM2612::Channel::Operator::serialize(serializer& s) -> void {
   s(keyOn);
+  s(keyLine);
   s(lfoEnable);
   s(detune);
   s(multiple);

ares/component/audio/ym2612/ym2612.cpp

@@ -27,7 +27,7 @@ auto YM2612::clock() -> array<i16[2]> {
     auto wave = [&](u32 n, u32 modulation) -> s32 {
       s32 x = (modulation >> 1) + (op[n].phase.value >> 10);
       s32 y = sine[x & 0x3ff] + op[n].outputLevel;
-      return y < 0x2000 ? pow2[y & 0x1ff] << 2 >> (y >> 9) : 0;
+      return y < 0x1a00 ? pow2[y & 0x1ff] << 2 >> (y >> 9) : 0; // -78 dB floor
     };
 
     s32 feedback = modMask & op[0].output + op[0].prior >> 9 - channel.feedback;
@@ -124,7 +124,7 @@ auto YM2612::clock() -> array<i16[2]> {
 
   if(++envelope.divider == 3) {
     envelope.divider = 0;
-    envelope.clock++;
+    if(!++envelope.clock) ++envelope.clock; // 12-bit counter: 1..4095 - zero-value is skipped (confirmed behavior)
   }
 
   for(auto& channel : channels) {

ares/component/audio/ym2612/ym2612.hpp

@@ -36,7 +36,7 @@ protected:
   } dac;
 
   struct Envelope {
-    n32 clock = 0;
+    n12 clock = 0;
     n32 divider = 0;
   } envelope;
 
@@ -91,7 +91,7 @@ protected:
       Operator(Channel& channel) : channel(channel), ym2612(channel.ym2612) {}
 
       //channel.cpp
-      auto trigger(bool) -> void;
+      auto updateKeyState(bool) -> void;
 
       auto runEnvelope() -> void;
       auto runPhase() -> void;
@@ -105,6 +105,7 @@ protected:
       auto serialize(serializer&) -> void;
 
       n1 keyOn = 0;
+      n1 keyLine = 0;
       n1 lfoEnable = 0;
       n3 detune = 0;
       n4 multiple = 0;

ares/component/eeprom/m24c/m24c.cpp

@@ -74,7 +74,7 @@ auto M24C::write() -> void {
   case Mode::Device:
     if(counter <= 8) {
       device = device << 1 | data();
-    } else if(!select()) {
+    } else if(select() != Acknowledge) {
       mode = Mode::Standby;
     } else if(device & 1) {
       mode = Mode::Read;

ares/component/processor/huc6280/algorithms.cpp

@@ -1,17 +1,15 @@
 auto HuC6280::algorithmADC(n8 i) -> n8 {
-  i16 o;
+  i16 o = A + i + C;
   if(!D) {
-    o = A + i + C;
+    C = o.bit(8);
     V = ~(A ^ i) & (A ^ o) & 0x80;
   } else {
     idle();
-    o = (A & 0x0f) + (i & 0x0f) + (C << 0);
-    if(o > 0x09) o += 0x06;
-    C = o > 0x0f;
-    o = (A & 0xf0) + (i & 0xf0) + (C << 4) + (o & 0x0f);
+    n8 l = (A & 0x0f) + (i & 0x0f) + C;
+    if(l > 0x09) o += 0x06;
     if(o > 0x9f) o += 0x60;
+    C = o > 0xff;
   }
-  C = o.bit(8);
   Z = n8(o) == 0;
   N = o.bit(7);
   return o;
@@ -125,19 +123,17 @@ auto HuC6280::algorithmROR(n8 i) -> n8 {
 
 auto HuC6280::algorithmSBC(n8 i) -> n8 {
   i ^= 0xff;
-  i16 o;
+  i16 o = A + i + C;
   if(!D) {
-    o = A + i + C;
+    C = o.bit(8);
     V = ~(A ^ i) & (A ^ o) & 0x80;
   } else {
     idle();
-    o = (A & 0x0f) + (i & 0x0f) + (C << 0);
-    if(o <= 0x0f) o -= 0x06;
-    C = o > 0x0f;
-    o = (A & 0xf0) + (i & 0xf0) + (C << 4) + (o & 0x0f);
+    n8 l = (A & 0x0f) + (i & 0x0f) + C;
+    C = o > 0xff;
     if(o <= 0xff) o -= 0x60;
+    if(l <= 0x0f) o -= 0x06;
   }
-  C = o.bit(8);
   Z = n8(o) == 0;
   N = o.bit(7);
   return o;

ares/component/processor/m68000/disassembler.cpp

@@ -50,7 +50,7 @@ template<u32 Size> auto M68000::_address(EffectiveAddress& ea) -> string {
   if(ea.mode ==  2) return {_addressRegister(AddressRegister{ea.reg})};
   if(ea.mode ==  5) return {"$", hex(_readDisplacement(read(AddressRegister{ea.reg})), 6L)};
   if(ea.mode ==  6) return {"$", hex(_readIndex(read(AddressRegister{ea.reg})), 6L)};
-  if(ea.mode ==  7) return {"$", hex((i16)_readPC<Word>(), 6L)};
+  if(ea.mode ==  7) { auto imm = _readPC<Word>(); return {"$", imm >= 0x8000 ? hex((i16)imm, 6L, 'f') : hex((i16)imm, 6L, '0') }; }
   if(ea.mode ==  8) return {"$", hex(_readPC<Long>(), 6L)};
   if(ea.mode ==  9) return {"$", hex(_pc + (i16)_readPC(), 6L)};
   if(ea.mode == 10) return {"$", hex(_readIndex(_pc), 6L)};
@@ -65,7 +65,7 @@ template<u32 Size> auto M68000::_effectiveAddress(EffectiveAddress& ea) -> strin
   if(ea.mode ==  4) return {"-(", _addressRegister(AddressRegister{ea.reg}), ")"};
   if(ea.mode ==  5) return {"($", hex(_readDisplacement(read(AddressRegister{ea.reg})), 6L), ")"};
   if(ea.mode ==  6) return {"($", hex(_readIndex(read(AddressRegister{ea.reg})), 6L), ")"};
-  if(ea.mode ==  7) return {"($", hex((i16)_readPC<Word>(), 6L), ")"};
+  if(ea.mode ==  7) { auto imm = _readPC<Word>(); return {"($", imm >= 0x8000 ? hex((i16)imm, 6L, 'f') : hex((i16)imm, 6L, '0'), ")"}; }
   if(ea.mode ==  8) return {"($", hex(_readPC<Long>(), 6L), ")"};
   if(ea.mode ==  9) return {"($", hex(_readDisplacement(_pc), 6L), ")"};
   if(ea.mode == 10) return {"($", hex(_readIndex(_pc), 6L), ")"};

ares/component/processor/m68000/instruction.cpp

@@ -414,6 +414,7 @@ M68000::M68000() {
   for(n3 mode : range(8))
   for(n3 reg  : range(8)) {
     auto opcode = pattern("1011 ---+ 11-- ----") | areg << 9 | mode << 3 | reg << 0;
+    if(mode == 7 && reg >= 5) continue;
 
     AddressRegister with{areg};
     EffectiveAddress from{mode, reg};
@@ -1256,7 +1257,7 @@ M68000::M68000() {
   for(n3 mode : range(8))
   for(n3 reg  : range(8)) {
     auto opcode = pattern("0100 1010 ++-- ----") | mode << 3 | reg << 0;
-    if(mode == 7 && reg >= 2) continue;
+    if(mode == 1 || mode == 7 && reg >= 2) continue;
 
     EffectiveAddress from{mode, reg};
     bind(opcode | 0 << 6, TST<Byte>, from);

ares/component/processor/m68000/instructions.cpp

@@ -1,7 +1,7 @@
 auto M68000::instructionABCD(EffectiveAddress from, EffectiveAddress with) -> void {
   if(from.mode == DataRegisterDirect) idle(2);
-  auto target = read<Byte, Hold, Fast>(with);
   auto source = read<Byte>(from);
+  auto target = read<Byte, Hold, Fast>(with);
   auto result = source + target + r.x;
   bool c = false;
   bool v = false;
@@ -98,8 +98,8 @@ template<u32 Size> auto M68000::instructionADDX(EffectiveAddress from, Effective
   if constexpr(Size == Long) {
     if(from.mode == DataRegisterDirect) idle(4);
   }
-  auto target = read<Size, Hold, Fast>(with);
   auto source = read<Size>(from);
+  auto target = read<Size, Hold, Fast>(with);
   auto result = ADD<Size, Extend>(source, target);
   prefetch();
   write<Size>(with, result);
@@ -332,14 +332,20 @@ auto M68000::instructionCHK(DataRegister compare, EffectiveAddress maximum) -> v
 
   r.z = clip<Word>(target) == 0;
   r.n = sign<Word>(target) < 0;
-  if(r.n) return exception(Exception::BoundsCheck, Vector::BoundsCheck);
+  if(r.n) {
+    prefetched();
+    return exception(Exception::BoundsCheck, Vector::BoundsCheck);
+  }
 
   auto result = (n64)target - source;
   r.c = sign<Word>(result >> 1) < 0;
   r.v = sign<Word>((target ^ source) & (target ^ result)) < 0;
   r.z = clip<Word>(result) == 0;
   r.n = sign<Word>(result) < 0;
-  if(r.n == r.v && !r.z) return exception(Exception::BoundsCheck, Vector::BoundsCheck);
+  if(r.n == r.v && !r.z) {
+    prefetched();
+    return exception(Exception::BoundsCheck, Vector::BoundsCheck);
+  }
   prefetch();
 }
 
@@ -390,22 +396,30 @@ template<u32 Size> auto M68000::instructionCMPM(EffectiveAddress from, Effective
 }
 
 auto M68000::instructionDBCC(n4 test, DataRegister with) -> void {
-  auto displacement = extension<Word>();
-  if(condition(test)) {
-    idle(4);
-  } else {
+  idle(2);
+  r.pc -= 2;
+  if(!condition(test)) { // cc false
+    auto disp = sign<Word>(r.irc);
+    // predict branch
+    r.pc += disp;
+    prefetch();
+    // decrement register
     n16 result = read<Word>(with);
     write<Word>(with, result - 1);
     if(result) {
-      idle(2);
-      r.pc -= 4;
-      r.pc += sign<Word>(displacement);
-      prefetch();
+      // branch taken
+      prefetch(); 
+      return;
     } else {
-      idle(2);
+      // branch not taken
+      r.pc -= disp;
     }
+  } else { // cc true
+    idle(2);
+    r.pc += 2;
   }
   prefetch();
+  prefetch();
 }
 
 auto M68000::instructionDIVS(EffectiveAddress from, DataRegister with) -> void {
@@ -413,6 +427,7 @@ auto M68000::instructionDIVS(EffectiveAddress from, DataRegister with) -> void {
   n32 divisor  = read<Word>(from) << 16, odivisor = divisor;
 
   if(divisor == 0) {
+    prefetched();
     return exception(Exception::DivisionByZero, Vector::DivisionByZero);
   }
 
@@ -476,8 +491,8 @@ auto M68000::instructionDIVS(EffectiveAddress from, DataRegister with) -> void {
     return;
   }
 
-  r.z = quotient == 0;
-  r.n = quotient < 0;
+  r.z = clip<Word>(quotient) == 0;
+  r.n = sign<Word>(quotient) < 0;
 
   idle(ticks);
   write<Long>(with, dividend | quotient);
@@ -489,6 +504,7 @@ auto M68000::instructionDIVU(EffectiveAddress from, DataRegister with) -> void {
   n32 divisor  = read<Word>(from) << 16;
 
   if(divisor == 0) {
+    prefetched();
     return exception(Exception::DivisionByZero, Vector::DivisionByZero);
   }
 
@@ -515,8 +531,8 @@ auto M68000::instructionDIVU(EffectiveAddress from, DataRegister with) -> void {
   ticks += force ? 6 : carry ? 4 : 2;
   quotient = quotient << 1 | carry;
 
-  r.z = quotient == 0;
-  r.n = quotient < 0;
+  r.z = clip<Word>(quotient) == 0;
+  r.n = sign<Word>(quotient) < 0;
 
   idle(ticks);
   write<Long>(with, dividend | quotient);
@@ -1118,8 +1134,8 @@ auto M68000::instructionRTS() -> void {
 
 auto M68000::instructionSBCD(EffectiveAddress from, EffectiveAddress with) -> void {
   if(from.mode == DataRegisterDirect) idle(2);
-  auto target = read<Byte, Hold, Fast>(with);
   auto source = read<Byte>(from);
+  auto target = read<Byte, Hold, Fast>(with);
   auto result = target - source - r.x;
   bool c = false;
   bool v = false;
@@ -1240,8 +1256,8 @@ template<u32 Size> auto M68000::instructionSUBX(EffectiveAddress from, Effective
   if constexpr(Size == Long) {
     if(from.mode == DataRegisterDirect) idle(4);
   }
-  auto target = read<Size, Hold, Fast>(with);
   auto source = read<Size>(from);
+  auto target = read<Size, Hold, Fast>(with);
   auto result = SUB<Size, Extend>(source, target);
   prefetch();
   write<Size>(with, result);
@@ -1262,7 +1278,7 @@ auto M68000::instructionSWAP(DataRegister with) -> void {
 auto M68000::instructionTAS(EffectiveAddress with) -> void {
   n32 data;
 
-  if(with.mode == DataRegisterDirect) {
+  if(lockable() || with.mode == DataRegisterDirect) {
     data = read<Byte, Hold>(with);
     prefetch();
     write<Byte>(with, data | 0x80);

ares/component/processor/m68000/m68000.cpp

@@ -50,21 +50,36 @@ auto M68000::supervisor() -> bool {
 
 auto M68000::exception(u32 exception, u32 vector, u32 priority) -> void {
   r.stop  = false;
-  idle(10);  //todo: not accurate
-
-  auto pc = r.pc;
-  auto sr = readSR();
 
+  // register setup (+6 cyc)
+  idle(6);
+  n32 pc = r.pc - 4;
+  n32 sr = readSR();
   if(!r.s) swap(r.a[7], r.sp);
-  r.i = priority;
   r.s = 1;
   r.t = 0;
 
-  push<Long>(pc - 4);
-  push<Word>(sr);
+  // push pc low (+4 cyc)
+  push<Word>(pc & 0x0000ffff);
 
+  // external interrupt handling
+  if(exception == Exception::Interrupt) {
+    // IACK vector number acquisition (+4 cyc normal or +10 to +18 cyc autovectored)
+    // & justify vector number (+4 cyc)
+    idle(16+4); // assuming autovector cycles (approximated for Megadrive)
+    r.i = priority;
+  }
+
+  // push pc high & sr (+8 cyc)
+  push<Long>(sr << 16 | pc >> 16);
+
+  // read vector address (+8 cyc)
   r.pc = read<Long>(vector << 2);
+  // todo: if the vector address read causes an address error, this exception should be escalated
+
+  // prefetch (+8 cyc)
   prefetch();
+  if(exception == Exception::Interrupt) idle(2); // dead cycles (+2 cyc)
   prefetch();
 }
 

ares/component/processor/m68000/m68000.hpp

@@ -9,8 +9,10 @@ struct M68000 {
   virtual auto wait(u32 clocks) -> void = 0;
   virtual auto read(n1 upper, n1 lower, n24 address, n16 data = 0) -> n16 = 0;
   virtual auto write(n1 upper, n1 lower, n24 address, n16 data) -> void = 0;
+  virtual auto lockable() -> bool { return true; }
 
   auto ird() const -> n16 { return r.ird; }
+  auto irc() const -> n16 { return r.irc; }
 
   enum : bool { User, Supervisor };
   enum : u32  { Byte, Word, Long };
@@ -70,8 +72,8 @@ struct M68000 {
   M68000();
   auto power() -> void;
   auto supervisor() -> bool;
-  auto exception(u32 exception, u32 vector, u32 priority = 7) -> void;
-  auto interrupt(u32 vector, u32 priority = 7) -> void;
+  auto exception(u32 exception, u32 vector, u32 priority = 0) -> void;
+  auto interrupt(u32 vector, u32 priority = 0) -> void;
 
   //registers.cpp
   struct DataRegister {

ares/component/processor/mos6502/algorithms.cpp

@@ -1,19 +1,32 @@
 auto MOS6502::algorithmADC(n8 i) -> n8 {
-  i16 o;
+  i16 o = A + i + C;
   if(!BCD || !D) {
-    o = A + i + C;
+    C = o.bit(8);
+    Z = n8(o) == 0;
+    N = o.bit(7);
     V = ~(A ^ i) & (A ^ o) & 0x80;
   } else {
     idle();
+    Z = n8(o) == 0;
     o = (A & 0x0f) + (i & 0x0f) + (C << 0);
     if(o > 0x09) o += 0x06;
     C = o > 0x0f;
     o = (A & 0xf0) + (i & 0xf0) + (C << 4) + (o & 0x0f);
+    N = o.bit(7);
+    V = ~(A ^ i) & (A ^ o) & 0x80;
     if(o > 0x9f) o += 0x60;
+    C = o > 0xff;
   }
-  C = o.bit(8);
-  Z = n8(o) == 0;
+  return o;
+}
+
+auto MOS6502::algorithmALR(n8 i) -> n8 {
+  n8 o = A & i;
+  C = o.bit(0);
+  o >>= 1;
+  Z = o == 0;
   N = o.bit(7);
+
   return o;
 }
 
@@ -24,6 +37,29 @@ auto MOS6502::algorithmAND(n8 i) -> n8 {
   return o;
 }
 
+auto MOS6502::algorithmANC(n8 i) -> n8 {
+  n8 o = A & i;
+  Z = o == 0;
+  N = o.bit(7);
+  C = N;
+  return o;
+}
+
+auto MOS6502::algorithmARR(n8 i) -> n8 {
+  n8 o = A & i;
+  Z = o == 0;
+  N = o.bit(7);
+
+  bool c = C;
+  o = c << 7 | o >> 1;
+  Z = o == 0;
+  N = o.bit(7);
+  C = o.bit(6);
+  V = o.bit(6) ^ o.bit(5);
+
+  return o;
+}
+
 auto MOS6502::algorithmASL(n8 i) -> n8 {
   C = i.bit(7);
   i <<= 1;
@@ -32,6 +68,23 @@ auto MOS6502::algorithmASL(n8 i) -> n8 {
   return i;
 }
 
+auto MOS6502::algorithmATX(n8 i) -> n8 {
+  A |= 0xff; //TODO: OR value may differ on non-NES platforms
+  n8 o = A & i;
+  X = o;
+  Z = o == 0;
+  N = o.bit(7);
+  return o;
+}
+
+auto MOS6502::algorithmAXS(n8 i) -> n8 {
+  n9 o = (A & X) - i;
+  C = !o.bit(8);
+  Z = n8(o) == 0;
+  N = o.bit(7);
+  return o;
+}
+
 auto MOS6502::algorithmBIT(n8 i) -> n8 {
   Z = (A & i) == 0;
   V = i.bit(6);
@@ -125,20 +178,35 @@ auto MOS6502::algorithmROR(n8 i) -> n8 {
 
 auto MOS6502::algorithmSBC(n8 i) -> n8 {
   i = ~i;
-  i16 o;
+  i16 o = A + i + C;
   if(!BCD || !D) {
-    o = A + i + C;
+    C = o.bit(8);
+    Z = n8(o) == 0;
+    N = o.bit(7);
     V = ~(A ^ i) & (A ^ o) & 0x80;
   } else {
     idle();
+    Z = n8(o) == 0;
     o = (A & 0x0f) + (i & 0x0f) + (C << 0);
     if(o <= 0x0f) o -= 0x06;
     C = o > 0x0f;
     o = (A & 0xf0) + (i & 0xf0) + (C << 4) + (o & 0x0f);
+    N = o.bit(7);
+    V = ~(A ^ i) & (A ^ o) & 0x80;
     if(o <= 0xff) o -= 0x60;
+    C = o > 0xff;
   }
-  C = o.bit(8);
-  Z = n8(o) == 0;
-  N = o.bit(7);
   return o;
 }
+
+auto MOS6502::algorithmSLO(n8 i) -> n8 {
+  C = i.bit(7);
+  i <<= 1;
+  Z = i == 0;
+  N = i.bit(7);
+
+  A |= i;
+  Z = A == 0;
+  N = A.bit(7);
+  return i;
+}

ares/component/processor/mos6502/disassembler.cpp

@@ -57,19 +57,24 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   switch(auto opcode = readDebugger(pc)) {
   op(0x00, "brk", immediate)
   op(0x01, "ora", indirectX)
+  op(0x04, "nop", zeroPage)
   op(0x05, "ora", zeroPage)
   op(0x06, "asl", zeroPage)
   op(0x08, "php", implied)
   op(0x09, "ora", immediate)
   op(0x0a, "asl", implied)
+  op(0x0c, "nop", absolute)
   op(0x0d, "ora", absolute)
   op(0x0e, "asl", absolute)
   op(0x10, "bpl", branch)
   op(0x11, "ora", indirectY)
+  op(0x14, "nop", zeroPageX)
   op(0x15, "ora", zeroPageX)
   op(0x16, "asl", zeroPageX)
   op(0x18, "clc", implied)
   op(0x19, "ora", absoluteY)
+  op(0x1a, "nop", implied)
+  op(0x1c, "nop", absoluteX)
   op(0x1d, "ora", absoluteX)
   op(0x1e, "asl", absoluteX)
   op(0x20, "jsr", absolute)
@@ -85,14 +90,18 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0x2e, "rol", absolute)
   op(0x30, "bmi", branch)
   op(0x31, "and", indirectY)
+  op(0x34, "nop", zeroPageX)
   op(0x35, "and", zeroPageX)
   op(0x36, "rol", zeroPageX)
   op(0x38, "sec", implied)
   op(0x39, "and", absoluteY)
+  op(0x3a, "nop", implied)
+  op(0x3c, "nop", absoluteX)
   op(0x3d, "and", absoluteX)
   op(0x3e, "rol", absoluteX)
   op(0x40, "rti", implied)
   op(0x41, "eor", indirectX)
+  op(0x44, "nop", zeroPage)
   op(0x45, "eor", zeroPage)
   op(0x46, "lsr", zeroPage)
   op(0x48, "pha", implied)
@@ -103,14 +112,18 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0x4e, "lsr", absolute)
   op(0x50, "bvc", branch)
   op(0x51, "eor", indirectY)
+  op(0x54, "nop", zeroPageX)
   op(0x55, "eor", zeroPageX)
   op(0x56, "lsr", zeroPageX)
   op(0x58, "cli", implied)
   op(0x59, "eor", absoluteY)
+  op(0x5a, "nop", implied)
+  op(0x5c, "nop", absoluteX)
   op(0x5d, "eor", absoluteX)
   op(0x5e, "lsr", absoluteX)
   op(0x60, "rts", implied)
   op(0x61, "adc", indirectX)
+  op(0x64, "nop", zeroPage)
   op(0x65, "adc", zeroPage)
   op(0x66, "ror", zeroPage)
   op(0x68, "pla", implied)
@@ -121,17 +134,23 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0x6e, "ror", absolute)
   op(0x70, "bvs", branch)
   op(0x71, "adc", indirectY)
+  op(0x74, "nop", zeroPageX)
   op(0x75, "adc", zeroPageX)
   op(0x76, "ror", zeroPageX)
   op(0x78, "sei", implied)
   op(0x79, "adc", absoluteY)
+  op(0x7a, "nop", implied)
+  op(0x7c, "nop", absoluteX)
   op(0x7d, "adc", absoluteX)
   op(0x7e, "ror", absoluteX)
+  op(0x80, "nop", immediate)
   op(0x81, "sta", indirectX)
+  op(0x82, "nop", immediate)
   op(0x84, "sty", zeroPage)
   op(0x85, "sta", zeroPage)
   op(0x86, "stx", zeroPage)
   op(0x88, "dey", implied)
+  op(0x89, "nop", immediate)
   op(0x8a, "txa", implied)
   op(0x8c, "sty", absolute)
   op(0x8d, "sta", absolute)
@@ -170,6 +189,7 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0xbe, "ldx", absoluteY)
   op(0xc0, "cpy", immediate)
   op(0xc1, "cmp", indirectX)
+  op(0xc2, "nop", immediate)
   op(0xc4, "cpy", zeroPage)
   op(0xc5, "cmp", zeroPage)
   op(0xc6, "dec", zeroPage)
@@ -181,14 +201,18 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0xce, "dec", absolute)
   op(0xd0, "bne", branch)
   op(0xd1, "cmp", indirectY)
+  op(0xd4, "nop", zeroPageX)
   op(0xd5, "cmp", zeroPageX)
   op(0xd6, "dec", zeroPageX)
   op(0xd8, "cld", implied)
   op(0xd9, "cmp", absoluteY)
+  op(0xda, "nop", implied)
+  op(0xdc, "nop", absoluteX)
   op(0xdd, "cmp", absoluteX)
   op(0xde, "dec", absoluteX)
   op(0xe0, "cpx", immediate)
   op(0xe1, "sbc", indirectX)
+  op(0xe2, "nop", immediate)
   op(0xe4, "cpx", zeroPage)
   op(0xe5, "sbc", zeroPage)
   op(0xe6, "inc", zeroPage)
@@ -200,10 +224,13 @@ auto MOS6502::disassembleInstruction(maybe<n16> _pc) -> string {
   op(0xee, "inc", absolute)
   op(0xf0, "beq", branch)
   op(0xf1, "sbc", indirectY)
+  op(0xf4, "nop", zeroPageX)
   op(0xf5, "sbc", zeroPageX)
   op(0xf6, "inc", zeroPageX)
   op(0xf8, "sed", implied)
   op(0xf9, "sbc", absoluteY)
+  op(0xfa, "nop", implied)
+  op(0xfc, "nop", absoluteX)
   op(0xfd, "sbc", absoluteX)
   op(0xfe, "inc", absoluteX)
 

ares/component/processor/mos6502/instruction.cpp

@@ -20,21 +20,32 @@ auto MOS6502::instruction() -> void {
   switch(code) {
   op(0x00, Break)
   op(0x01, IndirectXRead, fp(ORA), A)
+  op(0x04, NoOperationZeroPage)
   op(0x05, ZeroPageRead, fp(ORA), A)
   op(0x06, ZeroPageModify, fp(ASL))
+  op(0x07, ZeroPageModify, fp(SLO))
   op(0x08, PushP)
   op(0x09, Immediate, fp(ORA), A)
   op(0x0a, Implied, fp(ASL), A)
+  op(0x0b, Immediate, fp(ANC), A)
+  op(0x0c, NoOperationAbsolute)
   op(0x0d, AbsoluteRead, fp(ORA), A)
   op(0x0e, AbsoluteModify, fp(ASL))
+  op(0x0f, AbsoluteModify, fp(SLO))
   op(0x10, Branch, N == 0)
   op(0x11, IndirectYRead, fp(ORA), A)
+  op(0x14, NoOperationZeroPage, X)
   op(0x15, ZeroPageRead, fp(ORA), A, X)
   op(0x16, ZeroPageModify, fp(ASL), X)
+  op(0x17, ZeroPageModify, fp(SLO), X)
   op(0x18, Clear, C)
   op(0x19, AbsoluteRead, fp(ORA), A, Y)
+  op(0x1a, NoOperation)
+  op(0x1b, AbsoluteModify, fp(SLO), Y)
+  op(0x1c, NoOperationAbsolute, X)
   op(0x1d, AbsoluteRead, fp(ORA), A, X)
   op(0x1e, AbsoluteModify, fp(ASL), X)
+  op(0x1f, AbsoluteModify, fp(SLO), X)
   op(0x20, CallAbsolute)
   op(0x21, IndirectXRead, fp(AND), A)
   op(0x24, ZeroPageRead, fp(BIT), A)
@@ -43,58 +54,75 @@ auto MOS6502::instruction() -> void {
   op(0x28, PullP)
   op(0x29, Immediate, fp(AND), A)
   op(0x2a, Implied, fp(ROL), A)
+  op(0x2b, Immediate, fp(ANC), A)
   op(0x2c, AbsoluteRead, fp(BIT), A)
   op(0x2d, AbsoluteRead, fp(AND), A)
   op(0x2e, AbsoluteModify, fp(ROL))
   op(0x30, Branch, N == 1)
   op(0x31, IndirectYRead, fp(AND), A)
+  op(0x34, NoOperationZeroPage, X)
   op(0x35, ZeroPageRead, fp(AND), A, X)
   op(0x36, ZeroPageModify, fp(ROL), X)
   op(0x38, Set, C)
   op(0x39, AbsoluteRead, fp(AND), A, Y)
+  op(0x3a, NoOperation)
+  op(0x3c, NoOperationAbsolute, X)
   op(0x3d, AbsoluteRead, fp(AND), A, X)
   op(0x3e, AbsoluteModify, fp(ROL), X)
   op(0x40, ReturnInterrupt)
   op(0x41, IndirectXRead, fp(EOR), A)
+  op(0x44, NoOperationZeroPage)
   op(0x45, ZeroPageRead, fp(EOR), A)
   op(0x46, ZeroPageModify, fp(LSR))
   op(0x48, Push, A)
   op(0x49, Immediate, fp(EOR), A)
   op(0x4a, Implied, fp(LSR), A)
+  op(0x4b, Immediate, fp(ALR), A)
   op(0x4c, JumpAbsolute)
   op(0x4d, AbsoluteRead, fp(EOR), A)
   op(0x4e, AbsoluteModify, fp(LSR))
   op(0x50, Branch, V == 0)
   op(0x51, IndirectYRead, fp(EOR), A)
+  op(0x54, NoOperationZeroPage, X)
   op(0x55, ZeroPageRead, fp(EOR), A, X)
   op(0x56, ZeroPageModify, fp(LSR), X)
   op(0x58, Clear, I)
   op(0x59, AbsoluteRead, fp(EOR), A, Y)
+  op(0x5a, NoOperation)
+  op(0x5c, NoOperationAbsolute, X)
   op(0x5d, AbsoluteRead, fp(EOR), A, X)
   op(0x5e, AbsoluteModify, fp(LSR), X)
   op(0x60, ReturnSubroutine)
   op(0x61, IndirectXRead, fp(ADC), A)
+  op(0x64, NoOperationZeroPage)
   op(0x65, ZeroPageRead, fp(ADC), A)
   op(0x66, ZeroPageModify, fp(ROR))
   op(0x68, Pull, A)
   op(0x69, Immediate, fp(ADC), A)
   op(0x6a, Implied, fp(ROR), A)
+  op(0x6b, Immediate, fp(ARR), A)
   op(0x6c, JumpIndirect)
   op(0x6d, AbsoluteRead, fp(ADC), A)
   op(0x6e, AbsoluteModify, fp(ROR))
   op(0x70, Branch, V == 1)
   op(0x71, IndirectYRead, fp(ADC), A)
+  op(0x74, NoOperationZeroPage, X)
   op(0x75, ZeroPageRead, fp(ADC), A, X)
   op(0x76, ZeroPageModify, fp(ROR), X)
   op(0x78, Set, I)
   op(0x79, AbsoluteRead, fp(ADC), A, Y)
+  op(0x7a, NoOperation)
+  op(0x7c, NoOperationAbsolute, X)
   op(0x7d, AbsoluteRead, fp(ADC), A, X)
   op(0x7e, AbsoluteModify, fp(ROR), X)
+  op(0x80, NoOperationImmediate)
   op(0x81, IndirectXWrite, A)
+  op(0x82, NoOperationImmediate)
   op(0x84, ZeroPageWrite, Y)
   op(0x85, ZeroPageWrite, A)
   op(0x86, ZeroPageWrite, X)
   op(0x88, Implied, fp(DEC), Y)
+  op(0x89, NoOperationImmediate)
   op(0x8a, Transfer, X, A, 1)
   op(0x8c, AbsoluteWrite, Y)
   op(0x8d, AbsoluteWrite, A)
@@ -117,11 +145,13 @@ auto MOS6502::instruction() -> void {
   op(0xa8, Transfer, A, Y, 1)
   op(0xa9, Immediate, fp(LD), A)
   op(0xaa, Transfer, A, X, 1)
+  op(0xab, Immediate, fp(ATX), A)
   op(0xac, AbsoluteRead, fp(LD), Y)
   op(0xad, AbsoluteRead, fp(LD), A)
   op(0xae, AbsoluteRead, fp(LD), X)
   op(0xb0, Branch, C == 1)
   op(0xb1, IndirectYRead, fp(LD), A)
+  op(0xb3, IndirectYRead, fp(LD), A, X)
   op(0xb4, ZeroPageRead, fp(LD), Y, X)
   op(0xb5, ZeroPageRead, fp(LD), A, X)
   op(0xb6, ZeroPageRead, fp(LD), X, Y)
@@ -133,45 +163,56 @@ auto MOS6502::instruction() -> void {
   op(0xbe, AbsoluteRead, fp(LD), X, Y)
   op(0xc0, Immediate, fp(CPY), Y)
   op(0xc1, IndirectXRead, fp(CMP), A)
+  op(0xc2, NoOperationImmediate)
   op(0xc4, ZeroPageRead, fp(CPY), Y)
   op(0xc5, ZeroPageRead, fp(CMP), A)
   op(0xc6, ZeroPageModify, fp(DEC))
   op(0xc8, Implied, fp(INC), Y)
   op(0xc9, Immediate, fp(CMP), A)
   op(0xca, Implied, fp(DEC), X)
+  op(0xcb, Immediate, fp(AXS), X)
   op(0xcc, AbsoluteRead, fp(CPY), Y)
   op(0xcd, AbsoluteRead, fp(CMP), A)
   op(0xce, AbsoluteModify, fp(DEC))
   op(0xd0, Branch, Z == 0)
   op(0xd1, IndirectYRead, fp(CMP), A)
+  op(0xd4, NoOperationZeroPage, X)
   op(0xd5, ZeroPageRead, fp(CMP), A, X)
   op(0xd6, ZeroPageModify, fp(DEC), X)
   op(0xd8, Clear, D)
   op(0xd9, AbsoluteRead, fp(CMP), A, Y)
+  op(0xda, NoOperation)
+  op(0xdc, NoOperationAbsolute, X)
   op(0xdd, AbsoluteRead, fp(CMP), A, X)
   op(0xde, AbsoluteModify, fp(DEC), X)
   op(0xe0, Immediate, fp(CPX), X)
   op(0xe1, IndirectXRead, fp(SBC), A)
+  op(0xe2, NoOperationImmediate)
   op(0xe4, ZeroPageRead, fp(CPX), X)
   op(0xe5, ZeroPageRead, fp(SBC), A)
   op(0xe6, ZeroPageModify, fp(INC))
   op(0xe8, Implied, fp(INC), X)
   op(0xe9, Immediate, fp(SBC), A)
   op(0xea, NoOperation)
+  op(0xeb, Immediate, fp(SBC), A)
   op(0xec, AbsoluteRead, fp(CPX), X)
   op(0xed, AbsoluteRead, fp(SBC), A)
   op(0xee, AbsoluteModify, fp(INC))
   op(0xf0, Branch, Z == 1)
   op(0xf1, IndirectYRead, fp(SBC), A)
+  op(0xf4, NoOperationZeroPage, X)
   op(0xf5, ZeroPageRead, fp(SBC), A, X)
   op(0xf6, ZeroPageModify, fp(INC), X)
   op(0xf8, Set, D)
   op(0xf9, AbsoluteRead, fp(SBC), A, Y)
+  op(0xfa, NoOperation)
+  op(0xfc, NoOperationAbsolute, X)
   op(0xfd, AbsoluteRead, fp(SBC), A, X)
   op(0xfe, AbsoluteModify, fp(INC), X)
   }
 
   //unimplemented instruction
+  debug(unusual, "[6502] unimplemented opcode ", hex(code));
   return instructionNoOperation();
 }
 

ares/component/processor/mos6502/instructions.cpp

@@ -112,6 +112,11 @@ auto MOS6502::instructionIndirectYRead(fp alu, n8& data) -> void {
 L data = ALU(read(absolute + Y));
 }
 
+auto MOS6502::instructionIndirectYRead(fp alu, n8& data, n8& data2) -> void {
+  instructionIndirectYRead(alu, data);
+  data2 = data;
+}
+
 auto MOS6502::instructionIndirectYWrite(n8& data) -> void {
   auto zeroPage = operand();
   n16 absolute = load(zeroPage + 0);
@@ -139,6 +144,34 @@ auto MOS6502::instructionNoOperation() -> void {
 L idle();
 }
 
+auto MOS6502::instructionNoOperationAbsolute() -> void {
+  n16 absolute = operand();
+  absolute |= operand() << 8;
+L read(absolute);
+}
+
+auto MOS6502::instructionNoOperationAbsolute(n8 index) -> void {
+  n16 absolute = operand();
+  absolute |= operand() << 8;
+  idlePageCrossed(absolute, absolute + index);
+L read(absolute + index);
+}
+
+auto MOS6502::instructionNoOperationImmediate() -> void {
+L operand();
+}
+
+auto MOS6502::instructionNoOperationZeroPage() -> void {
+  auto zeroPage = operand();
+L load(zeroPage);
+}
+
+auto MOS6502::instructionNoOperationZeroPage(n8 index) -> void {
+  auto zeroPage = operand();
+  load(zeroPage);
+L load(zeroPage + index);
+}
+
 auto MOS6502::instructionPull(n8& data) -> void {
   idle();
   idle();

ares/component/processor/mos6502/mos6502.hpp

@@ -31,7 +31,12 @@ struct MOS6502 {
   using fp = auto (MOS6502::*)(n8) -> n8;
   auto algorithmADC(n8) -> n8;
   auto algorithmAND(n8) -> n8;
+  auto algorithmANC(n8) -> n8;
+  auto algorithmARR(n8) -> n8;
   auto algorithmASL(n8) -> n8;
+  auto algorithmATX(n8) -> n8;
+  auto algorithmALR(n8) -> n8;
+  auto algorithmAXS(n8) -> n8;
   auto algorithmBIT(n8) -> n8;
   auto algorithmCMP(n8) -> n8;
   auto algorithmCPX(n8) -> n8;
@@ -45,6 +50,7 @@ struct MOS6502 {
   auto algorithmROL(n8) -> n8;
   auto algorithmROR(n8) -> n8;
   auto algorithmSBC(n8) -> n8;
+  auto algorithmSLO(n8) -> n8;
 
   //instruction.cpp
   auto interrupt() -> void;
@@ -66,10 +72,16 @@ struct MOS6502 {
   auto instructionIndirectXRead(fp alu, n8& data) -> void;
   auto instructionIndirectXWrite(n8& data) -> void;
   auto instructionIndirectYRead(fp alu, n8& data) -> void;
+  auto instructionIndirectYRead(fp, n8& data, n8& data2) -> void;
   auto instructionIndirectYWrite(n8& data) -> void;
   auto instructionJumpAbsolute() -> void;
   auto instructionJumpIndirect() -> void;
   auto instructionNoOperation() -> void;
+  auto instructionNoOperationAbsolute() -> void;
+  auto instructionNoOperationAbsolute(n8 index) -> void;
+  auto instructionNoOperationImmediate() -> void;
+  auto instructionNoOperationZeroPage() -> void;
+  auto instructionNoOperationZeroPage(n8 index) -> void;
   auto instructionPull(n8& data) -> void;
   auto instructionPullP() -> void;
   auto instructionPush(n8& data) -> void;

ares/component/processor/sh2/cached.cpp

@@ -1,3 +1,11 @@
+auto SH2::Recompiler::invalidate(u32 address) -> void {
+  auto pool = pools[address >> 8 & 0xffffff];
+  if(!pool) return;
+  memory::jitprotect(false);
+  pool->blocks[address >> 1 & 0x7f] = nullptr;
+  memory::jitprotect(true);
+}
+
 auto SH2::Recompiler::pool(u32 address) -> Pool* {
   auto& pool = pools[address >> 8 & 0xffffff];
   if(!pool) pool = (Pool*)allocator.acquire(sizeof(Pool));
@@ -7,7 +15,9 @@ auto SH2::Recompiler::pool(u32 address) -> Pool* {
 auto SH2::Recompiler::block(u32 address) -> Block* {
   if(auto block = pool(address)->blocks[address >> 1 & 0x7f]) return block;
   auto block = emit(address);
-  return pool(address)->blocks[address >> 1 & 0x7f] = block;
+  pool(address)->blocks[address >> 1 & 0x7f] = block;
+  memory::jitprotect(true);
+  return block;
 }
 
 alwaysinline auto SH2::Recompiler::emitInstruction(u16 opcode) -> bool {
@@ -25,34 +35,37 @@ alwaysinline auto SH2::Recompiler::emitInstruction(u16 opcode) -> bool {
   return 0;
 }
 
+#define CCR mem(sreg(1), offsetof(Registers, CCR))
+
 auto SH2::Recompiler::emit(u32 address) -> Block* {
   if(unlikely(allocator.available() < 1_MiB)) {
     print("SH2 allocator flush\n");
+    memory::jitprotect(false);
     allocator.release(bump_allocator::zero_fill);
+    memory::jitprotect(true);
     reset();
   }
 
   auto block = (Block*)allocator.acquire(sizeof(Block));
-  block->code = allocator.acquire();
-  bind({block->code, allocator.available()});
+  beginFunction(2);
 
   bool hasBranched = 0;
   while(true) {
     u16 instruction = self.readWord(address);
     bool branched = emitInstruction(instruction);
-    mov(rax, mem64(&self.CCR));
-    inc(rax);
-    mov(mem64(&self.CCR), rax);
-    call(&SH2::instructionEpilogue, &self);
+    add64(CCR, CCR, imm(1));
+    call(&SH2::instructionEpilogue);
     address += 2;
     if(hasBranched || (address & 0xfe) == 0) break;  //block boundary
     hasBranched = branched;
-    test(al, al);
-    jz(imm8(1));
-    ret();
+    testJumpEpilog();
   }
-  ret();
+  jumpEpilog();
+
+  memory::jitprotect(false);
+  block->code = endFunction();
 
-  allocator.reserve(size());
   return block;
 }
+
+#undef CCR

ares/component/processor/sh2/disassembler.cpp

@@ -1,8 +1,8 @@
 template<typename... P>
 auto SH2::hint(P&&... p) const -> string {
   if(1) return {};
-  if(1) return {"\e[0m\e[37m", forward<P>(p)..., "\e[0m"};
-  return {forward<P>(p)...};
+  if(1) return {"\e[0m\e[37m", std::forward<P>(p)..., "\e[0m"};
+  return {std::forward<P>(p)...};
 }
 
 auto SH2::disassembleInstruction() -> string {

ares/component/processor/sh2/exceptions.cpp

@@ -34,22 +34,16 @@ auto SH2::push(u32 data) -> void {
 
 auto SH2::interrupt(u8 level, u8 vector) -> void {
   push(SR);
-  push(PC);
+  push(PC - 4);
   jump(readLong(VBR + vector * 4) + 4);
   SR.I = level;
 }
 
-auto SH2::exception(u8 vector) -> void {
-  push(SR);
-  push(PC - 2);
-  jump(readLong(VBR + vector * 4) + 4);
-}
-
 auto SH2::addressErrorCPU() -> void {
   static constexpr u8 vector = 0x09;
   SP &= ~3;  //not accurate; but prevents infinite recursion
   push(SR);
-  push(PC);
+  push(PC - 4);
   jump(readLong(VBR + vector * 4) + 4);
 }
 
@@ -57,23 +51,23 @@ auto SH2::addressErrorDMA() -> void {
   static constexpr u8 vector = 0x0a;
   SP &= ~3;  //not accurate; but prevents infinite recursion
   push(SR);
-  push(PC);
+  push(PC - 4);
   jump(readLong(VBR + vector * 4) + 4);
 }
 
 auto SH2::illegalInstruction() -> void {
   if(inDelaySlot()) return illegalSlotInstruction();
+  debug(unusual, "[SH2] illegal instruction: 0x", hex(busReadWord(PC - 4), 4L), " @ 0x", hex(PC - 4));
   static constexpr u8 vector = 0x04;
   push(SR);
-  push(PC);
+  push(PC - 4);
   jump(readLong(VBR + vector * 4) + 4);
-  debug(unusual, "[SH2] illegal instruction: 0x", hex(busReadWord(PC - 4), 4L));
 }
 
 auto SH2::illegalSlotInstruction() -> void {
+  debug(unusual, "[SH2] illegal slot instruction: 0x", hex(busReadWord(PC - 4), 4L));
   static constexpr u8 vector = 0x06;
   push(SR);
-  push(PC - 2);
+  push(PPC - 4);
   jump(readLong(VBR + vector * 4) + 4);
-  debug(unusual, "[SH2] illegal slot instruction: 0x", hex(busReadWord(PC - 4), 4L));
 }

ares/component/processor/sh2/instruction.cpp

@@ -37,7 +37,7 @@ auto SH2::instruction() -> void {
     do {
       auto block = recompiler.block(PC - 4);
       block->execute(*this);
-    } while (CCR <= recompiler.min_cycles);
+    } while (CCR < cyclesUntilSync);
 
     step(CCR);
     CCR = 0;
@@ -45,7 +45,7 @@ auto SH2::instruction() -> void {
   }
 }
 
-auto SH2::instructionEpilogue() -> bool {
+auto SH2::instructionEpilogue() -> s32 {
   switch(PPM) {
   case Branch::Step: PC = PC + 2; return 0;
   case Branch::Slot: PC = PC + 2; PPM = Branch::Take; return 0;

ares/component/processor/sh2/instructions.cpp

@@ -5,9 +5,10 @@ auto SH2::ADD(u32 m, u32 n) -> void {
 
 //ADDC Rm,Rn
 auto SH2::ADDC(u32 m, u32 n) -> void {
-  bool c = (u64)R[n] + R[m] >> 32;
-  R[n] = R[n] + R[m] + SR.T;
-  SR.T = c;
+  u32 a = R[n];
+  u32 b = a + R[m];
+  R[n] = b + SR.T;
+  SR.T = a > b || b > R[n];
 }
 
 //ADD #imm,Rn
@@ -17,9 +18,11 @@ auto SH2::ADDI(u32 i, u32 n) -> void {
 
 //ADDV Rm,Rn
 auto SH2::ADDV(u32 m, u32 n) -> void {
-  bool v = ~(R[n] ^ R[m]) & (R[n] ^ R[n] + R[m]) >> 31;
+  u32 d = ((s32)R[n] < 0);
+  u32 s = ((s32)R[m] < 0) + d;
   R[n] += R[m];
-  SR.T = v;
+  u32 r = ((s32)R[n] < 0) + d;
+  SR.T = s != 1 ? r == 1 : 0;
 }
 
 //AND Rm,Rn
@@ -140,7 +143,11 @@ auto SH2::CMPPZ(u32 n) -> void {
 
 //CMP/STR Rm,Rn
 auto SH2::CMPSTR(u32 m, u32 n) -> void {
-  SR.T = bool(R[n] ^ R[m]);
+  u32 t = R[n] ^ R[m];
+  SR.T = !(t & 0xff << 24)
+      || !(t & 0xff << 16)
+      || !(t & 0xff << 8)
+      || !(t & 0xff);
 }
 
 //DIV0S Rm,Rn
@@ -353,8 +360,9 @@ auto SH2::MOVBLG(u32 d) -> void {
 
 //MOV.B Rm,@-Rn
 auto SH2::MOVBM(u32 m, u32 n) -> void {
+  u32 t = R[m];
   R[n] -= 1;
-  writeByte(R[n], R[m]);
+  writeByte(R[n], t);
 }
 
 //MOV.B @Rm+,Rn
@@ -405,8 +413,9 @@ auto SH2::MOVWLG(u32 d) -> void {
 
 //MOV.W Rm,@-Rn
 auto SH2::MOVWM(u32 m, u32 n) -> void {
+  u32 t = R[m];
   R[n] -= 2;
-  writeWord(R[n], R[m]);
+  writeWord(R[n], t);
 }
 
 //MOV.W @Rm+,Rn
@@ -457,8 +466,9 @@ auto SH2::MOVLLG(u32 d) -> void {
 
 //MOV.L Rm,@-Rn
 auto SH2::MOVLM(u32 m, u32 n) -> void {
+  u32 t = R[m];
   R[n] -= 4;
-  writeLong(R[n], R[m]);
+  writeLong(R[n], t);
 }
 
 //MOV.L @Rm+,Rn
@@ -493,18 +503,21 @@ auto SH2::MOVI(u32 i, u32 n) -> void {
 }
 
 //MOV.W @(disp,PC),Rn
-auto SH2::MOVWI(u32 d, u32 n) -> void {
-  R[n] = (s16)readWord(PC + d * 2);
+noinline auto SH2::MOVWI(u32 d, u32 n) -> void {
+  u32 pc = inDelaySlot() ? PPC - 2 : PC;
+  R[n] = (s16)readWord(pc + d * 2);
 }
 
 //MOV.L @(disp,PC),Rn
-auto SH2::MOVLI(u32 d, u32 n) -> void {
-  R[n] = readLong((PC & ~3) + d * 4);
+noinline auto SH2::MOVLI(u32 d, u32 n) -> void {
+  u32 pc = inDelaySlot() ? PPC - 2 : PC;
+  R[n] = readLong((pc & ~3) + d * 4);
 }
 
 //MOVA @(disp,PC),R0
-auto SH2::MOVA(u32 d) -> void {
-  R[0] = (PC & ~3) + d * 4 - inDelaySlot() * 2;
+noinline auto SH2::MOVA(u32 d) -> void {
+  u32 pc = inDelaySlot() ? PPC - 2 : PC;
+  R[0] = (pc & ~3) + d * 4;
 }
 
 //MOVT Rn
@@ -536,7 +549,7 @@ auto SH2::NEG(u32 m, u32 n) -> void {
 auto SH2::NEGC(u32 m, u32 n) -> void {
   u32 t = 0 - R[m];
   R[n] = t - SR.T;
-  SR.T = (s32)t >= 0 || t < R[n];
+  SR.T = 0 < t || t < R[n];
 }
 
 //NOP
@@ -592,7 +605,7 @@ auto SH2::ROTR(u32 n) -> void {
 
 //RTE
 auto SH2::RTE() -> void {
-  delaySlot(readLong(SP + 0));
+  delaySlot(readLong(SP + 0) + 4);
   SR  = readLong(SP + 4);
   SP += 8;
 }
@@ -790,10 +803,7 @@ auto SH2::TAS(u32 n) -> void {
     ) return (void)(exceptions |= AddressErrorCPU);
   }
 
-  auto cacheEnable = cache.enable;
-  cache.enable = 0;
-  u8 b = readByte(R[n]);
-  cache.enable = cacheEnable;
+  u8 b = readByte<Bus::Internal>(R[n]);
   SR.T = b == 0;
   writeByte(R[n], b | 0x80);
 }
@@ -801,7 +811,7 @@ auto SH2::TAS(u32 n) -> void {
 //TRAPA #imm
 auto SH2::TRAPA(u32 i) -> void {
   push(SR);
-  push(PC + 2);
+  push(PC - 2);
   branch(readLong(VBR + i * 4) + 4);
 }