pax_global_header00006660000000000000000000000064146450674210014523gustar00rootroot0000000000000052 comment=2798e65d6fc89d89aace0b0d779af6406809b940 mupen64plus-rsp-hle-src-2.6.0/000077500000000000000000000000001464506742100160675ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/.gitattributes000066400000000000000000000007661464506742100207730ustar00rootroot00000000000000* text=auto # normal text files *.6 text AUTHORS text *.c text *.cfg text *.cht text *.conf text COPYING text *.cpp text *.def text *-license text *.h text *.html text *.ini text INSTALL text LICENSES text Makefile text *.py text README text RELEASE text *.S text *.sh text *.txt text *.ver text # windows specific text files *.sln text eol=crlf *.vcproj text eol=crlf *.vcxproj text eol=crlf *.vcxproj.filters text eol=crlf # binary files *.gz binary *.ttf binary cursor.tex binary font.tex binary mupen64plus-rsp-hle-src-2.6.0/.github/000077500000000000000000000000001464506742100174275ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/.github/workflows/000077500000000000000000000000001464506742100214645ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/.github/workflows/build.yml000066400000000000000000000112421464506742100233060ustar00rootroot00000000000000name: HLE RSP on: push: paths-ignore: - '.{gitattributes,gitignore,travis.yml}' - '*.md,appveyor.yml,README' pull_request: paths-ignore: - '.{gitattributes,gitignore,travis.yml}' - '*.md,appveyor.yml,README' workflow_dispatch: jobs: Linux: strategy: fail-fast: false matrix: include: - cc: GCC arch: x64 - cc: GCC arch: x86 - cc: Clang arch: x64 - cc: Clang arch: x86 name: Linux / ${{ matrix.cc }} / ${{ matrix.arch }} runs-on: ubuntu-22.04 steps: - uses: actions/checkout@v4 - name: Get build dependencies and arrange the environment run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core export C_CLANG_SUFFIX="-15" C_GCC_SUFFIX="-12" ./../mupen64plus-core/.github/workflows/scripts/ci_install_ubuntu_deps.sh ${{ matrix.arch }} ${{ matrix.cc }} - name: Build and related stuff, backup binaries run: | export C_CLANG_SUFFIX="-15" C_GCC_SUFFIX="-12" ./../mupen64plus-core/.github/workflows/scripts/ci_build.sh ${{ matrix.arch }} ${{ matrix.cc }} makepkg - name: Upload artifact if: matrix.cc == 'GCC' uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/*.tar.gz MSYS2: strategy: fail-fast: false matrix: include: - cc: GCC arch: x64 cross: x86_64 env: MINGW64 - cc: GCC arch: x86 cross: i686 env: MINGW32 name: Windows / MSYS2 ${{ matrix.cc }} / ${{ matrix.arch }} runs-on: windows-2022 defaults: run: shell: msys2 {0} steps: - uses: actions/checkout@v4 - uses: msys2/setup-msys2@v2 with: msystem: ${{ matrix.env }} update: true install: >- git libtool make mingw-w64-${{ matrix.cross }}-gcc mingw-w64-${{ matrix.cross }}-toolchain mingw-w64-${{ matrix.cross }}-ntldd - name: Build and related stuff, backup binaries run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core ./../mupen64plus-core/.github/workflows/scripts/ci_build.sh ${{ matrix.arch }} ${{ matrix.cc }} - name: Backup dependencies, etc... run: | ./../mupen64plus-core/.github/workflows/scripts/ci_backup_mingw_deps.sh ${{ matrix.env }} - name: Upload artifact uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/* MSVC: strategy: fail-fast: false matrix: include: - toolset: v143 arch: x64 vs: 2022 - toolset: v141_xp arch: x86 vs: 2019 name: Windows / MSVC with ${{ matrix.toolset }} / ${{ matrix.arch }} runs-on: windows-${{ matrix.vs }} defaults: run: shell: cmd steps: - uses: actions/checkout@v4 - uses: microsoft/setup-msbuild@v2 - name: Build and related stuff, backup binaries run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ..\mupen64plus-core set TOOLSET=${{ matrix.toolset }} .\..\mupen64plus-core\.github\workflows\scripts\ci_msvc_build.cmd ${{ matrix.arch }} - name: Backup dependencies, etc... run: | .\..\mupen64plus-core\.github\workflows\scripts\ci_backup_msvc_deps.cmd ${{ matrix.arch }} - name: Upload artifact uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/* Nightly-build: runs-on: ubuntu-latest if: github.ref_name == 'master' needs: [Linux, MSYS2, MSVC] steps: - uses: actions/checkout@v4 - name: Download artifacts uses: actions/download-artifact@v4 with: path: binaries - name: Get some tools run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core sudo apt-get update sudo apt-get -y install hashdeep - name: Creating new artifacts and update nightly-build run: | ./../mupen64plus-core/.github/workflows/scripts/ci_nightly_artifacts.sh - name: Nightly-build uses: ncipollo/release-action@v1 with: prerelease: true allowUpdates: true removeArtifacts: true replacesArtifacts: false tag: nightly-build artifacts: pkg/* mupen64plus-rsp-hle-src-2.6.0/.github/workflows/schedule.yml000066400000000000000000000112331464506742100240030ustar00rootroot00000000000000name: HLE RSP - Scheduled on: schedule: - cron: '30 14 21 * *' jobs: Linux: strategy: fail-fast: false matrix: include: - cc: GCC arch: x64 - cc: GCC arch: x86 - cc: Clang arch: x64 - cc: Clang arch: x86 name: Linux / ${{ matrix.cc }} / ${{ matrix.arch }} runs-on: ubuntu-22.04 if: github.repository == 'mupen64plus/mupen64plus-rsp-hle' steps: - uses: actions/checkout@v4 - name: Get build dependencies and arrange the environment run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core export C_CLANG_SUFFIX="-15" C_GCC_SUFFIX="-12" ./../mupen64plus-core/.github/workflows/scripts/ci_install_ubuntu_deps.sh ${{ matrix.arch }} ${{ matrix.cc }} - name: Build and related stuff, backup binaries run: | export C_CLANG_SUFFIX="-15" C_GCC_SUFFIX="-12" ./../mupen64plus-core/.github/workflows/scripts/ci_build.sh ${{ matrix.arch }} ${{ matrix.cc }} makepkg - name: Upload artifact if: matrix.cc == 'GCC' uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/*.tar.gz MSYS2: strategy: fail-fast: false matrix: include: - cc: GCC arch: x64 cross: x86_64 env: MINGW64 - cc: GCC arch: x86 cross: i686 env: MINGW32 name: Windows / MSYS2 ${{ matrix.cc }} / ${{ matrix.arch }} runs-on: windows-2022 if: github.repository == 'mupen64plus/mupen64plus-rsp-hle' defaults: run: shell: msys2 {0} steps: - uses: actions/checkout@v4 - uses: msys2/setup-msys2@v2 with: msystem: ${{ matrix.env }} update: true install: >- git libtool make mingw-w64-${{ matrix.cross }}-gcc mingw-w64-${{ matrix.cross }}-toolchain mingw-w64-${{ matrix.cross }}-ntldd - name: Build and related stuff, backup binaries run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core ./../mupen64plus-core/.github/workflows/scripts/ci_build.sh ${{ matrix.arch }} ${{ matrix.cc }} - name: Backup dependencies, etc... run: | ./../mupen64plus-core/.github/workflows/scripts/ci_backup_mingw_deps.sh ${{ matrix.env }} - name: Upload artifact uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/* MSVC: strategy: fail-fast: false matrix: include: - toolset: v143 arch: x64 vs: 2022 - toolset: v141_xp arch: x86 vs: 2019 name: Windows / MSVC with ${{ matrix.toolset }} / ${{ matrix.arch }} runs-on: windows-${{ matrix.vs }} if: github.repository == 'mupen64plus/mupen64plus-rsp-hle' defaults: run: shell: cmd steps: - uses: actions/checkout@v4 - uses: microsoft/setup-msbuild@v2 - name: Build and related stuff, backup binaries run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ..\mupen64plus-core set TOOLSET=${{ matrix.toolset }} .\..\mupen64plus-core\.github\workflows\scripts\ci_msvc_build.cmd ${{ matrix.arch }} - name: Backup dependencies, etc... run: | .\..\mupen64plus-core\.github\workflows\scripts\ci_backup_msvc_deps.cmd ${{ matrix.arch }} - name: Upload artifact uses: actions/upload-artifact@v4 with: name: ${{ env.PKG_NAME }} path: pkg/* Nightly-build: runs-on: ubuntu-latest if: github.ref_name == 'master' needs: [Linux, MSYS2, MSVC] steps: - uses: actions/checkout@v4 - name: Download artifacts uses: actions/download-artifact@v4 with: path: binaries - name: Get some tools run: | git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ../mupen64plus-core sudo apt-get update sudo apt-get -y install hashdeep - name: Creating new artifacts and update nightly-build run: | ./../mupen64plus-core/.github/workflows/scripts/ci_nightly_artifacts.sh - name: Nightly-build uses: ncipollo/release-action@v1 with: prerelease: true allowUpdates: true removeArtifacts: true replacesArtifacts: false tag: nightly-build artifacts: pkg/* mupen64plus-rsp-hle-src-2.6.0/.gitignore000066400000000000000000000000751464506742100200610ustar00rootroot00000000000000/projects/unix/_obj*/ /projects/unix/mupen64plus-rsp-hle*.so mupen64plus-rsp-hle-src-2.6.0/.travis.yml000066400000000000000000000052041464506742100202010ustar00rootroot00000000000000sudo: required dist: xenial language: cpp compiler: - gcc - clang addons: apt: packages: - git before_install: - git clone --depth=1 --branch=master git://github.com/mupen64plus/mupen64plus-core.git deps/mupen64plus-core env: - DUMP=0 - DUMP=1 script: - make -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 clean && LDFLAGS="-Wl,--no-add-needed -Wl,--no-undefined" OPTFLAGS="-O2" make CC="${CC}" CXX="${CXX}" -j$(nproc) -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 all # extra mxe build entries matrix: include: - env: - MXE_CPU=i686 - PATH="/usr/lib/mxe/usr/bin/:$PATH" before_install: - curl -sSL "https://mirror.mxe.cc/repos/apt/client-conf/mxeapt.gpg" | sudo -E apt-key add - - echo "deb https://mirror.mxe.cc/repos/apt xenial main" | sudo tee -a /etc/apt/sources.list - sudo apt-get update -qq - sudo apt-get -y --allow-unauthenticated install mxe-i686-w64-mingw32.shared-gcc - git clone --depth=1 --branch=master git://github.com/mupen64plus/mupen64plus-core.git deps/mupen64plus-core script: - make UNAME=MINGW CROSS_COMPILE="${MXE_CPU}-w64-mingw32.shared-" CC="${MXE_CPU}-w64-mingw32.shared-gcc" CXX="${MXE_CPU}-w64-mingw32.shared-g++" HOST_CPU="${MXE_CPU}" -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 clean && make UNAME=MINGW CROSS_COMPILE="${MXE_CPU}-w64-mingw32.shared-" CC="${MXE_CPU}-w64-mingw32.shared-gcc" CXX="${MXE_CPU}-w64-mingw32.shared-g++" HOST_CPU="${MXE_CPU}" -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 -j$(nproc) all - env: - MXE_CPU=x86_64 - PATH="/usr/lib/mxe/usr/bin/:$PATH" before_install: - curl -sSL "https://mirror.mxe.cc/repos/apt/client-conf/mxeapt.gpg" | sudo -E apt-key add - - echo "deb https://mirror.mxe.cc/repos/apt xenial main" | sudo tee -a /etc/apt/sources.list - sudo apt-get update -qq - sudo apt-get -y --allow-unauthenticated install mxe-x86-64-w64-mingw32.shared-gcc - git clone --depth=1 --branch=master git://github.com/mupen64plus/mupen64plus-core.git deps/mupen64plus-core script: - make UNAME=MINGW CROSS_COMPILE="${MXE_CPU}-w64-mingw32.shared-" CC="${MXE_CPU}-w64-mingw32.shared-gcc" CXX="${MXE_CPU}-w64-mingw32.shared-g++" HOST_CPU="${MXE_CPU}" -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 clean && make UNAME=MINGW CROSS_COMPILE="${MXE_CPU}-w64-mingw32.shared-" CC="${MXE_CPU}-w64-mingw32.shared-gcc" CXX="${MXE_CPU}-w64-mingw32.shared-g++" HOST_CPU="${MXE_CPU}" -C projects/unix APIDIR="$(pwd)/deps/mupen64plus-core/src/api/" V=1 -j$(nproc) all mupen64plus-rsp-hle-src-2.6.0/INSTALL000066400000000000000000000014571464506742100171270ustar00rootroot00000000000000Mupen64Plus-RSP-HLE INSTALL --------------------------- This text file was written to explain the installation process of the Mupen64Plus-RSP-HLE plugin. If this module is part of a Mupen64Plus source code bundle, the user should run the "m64p_install.sh" script in the root of the unzipped bundle to install all of the included modules in the bundle. If this module is a standalone source code release, you should build the library from source code and install it via the makefile, like this: $ cd projects/unix $ make all $ sudo make install If you want to build the Mupen64Plus-RSP-HLE module for installation in a home folder for a single user, you may build it like this (replacing with your desired local installation path): $ cd projects/unix $ make all $ make install LIBDIR= mupen64plus-rsp-hle-src-2.6.0/LICENSES000066400000000000000000000456311464506742100172300ustar00rootroot00000000000000Mupen64Plus-rsp-hle LICENSE --------------------------- Mupen64Plus-rsp-hle is licensed under the GNU General Public License version 2. The authors of Mupen64Plus-rsp-hle are: * Richard Goedeken (Richard42) * Bobby Smiles (bsmiles32) * Sven Eckelmann (ecsv) * John Chadwick (NMN) * James Hood (Ebenblues) * Scott Gorman (okaygo) * Scott Knauert (Tillin9) * Jesse Dean (DarkJezter) * Louai Al-Khanji (slougi) * Bob Forder (orbitaldecay) * Jason Espinosa (hasone) * HyperHacker * and others. Mupen64Plus is based on GPL-licensed source code from Mupen64 v0.5, originally written by: * Hacktarux * Dave2001 * Zilmar * Gregor Anich (Blight) * Juha Luotio (JttL) * and others. GNU GENERAL PUBLIC LICENSE Version 2, June 1991 Copyright (C) 1989, 1991 Free Software Foundation, Inc. 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Library General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. GNU GENERAL PUBLIC LICENSE TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does. 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. 8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. 9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. NO WARRANTY 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. END OF TERMS AND CONDITIONS Appendix: How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. Copyright (C) 19yy This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: Gnomovision version 69, Copyright (C) 19yy name of author Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than `show w' and `show c'; they could even be mouse-clicks or menu items--whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names: Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomovision' (which makes passes at compilers) written by James Hacker. , 1 April 1989 Ty Coon, President of Vice This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Library General Public License instead of this License. mupen64plus-rsp-hle-src-2.6.0/RELEASE000066400000000000000000000167451464506742100171070ustar00rootroot00000000000000RSP High-Level Emulation plugin for Mupen64Plus ----------------------------------------------- -------------------------------------------- # Mupen64Plus-rsp-hle v2.6.0 - July 14, 2024 ## Merged PRs * [90](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/90): CI/CD: Update to generic scripts, independent schedule, etc * [89](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/89): Fix check for M64P_BIG_ENDIAN * [88](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/88): CI/CD: Add BLAKE2 hashfile * [87](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/87): CI/CD: Integrate scheduled builds and other... * [86](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/86): CI/CD: Update MSVC * [85](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/85): CI/CD: Update * [84](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/84): Added overload distortion effect, used by a few tracks in CBFD * [83](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/83): Fix out of bounds memory access encountered in CBFD * [82](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/82): CI/CD: Implement GitHub Actions and public nightly builds * [81](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/81): Add RISC-V Support * [80](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/80): build: Allow out of tree builds. * [79](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/79): Fix undefined behavior * [77](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/77): Allow non-default compilers without resorting to symbolic links * [76](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/76): Allow to set custom platform toolset from commands * [75](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/75): AppVeyor with artifact packaging * [74](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/74): Fix HQVM implementation * [73](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/73): Rework task dispatching implementation * [72](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/72): Changes to have more generic project files * [70](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/70): Implement HVQM ucode * [69](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/69): Integrate Hvqm preliminary work (not fully functional yet) * [68](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/68): Added ppc64le support ## Top Contributors (2 or more commits) 1. Jj0YzL5nvJ 2. richard42 3. Gillou68310 4. derselbst ------------------------------------------------ # Mupen64Plus-rsp-hle v2.5.9 - February 10, 2019 ## Merged PRs * [67](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/67): Remove use of ConfigSaveSection * [66](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/66): Makefile aarch64 * [65](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/65): Indi and Naboo now have an HLE implementation * [64](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/64): Remove WDC and Stunt Racer from LLE list * [63](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/63): Update homepage links * [62](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/62): Don't set MI_INTR_SP if task is not finished yet * [61](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/61): Implement RE2 fill video double buffer ucode * [60](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/60): Update jpeg.c * [59](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/59): If DEBUG not set, set NDEBUG * [58](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/58): Build MXE targets as additional tests in travis build matrix * [54](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/54): Implement RSP fallback. * [53](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/53): Implement RE2 video frame decoding ucode * [52](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/52): specify audio ucode flags explicitly * [51](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/51): Let each ucode specify how sp status should be modified. * [47](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/47): Prevent yet one more illegal access to alist_buffer * [46](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/46): Add support for Re2 bilinear resizing * [45](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/45): Prevent more illegal accesses of alist_buffer * [44](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/44): Prevent CLEARBUFF for naudio and nead from deriving a count too big for alist_buffer * [43](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/43): Backport fixes from Project64 * [42](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/42): Use faster container based Travis CI * [41](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/41): Use Ubuntu Trusty as base system for Travis CI * [40](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/40): Backports from project64 * [38](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/38): Added x64 configuration to VisualStudio2013 project file * [37](https://github.com/mupen64plus/mupen64plus-rsp-hle/pull/37): Remove unnecessary OpenBSD warning. Mupen64Plus-rsp-hle v2.5 - April 26, 2015 ------------------------------------------------- - Game-specific fixes: Bottom of the 9th, IndianaJones, BattleForNaboo, Conkers Bad Fur Day - Support for MusyX microcodes - Improve audio microcode identification - Huge quantity of code cleanups and refactoring to improve organization - Add support for additional audio commands: #16, POLEF, RESAMPLE_ZOH - Multiple bugfixes in audio processing code - Move global variables into a struct so code is re-entrant - bugfix: microcode detection could sometimes fail after reset Mupen64Plus-rsp-hle v2.0 - July 4, 2013 ------------------------------------------------- - Add support for MusyX ucode detection - support JPEG decoding used in Pokemon Stadium Japan - lots of refactoring to clean up code - Project files for Visual Studio 2012 - Makefile changes - add support for PowerPC and MinGW32 builds - add cross-compiling support to build Win32 executables (MXE) under Linux Mupen64Plus-rsp-hle v1.99.5 - March 10, 2012 ------------------------------------------------- - Handle JPEG decompression, used in Ogre Battle 64 and Pokemon Stadium - updated RSP plugin for new Mupen64plus 2.0 API versioning scheme - bugfix: #102 - Missing backgrounds in Ogre Battle 64 - many makefile fixes and improvements Mupen64Plus-rsp-hle v1.99.4 - November 22, 2010 ------------------------------------------------- - merged all big-endian fixes from mupen64gc project - makefile fixes and improvements Mupen64Plus-rsp-hle v1.99.3 - February 13, 2010 ------------------------------------------------- - Makefile improvement: added OS type GNU/kFreeBSD Mupen64Plus-rsp-hle v1.99.2 - January 6, 2010 ------------------------------------------------- - new feature: added MSVC8 project file for RSP-HLE plugin, fixed a few minor incompatibilities - Makefile improvements: - throw error if OS/CPU not supported - use DESTDIR in install/uninstall paths - Allow user-specified CC/CXX/LD paths - use C++ compiler to link instead of LD, because the compiler knows where the standard C++ libs are Mupen64Plus-rsp-hle v1.99.1 - December 14, 2009 ------------------------------------------------- - Converted to new Mupen64Plus 2.0 API - Refactored build system to separate source and object files - Refactored all code to remove win32-specific things, unnecessary functions, and clean up mupen64plus-rsp-hle-src-2.6.0/appveyor.yml000066400000000000000000000015271464506742100204640ustar00rootroot00000000000000version: 1.0.{build} image: Visual Studio 2022 skip_tags: true skip_commits: files: - '**/*.md' - .github/ - data/* - .gitattributes - .gitignore - .travis.yml - README configuration: - Release platform: - Win32 - x64 before_build: - git tag --delete nightly-build - git clone --depth 1 https://github.com/mupen64plus/mupen64plus-core.git ..\mupen64plus-core build_script: - msbuild projects\msvc\mupen64plus-rsp-hle.vcxproj /p:Configuration=%configuration%;Platform=%platform% after_build: - ps: $env:rev1 = git describe --tags - set rev2=%platform% - if "%rev2%"=="Win32" set rev2=x86 - set filepkg=mupen64plus-rsp-hle_v%rev1%_%rev2% - cd projects\msvc\%platform%\%configuration% - 7z a -t7z ..\..\..\..\build\%filepkg%.7z *.dll artifacts: - path: build\$(filepkg).7z name: $(filepkg) mupen64plus-rsp-hle-src-2.6.0/projects/000077500000000000000000000000001464506742100177205ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/projects/msvc/000077500000000000000000000000001464506742100206705ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/projects/msvc/mupen64plus-rsp-hle.vcxproj000066400000000000000000000223361464506742100260650ustar00rootroot00000000000000 Debug Win32 Debug x64 Release Win32 Release x64 {2EC7CEE3-C7A7-4F2E-B2C8-4DF6AFEC3E9A} mupen64plusrsphle $([Microsoft.Build.Utilities.ToolLocationHelper]::GetLatestSDKTargetPlatformVersion('Windows', '10.0')) $(LatestTargetPlatformVersion) $(WindowsTargetPlatformVersion) $(DefaultPlatformToolset) DynamicLibrary MultiByte true $(SolutionDir)$(Platform)\$(Configuration)\ $(Platform)\$(Configuration)\ DynamicLibrary MultiByte true $(SolutionDir)$(Platform)\$(Configuration)\ $(Platform)\$(Configuration)\ DynamicLibrary MultiByte $(SolutionDir)$(Platform)\$(Configuration)\ $(Platform)\$(Configuration)\ DynamicLibrary MultiByte $(SolutionDir)$(Platform)\$(Configuration)\ $(Platform)\$(Configuration)\ Disabled ..\..\..\mupen64plus-core\src\api;%(AdditionalIncludeDirectories) WIN32;_DEBUG;_WINDOWS;_USRDLL;_CRT_SECURE_NO_DEPRECATE;inline=__inline;%(PreprocessorDefinitions) EnableFastChecks MultiThreadedDebugDLL Level3 true Windows Disabled ..\..\..\mupen64plus-core\src\api;%(AdditionalIncludeDirectories) WIN32;_DEBUG;_WINDOWS;_USRDLL;_CRT_SECURE_NO_DEPRECATE;inline=__inline;%(PreprocessorDefinitions) EnableFastChecks MultiThreadedDebugDLL Level3 true Windows ..\..\..\mupen64plus-core\src\api;%(AdditionalIncludeDirectories) WIN32;NDEBUG;_WINDOWS;_USRDLL;_CRT_SECURE_NO_DEPRECATE;inline=__inline;%(PreprocessorDefinitions) MultiThreadedDLL Level3 MaxSpeed true true true Windows true true ..\..\..\mupen64plus-core\src\api;%(AdditionalIncludeDirectories) WIN32;NDEBUG;_WINDOWS;_USRDLL;_CRT_SECURE_NO_DEPRECATE;inline=__inline;%(PreprocessorDefinitions) MultiThreadedDLL Level3 MaxSpeed true true true Windows true true mupen64plus-rsp-hle-src-2.6.0/projects/unix/000077500000000000000000000000001464506742100207035ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/projects/unix/Makefile000077500000000000000000000242411464506742100223510ustar00rootroot00000000000000#/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * # * mupen64plus-rsp-hle - Makefile * # * Mupen64Plus homepage: https://mupen64plus.org/ * # * Copyright (C) 2008-2009 Richard Goedeken * # * Copyright (C) 2007-2008 DarkJeztr Tillin9 * # * * # * This program is free software; you can redistribute it and/or modify * # * it under the terms of the GNU General Public License as published by * # * the Free Software Foundation; either version 2 of the License, or * # * (at your option) any later version. * # * * # * This program is distributed in the hope that it will be useful, * # * but WITHOUT ANY WARRANTY; without even the implied warranty of * # * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * # * GNU General Public License for more details. * # * * # * You should have received a copy of the GNU General Public License * # * along with this program; if not, write to the * # * Free Software Foundation, Inc., * # * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * # * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ # Makefile for Mupen64 HLE RSP plugin in Mupen64plus. # detect operating system UNAME ?= $(shell uname -s) OS := NONE ifeq ("$(UNAME)","Linux") OS = LINUX SO_EXTENSION = so SHARED = -shared endif ifeq ("$(UNAME)","linux") OS = LINUX SO_EXTENSION = so SHARED = -shared endif ifneq ("$(filter GNU hurd,$(UNAME))","") OS = LINUX SO_EXTENSION = so SHARED = -shared endif ifeq ("$(UNAME)","Darwin") OS = OSX SO_EXTENSION = dylib SHARED = -bundle endif ifeq ("$(UNAME)","FreeBSD") OS = FREEBSD SO_EXTENSION = so SHARED = -shared endif ifeq ("$(UNAME)","OpenBSD") OS = FREEBSD SO_EXTENSION = so SHARED = -shared endif ifneq ("$(filter GNU/kFreeBSD kfreebsd,$(UNAME))","") OS = LINUX SO_EXTENSION = so SHARED = -shared endif ifeq ("$(patsubst MINGW%,MINGW,$(UNAME))","MINGW") OS = MINGW SO_EXTENSION = dll SHARED = -shared PIC = 0 endif ifeq ("$(OS)","NONE") $(error OS type "$(UNAME)" not supported. Please file bug report at 'https://github.com/mupen64plus/mupen64plus-core/issues') endif # detect system architecture HOST_CPU ?= $(shell uname -m) NO_ASM ?= 1 CPU := NONE ifneq ("$(filter x86_64 amd64,$(HOST_CPU))","") CPU := X86 ifeq ("$(BITS)", "32") ARCH_DETECTED := 64BITS_32 PIC ?= 0 else ARCH_DETECTED := 64BITS PIC ?= 1 endif endif ifneq ("$(filter pentium i%86,$(HOST_CPU))","") CPU := X86 ARCH_DETECTED := 32BITS PIC ?= 0 endif ifneq ("$(filter ppc macppc socppc powerpc,$(HOST_CPU))","") CPU := PPC ARCH_DETECTED := 32BITS BIG_ENDIAN := 1 PIC ?= 1 $(warning Architecture "$(HOST_CPU)" not officially supported.') endif ifneq ("$(filter ppc64 powerpc64,$(HOST_CPU))","") CPU := PPC ARCH_DETECTED := 64BITS BIG_ENDIAN := 1 PIC ?= 1 $(warning Architecture "$(HOST_CPU)" not officially supported.') endif ifneq ("$(filter ppc64le powerpc64le,$(HOST_CPU))","") CPU := PPC ARCH_DETECTED := 64BITS BIG_ENDIAN := 0 PIC ?= 1 $(warning Architecture "$(HOST_CPU)" not officially supported.') endif ifneq ("$(filter arm%,$(HOST_CPU))","") ifeq ("$(filter arm%b,$(HOST_CPU))","") CPU := ARM ARCH_DETECTED := 32BITS PIC ?= 1 $(warning Architecture "$(HOST_CPU)" not officially supported.') endif endif ifneq ("$(filter mips,$(HOST_CPU))","") CPU := MIPS ARCH_DETECTED := 32BITS PIC ?= 1 $(warning Architecture "$(HOST_CPU)" not officially supported.') endif ifneq ("$(filter aarch64,$(HOST_CPU))","") CPU := AARCH ARCH_DETECTED := 64BITS PIC ?= 1 NEW_DYNAREC := 1 NO_ASM := 1 endif ifneq ("$(filter riscv64,$(HOST_CPU))","") CPU := RISCV64 ARCH_DETECTED := 64BITS PIC ?= 1 NO_ASM := 1 $(warning Architecture "$(HOST_CPU)" not officially supported.) endif ifeq ("$(CPU)","NONE") $(error CPU type "$(HOST_CPU)" not supported. Please file bug report at 'https://github.com/mupen64plus/mupen64plus-core/issues') endif SRCDIR = ../../src OBJDIR = _obj$(POSTFIX) # base CFLAGS, LDLIBS, and LDFLAGS OPTFLAGS ?= -O3 -flto WARNFLAGS ?= -Wall CFLAGS += $(OPTFLAGS) $(WARNFLAGS) -ffast-math -fno-strict-aliasing -fvisibility=hidden -I$(SRCDIR) LDFLAGS += $(SHARED) # Since we are building a shared library, we must compile with -fPIC on some architectures # On 32-bit x86 systems we do not want to use -fPIC because we don't have to and it has a big performance penalty on this arch ifeq ($(PIC), 1) CFLAGS += -fPIC else CFLAGS += -fno-PIC endif ifeq ($(BIG_ENDIAN), 1) CFLAGS += -DM64P_BIG_ENDIAN endif # tweak flags for 32-bit build on 64-bit system ifeq ($(ARCH_DETECTED), 64BITS_32) ifeq ($(OS), FREEBSD) $(error Do not use the BITS=32 option with FreeBSD, use -m32 and -m elf_i386) endif ifneq ($(OS), OSX) ifeq ($(OS), MINGW) LDFLAGS += -Wl,-m,i386pe else CFLAGS += -m32 LDFLAGS += -Wl,-m,elf_i386 endif endif endif ifeq ($(ARCH_DETECTED), 64BITS) ifeq ($(OS), MINGW) LDFLAGS += -Wl,-m,i386pep endif endif # set special flags per-system ifeq ($(OS), LINUX) # only export api symbols LDFLAGS += -Wl,-version-script,$(SRCDIR)/rsp_api_export.ver LDLIBS += -ldl endif ifeq ($(OS), OSX) OSX_SDK_PATH = $(shell xcrun --sdk macosx --show-sdk-path) ifeq ($(CPU), X86) ifeq ($(ARCH_DETECTED), 64BITS) CFLAGS += -pipe -arch x86_64 -mmacosx-version-min=10.9 -isysroot $(OSX_SDK_PATH) else CFLAGS += -pipe -mmmx -msse -fomit-frame-pointer -arch i686 -mmacosx-version-min=10.9 -isysroot $(OSX_SDK_PATH) LDFLAGS += -read_only_relocs suppress endif endif endif # set mupen64plus core API header path ifneq ("$(APIDIR)","") CFLAGS += "-I$(APIDIR)" else TRYDIR = ../../../mupen64plus-core/src/api ifneq ("$(wildcard $(TRYDIR)/m64p_types.h)","") CFLAGS += -I$(TRYDIR) else TRYDIR = /usr/local/include/mupen64plus ifneq ("$(wildcard $(TRYDIR)/m64p_types.h)","") CFLAGS += -I$(TRYDIR) else TRYDIR = /usr/include/mupen64plus ifneq ("$(wildcard $(TRYDIR)/m64p_types.h)","") CFLAGS += -I$(TRYDIR) else $(error Mupen64Plus API header files not found! Use makefile parameter APIDIR to force a location.) endif endif endif endif # reduced compile output when running make without V=1 ifneq ($(findstring $(MAKEFLAGS),s),s) ifndef V Q_CC = @echo ' CC '$@; Q_LD = @echo ' LD '$@; endif endif # set base program pointers and flags CC = $(CROSS_COMPILE)gcc RM ?= rm -f INSTALL ?= install MKDIR ?= mkdir -p COMPILE.c = $(Q_CC)$(CC) $(CFLAGS) $(CPPFLAGS) $(TARGET_ARCH) -c LINK.o = $(Q_LD)$(CC) $(CFLAGS) $(LDFLAGS) $(TARGET_ARCH) # set special flags for given Makefile parameters ifeq ($(DEBUG),1) CFLAGS += -g INSTALL_STRIP_FLAG ?= else CFLAGS += -DNDEBUG ifneq ($(OS),OSX) INSTALL_STRIP_FLAG ?= -s endif endif # set installation options ifeq ($(PREFIX),) PREFIX := /usr/local endif ifeq ($(LIBDIR),) LIBDIR := $(PREFIX)/lib endif ifeq ($(PLUGINDIR),) PLUGINDIR := $(LIBDIR)/mupen64plus endif # enable/disable task dumping support ifeq ($(DUMP), 1) CFLAGS += -DENABLE_TASK_DUMP endif # list of source files to compile SOURCE = \ $(SRCDIR)/alist.c \ $(SRCDIR)/alist_audio.c \ $(SRCDIR)/alist_naudio.c \ $(SRCDIR)/alist_nead.c \ $(SRCDIR)/audio.c \ $(SRCDIR)/cicx105.c \ $(SRCDIR)/hle.c \ $(SRCDIR)/hvqm.c \ $(SRCDIR)/jpeg.c \ $(SRCDIR)/memory.c \ $(SRCDIR)/mp3.c \ $(SRCDIR)/musyx.c \ $(SRCDIR)/re2.c \ $(SRCDIR)/plugin.c ifeq ($(OS), MINGW) SOURCE += \ $(SRCDIR)/osal_dynamiclib_win32.c else SOURCE += \ $(SRCDIR)/osal_dynamiclib_unix.c endif # generate a list of object files build, make a temporary directory for them OBJECTS := $(patsubst $(SRCDIR)/%.c, $(OBJDIR)/%.o, $(filter %.c, $(SOURCE))) OBJDIRS = $(dir $(OBJECTS)) $(shell $(MKDIR) $(OBJDIRS)) # build targets TARGET = mupen64plus-rsp-hle$(POSTFIX).$(SO_EXTENSION) targets: @echo "Mupen64Plus-rsp-hle makefile. " @echo " Targets:" @echo " all == Build Mupen64Plus rsp-hle plugin" @echo " clean == remove object files" @echo " rebuild == clean and re-build all" @echo " install == Install Mupen64Plus rsp-hle plugin" @echo " uninstall == Uninstall Mupen64Plus rsp-hle plugin" @echo " Options:" @echo " BITS=32 == build 32-bit binaries on 64-bit machine" @echo " APIDIR=path == path to find Mupen64Plus Core headers" @echo " OPTFLAGS=flag == compiler optimization (default: -O3 -flto)" @echo " WARNFLAGS=flag == compiler warning levels (default: -Wall)" @echo " PIC=(1|0) == Force enable/disable of position independent code" @echo " POSTFIX=name == String added to the name of the the build (default: '')" @echo " DUMP=(1|0) == Enable/Disable unknown task dumping (default: 0)" @echo " Install Options:" @echo " PREFIX=path == install/uninstall prefix (default: /usr/local)" @echo " LIBDIR=path == library prefix (default: PREFIX/lib)" @echo " PLUGINDIR=path == path to install plugin libraries (default: LIBDIR/mupen64plus)" @echo " DESTDIR=path == path to prepend to all installation paths (only for packagers)" @echo " Debugging Options:" @echo " DEBUG=1 == add debugging symbols" @echo " V=1 == show verbose compiler output" all: $(TARGET) install: $(TARGET) $(INSTALL) -d "$(DESTDIR)$(PLUGINDIR)" $(INSTALL) -m 0644 $(INSTALL_STRIP_FLAG) $(TARGET) "$(DESTDIR)$(PLUGINDIR)" uninstall: $(RM) "$(DESTDIR)$(PLUGINDIR)/$(TARGET)" clean: $(RM) -r $(OBJDIR) $(TARGET) rebuild: clean all # build dependency files CFLAGS += -MD -MP -include $(OBJECTS:.o=.d) # standard build rules $(OBJDIR)/%.o: $(SRCDIR)/%.c $(COMPILE.c) -o $@ $< $(TARGET): $(OBJECTS) $(LINK.o) $^ $(LOADLIBES) $(LDLIBS) -o $@ .PHONY: all clean install uninstall targets mupen64plus-rsp-hle-src-2.6.0/src/000077500000000000000000000000001464506742100166565ustar00rootroot00000000000000mupen64plus-rsp-hle-src-2.6.0/src/alist.c000066400000000000000000000763501464506742100201510ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - alist.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include #include "alist.h" #include "arithmetics.h" #include "audio.h" #include "hle_external.h" #include "hle_internal.h" #include "memory.h" struct ramp_t { int64_t value; int64_t step; int64_t target; }; /* local functions */ static void swap(int16_t **a, int16_t **b) { int16_t* tmp = *b; *b = *a; *a = tmp; } static int16_t* sample(struct hle_t* hle, unsigned pos) { return (int16_t*)hle->alist_buffer + ((pos ^ S) & 0xfff); } static uint8_t* alist_u8(struct hle_t* hle, uint16_t dmem) { return (uint8_t*)(hle->alist_buffer + ((dmem ^ S8) & 0xfff)); } static int16_t* alist_s16(struct hle_t* hle, uint16_t dmem) { return (int16_t*)(hle->alist_buffer + ((dmem ^ S16) & 0xfff)); } static void sample_mix(int16_t* dst, int16_t src, int16_t gain) { *dst = clamp_s16(*dst + ((src * gain) >> 15)); } static void alist_envmix_mix(size_t n, int16_t** dst, const int16_t* gains, int16_t src) { size_t i; for(i = 0; i < n; ++i) sample_mix(dst[i], src, gains[i]); } static int16_t ramp_step(struct ramp_t* ramp) { bool target_reached; ramp->value += ramp->step; target_reached = (ramp->step <= 0) ? (ramp->value <= ramp->target) : (ramp->value >= ramp->target); if (target_reached) { ramp->value = ramp->target; ramp->step = 0; } return (int16_t)(ramp->value >> 16); } /* global functions */ void alist_process(struct hle_t* hle, const acmd_callback_t abi[], unsigned int abi_size) { uint32_t w1, w2; unsigned int acmd; const uint32_t *alist = dram_u32(hle, *dmem_u32(hle, TASK_DATA_PTR)); const uint32_t *const alist_end = alist + (*dmem_u32(hle, TASK_DATA_SIZE) >> 2); while (alist != alist_end) { w1 = *(alist++); w2 = *(alist++); acmd = (w1 >> 24) & 0x7f; if (acmd < abi_size) (*abi[acmd])(hle, w1, w2); else HleWarnMessage(hle->user_defined, "Invalid ABI command %u", acmd); } } uint32_t alist_get_address(struct hle_t* hle, uint32_t so, const uint32_t *segments, size_t n) { uint8_t segment = (so >> 24) & 0x3f; uint32_t offset = (so & 0xffffff); if (segment >= n) { HleWarnMessage(hle->user_defined, "Invalid segment %u", segment); return offset; } return segments[segment] + offset; } void alist_set_address(struct hle_t* hle, uint32_t so, uint32_t *segments, size_t n) { uint8_t segment = (so >> 24) & 0x3f; uint32_t offset = (so & 0xffffff); if (segment >= n) { HleWarnMessage(hle->user_defined, "Invalid segment %u", segment); return; } segments[segment] = offset; } void alist_clear(struct hle_t* hle, uint16_t dmem, uint16_t count) { while(count != 0) { *alist_u8(hle, dmem++) = 0; --count; } } void alist_load(struct hle_t* hle, uint16_t dmem, uint32_t address, uint16_t count) { /* enforce DMA alignment constraints */ dmem &= ~3; address &= ~7; count = align(count, 8); memcpy(hle->alist_buffer + dmem, hle->dram + address, count); } void alist_save(struct hle_t* hle, uint16_t dmem, uint32_t address, uint16_t count) { /* enforce DMA alignment constraints */ dmem &= ~3; address &= ~7; count = align(count, 8); memcpy(hle->dram + address, hle->alist_buffer + dmem, count); } void alist_move(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count) { while (count != 0) { *alist_u8(hle, dmemo++) = *alist_u8(hle, dmemi++); --count; } } void alist_copy_every_other_sample(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count) { while (count != 0) { *alist_s16(hle, dmemo) = *alist_s16(hle, dmemi); dmemo += 2; dmemi += 4; --count; } } void alist_repeat64(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint8_t count) { uint16_t buffer[64]; memcpy(buffer, hle->alist_buffer + dmemi, 128); while(count != 0) { memcpy(hle->alist_buffer + dmemo, buffer, 128); dmemo += 128; --count; } } void alist_copy_blocks(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t block_size, uint8_t count) { int block_left = count; do { int bytes_left = block_size; do { memcpy(hle->alist_buffer + dmemo, hle->alist_buffer + dmemi, 0x20); bytes_left -= 0x20; dmemi += 0x20; dmemo += 0x20; } while(bytes_left > 0); --block_left; } while(block_left > 0); } void alist_interleave(struct hle_t* hle, uint16_t dmemo, uint16_t left, uint16_t right, uint16_t count) { uint16_t *dst = (uint16_t*)(hle->alist_buffer + dmemo); const uint16_t *srcL = (uint16_t*)(hle->alist_buffer + left); const uint16_t *srcR = (uint16_t*)(hle->alist_buffer + right); count >>= 2; while(count != 0) { uint16_t l1 = *(srcL++); uint16_t l2 = *(srcL++); uint16_t r1 = *(srcR++); uint16_t r2 = *(srcR++); #ifdef M64P_BIG_ENDIAN *(dst++) = l1; *(dst++) = r1; *(dst++) = l2; *(dst++) = r2; #else *(dst++) = r2; *(dst++) = l2; *(dst++) = r1; *(dst++) = l1; #endif --count; } } void alist_envmix_exp( struct hle_t* hle, bool init, bool aux, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address) { size_t n = (aux) ? 4 : 2; const int16_t* const in = (int16_t*)(hle->alist_buffer + dmemi); int16_t* const dl = (int16_t*)(hle->alist_buffer + dmem_dl); int16_t* const dr = (int16_t*)(hle->alist_buffer + dmem_dr); int16_t* const wl = (int16_t*)(hle->alist_buffer + dmem_wl); int16_t* const wr = (int16_t*)(hle->alist_buffer + dmem_wr); struct ramp_t ramps[2]; int32_t exp_seq[2]; int32_t exp_rates[2]; uint32_t ptr = 0; int x, y; short save_buffer[40]; memcpy((uint8_t *)save_buffer, (hle->dram + address), sizeof(save_buffer)); if (init) { ramps[0].value = (vol[0] << 16); ramps[1].value = (vol[1] << 16); ramps[0].target = (target[0] << 16); ramps[1].target = (target[1] << 16); exp_rates[0] = rate[0]; exp_rates[1] = rate[1]; exp_seq[0] = (vol[0] * rate[0]); exp_seq[1] = (vol[1] * rate[1]); } else { wet = *(int16_t *)(save_buffer + 0); /* 0-1 */ dry = *(int16_t *)(save_buffer + 2); /* 2-3 */ ramps[0].target = *(int32_t *)(save_buffer + 4); /* 4-5 */ ramps[1].target = *(int32_t *)(save_buffer + 6); /* 6-7 */ exp_rates[0] = *(int32_t *)(save_buffer + 8); /* 8-9 (save_buffer is a 16bit pointer) */ exp_rates[1] = *(int32_t *)(save_buffer + 10); /* 10-11 */ exp_seq[0] = *(int32_t *)(save_buffer + 12); /* 12-13 */ exp_seq[1] = *(int32_t *)(save_buffer + 14); /* 14-15 */ ramps[0].value = *(int32_t *)(save_buffer + 16); /* 12-13 */ ramps[1].value = *(int32_t *)(save_buffer + 18); /* 14-15 */ } /* init which ensure ramp.step != 0 iff ramp.value == ramp.target */ ramps[0].step = ramps[0].target - ramps[0].value; ramps[1].step = ramps[1].target - ramps[1].value; for (y = 0; y < count; y += 16) { if (ramps[0].step != 0) { exp_seq[0] = ((int64_t)exp_seq[0]*(int64_t)exp_rates[0]) >> 16; ramps[0].step = (exp_seq[0] - ramps[0].value) >> 3; } if (ramps[1].step != 0) { exp_seq[1] = ((int64_t)exp_seq[1]*(int64_t)exp_rates[1]) >> 16; ramps[1].step = (exp_seq[1] - ramps[1].value) >> 3; } for (x = 0; x < 8; ++x) { int16_t gains[4]; int16_t* buffers[4]; int16_t l_vol = ramp_step(&ramps[0]); int16_t r_vol = ramp_step(&ramps[1]); buffers[0] = dl + (ptr^S); buffers[1] = dr + (ptr^S); buffers[2] = wl + (ptr^S); buffers[3] = wr + (ptr^S); gains[0] = clamp_s16((l_vol * dry + 0x4000) >> 15); gains[1] = clamp_s16((r_vol * dry + 0x4000) >> 15); gains[2] = clamp_s16((l_vol * wet + 0x4000) >> 15); gains[3] = clamp_s16((r_vol * wet + 0x4000) >> 15); alist_envmix_mix(n, buffers, gains, in[ptr^S]); ++ptr; } } *(int16_t *)(save_buffer + 0) = wet; /* 0-1 */ *(int16_t *)(save_buffer + 2) = dry; /* 2-3 */ *(int32_t *)(save_buffer + 4) = (int32_t)ramps[0].target; /* 4-5 */ *(int32_t *)(save_buffer + 6) = (int32_t)ramps[1].target; /* 6-7 */ *(int32_t *)(save_buffer + 8) = exp_rates[0]; /* 8-9 (save_buffer is a 16bit pointer) */ *(int32_t *)(save_buffer + 10) = exp_rates[1]; /* 10-11 */ *(int32_t *)(save_buffer + 12) = exp_seq[0]; /* 12-13 */ *(int32_t *)(save_buffer + 14) = exp_seq[1]; /* 14-15 */ *(int32_t *)(save_buffer + 16) = (int32_t)ramps[0].value; /* 12-13 */ *(int32_t *)(save_buffer + 18) = (int32_t)ramps[1].value; /* 14-15 */ memcpy(hle->dram + address, (uint8_t *)save_buffer, sizeof(save_buffer)); } void alist_envmix_ge( struct hle_t* hle, bool init, bool aux, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address) { unsigned k; size_t n = (aux) ? 4 : 2; const int16_t* const in = (int16_t*)(hle->alist_buffer + dmemi); int16_t* const dl = (int16_t*)(hle->alist_buffer + dmem_dl); int16_t* const dr = (int16_t*)(hle->alist_buffer + dmem_dr); int16_t* const wl = (int16_t*)(hle->alist_buffer + dmem_wl); int16_t* const wr = (int16_t*)(hle->alist_buffer + dmem_wr); struct ramp_t ramps[2]; short save_buffer[40]; memcpy((uint8_t *)save_buffer, (hle->dram + address), 80); if (init) { ramps[0].value = (vol[0] << 16); ramps[1].value = (vol[1] << 16); ramps[0].target = (target[0] << 16); ramps[1].target = (target[1] << 16); ramps[0].step = rate[0] / 8; ramps[1].step = rate[1] / 8; } else { wet = *(int16_t *)(save_buffer + 0); /* 0-1 */ dry = *(int16_t *)(save_buffer + 2); /* 2-3 */ ramps[0].target = *(int32_t *)(save_buffer + 4); /* 4-5 */ ramps[1].target = *(int32_t *)(save_buffer + 6); /* 6-7 */ ramps[0].step = *(int32_t *)(save_buffer + 8); /* 8-9 (save_buffer is a 16bit pointer) */ ramps[1].step = *(int32_t *)(save_buffer + 10); /* 10-11 */ /* *(int32_t *)(save_buffer + 12);*/ /* 12-13 */ /* *(int32_t *)(save_buffer + 14);*/ /* 14-15 */ ramps[0].value = *(int32_t *)(save_buffer + 16); /* 12-13 */ ramps[1].value = *(int32_t *)(save_buffer + 18); /* 14-15 */ } count >>= 1; for (k = 0; k < count; ++k) { int16_t gains[4]; int16_t* buffers[4]; int16_t l_vol = ramp_step(&ramps[0]); int16_t r_vol = ramp_step(&ramps[1]); buffers[0] = dl + (k^S); buffers[1] = dr + (k^S); buffers[2] = wl + (k^S); buffers[3] = wr + (k^S); gains[0] = clamp_s16((l_vol * dry + 0x4000) >> 15); gains[1] = clamp_s16((r_vol * dry + 0x4000) >> 15); gains[2] = clamp_s16((l_vol * wet + 0x4000) >> 15); gains[3] = clamp_s16((r_vol * wet + 0x4000) >> 15); alist_envmix_mix(n, buffers, gains, in[k^S]); } *(int16_t *)(save_buffer + 0) = wet; /* 0-1 */ *(int16_t *)(save_buffer + 2) = dry; /* 2-3 */ *(int32_t *)(save_buffer + 4) = (int32_t)ramps[0].target; /* 4-5 */ *(int32_t *)(save_buffer + 6) = (int32_t)ramps[1].target; /* 6-7 */ *(int32_t *)(save_buffer + 8) = (int32_t)ramps[0].step; /* 8-9 (save_buffer is a 16bit pointer) */ *(int32_t *)(save_buffer + 10) = (int32_t)ramps[1].step; /* 10-11 */ /**(int32_t *)(save_buffer + 12);*/ /* 12-13 */ /**(int32_t *)(save_buffer + 14);*/ /* 14-15 */ *(int32_t *)(save_buffer + 16) = (int32_t)ramps[0].value; /* 12-13 */ *(int32_t *)(save_buffer + 18) = (int32_t)ramps[1].value; /* 14-15 */ memcpy(hle->dram + address, (uint8_t *)save_buffer, 80); } void alist_envmix_lin( struct hle_t* hle, bool init, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address) { size_t k; struct ramp_t ramps[2]; int16_t save_buffer[40]; const int16_t * const in = (int16_t*)(hle->alist_buffer + dmemi); int16_t* const dl = (int16_t*)(hle->alist_buffer + dmem_dl); int16_t* const dr = (int16_t*)(hle->alist_buffer + dmem_dr); int16_t* const wl = (int16_t*)(hle->alist_buffer + dmem_wl); int16_t* const wr = (int16_t*)(hle->alist_buffer + dmem_wr); memcpy((uint8_t *)save_buffer, hle->dram + address, 80); if (init) { ramps[0].step = rate[0] / 8; ramps[0].value = (vol[0] << 16); ramps[0].target = (target[0] << 16); ramps[1].step = rate[1] / 8; ramps[1].value = (vol[1] << 16); ramps[1].target = (target[1] << 16); } else { wet = *(int16_t *)(save_buffer + 0); /* 0-1 */ dry = *(int16_t *)(save_buffer + 2); /* 2-3 */ ramps[0].target = *(int16_t *)(save_buffer + 4) << 16; /* 4-5 */ ramps[1].target = *(int16_t *)(save_buffer + 6) << 16; /* 6-7 */ ramps[0].step = *(int32_t *)(save_buffer + 8); /* 8-9 (save_buffer is a 16bit pointer) */ ramps[1].step = *(int32_t *)(save_buffer + 10); /* 10-11 */ ramps[0].value = *(int32_t *)(save_buffer + 16); /* 16-17 */ ramps[1].value = *(int32_t *)(save_buffer + 18); /* 16-17 */ } count >>= 1; for(k = 0; k < count; ++k) { int16_t gains[4]; int16_t* buffers[4]; int16_t l_vol = ramp_step(&ramps[0]); int16_t r_vol = ramp_step(&ramps[1]); buffers[0] = dl + (k^S); buffers[1] = dr + (k^S); buffers[2] = wl + (k^S); buffers[3] = wr + (k^S); gains[0] = clamp_s16((l_vol * dry + 0x4000) >> 15); gains[1] = clamp_s16((r_vol * dry + 0x4000) >> 15); gains[2] = clamp_s16((l_vol * wet + 0x4000) >> 15); gains[3] = clamp_s16((r_vol * wet + 0x4000) >> 15); alist_envmix_mix(4, buffers, gains, in[k^S]); } *(int16_t *)(save_buffer + 0) = wet; /* 0-1 */ *(int16_t *)(save_buffer + 2) = dry; /* 2-3 */ *(int16_t *)(save_buffer + 4) = (int16_t)(ramps[0].target >> 16); /* 4-5 */ *(int16_t *)(save_buffer + 6) = (int16_t)(ramps[1].target >> 16); /* 6-7 */ *(int32_t *)(save_buffer + 8) = (int32_t)ramps[0].step; /* 8-9 (save_buffer is a 16bit pointer) */ *(int32_t *)(save_buffer + 10) = (int32_t)ramps[1].step; /* 10-11 */ *(int32_t *)(save_buffer + 16) = (int32_t)ramps[0].value; /* 16-17 */ *(int32_t *)(save_buffer + 18) = (int32_t)ramps[1].value; /* 18-19 */ memcpy(hle->dram + address, (uint8_t *)save_buffer, 80); } void alist_envmix_nead( struct hle_t* hle, bool swap_wet_LR, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, unsigned count, uint16_t *env_values, uint16_t *env_steps, const int16_t *xors) { int16_t *in = (int16_t*)(hle->alist_buffer + dmemi); int16_t *dl = (int16_t*)(hle->alist_buffer + dmem_dl); int16_t *dr = (int16_t*)(hle->alist_buffer + dmem_dr); int16_t *wl = (int16_t*)(hle->alist_buffer + dmem_wl); int16_t *wr = (int16_t*)(hle->alist_buffer + dmem_wr); /* make sure count is a multiple of 8 */ count = align(count, 8); if (swap_wet_LR) swap(&wl, &wr); while (count != 0) { size_t i; for(i = 0; i < 8; ++i) { int16_t l = (((int32_t)in[i^S] * (uint32_t)env_values[0]) >> 16) ^ xors[0]; int16_t r = (((int32_t)in[i^S] * (uint32_t)env_values[1]) >> 16) ^ xors[1]; int16_t l2 = (((int32_t)l * (uint32_t)env_values[2]) >> 16) ^ xors[2]; int16_t r2 = (((int32_t)r * (uint32_t)env_values[2]) >> 16) ^ xors[3]; dl[i^S] = clamp_s16(dl[i^S] + l); dr[i^S] = clamp_s16(dr[i^S] + r); wl[i^S] = clamp_s16(wl[i^S] + l2); wr[i^S] = clamp_s16(wr[i^S] + r2); } env_values[0] += env_steps[0]; env_values[1] += env_steps[1]; env_values[2] += env_steps[2]; dl += 8; dr += 8; wl += 8; wr += 8; in += 8; count -= 8; } } void alist_mix(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count, int16_t gain) { int16_t *dst = (int16_t*)(hle->alist_buffer + dmemo); const int16_t *src = (int16_t*)(hle->alist_buffer + dmemi); count >>= 1; while(count != 0) { sample_mix(dst, *src, gain); ++dst; ++src; --count; } } void alist_multQ44(struct hle_t* hle, uint16_t dmem, uint16_t count, int8_t gain) { int16_t *dst = (int16_t*)(hle->alist_buffer + dmem); count >>= 1; while(count != 0) { *dst = clamp_s16(*dst * gain >> 4); ++dst; --count; } } void alist_add(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count) { int16_t *dst = (int16_t*)(hle->alist_buffer + dmemo); const int16_t *src = (int16_t*)(hle->alist_buffer + dmemi); count >>= 1; while(count != 0) { *dst = clamp_s16(*dst + *src); ++dst; ++src; --count; } } static void alist_resample_reset(struct hle_t* hle, uint16_t pos, uint32_t* pitch_accu) { unsigned k; for(k = 0; k < 4; ++k) *sample(hle, pos + k) = 0; *pitch_accu = 0; } static void alist_resample_load(struct hle_t* hle, uint32_t address, uint16_t pos, uint32_t* pitch_accu) { *sample(hle, pos + 0) = *dram_u16(hle, address + 0); *sample(hle, pos + 1) = *dram_u16(hle, address + 2); *sample(hle, pos + 2) = *dram_u16(hle, address + 4); *sample(hle, pos + 3) = *dram_u16(hle, address + 6); *pitch_accu = *dram_u16(hle, address + 8); } static void alist_resample_save(struct hle_t* hle, uint32_t address, uint16_t pos, uint32_t pitch_accu) { *dram_u16(hle, address + 0) = *sample(hle, pos + 0); *dram_u16(hle, address + 2) = *sample(hle, pos + 1); *dram_u16(hle, address + 4) = *sample(hle, pos + 2); *dram_u16(hle, address + 6) = *sample(hle, pos + 3); *dram_u16(hle, address + 8) = pitch_accu; } void alist_resample( struct hle_t* hle, bool init, bool flag2, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint32_t pitch, /* Q16.16 */ uint32_t address) { uint32_t pitch_accu; uint16_t ipos = dmemi >> 1; uint16_t opos = dmemo >> 1; count >>= 1; ipos -= 4; if (flag2) HleWarnMessage(hle->user_defined, "alist_resample: flag2 is not implemented"); if (init) alist_resample_reset(hle, ipos, &pitch_accu); else alist_resample_load(hle, address, ipos, &pitch_accu); while (count != 0) { const int16_t* lut = RESAMPLE_LUT + ((pitch_accu & 0xfc00) >> 8); *sample(hle, opos++) = clamp_s16( ( (*sample(hle, ipos ) * lut[0]) + (*sample(hle, ipos + 1) * lut[1]) + (*sample(hle, ipos + 2) * lut[2]) + (*sample(hle, ipos + 3) * lut[3]) ) >> 15); pitch_accu += pitch; ipos += (pitch_accu >> 16); pitch_accu &= 0xffff; --count; } alist_resample_save(hle, address, ipos, pitch_accu); } void alist_resample_zoh( struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint32_t pitch, uint32_t pitch_accu) { uint16_t ipos = dmemi >> 1; uint16_t opos = dmemo >> 1; count >>= 1; while(count != 0) { *sample(hle, opos++) = *sample(hle, ipos); pitch_accu += pitch; ipos += (pitch_accu >> 16); pitch_accu &= 0xffff; --count; } } typedef unsigned int (*adpcm_predict_frame_t)(struct hle_t* hle, int16_t* dst, uint16_t dmemi, unsigned char scale); static unsigned int adpcm_predict_frame_4bits(struct hle_t* hle, int16_t* dst, uint16_t dmemi, unsigned char scale) { unsigned int i; unsigned int rshift = (scale < 12) ? 12 - scale : 0; for(i = 0; i < 8; ++i) { uint8_t byte = *alist_u8(hle, dmemi++); *(dst++) = adpcm_predict_sample(byte, 0xf0, 8, rshift); *(dst++) = adpcm_predict_sample(byte, 0x0f, 12, rshift); } return 8; } static unsigned int adpcm_predict_frame_2bits(struct hle_t* hle, int16_t* dst, uint16_t dmemi, unsigned char scale) { unsigned int i; unsigned int rshift = (scale < 14) ? 14 - scale : 0; for(i = 0; i < 4; ++i) { uint8_t byte = *alist_u8(hle, dmemi++); *(dst++) = adpcm_predict_sample(byte, 0xc0, 8, rshift); *(dst++) = adpcm_predict_sample(byte, 0x30, 10, rshift); *(dst++) = adpcm_predict_sample(byte, 0x0c, 12, rshift); *(dst++) = adpcm_predict_sample(byte, 0x03, 14, rshift); } return 4; } void alist_adpcm( struct hle_t* hle, bool init, bool loop, bool two_bit_per_sample, uint16_t dmemo, uint16_t dmemi, uint16_t count, const int16_t* codebook, uint32_t loop_address, uint32_t last_frame_address) { int16_t last_frame[16]; size_t i; adpcm_predict_frame_t predict_frame = (two_bit_per_sample) ? adpcm_predict_frame_2bits : adpcm_predict_frame_4bits; assert((count & 0x1f) == 0); if (init) memset(last_frame, 0, 16*sizeof(last_frame[0])); else dram_load_u16(hle, (uint16_t*)last_frame, (loop) ? loop_address : last_frame_address, 16); for(i = 0; i < 16; ++i, dmemo += 2) *alist_s16(hle, dmemo) = last_frame[i]; while (count != 0) { int16_t frame[16]; uint8_t code = *alist_u8(hle, dmemi++); unsigned char scale = (code & 0xf0) >> 4; const int16_t* const cb_entry = codebook + ((code & 0xf) << 4); dmemi += predict_frame(hle, frame, dmemi, scale); adpcm_compute_residuals(last_frame , frame , cb_entry, last_frame + 14, 8); adpcm_compute_residuals(last_frame + 8, frame + 8, cb_entry, last_frame + 6 , 8); for(i = 0; i < 16; ++i, dmemo += 2) *alist_s16(hle, dmemo) = last_frame[i]; count -= 32; } dram_store_u16(hle, (uint16_t*)last_frame, last_frame_address, 16); } void alist_filter( struct hle_t* hle, uint16_t dmem, uint16_t count, uint32_t address, const uint32_t* lut_address) { int x; int16_t outbuff[0x3c0]; int16_t *outp = outbuff; int16_t* const lutt6 = (int16_t*)(hle->dram + lut_address[0]); int16_t* const lutt5 = (int16_t*)(hle->dram + lut_address[1]); int16_t* in1 = (int16_t*)(hle->dram + address); int16_t* in2 = (int16_t*)(hle->alist_buffer + dmem); for (x = 0; x < 8; ++x) { int32_t v = (lutt5[x] + lutt6[x]) >> 1; lutt5[x] = lutt6[x] = v; } for (x = 0; x < count; x += 16) { int32_t v[8]; v[1] = in1[0] * lutt6[6]; v[1] += in1[3] * lutt6[7]; v[1] += in1[2] * lutt6[4]; v[1] += in1[5] * lutt6[5]; v[1] += in1[4] * lutt6[2]; v[1] += in1[7] * lutt6[3]; v[1] += in1[6] * lutt6[0]; v[1] += in2[1] * lutt6[1]; /* 1 */ v[0] = in1[3] * lutt6[6]; v[0] += in1[2] * lutt6[7]; v[0] += in1[5] * lutt6[4]; v[0] += in1[4] * lutt6[5]; v[0] += in1[7] * lutt6[2]; v[0] += in1[6] * lutt6[3]; v[0] += in2[1] * lutt6[0]; v[0] += in2[0] * lutt6[1]; v[3] = in1[2] * lutt6[6]; v[3] += in1[5] * lutt6[7]; v[3] += in1[4] * lutt6[4]; v[3] += in1[7] * lutt6[5]; v[3] += in1[6] * lutt6[2]; v[3] += in2[1] * lutt6[3]; v[3] += in2[0] * lutt6[0]; v[3] += in2[3] * lutt6[1]; v[2] = in1[5] * lutt6[6]; v[2] += in1[4] * lutt6[7]; v[2] += in1[7] * lutt6[4]; v[2] += in1[6] * lutt6[5]; v[2] += in2[1] * lutt6[2]; v[2] += in2[0] * lutt6[3]; v[2] += in2[3] * lutt6[0]; v[2] += in2[2] * lutt6[1]; v[5] = in1[4] * lutt6[6]; v[5] += in1[7] * lutt6[7]; v[5] += in1[6] * lutt6[4]; v[5] += in2[1] * lutt6[5]; v[5] += in2[0] * lutt6[2]; v[5] += in2[3] * lutt6[3]; v[5] += in2[2] * lutt6[0]; v[5] += in2[5] * lutt6[1]; v[4] = in1[7] * lutt6[6]; v[4] += in1[6] * lutt6[7]; v[4] += in2[1] * lutt6[4]; v[4] += in2[0] * lutt6[5]; v[4] += in2[3] * lutt6[2]; v[4] += in2[2] * lutt6[3]; v[4] += in2[5] * lutt6[0]; v[4] += in2[4] * lutt6[1]; v[7] = in1[6] * lutt6[6]; v[7] += in2[1] * lutt6[7]; v[7] += in2[0] * lutt6[4]; v[7] += in2[3] * lutt6[5]; v[7] += in2[2] * lutt6[2]; v[7] += in2[5] * lutt6[3]; v[7] += in2[4] * lutt6[0]; v[7] += in2[7] * lutt6[1]; v[6] = in2[1] * lutt6[6]; v[6] += in2[0] * lutt6[7]; v[6] += in2[3] * lutt6[4]; v[6] += in2[2] * lutt6[5]; v[6] += in2[5] * lutt6[2]; v[6] += in2[4] * lutt6[3]; v[6] += in2[7] * lutt6[0]; v[6] += in2[6] * lutt6[1]; outp[1] = ((v[1] + 0x4000) >> 15); outp[0] = ((v[0] + 0x4000) >> 15); outp[3] = ((v[3] + 0x4000) >> 15); outp[2] = ((v[2] + 0x4000) >> 15); outp[5] = ((v[5] + 0x4000) >> 15); outp[4] = ((v[4] + 0x4000) >> 15); outp[7] = ((v[7] + 0x4000) >> 15); outp[6] = ((v[6] + 0x4000) >> 15); in1 = in2; in2 += 8; outp += 8; } memcpy(hle->dram + address, in2 - 8, 16); memcpy(hle->alist_buffer + dmem, outbuff, count); } void alist_polef( struct hle_t* hle, bool init, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint16_t gain, int16_t* table, uint32_t address) { int16_t *dst = (int16_t*)(hle->alist_buffer + dmemo); const int16_t* const h1 = table; int16_t* const h2 = table + 8; unsigned i; int16_t l1, l2; int16_t h2_before[8]; count = align(count, 16); if (init) { l1 = 0; l2 = 0; } else { l1 = *dram_u16(hle, address + 4); l2 = *dram_u16(hle, address + 6); } for(i = 0; i < 8; ++i) { h2_before[i] = h2[i]; h2[i] = (((int32_t)h2[i] * gain) >> 14); } do { int16_t frame[8]; for(i = 0; i < 8; ++i, dmemi += 2) frame[i] = *alist_s16(hle, dmemi); for(i = 0; i < 8; ++i) { int32_t accu = frame[i] * gain; accu += h1[i]*l1 + h2_before[i]*l2 + rdot(i, h2, frame); dst[i^S] = clamp_s16(accu >> 14); } l1 = dst[6^S]; l2 = dst[7^S]; dst += 8; count -= 16; } while (count != 0); dram_store_u32(hle, (uint32_t*)(dst - 4), address, 2); } void alist_iirf( struct hle_t* hle, bool init, uint16_t dmemo, uint16_t dmemi, uint16_t count, int16_t* table, uint32_t address) { int16_t *dst = (int16_t*)(hle->alist_buffer + dmemo); int32_t i, prev; int16_t frame[8]; int16_t ibuf[4]; uint16_t index = 7; count = align(count, 16); if(init) { for(i = 0; i < 8; ++i) frame[i] = 0; ibuf[1] = 0; ibuf[2] = 0; } else { frame[6] = *dram_u16(hle, address + 4); frame[7] = *dram_u16(hle, address + 6); ibuf[1] = (int16_t)*dram_u16(hle, address + 8); ibuf[2] = (int16_t)*dram_u16(hle, address + 10); } prev = vmulf(table[9], frame[6]) * 2; do { for(i = 0; i < 8; ++i) { int32_t accu; ibuf[index&3] = *alist_s16(hle, dmemi); accu = prev + vmulf(table[0], ibuf[index&3]) + vmulf(table[1], ibuf[(index-1)&3]) + vmulf(table[0], ibuf[(index-2)&3]); accu += vmulf(table[8], frame[index]) * 2; prev = vmulf(table[9], frame[index]) * 2; dst[i^S] = frame[i] = accu; index=(index+1)&7; dmemi += 2; } dst += 8; count -= 0x10; } while (count > 0); dram_store_u16(hle, (uint16_t*)&frame[6], address + 4, 2); dram_store_u16(hle, (uint16_t*)&ibuf[(index-2)&3], address+8, 1); dram_store_u16(hle, (uint16_t*)&ibuf[(index-1)&3], address+10, 1); } /* Perform a clamped gain, then attenuate it back by an amount */ void alist_overload(struct hle_t* hle, uint16_t dmem, int16_t count, int16_t gain, uint16_t attenuation) { int16_t accu; int16_t * sample = (int16_t*)(hle->alist_buffer + dmem); while (count != 0) { accu = clamp_s16(*sample * gain); *sample = (accu * attenuation) >> 16; sample++; count --; } } mupen64plus-rsp-hle-src-2.6.0/src/alist.h000066400000000000000000000140071464506742100201450ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - alist.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef ALIST_INTERNAL_H #define ALIST_INTERNAL_H #include #include #include struct hle_t; typedef void (*acmd_callback_t)(struct hle_t* hle, uint32_t w1, uint32_t w2); void alist_process(struct hle_t* hle, const acmd_callback_t abi[], unsigned int abi_size); uint32_t alist_get_address(struct hle_t* hle, uint32_t so, const uint32_t *segments, size_t n); void alist_set_address(struct hle_t* hle, uint32_t so, uint32_t *segments, size_t n); void alist_clear(struct hle_t* hle, uint16_t dmem, uint16_t count); void alist_load(struct hle_t* hle, uint16_t dmem, uint32_t address, uint16_t count); void alist_save(struct hle_t* hle, uint16_t dmem, uint32_t address, uint16_t count); void alist_move(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count); void alist_copy_every_other_sample(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count); void alist_repeat64(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint8_t count); void alist_copy_blocks(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t block_size, uint8_t count); void alist_interleave(struct hle_t* hle, uint16_t dmemo, uint16_t left, uint16_t right, uint16_t count); void alist_envmix_exp( struct hle_t* hle, bool init, bool aux, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address); void alist_envmix_ge( struct hle_t* hle, bool init, bool aux, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address); void alist_envmix_lin( struct hle_t* hle, bool init, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, uint16_t count, int16_t dry, int16_t wet, const int16_t *vol, const int16_t *target, const int32_t *rate, uint32_t address); void alist_envmix_nead( struct hle_t* hle, bool swap_wet_LR, uint16_t dmem_dl, uint16_t dmem_dr, uint16_t dmem_wl, uint16_t dmem_wr, uint16_t dmemi, unsigned count, uint16_t *env_values, uint16_t *env_steps, const int16_t *xors); void alist_mix(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count, int16_t gain); void alist_multQ44(struct hle_t* hle, uint16_t dmem, uint16_t count, int8_t gain); void alist_add(struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count); void alist_adpcm( struct hle_t* hle, bool init, bool loop, bool two_bit_per_sample, uint16_t dmemo, uint16_t dmemi, uint16_t count, const int16_t* codebook, uint32_t loop_address, uint32_t last_frame_address); void alist_resample( struct hle_t* hle, bool init, bool flag2, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint32_t pitch, uint32_t address); void alist_resample_zoh( struct hle_t* hle, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint32_t pitch, uint32_t pitch_accu); void alist_filter( struct hle_t* hle, uint16_t dmem, uint16_t count, uint32_t address, const uint32_t* lut_address); void alist_polef( struct hle_t* hle, bool init, uint16_t dmemo, uint16_t dmemi, uint16_t count, uint16_t gain, int16_t* table, uint32_t address); void alist_iirf( struct hle_t* hle, bool init, uint16_t dmemo, uint16_t dmemi, uint16_t count, int16_t* table, uint32_t address); void alist_overload( struct hle_t* hle, uint16_t dmem, int16_t count, int16_t gain, uint16_t attenuation); /* * Audio flags */ #define A_INIT 0x01 #define A_CONTINUE 0x00 #define A_LOOP 0x02 #define A_OUT 0x02 #define A_LEFT 0x02 #define A_RIGHT 0x00 #define A_VOL 0x04 #define A_RATE 0x00 #define A_AUX 0x08 #define A_NOAUX 0x00 #define A_MAIN 0x00 #define A_MIX 0x10 #endif mupen64plus-rsp-hle-src-2.6.0/src/alist_audio.c000066400000000000000000000225111464506742100213200ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - alist_audio.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "alist.h" #include "common.h" #include "hle_internal.h" #include "memory.h" #include "ucodes.h" enum { DMEM_BASE = 0x5c0 }; /* helper functions */ static uint32_t get_address(struct hle_t* hle, uint32_t so) { return alist_get_address(hle, so, hle->alist_audio.segments, N_SEGMENTS); } static void set_address(struct hle_t* hle, uint32_t so) { alist_set_address(hle, so, hle->alist_audio.segments, N_SEGMENTS); } static void clear_segments(struct hle_t* hle) { memset(hle->alist_audio.segments, 0, N_SEGMENTS*sizeof(hle->alist_audio.segments[0])); } /* audio commands definition */ static void SPNOOP(struct hle_t* UNUSED(hle), uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { } static void CLEARBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmem = w1 + DMEM_BASE; uint16_t count = w2 & 0xfff; if (count == 0) return; alist_clear(hle, dmem, align(count, 16)); } static void ENVMIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = get_address(hle, w2); alist_envmix_exp( hle, flags & A_INIT, flags & A_AUX, hle->alist_audio.out, hle->alist_audio.dry_right, hle->alist_audio.wet_left, hle->alist_audio.wet_right, hle->alist_audio.in, hle->alist_audio.count, hle->alist_audio.dry, hle->alist_audio.wet, hle->alist_audio.vol, hle->alist_audio.target, hle->alist_audio.rate, address); } static void ENVMIXER_GE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = get_address(hle, w2); alist_envmix_ge( hle, flags & A_INIT, flags & A_AUX, hle->alist_audio.out, hle->alist_audio.dry_right, hle->alist_audio.wet_left, hle->alist_audio.wet_right, hle->alist_audio.in, hle->alist_audio.count, hle->alist_audio.dry, hle->alist_audio.wet, hle->alist_audio.vol, hle->alist_audio.target, hle->alist_audio.rate, address); } static void RESAMPLE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint16_t pitch = w1; uint32_t address = get_address(hle, w2); alist_resample( hle, flags & A_INIT, flags & 0x2, hle->alist_audio.out, hle->alist_audio.in, align(hle->alist_audio.count, 16), pitch << 1, address); } static void SETVOL(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); if (flags & A_AUX) { hle->alist_audio.dry = w1; hle->alist_audio.wet = w2; } else { unsigned lr = (flags & A_LEFT) ? 0 : 1; if (flags & A_VOL) hle->alist_audio.vol[lr] = w1; else { hle->alist_audio.target[lr] = w1; hle->alist_audio.rate[lr] = w2; } } } static void SETLOOP(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { hle->alist_audio.loop = get_address(hle, w2); } static void ADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = get_address(hle, w2); alist_adpcm( hle, flags & A_INIT, flags & A_LOOP, false, /* unsupported in this ucode */ hle->alist_audio.out, hle->alist_audio.in, align(hle->alist_audio.count, 32), hle->alist_audio.table, hle->alist_audio.loop, address); } static void LOADBUFF(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { uint32_t address = get_address(hle, w2); if (hle->alist_audio.count == 0) return; alist_load(hle, hle->alist_audio.in, address, hle->alist_audio.count); } static void SAVEBUFF(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { uint32_t address = get_address(hle, w2); if (hle->alist_audio.count == 0) return; alist_save(hle, hle->alist_audio.out, address, hle->alist_audio.count); } static void SETBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); if (flags & A_AUX) { hle->alist_audio.dry_right = w1 + DMEM_BASE; hle->alist_audio.wet_left = (w2 >> 16) + DMEM_BASE; hle->alist_audio.wet_right = w2 + DMEM_BASE; } else { hle->alist_audio.in = w1 + DMEM_BASE; hle->alist_audio.out = (w2 >> 16) + DMEM_BASE; hle->alist_audio.count = w2; } } static void DMEMMOVE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmemi = w1 + DMEM_BASE; uint16_t dmemo = (w2 >> 16) + DMEM_BASE; uint16_t count = w2; if (count == 0) return; alist_move(hle, dmemo, dmemi, align(count, 16)); } static void LOADADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = w1; uint32_t address = get_address(hle, w2); dram_load_u16(hle, (uint16_t*)hle->alist_audio.table, address, align(count, 8) >> 1); } static void INTERLEAVE(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { uint16_t left = (w2 >> 16) + DMEM_BASE; uint16_t right = w2 + DMEM_BASE; if (hle->alist_audio.count == 0) return; alist_interleave(hle, hle->alist_audio.out, left, right, align(hle->alist_audio.count, 16)); } static void MIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { int16_t gain = w1; uint16_t dmemi = (w2 >> 16) + DMEM_BASE; uint16_t dmemo = w2 + DMEM_BASE; if (hle->alist_audio.count == 0) return; alist_mix(hle, dmemo, dmemi, align(hle->alist_audio.count, 32), gain); } static void SEGMENT(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { set_address(hle, w2); } static void POLEF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint16_t gain = w1; uint32_t address = get_address(hle, w2); if (hle->alist_audio.count == 0) return; alist_polef( hle, flags & A_INIT, hle->alist_audio.out, hle->alist_audio.in, align(hle->alist_audio.count, 16), gain, hle->alist_audio.table, address); } /* global functions */ void alist_process_audio(struct hle_t* hle) { static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM , CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, SEGMENT, SETBUFF, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, POLEF, SETLOOP }; clear_segments(hle); alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_audio_ge(struct hle_t* hle) { static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM , CLEARBUFF, ENVMIXER_GE, LOADBUFF, RESAMPLE, SAVEBUFF, SEGMENT, SETBUFF, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, POLEF, SETLOOP }; clear_segments(hle); alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_audio_bc(struct hle_t* hle) { static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM , CLEARBUFF, ENVMIXER_GE, LOADBUFF, RESAMPLE, SAVEBUFF, SEGMENT, SETBUFF, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, POLEF, SETLOOP }; clear_segments(hle); alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } mupen64plus-rsp-hle-src-2.6.0/src/alist_naudio.c000066400000000000000000000261371464506742100215060ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - alist_naudio.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include "alist.h" #include "common.h" #include "hle_external.h" #include "hle_internal.h" #include "memory.h" #include "ucodes.h" enum { NAUDIO_COUNT = 0x170 }; /* ie 184 samples */ enum { NAUDIO_MAIN = 0x4f0, NAUDIO_MAIN2 = 0x660, NAUDIO_DRY_LEFT = 0x9d0, NAUDIO_DRY_RIGHT = 0xb40, NAUDIO_WET_LEFT = 0xcb0, NAUDIO_WET_RIGHT = 0xe20 }; /* audio commands definition */ static void UNKNOWN(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t acmd = (w1 >> 24); HleWarnMessage(hle->user_defined, "Unknown audio command %d: %08x %08x", acmd, w1, w2); } static void SPNOOP(struct hle_t* UNUSED(hle), uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { } static void NAUDIO_0000(struct hle_t* hle, uint32_t w1, uint32_t w2) { /* ??? */ UNKNOWN(hle, w1, w2); } static void NAUDIO_02B0(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { /* emulate code at 0x12b0 (inside SETVOL), because PC always execute in IMEM */ hle->alist_naudio.rate[1] &= ~0xffff; hle->alist_naudio.rate[1] |= (w2 & 0xffff); } static void NAUDIO_14(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint16_t gain = w1; uint8_t select_main = (w2 >> 24); uint32_t address = (w2 & 0xffffff); uint16_t dmem = (select_main == 0) ? NAUDIO_MAIN : NAUDIO_MAIN2; if (hle->alist_naudio.table[0] == 0 && hle->alist_naudio.table[1] == 0) { alist_polef( hle, flags & A_INIT, dmem, dmem, NAUDIO_COUNT, gain, hle->alist_naudio.table, address); } else { alist_iirf( hle, flags & A_INIT, dmem, dmem, NAUDIO_COUNT, hle->alist_naudio.table, address); } } static void SETVOL(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); if (flags & A_VOL) { if (flags & A_LEFT) { hle->alist_naudio.vol[0] = w1; hle->alist_naudio.dry = (w2 >> 16); hle->alist_naudio.wet = w2; } else { /* A_RIGHT */ hle->alist_naudio.target[1] = w1; hle->alist_naudio.rate[1] = w2; } } else { /* A_RATE */ hle->alist_naudio.target[0] = w1; hle->alist_naudio.rate[0] = w2; } } static void ENVMIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = (w2 & 0xffffff); hle->alist_naudio.vol[1] = w1; alist_envmix_lin( hle, flags & A_INIT, NAUDIO_DRY_LEFT, NAUDIO_DRY_RIGHT, NAUDIO_WET_LEFT, NAUDIO_WET_RIGHT, NAUDIO_MAIN, NAUDIO_COUNT, hle->alist_naudio.dry, hle->alist_naudio.wet, hle->alist_naudio.vol, hle->alist_naudio.target, hle->alist_naudio.rate, address); } static void CLEARBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmem = w1 + NAUDIO_MAIN; uint16_t count = w2 & 0xfff; alist_clear(hle, dmem, count); } static void MIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { int16_t gain = w1; uint16_t dmemi = (w2 >> 16) + NAUDIO_MAIN; uint16_t dmemo = w2 + NAUDIO_MAIN; alist_mix(hle, dmemo, dmemi, NAUDIO_COUNT, gain); } static void LOADBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xfff; uint16_t dmem = (w1 & 0xfff) + NAUDIO_MAIN; uint32_t address = (w2 & 0xffffff); alist_load(hle, dmem, address, count); } static void SAVEBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xfff; uint16_t dmem = (w1 & 0xfff) + NAUDIO_MAIN; uint32_t address = (w2 & 0xffffff); alist_save(hle, dmem, address, count); } static void LOADADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = w1; uint32_t address = (w2 & 0xffffff); dram_load_u16(hle, (uint16_t*)hle->alist_naudio.table, address, count >> 1); } static void DMEMMOVE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmemi = w1 + NAUDIO_MAIN; uint16_t dmemo = (w2 >> 16) + NAUDIO_MAIN; uint16_t count = w2; alist_move(hle, dmemo, dmemi, (count + 3) & ~3); } static void SETLOOP(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { hle->alist_naudio.loop = (w2 & 0xffffff); } static void ADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint32_t address = (w1 & 0xffffff); uint8_t flags = (w2 >> 28); uint16_t count = (w2 >> 16) & 0xfff; uint16_t dmemi = ((w2 >> 12) & 0xf) + NAUDIO_MAIN; uint16_t dmemo = (w2 & 0xfff) + NAUDIO_MAIN; alist_adpcm( hle, flags & A_INIT, flags & A_LOOP, false, /* unsuported by this ucode */ dmemo, dmemi, (count + 0x1f) & ~0x1f, hle->alist_naudio.table, hle->alist_naudio.loop, address); } static void RESAMPLE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint32_t address = (w1 & 0xffffff); uint8_t flags = (w2 >> 30); uint16_t pitch = (w2 >> 14); uint16_t dmemi = ((w2 >> 2) & 0xfff) + NAUDIO_MAIN; uint16_t dmemo = (w2 & 0x3) ? NAUDIO_MAIN2 : NAUDIO_MAIN; alist_resample( hle, flags & A_INIT, false, /* TODO: check which ABI supports it */ dmemo, dmemi, NAUDIO_COUNT, pitch << 1, address); } static void INTERLEAVE(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { alist_interleave(hle, NAUDIO_MAIN, NAUDIO_DRY_LEFT, NAUDIO_DRY_RIGHT, NAUDIO_COUNT); } static void MP3ADDY(struct hle_t* UNUSED(hle), uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { } static void MP3(struct hle_t* hle, uint32_t w1, uint32_t w2) { unsigned index = (w1 & 0x1e); uint32_t address = (w2 & 0xffffff); mp3_task(hle, index, address); } static void OVERLOAD(struct hle_t* hle, uint32_t w1, uint32_t w2) { /* Overload distortion effect for Conker's Bad Fur Day */ uint16_t dmem = (w1 & 0xfff) + NAUDIO_MAIN; int16_t gain = (int16_t)(uint16_t)w2; uint16_t attenuation = w2 >> 16; alist_overload(hle, dmem, NAUDIO_COUNT, gain, attenuation); } /* global functions */ void alist_process_naudio(struct hle_t* hle) { static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM, CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, NAUDIO_0000, NAUDIO_0000, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, NAUDIO_02B0, SETLOOP }; alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_naudio_bk(struct hle_t* hle) { /* TODO: see what differs from alist_process_naudio */ static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM, CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, NAUDIO_0000, NAUDIO_0000, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, NAUDIO_02B0, SETLOOP }; alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_naudio_dk(struct hle_t* hle) { /* TODO: see what differs from alist_process_naudio */ static const acmd_callback_t ABI[0x10] = { SPNOOP, ADPCM, CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, MIXER, MIXER, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, NAUDIO_02B0, SETLOOP }; alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_naudio_mp3(struct hle_t* hle) { static const acmd_callback_t ABI[0x10] = { OVERLOAD, ADPCM, CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, MP3, MP3ADDY, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, NAUDIO_14, SETLOOP }; alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_naudio_cbfd(struct hle_t* hle) { /* What differs from alist_process_naudio_mp3? * * JoshW: It appears that despite being a newer game, CBFD appears to have a slightly older ucode version * compared to JFG, B.T. et al. * For naudio_mp3, the functions DMEM parameters have an additional protective AND on them * (basically dmem & 0xffff). * But there are minor differences are in the RESAMPLE and ENVMIXER functions. * I don't think it is making any noticeable difference, as it could be just a simplification of the logic. * * bsmiles32: The only difference I could remember between mp3 and cbfd variants is in the MP3ADDY command. * And the MP3 overlay is also different. */ static const acmd_callback_t ABI[0x10] = { OVERLOAD, ADPCM, CLEARBUFF, ENVMIXER, LOADBUFF, RESAMPLE, SAVEBUFF, MP3, MP3ADDY, SETVOL, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, NAUDIO_14, SETLOOP }; alist_process(hle, ABI, 0x10); rsp_break(hle, SP_STATUS_TASKDONE); } mupen64plus-rsp-hle-src-2.6.0/src/alist_nead.c000066400000000000000000000426311464506742100211330ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - alist_nead.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include "alist.h" #include "common.h" #include "hle_external.h" #include "hle_internal.h" #include "memory.h" #include "ucodes.h" /* remove windows define to 0x06 */ #ifdef DUPLICATE #undef DUPLICATE #endif /* audio commands definition */ static void UNKNOWN(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t acmd = (w1 >> 24); HleWarnMessage(hle->user_defined, "Unknown audio command %d: %08x %08x", acmd, w1, w2); } static void SPNOOP(struct hle_t* UNUSED(hle), uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { } static void LOADADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = w1; uint32_t address = (w2 & 0xffffff); dram_load_u16(hle, (uint16_t*)hle->alist_nead.table, address, count >> 1); } static void SETLOOP(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { hle->alist_nead.loop = w2 & 0xffffff; } static void SETBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { hle->alist_nead.in = w1; hle->alist_nead.out = (w2 >> 16); hle->alist_nead.count = w2; } static void ADPCM(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = (w2 & 0xffffff); alist_adpcm( hle, flags & 0x1, flags & 0x2, flags & 0x4, hle->alist_nead.out, hle->alist_nead.in, (hle->alist_nead.count + 0x1f) & ~0x1f, hle->alist_nead.table, hle->alist_nead.loop, address); } static void CLEARBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmem = w1; uint16_t count = w2 & 0xfff; if (count == 0) return; alist_clear(hle, dmem, count); } static void LOADBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xfff; uint16_t dmem = (w1 & 0xfff); uint32_t address = (w2 & 0xffffff); alist_load(hle, dmem, address, count); } static void SAVEBUFF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xfff; uint16_t dmem = (w1 & 0xfff); uint32_t address = (w2 & 0xffffff); alist_save(hle, dmem, address, count); } static void MIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xff0; int16_t gain = w1; uint16_t dmemi = (w2 >> 16); uint16_t dmemo = w2; alist_mix(hle, dmemo, dmemi, count, gain); } static void RESAMPLE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint16_t pitch = w1; uint32_t address = (w2 & 0xffffff); alist_resample( hle, flags & 0x1, false, /* TODO: check which ABI supports it */ hle->alist_nead.out, hle->alist_nead.in, (hle->alist_nead.count + 0xf) & ~0xf, pitch << 1, address); } static void RESAMPLE_ZOH(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t pitch = w1; uint16_t pitch_accu = w2; alist_resample_zoh( hle, hle->alist_nead.out, hle->alist_nead.in, hle->alist_nead.count, pitch << 1, pitch_accu); } static void DMEMMOVE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t dmemi = w1; uint16_t dmemo = (w2 >> 16); uint16_t count = w2; if (count == 0) return; alist_move(hle, dmemo, dmemi, (count + 3) & ~3); } static void ENVSETUP1_MK(struct hle_t* hle, uint32_t w1, uint32_t w2) { hle->alist_nead.env_values[2] = (w1 >> 8) & 0xff00; hle->alist_nead.env_steps[2] = 0; hle->alist_nead.env_steps[0] = (w2 >> 16); hle->alist_nead.env_steps[1] = w2; } static void ENVSETUP1(struct hle_t* hle, uint32_t w1, uint32_t w2) { hle->alist_nead.env_values[2] = (w1 >> 8) & 0xff00; hle->alist_nead.env_steps[2] = w1; hle->alist_nead.env_steps[0] = (w2 >> 16); hle->alist_nead.env_steps[1] = w2; } static void ENVSETUP2(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { hle->alist_nead.env_values[0] = (w2 >> 16); hle->alist_nead.env_values[1] = w2; } static void ENVMIXER_MK(struct hle_t* hle, uint32_t w1, uint32_t w2) { int16_t xors[4]; uint16_t dmemi = (w1 >> 12) & 0xff0; uint8_t count = (w1 >> 8) & 0xff; uint16_t dmem_dl = (w2 >> 20) & 0xff0; uint16_t dmem_dr = (w2 >> 12) & 0xff0; uint16_t dmem_wl = (w2 >> 4) & 0xff0; uint16_t dmem_wr = (w2 << 4) & 0xff0; xors[2] = 0; /* unsupported by this ucode */ xors[3] = 0; /* unsupported by this ucode */ xors[0] = 0 - (int16_t)((w1 & 0x2) >> 1); xors[1] = 0 - (int16_t)((w1 & 0x1) ); alist_envmix_nead( hle, false, /* unsupported by this ucode */ dmem_dl, dmem_dr, dmem_wl, dmem_wr, dmemi, count, hle->alist_nead.env_values, hle->alist_nead.env_steps, xors); } static void ENVMIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { int16_t xors[4]; uint16_t dmemi = (w1 >> 12) & 0xff0; uint8_t count = (w1 >> 8) & 0xff; bool swap_wet_LR = (w1 >> 4) & 0x1; uint16_t dmem_dl = (w2 >> 20) & 0xff0; uint16_t dmem_dr = (w2 >> 12) & 0xff0; uint16_t dmem_wl = (w2 >> 4) & 0xff0; uint16_t dmem_wr = (w2 << 4) & 0xff0; xors[2] = 0 - (int16_t)((w1 & 0x8) >> 1); xors[3] = 0 - (int16_t)((w1 & 0x4) >> 1); xors[0] = 0 - (int16_t)((w1 & 0x2) >> 1); xors[1] = 0 - (int16_t)((w1 & 0x1) ); alist_envmix_nead( hle, swap_wet_LR, dmem_dl, dmem_dr, dmem_wl, dmem_wr, dmemi, count, hle->alist_nead.env_values, hle->alist_nead.env_steps, xors); } static void DUPLICATE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t count = (w1 >> 16); uint16_t dmemi = w1; uint16_t dmemo = (w2 >> 16); alist_repeat64(hle, dmemo, dmemi, count); } static void INTERL(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = w1; uint16_t dmemi = (w2 >> 16); uint16_t dmemo = w2; alist_copy_every_other_sample(hle, dmemo, dmemi, count); } static void INTERLEAVE_MK(struct hle_t* hle, uint32_t UNUSED(w1), uint32_t w2) { uint16_t left = (w2 >> 16); uint16_t right = w2; if (hle->alist_nead.count == 0) return; alist_interleave(hle, hle->alist_nead.out, left, right, hle->alist_nead.count); } static void INTERLEAVE(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = ((w1 >> 12) & 0xff0); uint16_t dmemo = w1; uint16_t left = (w2 >> 16); uint16_t right = w2; alist_interleave(hle, dmemo, left, right, count); } static void ADDMIXER(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint16_t count = (w1 >> 12) & 0xff0; uint16_t dmemi = (w2 >> 16); uint16_t dmemo = w2; alist_add(hle, dmemo, dmemi, count); } static void HILOGAIN(struct hle_t* hle, uint32_t w1, uint32_t w2) { int8_t gain = (w1 >> 16); /* Q4.4 signed */ uint16_t count = w1 & 0xfff; uint16_t dmem = (w2 >> 16); alist_multQ44(hle, dmem, count, gain); } static void FILTER(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint32_t address = (w2 & 0xffffff); if (flags > 1) { hle->alist_nead.filter_count = w1; hle->alist_nead.filter_lut_address[0] = address; /* t6 */ } else { uint16_t dmem = w1; hle->alist_nead.filter_lut_address[1] = address + 0x10; /* t5 */ alist_filter(hle, dmem, hle->alist_nead.filter_count, address, hle->alist_nead.filter_lut_address); } } static void SEGMENT(struct hle_t* UNUSED(hle), uint32_t UNUSED(w1), uint32_t UNUSED(w2)) { } static void NEAD_16(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t count = (w1 >> 16); uint16_t dmemi = w1; uint16_t dmemo = (w2 >> 16); uint16_t block_size = w2; alist_copy_blocks(hle, dmemo, dmemi, block_size, count); } static void POLEF(struct hle_t* hle, uint32_t w1, uint32_t w2) { uint8_t flags = (w1 >> 16); uint16_t gain = w1; uint32_t address = (w2 & 0xffffff); if (hle->alist_nead.count == 0) return; alist_polef( hle, flags & A_INIT, hle->alist_nead.out, hle->alist_nead.in, hle->alist_nead.count, gain, hle->alist_nead.table, address); } void alist_process_nead_mk(struct hle_t* hle) { static const acmd_callback_t ABI[0x20] = { SPNOOP, ADPCM, CLEARBUFF, SPNOOP, SPNOOP, RESAMPLE, SPNOOP, SEGMENT, SETBUFF, SPNOOP, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE_MK, POLEF, SETLOOP, NEAD_16, INTERL, ENVSETUP1_MK, ENVMIXER_MK, LOADBUFF, SAVEBUFF, ENVSETUP2, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP }; alist_process(hle, ABI, 0x20); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_sf(struct hle_t* hle) { static const acmd_callback_t ABI[0x20] = { SPNOOP, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, SPNOOP, SETBUFF, SPNOOP, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE_MK, POLEF, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, SPNOOP, HILOGAIN, UNKNOWN, DUPLICATE, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP }; alist_process(hle, ABI, 0x20); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_sfj(struct hle_t* hle) { static const acmd_callback_t ABI[0x20] = { SPNOOP, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, SPNOOP, SETBUFF, SPNOOP, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE_MK, POLEF, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN, HILOGAIN, UNKNOWN, DUPLICATE, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP }; alist_process(hle, ABI, 0x20); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_fz(struct hle_t* hle) { static const acmd_callback_t ABI[0x20] = { UNKNOWN, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, SPNOOP, SPNOOP, SETBUFF, SPNOOP, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, SPNOOP, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN, SPNOOP, UNKNOWN, DUPLICATE, SPNOOP, SPNOOP, SPNOOP, SPNOOP, SPNOOP }; alist_process(hle, ABI, 0x20); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_wrjb(struct hle_t* hle) { static const acmd_callback_t ABI[0x20] = { SPNOOP, ADPCM, CLEARBUFF, UNKNOWN, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, SPNOOP, SETBUFF, SPNOOP, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, SPNOOP, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN, HILOGAIN, UNKNOWN, DUPLICATE, FILTER, SPNOOP, SPNOOP, SPNOOP, SPNOOP }; alist_process(hle, ABI, 0x20); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_ys(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { UNKNOWN, ADPCM, CLEARBUFF, UNKNOWN, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_1080(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { UNKNOWN, ADPCM, CLEARBUFF, UNKNOWN, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_oot(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { UNKNOWN, ADPCM, CLEARBUFF, UNKNOWN, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_mm(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { UNKNOWN, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_mmb(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { SPNOOP, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_ac(struct hle_t* hle) { static const acmd_callback_t ABI[0x18] = { UNKNOWN, ADPCM, CLEARBUFF, SPNOOP, ADDMIXER, RESAMPLE, RESAMPLE_ZOH, FILTER, SETBUFF, DUPLICATE, DMEMMOVE, LOADADPCM, MIXER, INTERLEAVE, HILOGAIN, SETLOOP, NEAD_16, INTERL, ENVSETUP1, ENVMIXER, LOADBUFF, SAVEBUFF, ENVSETUP2, UNKNOWN }; alist_process(hle, ABI, 0x18); rsp_break(hle, SP_STATUS_TASKDONE); } void alist_process_nead_mats(struct hle_t* hle) { /* FIXME: implement proper ucode * Forward the task if possible, * otherwise better to have no sound than garbage sound */ if (HleForwardTask(hle->user_defined) != 0) { rsp_break(hle, SP_STATUS_TASKDONE); } } void alist_process_nead_efz(struct hle_t* hle) { /* FIXME: implement proper ucode * Forward the task if possible, * otherwise use FZero ucode which should be very similar */ if (HleForwardTask(hle->user_defined) != 0) { alist_process_nead_fz(hle); } } mupen64plus-rsp-hle-src-2.6.0/src/arithmetics.h000066400000000000000000000035471464506742100213540ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - arithmetics.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef ARITHMETICS_H #define ARITHMETICS_H #include #include "common.h" static inline int16_t clamp_s16(int_fast32_t x) { x = (x < INT16_MIN) ? INT16_MIN: x; x = (x > INT16_MAX) ? INT16_MAX: x; return x; } static inline int32_t vmulf(int16_t x, int16_t y) { return (((int32_t)(x))*((int32_t)(y))+0x4000)>>15; } #endif mupen64plus-rsp-hle-src-2.6.0/src/audio.c000066400000000000000000000155471464506742100201370ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - audio.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "arithmetics.h" const int16_t RESAMPLE_LUT[64 * 4] = { (int16_t)0x0c39, (int16_t)0x66ad, (int16_t)0x0d46, (int16_t)0xffdf, (int16_t)0x0b39, (int16_t)0x6696, (int16_t)0x0e5f, (int16_t)0xffd8, (int16_t)0x0a44, (int16_t)0x6669, (int16_t)0x0f83, (int16_t)0xffd0, (int16_t)0x095a, (int16_t)0x6626, (int16_t)0x10b4, (int16_t)0xffc8, (int16_t)0x087d, (int16_t)0x65cd, (int16_t)0x11f0, (int16_t)0xffbf, (int16_t)0x07ab, (int16_t)0x655e, (int16_t)0x1338, (int16_t)0xffb6, (int16_t)0x06e4, (int16_t)0x64d9, (int16_t)0x148c, (int16_t)0xffac, (int16_t)0x0628, (int16_t)0x643f, (int16_t)0x15eb, (int16_t)0xffa1, (int16_t)0x0577, (int16_t)0x638f, (int16_t)0x1756, (int16_t)0xff96, (int16_t)0x04d1, (int16_t)0x62cb, (int16_t)0x18cb, (int16_t)0xff8a, (int16_t)0x0435, (int16_t)0x61f3, (int16_t)0x1a4c, (int16_t)0xff7e, (int16_t)0x03a4, (int16_t)0x6106, (int16_t)0x1bd7, (int16_t)0xff71, (int16_t)0x031c, (int16_t)0x6007, (int16_t)0x1d6c, (int16_t)0xff64, (int16_t)0x029f, (int16_t)0x5ef5, (int16_t)0x1f0b, (int16_t)0xff56, (int16_t)0x022a, (int16_t)0x5dd0, (int16_t)0x20b3, (int16_t)0xff48, (int16_t)0x01be, (int16_t)0x5c9a, (int16_t)0x2264, (int16_t)0xff3a, (int16_t)0x015b, (int16_t)0x5b53, (int16_t)0x241e, (int16_t)0xff2c, (int16_t)0x0101, (int16_t)0x59fc, (int16_t)0x25e0, (int16_t)0xff1e, (int16_t)0x00ae, (int16_t)0x5896, (int16_t)0x27a9, (int16_t)0xff10, (int16_t)0x0063, (int16_t)0x5720, (int16_t)0x297a, (int16_t)0xff02, (int16_t)0x001f, (int16_t)0x559d, (int16_t)0x2b50, (int16_t)0xfef4, (int16_t)0xffe2, (int16_t)0x540d, (int16_t)0x2d2c, (int16_t)0xfee8, (int16_t)0xffac, (int16_t)0x5270, (int16_t)0x2f0d, (int16_t)0xfedb, (int16_t)0xff7c, (int16_t)0x50c7, (int16_t)0x30f3, (int16_t)0xfed0, (int16_t)0xff53, (int16_t)0x4f14, (int16_t)0x32dc, (int16_t)0xfec6, (int16_t)0xff2e, (int16_t)0x4d57, (int16_t)0x34c8, (int16_t)0xfebd, (int16_t)0xff0f, (int16_t)0x4b91, (int16_t)0x36b6, (int16_t)0xfeb6, (int16_t)0xfef5, (int16_t)0x49c2, (int16_t)0x38a5, (int16_t)0xfeb0, (int16_t)0xfedf, (int16_t)0x47ed, (int16_t)0x3a95, (int16_t)0xfeac, (int16_t)0xfece, (int16_t)0x4611, (int16_t)0x3c85, (int16_t)0xfeab, (int16_t)0xfec0, (int16_t)0x4430, (int16_t)0x3e74, (int16_t)0xfeac, (int16_t)0xfeb6, (int16_t)0x424a, (int16_t)0x4060, (int16_t)0xfeaf, (int16_t)0xfeaf, (int16_t)0x4060, (int16_t)0x424a, (int16_t)0xfeb6, (int16_t)0xfeac, (int16_t)0x3e74, (int16_t)0x4430, (int16_t)0xfec0, (int16_t)0xfeab, (int16_t)0x3c85, (int16_t)0x4611, (int16_t)0xfece, (int16_t)0xfeac, (int16_t)0x3a95, (int16_t)0x47ed, (int16_t)0xfedf, (int16_t)0xfeb0, (int16_t)0x38a5, (int16_t)0x49c2, (int16_t)0xfef5, (int16_t)0xfeb6, (int16_t)0x36b6, (int16_t)0x4b91, (int16_t)0xff0f, (int16_t)0xfebd, (int16_t)0x34c8, (int16_t)0x4d57, (int16_t)0xff2e, (int16_t)0xfec6, (int16_t)0x32dc, (int16_t)0x4f14, (int16_t)0xff53, (int16_t)0xfed0, (int16_t)0x30f3, (int16_t)0x50c7, (int16_t)0xff7c, (int16_t)0xfedb, (int16_t)0x2f0d, (int16_t)0x5270, (int16_t)0xffac, (int16_t)0xfee8, (int16_t)0x2d2c, (int16_t)0x540d, (int16_t)0xffe2, (int16_t)0xfef4, (int16_t)0x2b50, (int16_t)0x559d, (int16_t)0x001f, (int16_t)0xff02, (int16_t)0x297a, (int16_t)0x5720, (int16_t)0x0063, (int16_t)0xff10, (int16_t)0x27a9, (int16_t)0x5896, (int16_t)0x00ae, (int16_t)0xff1e, (int16_t)0x25e0, (int16_t)0x59fc, (int16_t)0x0101, (int16_t)0xff2c, (int16_t)0x241e, (int16_t)0x5b53, (int16_t)0x015b, (int16_t)0xff3a, (int16_t)0x2264, (int16_t)0x5c9a, (int16_t)0x01be, (int16_t)0xff48, (int16_t)0x20b3, (int16_t)0x5dd0, (int16_t)0x022a, (int16_t)0xff56, (int16_t)0x1f0b, (int16_t)0x5ef5, (int16_t)0x029f, (int16_t)0xff64, (int16_t)0x1d6c, (int16_t)0x6007, (int16_t)0x031c, (int16_t)0xff71, (int16_t)0x1bd7, (int16_t)0x6106, (int16_t)0x03a4, (int16_t)0xff7e, (int16_t)0x1a4c, (int16_t)0x61f3, (int16_t)0x0435, (int16_t)0xff8a, (int16_t)0x18cb, (int16_t)0x62cb, (int16_t)0x04d1, (int16_t)0xff96, (int16_t)0x1756, (int16_t)0x638f, (int16_t)0x0577, (int16_t)0xffa1, (int16_t)0x15eb, (int16_t)0x643f, (int16_t)0x0628, (int16_t)0xffac, (int16_t)0x148c, (int16_t)0x64d9, (int16_t)0x06e4, (int16_t)0xffb6, (int16_t)0x1338, (int16_t)0x655e, (int16_t)0x07ab, (int16_t)0xffbf, (int16_t)0x11f0, (int16_t)0x65cd, (int16_t)0x087d, (int16_t)0xffc8, (int16_t)0x10b4, (int16_t)0x6626, (int16_t)0x095a, (int16_t)0xffd0, (int16_t)0x0f83, (int16_t)0x6669, (int16_t)0x0a44, (int16_t)0xffd8, (int16_t)0x0e5f, (int16_t)0x6696, (int16_t)0x0b39, (int16_t)0xffdf, (int16_t)0x0d46, (int16_t)0x66ad, (int16_t)0x0c39 }; int32_t rdot(size_t n, const int16_t *x, const int16_t *y) { int32_t accu = 0; y += n; while (n != 0) { accu += *(x++) * *(--y); --n; } return accu; } void adpcm_compute_residuals(int16_t* dst, const int16_t* src, const int16_t* cb_entry, const int16_t* last_samples, size_t count) { const int16_t* const book1 = cb_entry; const int16_t* const book2 = cb_entry + 8; const int16_t l1 = last_samples[0]; const int16_t l2 = last_samples[1]; size_t i; assert(count <= 8); for(i = 0; i < count; ++i) { int32_t accu = (int32_t)src[i] << 11; accu += book1[i]*l1 + book2[i]*l2 + rdot(i, book2, src); dst[i] = clamp_s16(accu >> 11); } } mupen64plus-rsp-hle-src-2.6.0/src/audio.h000066400000000000000000000041071464506742100201320ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - audio.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef AUDIO_H #define AUDIO_H #include #include #include "common.h" extern const int16_t RESAMPLE_LUT[64 * 4]; int32_t rdot(size_t n, const int16_t *x, const int16_t *y); static inline int16_t adpcm_predict_sample(uint8_t byte, uint8_t mask, unsigned lshift, unsigned rshift) { int16_t sample = (uint16_t)(byte & mask) << lshift; sample >>= rshift; /* signed */ return sample; } void adpcm_compute_residuals(int16_t* dst, const int16_t* src, const int16_t* cb_entry, const int16_t* last_samples, size_t count); #endif mupen64plus-rsp-hle-src-2.6.0/src/cicx105.c000066400000000000000000000047221464506742100202030ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - cicx105.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2012 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include "hle_internal.h" /** * During IPL3 stage of CIC x105 games, the RSP performs some checks and transactions * necessary for booting the game. * * We only implement the needed DMA transactions for booting. * * Found in Banjo-Tooie, Zelda, Perfect Dark, ...) **/ void cicx105_ucode(struct hle_t* hle) { /* memcpy is okay to use because access constrains are met (alignment, size) */ unsigned int i; unsigned char *dst = hle->dram + 0x2fb1f0; unsigned char *src = hle->imem + 0x120; /* dma_read(0x1120, 0x1e8, 0x1e8) */ memcpy(hle->imem + 0x120, hle->dram + 0x1e8, 0x1f0); /* dma_write(0x1120, 0x2fb1f0, 0xfe817000) */ for (i = 0; i < 24; ++i) { memcpy(dst, src, 8); dst += 0xff0; src += 0x8; } rsp_break(hle, 0); } mupen64plus-rsp-hle-src-2.6.0/src/common.h000066400000000000000000000034601464506742100203220ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - common.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef COMMON_H #define COMMON_H /* macro for unused variable warning suppression */ #ifdef __GNUC__ # define UNUSED(x) UNUSED_ ## x __attribute__((__unused__)) #else # define UNUSED(x) UNUSED_ ## x #endif /* macro for inline keyword */ #ifdef _MSC_VER #define inline __inline #endif #endif mupen64plus-rsp-hle-src-2.6.0/src/hle.c000066400000000000000000000434401464506742100175770ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - hle.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2012 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #ifdef ENABLE_TASK_DUMP #include #endif #include "hle_external.h" #include "hle_internal.h" #include "memory.h" #include "ucodes.h" #define min(a,b) (((a) < (b)) ? (a) : (b)) /* some rdp status flags */ #define DP_STATUS_FREEZE 0x2 /* helper functions prototypes */ static unsigned int sum_bytes(const unsigned char *bytes, unsigned int size); static bool is_task(struct hle_t* hle); static void send_dlist_to_gfx_plugin(struct hle_t* hle); static ucode_func_t try_audio_task_detection(struct hle_t* hle); static ucode_func_t try_normal_task_detection(struct hle_t* hle); static ucode_func_t non_task_detection(struct hle_t* hle); static ucode_func_t task_detection(struct hle_t* hle); #ifdef ENABLE_TASK_DUMP static void dump_binary(struct hle_t* hle, const char *const filename, const unsigned char *const bytes, unsigned int size); static void dump_task(struct hle_t* hle, const char *const filename); static void dump_unknown_task(struct hle_t* hle, unsigned int uc_start); static void dump_unknown_non_task(struct hle_t* hle, unsigned int uc_start); #endif /* Global functions */ void hle_init(struct hle_t* hle, unsigned char* dram, unsigned char* dmem, unsigned char* imem, unsigned int* mi_intr, unsigned int* sp_mem_addr, unsigned int* sp_dram_addr, unsigned int* sp_rd_length, unsigned int* sp_wr_length, unsigned int* sp_status, unsigned int* sp_dma_full, unsigned int* sp_dma_busy, unsigned int* sp_pc, unsigned int* sp_semaphore, unsigned int* dpc_start, unsigned int* dpc_end, unsigned int* dpc_current, unsigned int* dpc_status, unsigned int* dpc_clock, unsigned int* dpc_bufbusy, unsigned int* dpc_pipebusy, unsigned int* dpc_tmem, void* user_defined) { hle->dram = dram; hle->dmem = dmem; hle->imem = imem; hle->mi_intr = mi_intr; hle->sp_mem_addr = sp_mem_addr; hle->sp_dram_addr = sp_dram_addr; hle->sp_rd_length = sp_rd_length; hle->sp_wr_length = sp_wr_length; hle->sp_status = sp_status; hle->sp_dma_full = sp_dma_full; hle->sp_dma_busy = sp_dma_busy; hle->sp_pc = sp_pc; hle->sp_semaphore = sp_semaphore; hle->dpc_start = dpc_start; hle->dpc_end = dpc_end; hle->dpc_current = dpc_current; hle->dpc_status = dpc_status; hle->dpc_clock = dpc_clock; hle->dpc_bufbusy = dpc_bufbusy; hle->dpc_pipebusy = dpc_pipebusy; hle->dpc_tmem = dpc_tmem; hle->user_defined = user_defined; } void hle_execute(struct hle_t* hle) { uint32_t uc_start = *dmem_u32(hle, TASK_UCODE); uint32_t uc_dstart = *dmem_u32(hle, TASK_UCODE_DATA); uint32_t uc_dsize = *dmem_u32(hle, TASK_UCODE_DATA_SIZE); bool match = false; struct cached_ucodes_t * cached_ucodes = &hle->cached_ucodes; struct ucode_info_t *info = NULL; if (cached_ucodes->count > 0) info = &cached_ucodes->infos[cached_ucodes->count-1]; for (int i = 0; i < cached_ucodes->count; i++) { if (info->uc_start == uc_start && info->uc_dstart == uc_dstart && info->uc_dsize == uc_dsize) { match = true; break; } info--; } if (!match) { info = &cached_ucodes->infos[cached_ucodes->count]; info->uc_start = uc_start; info->uc_dstart = uc_dstart; info->uc_dsize = uc_dsize; info->uc_pfunc = task_detection(hle); cached_ucodes->count++; assert(cached_ucodes->count <= CACHED_UCODES_MAX_SIZE); assert(info->uc_pfunc != NULL); } info->uc_pfunc(hle); } /* local functions */ static unsigned int sum_bytes(const unsigned char *bytes, unsigned int size) { unsigned int sum = 0; const unsigned char *const bytes_end = bytes + size; while (bytes != bytes_end) sum += *bytes++; return sum; } /** * Try to figure if the RSP was launched using osSpTask* functions * and not run directly (in which case DMEM[0xfc0-0xfff] is meaningless). * * Previously, the ucode_size field was used to determine this, * but it is not robust enough (hi Pokemon Stadium !) because games could write anything * in this field : most ucode_boot discard the value and just use 0xf7f anyway. * * Using ucode_boot_size should be more robust in this regard. **/ static bool is_task(struct hle_t* hle) { return (*dmem_u32(hle, TASK_UCODE_BOOT_SIZE) <= 0x1000); } void rsp_break(struct hle_t* hle, unsigned int setbits) { *hle->sp_status |= setbits | SP_STATUS_BROKE | SP_STATUS_HALT; if ((*hle->sp_status & SP_STATUS_INTR_ON_BREAK)) { *hle->mi_intr |= MI_INTR_SP; HleCheckInterrupts(hle->user_defined); } } static void send_alist_to_audio_plugin(struct hle_t* hle) { HleProcessAlistList(hle->user_defined); rsp_break(hle, SP_STATUS_TASKDONE); } static void send_dlist_to_gfx_plugin(struct hle_t* hle) { /* Since GFX_INFO version 2, these bits are set before calling the ProcessDlistList function. * And the GFX plugin is responsible to unset them if needed. * For GFX_INFO version < 2, the GFX plugin didn't have access to sp_status so * it doesn't matter if we set these bits before calling ProcessDlistList function. */ *hle->sp_status |= SP_STATUS_TASKDONE | SP_STATUS_BROKE | SP_STATUS_HALT; HleProcessDlistList(hle->user_defined); if ((*hle->sp_status & SP_STATUS_INTR_ON_BREAK) && (*hle->sp_status & (SP_STATUS_TASKDONE | SP_STATUS_BROKE | SP_STATUS_HALT))) { *hle->mi_intr |= MI_INTR_SP; HleCheckInterrupts(hle->user_defined); } } static void task_done(struct hle_t* hle) { rsp_break(hle, SP_STATUS_TASKDONE); } static void unknown_ucode(struct hle_t* hle) { /* Forward task to RSP Fallback. * If task is not forwarded, use the regular "unknown ucode" path */ if (HleForwardTask(hle->user_defined) != 0) { uint32_t uc_start = *dmem_u32(hle, TASK_UCODE); HleWarnMessage(hle->user_defined, "unknown RSP code: uc_start: %x PC:%x", uc_start, *hle->sp_pc); #ifdef ENABLE_TASK_DUMP dump_unknown_non_task(hle, uc_start); #endif } } static void unknown_task(struct hle_t* hle) { /* Forward task to RSP Fallback. * If task is not forwarded, use the regular "unknown task" path */ if (HleForwardTask(hle->user_defined) != 0) { /* Send task_done signal for unknown ucodes to allow further processings */ rsp_break(hle, SP_STATUS_TASKDONE); uint32_t uc_start = *dmem_u32(hle, TASK_UCODE); HleWarnMessage(hle->user_defined, "unknown OSTask: uc_start: %x PC:%x", uc_start, *hle->sp_pc); #ifdef ENABLE_TASK_DUMP dump_unknown_task(hle, uc_start); #endif } } static ucode_func_t try_audio_task_detection(struct hle_t* hle) { /* identify audio ucode by using the content of ucode_data */ uint32_t ucode_data = *dmem_u32(hle, TASK_UCODE_DATA); uint32_t v; if (*dram_u32(hle, ucode_data) == 0x00000001) { if (*dram_u32(hle, ucode_data + 0x30) == 0xf0000f00) { v = *dram_u32(hle, ucode_data + 0x28); switch(v) { case 0x1e24138c: /* audio ABI (most common) */ return &alist_process_audio; case 0x1dc8138c: /* GoldenEye */ return &alist_process_audio_ge; case 0x1e3c1390: /* BlastCorp, DiddyKongRacing */ return &alist_process_audio_bc; default: HleWarnMessage(hle->user_defined, "ABI1 identification regression: v=%08x", v); } } else { v = *dram_u32(hle, ucode_data + 0x10); switch(v) { case 0x11181350: /* MarioKart, WaveRace (E) */ return &alist_process_nead_mk; case 0x111812e0: /* StarFox (J) */ return &alist_process_nead_sfj; case 0x110412ac: /* WaveRace (J RevB) */ return &alist_process_nead_wrjb; case 0x110412cc: /* StarFox/LylatWars (except J) */ return &alist_process_nead_sf; case 0x1cd01250: /* FZeroX */ return &alist_process_nead_fz; case 0x1f08122c: /* YoshisStory */ return &alist_process_nead_ys; case 0x1f38122c: /* 1080° Snowboarding */ return &alist_process_nead_1080; case 0x1f681230: /* Zelda OoT / Zelda MM (J, J RevA) */ return &alist_process_nead_oot; case 0x1f801250: /* Zelda MM (except J, J RevA, E Beta), PokemonStadium 2 */ return &alist_process_nead_mm; case 0x109411f8: /* Zelda MM (E Beta) */ return &alist_process_nead_mmb; case 0x1eac11b8: /* AnimalCrossing */ return &alist_process_nead_ac; case 0x00010010: /* MusyX v2 (IndianaJones, BattleForNaboo) */ return &musyx_v2_task; case 0x1f701238: /* Mario Artist Talent Studio */ return &alist_process_nead_mats; case 0x1f4c1230: /* FZeroX Expansion */ return &alist_process_nead_efz; default: HleWarnMessage(hle->user_defined, "ABI2 identification regression: v=%08x", v); } } } else { v = *dram_u32(hle, ucode_data + 0x10); switch(v) { case 0x00000001: /* MusyX v1 RogueSquadron, ResidentEvil2, PolarisSnoCross, TheWorldIsNotEnough, RugratsInParis, NBAShowTime, HydroThunder, Tarzan, GauntletLegend, Rush2049 */ return &musyx_v1_task; case 0x0000127c: /* naudio (many games) */ return &alist_process_naudio; case 0x00001280: /* BanjoKazooie */ return &alist_process_naudio_bk; case 0x1c58126c: /* DonkeyKong */ return &alist_process_naudio_dk; case 0x1ae8143c: /* BanjoTooie, JetForceGemini, MickeySpeedWayUSA, PerfectDark */ return &alist_process_naudio_mp3; case 0x1ab0140c: /* ConkerBadFurDay */ return &alist_process_naudio_cbfd; default: HleWarnMessage(hle->user_defined, "ABI3 identification regression: v=%08x", v); } } return NULL; } static ucode_func_t try_normal_task_detection(struct hle_t* hle) { unsigned int sum = sum_bytes((void*)dram_u32(hle, *dmem_u32(hle, TASK_UCODE)), min(*dmem_u32(hle, TASK_UCODE_SIZE), 0xf80) >> 1); switch (sum) { /* StoreVe12: found in Zelda Ocarina of Time [misleading task->type == 4] */ case 0x278: /* Nothing to emulate */ return &task_done; /* GFX: Twintris [misleading task->type == 0] */ case 0x212ee: if (hle->hle_gfx) { return &send_dlist_to_gfx_plugin; } return NULL; /* JPEG: found in Pokemon Stadium J */ case 0x2c85a: return &jpeg_decode_PS0; /* JPEG: found in Zelda Ocarina of Time, Pokemon Stadium 1, Pokemon Stadium 2 */ case 0x2caa6: return &jpeg_decode_PS; /* JPEG: found in Ogre Battle, Bottom of the 9th */ case 0x130de: case 0x278b0: return &jpeg_decode_OB; } /* Resident Evil 2 */ sum = sum_bytes((void*)dram_u32(hle, *dmem_u32(hle, TASK_UCODE)), 256); switch (sum) { case 0x450f: return &resize_bilinear_task; case 0x3b44: return &decode_video_frame_task; case 0x3d84: return &fill_video_double_buffer_task; } /* HVQM */ sum = sum_bytes((void*)dram_u32(hle, *dmem_u32(hle, TASK_UCODE)), 1488); switch (sum) { case 0x19495: return &hvqm2_decode_sp1_task; case 0x19728: return &hvqm2_decode_sp2_task; } return NULL; } static ucode_func_t non_task_detection(struct hle_t* hle) { const unsigned int sum = sum_bytes(hle->imem, 44); if (sum == 0x9e2) { /* CIC x105 ucode (used during boot of CIC x105 games) */ return &cicx105_ucode; } return &unknown_ucode; } static ucode_func_t task_detection(struct hle_t* hle) { if (is_task(hle)) { ucode_func_t uc_pfunc; uint32_t type = *dmem_u32(hle, TASK_TYPE); if (type == 2) { if (hle->hle_aud) { return &send_alist_to_audio_plugin; } uc_pfunc = try_audio_task_detection(hle); if (uc_pfunc) return uc_pfunc; } uc_pfunc = try_normal_task_detection(hle); if (uc_pfunc) return uc_pfunc; if (type == 1) { if (hle->hle_gfx) { return &send_dlist_to_gfx_plugin; } } return &unknown_task; } else { return non_task_detection(hle); } } #ifdef ENABLE_TASK_DUMP static void dump_unknown_task(struct hle_t* hle, unsigned int uc_start) { char filename[256]; uint32_t ucode = *dmem_u32(hle, TASK_UCODE); uint32_t ucode_data = *dmem_u32(hle, TASK_UCODE_DATA); uint32_t data_ptr = *dmem_u32(hle, TASK_DATA_PTR); sprintf(&filename[0], "task_%x.log", uc_start); dump_task(hle, filename); /* dump ucode_boot */ sprintf(&filename[0], "ucode_boot_%x.bin", uc_start); dump_binary(hle, filename, (void*)dram_u32(hle, *dmem_u32(hle, TASK_UCODE_BOOT)), *dmem_u32(hle, TASK_UCODE_BOOT_SIZE)); /* dump ucode */ if (ucode != 0) { sprintf(&filename[0], "ucode_%x.bin", uc_start); dump_binary(hle, filename, (void*)dram_u32(hle, ucode), 0xf80); } /* dump ucode_data */ if (ucode_data != 0) { sprintf(&filename[0], "ucode_data_%x.bin", uc_start); dump_binary(hle, filename, (void*)dram_u32(hle, ucode_data), *dmem_u32(hle, TASK_UCODE_DATA_SIZE)); } /* dump data */ if (data_ptr != 0) { sprintf(&filename[0], "data_%x.bin", uc_start); dump_binary(hle, filename, (void*)dram_u32(hle, data_ptr), *dmem_u32(hle, TASK_DATA_SIZE)); } } static void dump_unknown_non_task(struct hle_t* hle, unsigned int uc_start) { char filename[256]; /* dump IMEM & DMEM for further analysis */ sprintf(&filename[0], "imem_%x.bin", uc_start); dump_binary(hle, filename, hle->imem, 0x1000); sprintf(&filename[0], "dmem_%x.bin", uc_start); dump_binary(hle, filename, hle->dmem, 0x1000); } static void dump_binary(struct hle_t* hle, const char *const filename, const unsigned char *const bytes, unsigned int size) { FILE *f; /* if file already exists, do nothing */ f = fopen(filename, "r"); if (f == NULL) { /* else we write bytes to the file */ f = fopen(filename, "wb"); if (f != NULL) { if (fwrite(bytes, 1, size, f) != size) HleErrorMessage(hle->user_defined, "Writing error on %s", filename); fclose(f); } else HleErrorMessage(hle->user_defined, "Couldn't open %s for writing !", filename); } else fclose(f); } static void dump_task(struct hle_t* hle, const char *const filename) { FILE *f; f = fopen(filename, "r"); if (f == NULL) { f = fopen(filename, "w"); fprintf(f, "type = %d\n" "flags = %d\n" "ucode_boot = %#08x size = %#x\n" "ucode = %#08x size = %#x\n" "ucode_data = %#08x size = %#x\n" "dram_stack = %#08x size = %#x\n" "output_buff = %#08x *size = %#x\n" "data = %#08x size = %#x\n" "yield_data = %#08x size = %#x\n", *dmem_u32(hle, TASK_TYPE), *dmem_u32(hle, TASK_FLAGS), *dmem_u32(hle, TASK_UCODE_BOOT), *dmem_u32(hle, TASK_UCODE_BOOT_SIZE), *dmem_u32(hle, TASK_UCODE), *dmem_u32(hle, TASK_UCODE_SIZE), *dmem_u32(hle, TASK_UCODE_DATA), *dmem_u32(hle, TASK_UCODE_DATA_SIZE), *dmem_u32(hle, TASK_DRAM_STACK), *dmem_u32(hle, TASK_DRAM_STACK_SIZE), *dmem_u32(hle, TASK_OUTPUT_BUFF), *dmem_u32(hle, TASK_OUTPUT_BUFF_SIZE), *dmem_u32(hle, TASK_DATA_PTR), *dmem_u32(hle, TASK_DATA_SIZE), *dmem_u32(hle, TASK_YIELD_DATA_PTR), *dmem_u32(hle, TASK_YIELD_DATA_SIZE)); fclose(f); } else fclose(f); } #endif mupen64plus-rsp-hle-src-2.6.0/src/hle.h000066400000000000000000000044151464506742100176030ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - hle.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef HLE_H #define HLE_H #include "hle_internal.h" void hle_init(struct hle_t* hle, unsigned char* dram, unsigned char* dmem, unsigned char* imem, unsigned int* mi_intr, unsigned int* sp_mem_addr, unsigned int* sp_dram_addr, unsigned int* sp_rd_length, unsigned int* sp_wr_length, unsigned int* sp_status, unsigned int* sp_dma_full, unsigned int* sp_dma_busy, unsigned int* sp_pc, unsigned int* sp_semaphore, unsigned int* dpc_start, unsigned int* dpc_end, unsigned int* dpc_current, unsigned int* dpc_status, unsigned int* dpc_clock, unsigned int* dpc_bufbusy, unsigned int* dpc_pipebusy, unsigned int* dpc_tmem, void* user_defined); void hle_execute(struct hle_t* hle); #endif mupen64plus-rsp-hle-src-2.6.0/src/hle_external.h000066400000000000000000000045541464506742100215110ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - hle_external.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef HLE_EXTERNAL_H #define HLE_EXTERNAL_H #if defined(__GNUC__) #define ATTR_FMT(fmtpos, attrpos) __attribute__ ((format (printf, fmtpos, attrpos))) #else #define ATTR_FMT(fmtpos, attrpos) #endif /* users of the hle core are expected to define these functions */ void HleVerboseMessage(void* user_defined, const char *message, ...) ATTR_FMT(2, 3); void HleInfoMessage(void* user_defined, const char *message, ...) ATTR_FMT(2, 3); void HleErrorMessage(void* user_defined, const char *message, ...) ATTR_FMT(2, 3); void HleWarnMessage(void* user_defined, const char *message, ...) ATTR_FMT(2, 3); void HleCheckInterrupts(void* user_defined); void HleProcessDlistList(void* user_defined); void HleProcessAlistList(void* user_defined); void HleProcessRdpList(void* user_defined); void HleShowCFB(void* user_defined); int HleForwardTask(void* user_defined); #endif mupen64plus-rsp-hle-src-2.6.0/src/hle_internal.h000066400000000000000000000060661464506742100215030ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - hle_internal.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef HLE_INTERNAL_H #define HLE_INTERNAL_H #include #include "ucodes.h" /* rsp hle internal state - internal usage only */ struct hle_t { unsigned char* dram; unsigned char* dmem; unsigned char* imem; unsigned int* mi_intr; unsigned int* sp_mem_addr; unsigned int* sp_dram_addr; unsigned int* sp_rd_length; unsigned int* sp_wr_length; unsigned int* sp_status; unsigned int* sp_dma_full; unsigned int* sp_dma_busy; unsigned int* sp_pc; unsigned int* sp_semaphore; unsigned int* dpc_start; unsigned int* dpc_end; unsigned int* dpc_current; unsigned int* dpc_status; unsigned int* dpc_clock; unsigned int* dpc_bufbusy; unsigned int* dpc_pipebusy; unsigned int* dpc_tmem; /* for user convenience, this will be passed to "external" functions */ void* user_defined; int hle_gfx; int hle_aud; /* alist.c */ uint8_t alist_buffer[0x1000]; /* alist_audio.c */ struct alist_audio_t alist_audio; /* alist_naudio.c */ struct alist_naudio_t alist_naudio; /* alist_nead.c */ struct alist_nead_t alist_nead; /* mp3.c */ uint8_t mp3_buffer[0x1000]; struct cached_ucodes_t cached_ucodes; }; /* some mips interface interrupt flags */ #define MI_INTR_SP 0x1 /* some rsp status flags */ #define SP_STATUS_HALT 0x1 #define SP_STATUS_BROKE 0x2 #define SP_STATUS_INTR_ON_BREAK 0x40 #define SP_STATUS_TASKDONE 0x200 void rsp_break(struct hle_t* hle, unsigned int setbits); #endif mupen64plus-rsp-hle-src-2.6.0/src/hvqm.c000066400000000000000000000250601464506742100200000ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - hvqm.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2020 Gilles Siberlin * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include #include #include "hle_external.h" #include "hle_internal.h" #include "memory.h" /* Nest size */ #define HVQM2_NESTSIZE_L 70 /* Number of elements on long side */ #define HVQM2_NESTSIZE_S 38 /* Number of elements on short side */ #define HVQM2_NESTSIZE (HVQM2_NESTSIZE_L * HVQM2_NESTSIZE_S) struct HVQM2Block { uint8_t nbase; uint8_t dc; uint8_t dc_l; uint8_t dc_r; uint8_t dc_u; uint8_t dc_d; }; struct HVQM2Basis { uint8_t sx; uint8_t sy; int16_t scale; uint16_t offset; uint16_t lineskip; }; struct HVQM2Arg { uint32_t info; uint32_t buf; uint16_t buf_width; uint8_t chroma_step_h; uint8_t chroma_step_v; uint16_t hmcus; uint16_t vmcus; uint8_t alpha; uint32_t nest; }; struct RGBA { uint8_t r; uint8_t g; uint8_t b; uint8_t a; }; static struct HVQM2Arg arg; static const int16_t constant[5][16] = { {0x0006,0x0008,0x0008,0x0006,0x0008,0x000A,0x000A,0x0008,0x0008,0x000A,0x000A,0x0008,0x0006,0x0008,0x0008,0x0006}, {0x0002,0x0000,0xFFFF,0xFFFF,0x0002,0x0000,0xFFFF,0xFFFF,0x0002,0x0000,0xFFFF,0xFFFF,0x0002,0x0000,0xFFFF,0xFFFF}, {0xFFFF,0xFFFF,0x0000,0x0002,0xFFFF,0xFFFF,0x0000,0x0002,0xFFFF,0xFFFF,0x0000,0x0002,0xFFFF,0xFFFF,0x0000,0x0002}, {0x0002,0x0002,0x0002,0x0002,0x0000,0x0000,0x0000,0x0000,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF}, {0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0xFFFF,0x0000,0x0000,0x0000,0x0000,0x0002,0x0002,0x0002,0x0002} }; static int process_info(struct hle_t* hle, uint8_t* base, int16_t* out) { struct HVQM2Block block; uint8_t nbase = *base; dram_load_u8(hle, (uint8_t*)&block, arg.info, sizeof(struct HVQM2Block)); arg.info += 8; *base = block.nbase & 0x7; if ((block.nbase & nbase) != 0) return 0; if (block.nbase == 0) { //LABEL8 for (int i = 0; i < 16; i++) { out[i] = constant[0][i] * block.dc; out[i] += constant[1][i] * block.dc_l; out[i] += constant[2][i] * block.dc_r; out[i] += constant[3][i] * block.dc_u; out[i] += constant[4][i] * block.dc_d; out[i] += 4; out[i] >>= 3; } } else if ((block.nbase & 0xf) == 0) { //LABEL7 for (int i = 0; i < 16; i++) { out[i] = *dram_u8(hle, arg.info); arg.info++; } } else if (*base == 0) { //LABEL6 for (int i = 0; i < 16; i++) { out[i] = *(int8_t*)dram_u8(hle, arg.info) + block.dc; arg.info++; } } else { //LABEL5 struct HVQM2Basis basis; for (int i = 0; i < 16; i++) out[i] = block.dc; for (; *base != 0; (*base)--) { basis.sx = *dram_u8(hle, arg.info); arg.info++; basis.sy = *dram_u8(hle, arg.info); arg.info++; basis.scale = *dram_u16(hle, arg.info); arg.info += 2; basis.offset = *dram_u16(hle, arg.info); arg.info += 2; basis.lineskip = *dram_u16(hle, arg.info); arg.info += 2; int16_t vec[16]; uint32_t addr = arg.nest + basis.offset; int shift = (basis.sx != 0) ? 1 : 0; //LABEL9 //LABEL10 for (int i = 0; i < 16; i += 4) { vec[i] = *dram_u8(hle, addr); vec[i + 1] = *dram_u8(hle, addr + (1 << shift)); vec[i + 2] = *dram_u8(hle, addr + (2 << shift)); vec[i + 3] = *dram_u8(hle, addr + (3 << shift)); addr += basis.lineskip; } //LABEL11 int16_t sum = 0x8; for (int i = 0; i < 16; i++) sum += vec[i]; sum >>= 4; int16_t max = 0; for (int i = 0; i < 16; i++) { vec[i] -= sum; max = (abs(vec[i]) > max) ? abs(vec[i]) : max; } double dmax = 0.0; if (max > 0) dmax = (double)(basis.scale << 2) / (double)max; for (int i = 0; i < 16; i++) out[i] += (vec[i] < 0) ? (int16_t)((double)vec[i] * dmax - 0.5) : (int16_t)((double)vec[i] * dmax + 0.5); block.nbase &= 8; } assert(block.nbase == 0); //if(block.nbase != 0) // LABEL6 } return 1; } #define SATURATE8(x) ((unsigned int) x <= 255 ? x : (x < 0 ? 0: 255)) static struct RGBA YCbCr_to_RGBA(int16_t Y, int16_t Cb, int16_t Cr, uint8_t alpha) { struct RGBA color; //Format S10.6 int r = (int)(((double)Y + 0.5) + (1.765625 * (double)(Cr - 128))); int g = (int)(((double)Y + 0.5) - (0.34375 * (double)(Cr - 128)) - (0.71875 * (double)(Cb - 128))); int b = (int)(((double)Y + 0.5) + (1.40625 * (double)(Cb - 128))); color.r = SATURATE8(r); color.g = SATURATE8(g); color.b = SATURATE8(b); color.a = alpha; return color; } void store_rgba5551(struct hle_t* hle, struct RGBA color, uint32_t * addr) { uint16_t pixel = ((color.b >> 3) << 11) | ((color.g >> 3) << 6) | ((color.r >> 3) << 1) | (color.a & 1); dram_store_u16(hle, &pixel, *addr, 1); *addr += 2; } void store_rgba8888(struct hle_t* hle, struct RGBA color, uint32_t * addr) { uint32_t pixel = (color.b << 24) | (color.g << 16) | (color.r << 8) | color.a; dram_store_u32(hle, &pixel, *addr, 1); *addr += 4; } typedef void(*store_pixel_t)(struct hle_t* hle, struct RGBA color, uint32_t * addr); static void hvqm2_decode(struct hle_t* hle, int is32) { //uint32_t uc_data_ptr = *dmem_u32(hle, TASK_UCODE_DATA); uint32_t data_ptr = *dmem_u32(hle, TASK_DATA_PTR); assert((*dmem_u32(hle, TASK_FLAGS) & 0x1) == 0); /* Fill HVQM2Arg struct */ arg.info = *dram_u32(hle, data_ptr); data_ptr += 4; arg.buf = *dram_u32(hle, data_ptr); data_ptr += 4; arg.buf_width = *dram_u16(hle, data_ptr); data_ptr += 2; arg.chroma_step_h = *dram_u8(hle, data_ptr); data_ptr++; arg.chroma_step_v = *dram_u8(hle, data_ptr); data_ptr++; arg.hmcus = *dram_u16(hle, data_ptr); data_ptr += 2; arg.vmcus = *dram_u16(hle, data_ptr); data_ptr += 2; arg.alpha = *dram_u8(hle, data_ptr); arg.nest = data_ptr + 1; assert(arg.chroma_step_h == 2); assert((arg.chroma_step_v == 1) || (arg.chroma_step_v == 2)); assert((*hle->sp_status & 0x80) == 0); //SP_STATUS_YIELD int length, skip; store_pixel_t store_pixel; if (is32) { length = 0x20; skip = arg.buf_width << 2; arg.buf_width <<= 4; store_pixel = &store_rgba8888; } else { length = 0x10; skip = arg.buf_width << 1; arg.buf_width <<= 3; store_pixel = &store_rgba5551; } if (arg.chroma_step_v == 2) arg.buf_width += arg.buf_width; for (int i = arg.vmcus; i != 0; i--) { uint32_t out; int j; for (j = arg.hmcus, out = arg.buf; j != 0; j--, out += length) { uint8_t base = 0x80; int16_t Cb[16], Cr[16], Y1[32], Y2[32]; int16_t* pCb = Cb; int16_t* pCr = Cr; int16_t* pY1 = Y1; int16_t* pY2 = Y2; if (arg.chroma_step_v == 2) { if (process_info(hle, &base, pY1) == 0) continue; if (process_info(hle, &base, pY2) == 0) continue; pY1 = &Y1[16]; pY2 = &Y2[16]; } if (process_info(hle, &base, pY1) == 0) continue; if (process_info(hle, &base, pY2) == 0) continue; if (process_info(hle, &base, Cr) == 0) continue; if (process_info(hle, &base, Cb) == 0) continue; pY1 = Y1; pY2 = Y2; uint32_t out_buf = out; for (int k = 0; k < 4; k++) { for (int m = 0; m < arg.chroma_step_v; m++) { uint32_t addr = out_buf; for (int l = 0; l < 4; l++) { struct RGBA color = YCbCr_to_RGBA(pY1[l], pCb[l >> 1], pCr[l >> 1], arg.alpha); store_pixel(hle, color, &addr); } for (int l = 0; l < 4; l++) { struct RGBA color = YCbCr_to_RGBA(pY2[l], pCb[(l + 4) >> 1], pCr[(l + 4) >> 1], arg.alpha); store_pixel(hle, color, &addr); } out_buf += skip; pY1 += 4; pY2 += 4; } pCr += 4; pCb += 4; } } arg.buf += arg.buf_width; } rsp_break(hle, SP_STATUS_TASKDONE); } void hvqm2_decode_sp1_task(struct hle_t* hle) { hvqm2_decode(hle, 0); } void hvqm2_decode_sp2_task(struct hle_t* hle) { hvqm2_decode(hle, 1); }mupen64plus-rsp-hle-src-2.6.0/src/jpeg.c000066400000000000000000000501221464506742100177470ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - jpeg.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2012 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "arithmetics.h" #include "hle_external.h" #include "hle_internal.h" #include "memory.h" #define SUBBLOCK_SIZE 64 typedef void (*tile_line_emitter_t)(struct hle_t* hle, const int16_t *y, const int16_t *u, uint32_t address); typedef void (*subblock_transform_t)(int16_t *dst, const int16_t *src); /* standard jpeg ucode decoder */ static void jpeg_decode_std(struct hle_t* hle, const char *const version, const subblock_transform_t transform_luma, const subblock_transform_t transform_chroma, const tile_line_emitter_t emit_line); /* helper functions */ static uint8_t clamp_u8(int16_t x); static int16_t clamp_s12(int16_t x); static uint16_t clamp_RGBA_component(int16_t x); /* pixel conversion & formatting */ static uint32_t GetUYVY(int16_t y1, int16_t y2, int16_t u, int16_t v); static uint16_t GetRGBA(int16_t y, int16_t u, int16_t v); /* tile line emitters */ static void EmitYUVTileLine(struct hle_t* hle, const int16_t *y, const int16_t *u, uint32_t address); static void EmitRGBATileLine(struct hle_t* hle, const int16_t *y, const int16_t *u, uint32_t address); /* macroblocks operations */ static void decode_macroblock_ob(int16_t *macroblock, int32_t *y_dc, int32_t *u_dc, int32_t *v_dc, const int16_t *qtable); static void decode_macroblock_std(const subblock_transform_t transform_luma, const subblock_transform_t transform_chroma, int16_t *macroblock, unsigned int subblock_count, const int16_t qtables[3][SUBBLOCK_SIZE]); static void EmitTilesMode0(struct hle_t* hle, const tile_line_emitter_t emit_line, const int16_t *macroblock, uint32_t address); static void EmitTilesMode2(struct hle_t* hle, const tile_line_emitter_t emit_line, const int16_t *macroblock, uint32_t address); /* subblocks operations */ static void TransposeSubBlock(int16_t *dst, const int16_t *src); static void ZigZagSubBlock(int16_t *dst, const int16_t *src); static void ReorderSubBlock(int16_t *dst, const int16_t *src, const unsigned int *table); static void MultSubBlocks(int16_t *dst, const int16_t *src1, const int16_t *src2, unsigned int shift); static void ScaleSubBlock(int16_t *dst, const int16_t *src, int16_t scale); static void RShiftSubBlock(int16_t *dst, const int16_t *src, unsigned int shift); static void InverseDCT1D(const float *const x, float *dst, unsigned int stride); static void InverseDCTSubBlock(int16_t *dst, const int16_t *src); static void RescaleYSubBlock(int16_t *dst, const int16_t *src); static void RescaleUVSubBlock(int16_t *dst, const int16_t *src); /* transposed dequantization table */ static const int16_t DEFAULT_QTABLE[SUBBLOCK_SIZE] = { 16, 12, 14, 14, 18, 24, 49, 72, 11, 12, 13, 17, 22, 35, 64, 92, 10, 14, 16, 22, 37, 55, 78, 95, 16, 19, 24, 29, 56, 64, 87, 98, 24, 26, 40, 51, 68, 81, 103, 112, 40, 58, 57, 87, 109, 104, 121, 100, 51, 60, 69, 80, 103, 113, 120, 103, 61, 55, 56, 62, 77, 92, 101, 99 }; /* zig-zag indices */ static const unsigned int ZIGZAG_TABLE[SUBBLOCK_SIZE] = { 0, 1, 5, 6, 14, 15, 27, 28, 2, 4, 7, 13, 16, 26, 29, 42, 3, 8, 12, 17, 25, 30, 41, 43, 9, 11, 18, 24, 31, 40, 44, 53, 10, 19, 23, 32, 39, 45, 52, 54, 20, 22, 33, 38, 46, 51, 55, 60, 21, 34, 37, 47, 50, 56, 59, 61, 35, 36, 48, 49, 57, 58, 62, 63 }; /* transposition indices */ static const unsigned int TRANSPOSE_TABLE[SUBBLOCK_SIZE] = { 0, 8, 16, 24, 32, 40, 48, 56, 1, 9, 17, 25, 33, 41, 49, 57, 2, 10, 18, 26, 34, 42, 50, 58, 3, 11, 19, 27, 35, 43, 51, 59, 4, 12, 20, 28, 36, 44, 52, 60, 5, 13, 21, 29, 37, 45, 53, 61, 6, 14, 22, 30, 38, 46, 54, 62, 7, 15, 23, 31, 39, 47, 55, 63 }; /* IDCT related constants * Cn = alpha * cos(n * PI / 16) (alpha is chosen such as C4 = 1) */ static const float IDCT_C3 = 1.175875602f; static const float IDCT_C6 = 0.541196100f; static const float IDCT_K[10] = { 0.765366865f, /* C2-C6 */ -1.847759065f, /* -C2-C6 */ -0.390180644f, /* C5-C3 */ -1.961570561f, /* -C5-C3 */ 1.501321110f, /* C1+C3-C5-C7 */ 2.053119869f, /* C1+C3-C5+C7 */ 3.072711027f, /* C1+C3+C5-C7 */ 0.298631336f, /* -C1+C3+C5-C7 */ -0.899976223f, /* C7-C3 */ -2.562915448f /* -C1-C3 */ }; /* global functions */ /*************************************************************************** * JPEG decoding ucode found in Japanese exclusive version of Pokemon Stadium. **************************************************************************/ void jpeg_decode_PS0(struct hle_t* hle) { jpeg_decode_std(hle, "PS0", RescaleYSubBlock, RescaleUVSubBlock, EmitYUVTileLine); rsp_break(hle, SP_STATUS_TASKDONE); } /*************************************************************************** * JPEG decoding ucode found in Ocarina of Time, Pokemon Stadium 1 and * Pokemon Stadium 2. **************************************************************************/ void jpeg_decode_PS(struct hle_t* hle) { jpeg_decode_std(hle, "PS", NULL, NULL, EmitRGBATileLine); rsp_break(hle, SP_STATUS_TASKDONE); } /*************************************************************************** * JPEG decoding ucode found in Ogre Battle and Bottom of the 9th. **************************************************************************/ void jpeg_decode_OB(struct hle_t* hle) { int16_t qtable[SUBBLOCK_SIZE]; unsigned int mb; int32_t y_dc = 0; int32_t u_dc = 0; int32_t v_dc = 0; uint32_t address = *dmem_u32(hle, TASK_DATA_PTR); const unsigned int macroblock_count = *dmem_u32(hle, TASK_DATA_SIZE); const int qscale = *dmem_u32(hle, TASK_YIELD_DATA_SIZE); HleVerboseMessage(hle->user_defined, "jpeg_decode_OB: *buffer=%x, #MB=%d, qscale=%d", address, macroblock_count, qscale); if (qscale != 0) { if (qscale > 0) ScaleSubBlock(qtable, DEFAULT_QTABLE, qscale); else RShiftSubBlock(qtable, DEFAULT_QTABLE, -qscale); } for (mb = 0; mb < macroblock_count; ++mb) { int16_t macroblock[6 * SUBBLOCK_SIZE]; dram_load_u16(hle, (uint16_t *)macroblock, address, 6 * SUBBLOCK_SIZE); decode_macroblock_ob(macroblock, &y_dc, &u_dc, &v_dc, (qscale != 0) ? qtable : NULL); EmitTilesMode2(hle, EmitYUVTileLine, macroblock, address); address += (2 * 6 * SUBBLOCK_SIZE); } rsp_break(hle, SP_STATUS_TASKDONE); } /* local functions */ static void jpeg_decode_std(struct hle_t* hle, const char *const version, const subblock_transform_t transform_luma, const subblock_transform_t transform_chroma, const tile_line_emitter_t emit_line) { int16_t qtables[3][SUBBLOCK_SIZE]; unsigned int mb; uint32_t address; uint32_t macroblock_count; uint32_t mode; uint32_t qtableY_ptr; uint32_t qtableU_ptr; uint32_t qtableV_ptr; unsigned int subblock_count; unsigned int macroblock_size; /* macroblock contains at most 6 subblocks */ int16_t macroblock[6 * SUBBLOCK_SIZE]; uint32_t data_ptr; if (*dmem_u32(hle, TASK_FLAGS) & 0x1) { HleWarnMessage(hle->user_defined, "jpeg_decode_%s: task yielding not implemented", version); return; } data_ptr = *dmem_u32(hle, TASK_DATA_PTR); address = *dram_u32(hle, data_ptr); macroblock_count = *dram_u32(hle, data_ptr + 4); mode = *dram_u32(hle, data_ptr + 8); qtableY_ptr = *dram_u32(hle, data_ptr + 12); qtableU_ptr = *dram_u32(hle, data_ptr + 16); qtableV_ptr = *dram_u32(hle, data_ptr + 20); HleVerboseMessage(hle->user_defined, "jpeg_decode_%s: *buffer=%x, #MB=%d, mode=%d, *Qy=%x, *Qu=%x, *Qv=%x", version, address, macroblock_count, mode, qtableY_ptr, qtableU_ptr, qtableV_ptr); if (mode != 0 && mode != 2) { HleWarnMessage(hle->user_defined, "jpeg_decode_%s: invalid mode %d", version, mode); return; } subblock_count = mode + 4; macroblock_size = subblock_count * SUBBLOCK_SIZE; dram_load_u16(hle, (uint16_t *)qtables[0], qtableY_ptr, SUBBLOCK_SIZE); dram_load_u16(hle, (uint16_t *)qtables[1], qtableU_ptr, SUBBLOCK_SIZE); dram_load_u16(hle, (uint16_t *)qtables[2], qtableV_ptr, SUBBLOCK_SIZE); for (mb = 0; mb < macroblock_count; ++mb) { dram_load_u16(hle, (uint16_t *)macroblock, address, macroblock_size); decode_macroblock_std(transform_luma, transform_chroma, macroblock, subblock_count, (const int16_t (*)[SUBBLOCK_SIZE])qtables); if (mode == 0) EmitTilesMode0(hle, emit_line, macroblock, address); else EmitTilesMode2(hle, emit_line, macroblock, address); address += 2 * macroblock_size; } } static uint8_t clamp_u8(int16_t x) { return (x & (0xff00)) ? ((-x) >> 15) & 0xff : x; } static int16_t clamp_s12(int16_t x) { if (x < -0x800) x = -0x800; else if (x > 0x7f0) x = 0x7f0; return x; } static uint16_t clamp_RGBA_component(int16_t x) { if (x > 0xff0) x = 0xff0; else if (x < 0) x = 0; return (x & 0xf80); } static uint32_t GetUYVY(int16_t y1, int16_t y2, int16_t u, int16_t v) { return (uint32_t)clamp_u8(u) << 24 | (uint32_t)clamp_u8(y1) << 16 | (uint32_t)clamp_u8(v) << 8 | (uint32_t)clamp_u8(y2); } static uint16_t GetRGBA(int16_t y, int16_t u, int16_t v) { const float fY = (float)y + 2048.0f; const float fU = (float)u; const float fV = (float)v; const uint16_t r = clamp_RGBA_component((int16_t)(fY + 1.4025 * fV)); const uint16_t g = clamp_RGBA_component((int16_t)(fY - 0.3443 * fU - 0.7144 * fV)); const uint16_t b = clamp_RGBA_component((int16_t)(fY + 1.7729 * fU)); return (r << 4) | (g >> 1) | (b >> 6) | 1; } static void EmitYUVTileLine(struct hle_t* hle, const int16_t *y, const int16_t *u, uint32_t address) { uint32_t uyvy[8]; const int16_t *const v = u + SUBBLOCK_SIZE; const int16_t *const y2 = y + SUBBLOCK_SIZE; uyvy[0] = GetUYVY(y[0], y[1], u[0], v[0]); uyvy[1] = GetUYVY(y[2], y[3], u[1], v[1]); uyvy[2] = GetUYVY(y[4], y[5], u[2], v[2]); uyvy[3] = GetUYVY(y[6], y[7], u[3], v[3]); uyvy[4] = GetUYVY(y2[0], y2[1], u[4], v[4]); uyvy[5] = GetUYVY(y2[2], y2[3], u[5], v[5]); uyvy[6] = GetUYVY(y2[4], y2[5], u[6], v[6]); uyvy[7] = GetUYVY(y2[6], y2[7], u[7], v[7]); dram_store_u32(hle, uyvy, address, 8); } static void EmitRGBATileLine(struct hle_t* hle, const int16_t *y, const int16_t *u, uint32_t address) { uint16_t rgba[16]; const int16_t *const v = u + SUBBLOCK_SIZE; const int16_t *const y2 = y + SUBBLOCK_SIZE; rgba[0] = GetRGBA(y[0], u[0], v[0]); rgba[1] = GetRGBA(y[1], u[0], v[0]); rgba[2] = GetRGBA(y[2], u[1], v[1]); rgba[3] = GetRGBA(y[3], u[1], v[1]); rgba[4] = GetRGBA(y[4], u[2], v[2]); rgba[5] = GetRGBA(y[5], u[2], v[2]); rgba[6] = GetRGBA(y[6], u[3], v[3]); rgba[7] = GetRGBA(y[7], u[3], v[3]); rgba[8] = GetRGBA(y2[0], u[4], v[4]); rgba[9] = GetRGBA(y2[1], u[4], v[4]); rgba[10] = GetRGBA(y2[2], u[5], v[5]); rgba[11] = GetRGBA(y2[3], u[5], v[5]); rgba[12] = GetRGBA(y2[4], u[6], v[6]); rgba[13] = GetRGBA(y2[5], u[6], v[6]); rgba[14] = GetRGBA(y2[6], u[7], v[7]); rgba[15] = GetRGBA(y2[7], u[7], v[7]); dram_store_u16(hle, rgba, address, 16); } static void EmitTilesMode0(struct hle_t* hle, const tile_line_emitter_t emit_line, const int16_t *macroblock, uint32_t address) { unsigned int i; unsigned int y_offset = 0; unsigned int u_offset = 2 * SUBBLOCK_SIZE; for (i = 0; i < 8; ++i) { emit_line(hle, ¯oblock[y_offset], ¯oblock[u_offset], address); y_offset += 8; u_offset += 8; address += 32; } } static void EmitTilesMode2(struct hle_t* hle, const tile_line_emitter_t emit_line, const int16_t *macroblock, uint32_t address) { unsigned int i; unsigned int y_offset = 0; unsigned int u_offset = 4 * SUBBLOCK_SIZE; for (i = 0; i < 8; ++i) { emit_line(hle, ¯oblock[y_offset], ¯oblock[u_offset], address); emit_line(hle, ¯oblock[y_offset + 8], ¯oblock[u_offset], address + 32); y_offset += (i == 3) ? SUBBLOCK_SIZE + 16 : 16; u_offset += 8; address += 64; } } static void decode_macroblock_ob(int16_t *macroblock, int32_t *y_dc, int32_t *u_dc, int32_t *v_dc, const int16_t *qtable) { int sb; for (sb = 0; sb < 6; ++sb) { int16_t tmp_sb[SUBBLOCK_SIZE]; /* update DC */ int32_t dc = (int32_t)macroblock[0]; switch (sb) { case 0: case 1: case 2: case 3: *y_dc += dc; macroblock[0] = *y_dc & 0xffff; break; case 4: *u_dc += dc; macroblock[0] = *u_dc & 0xffff; break; case 5: *v_dc += dc; macroblock[0] = *v_dc & 0xffff; break; } ZigZagSubBlock(tmp_sb, macroblock); if (qtable != NULL) MultSubBlocks(tmp_sb, tmp_sb, qtable, 0); TransposeSubBlock(macroblock, tmp_sb); InverseDCTSubBlock(macroblock, macroblock); macroblock += SUBBLOCK_SIZE; } } static void decode_macroblock_std(const subblock_transform_t transform_luma, const subblock_transform_t transform_chroma, int16_t *macroblock, unsigned int subblock_count, const int16_t qtables[3][SUBBLOCK_SIZE]) { unsigned int sb; unsigned int q = 0; for (sb = 0; sb < subblock_count; ++sb) { int16_t tmp_sb[SUBBLOCK_SIZE]; const int isChromaSubBlock = (subblock_count - sb <= 2); if (isChromaSubBlock) ++q; MultSubBlocks(macroblock, macroblock, qtables[q], 4); ZigZagSubBlock(tmp_sb, macroblock); InverseDCTSubBlock(macroblock, tmp_sb); if (isChromaSubBlock) { if (transform_chroma != NULL) transform_chroma(macroblock, macroblock); } else { if (transform_luma != NULL) transform_luma(macroblock, macroblock); } macroblock += SUBBLOCK_SIZE; } } static void TransposeSubBlock(int16_t *dst, const int16_t *src) { ReorderSubBlock(dst, src, TRANSPOSE_TABLE); } static void ZigZagSubBlock(int16_t *dst, const int16_t *src) { ReorderSubBlock(dst, src, ZIGZAG_TABLE); } static void ReorderSubBlock(int16_t *dst, const int16_t *src, const unsigned int *table) { unsigned int i; /* source and destination sublocks cannot overlap */ assert(labs(dst - src) > SUBBLOCK_SIZE); for (i = 0; i < SUBBLOCK_SIZE; ++i) dst[i] = src[table[i]]; } static void MultSubBlocks(int16_t *dst, const int16_t *src1, const int16_t *src2, unsigned int shift) { unsigned int i; for (i = 0; i < SUBBLOCK_SIZE; ++i) { int32_t v = src1[i] * src2[i]; dst[i] = clamp_s16(v) << shift; } } static void ScaleSubBlock(int16_t *dst, const int16_t *src, int16_t scale) { unsigned int i; for (i = 0; i < SUBBLOCK_SIZE; ++i) { int32_t v = src[i] * scale; dst[i] = clamp_s16(v); } } static void RShiftSubBlock(int16_t *dst, const int16_t *src, unsigned int shift) { unsigned int i; for (i = 0; i < SUBBLOCK_SIZE; ++i) dst[i] = src[i] >> shift; } /*************************************************************************** * Fast 2D IDCT using separable formulation and normalization * Computations use single precision floats * Implementation based on Wikipedia : * http://fr.wikipedia.org/wiki/Transform%C3%A9e_en_cosinus_discr%C3%A8te **************************************************************************/ static void InverseDCT1D(const float *const x, float *dst, unsigned int stride) { float e[4]; float f[4]; float x26, x1357, x15, x37, x17, x35; x15 = IDCT_K[2] * (x[1] + x[5]); x37 = IDCT_K[3] * (x[3] + x[7]); x17 = IDCT_K[8] * (x[1] + x[7]); x35 = IDCT_K[9] * (x[3] + x[5]); x1357 = IDCT_C3 * (x[1] + x[3] + x[5] + x[7]); x26 = IDCT_C6 * (x[2] + x[6]); f[0] = x[0] + x[4]; f[1] = x[0] - x[4]; f[2] = x26 + IDCT_K[0] * x[2]; f[3] = x26 + IDCT_K[1] * x[6]; e[0] = x1357 + x15 + IDCT_K[4] * x[1] + x17; e[1] = x1357 + x37 + IDCT_K[6] * x[3] + x35; e[2] = x1357 + x15 + IDCT_K[5] * x[5] + x35; e[3] = x1357 + x37 + IDCT_K[7] * x[7] + x17; *dst = f[0] + f[2] + e[0]; dst += stride; *dst = f[1] + f[3] + e[1]; dst += stride; *dst = f[1] - f[3] + e[2]; dst += stride; *dst = f[0] - f[2] + e[3]; dst += stride; *dst = f[0] - f[2] - e[3]; dst += stride; *dst = f[1] - f[3] - e[2]; dst += stride; *dst = f[1] + f[3] - e[1]; dst += stride; *dst = f[0] + f[2] - e[0]; } static void InverseDCTSubBlock(int16_t *dst, const int16_t *src) { float x[8]; float block[SUBBLOCK_SIZE]; unsigned int i, j; /* idct 1d on rows (+transposition) */ for (i = 0; i < 8; ++i) { for (j = 0; j < 8; ++j) x[j] = (float)src[i * 8 + j]; InverseDCT1D(x, &block[i], 8); } /* idct 1d on columns (thanks to previous transposition) */ for (i = 0; i < 8; ++i) { InverseDCT1D(&block[i * 8], x, 1); /* C4 = 1 normalization implies a division by 8 */ for (j = 0; j < 8; ++j) dst[i + j * 8] = (int16_t)x[j] >> 3; } } static void RescaleYSubBlock(int16_t *dst, const int16_t *src) { unsigned int i; for (i = 0; i < SUBBLOCK_SIZE; ++i) dst[i] = (((uint32_t)(clamp_s12(src[i]) + 0x800) * 0xdb0) >> 16) + 0x10; } static void RescaleUVSubBlock(int16_t *dst, const int16_t *src) { unsigned int i; for (i = 0; i < SUBBLOCK_SIZE; ++i) dst[i] = (((int)clamp_s12(src[i]) * 0xe00) >> 16) + 0x80; } mupen64plus-rsp-hle-src-2.6.0/src/memory.c000066400000000000000000000054101464506742100203320ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - memory.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include "memory.h" /* Global functions */ void load_u8(uint8_t* dst, const unsigned char* buffer, unsigned address, size_t count) { while (count != 0) { *(dst++) = *u8(buffer, address); address += 1; --count; } } void load_u16(uint16_t* dst, const unsigned char* buffer, unsigned address, size_t count) { while (count != 0) { *(dst++) = *u16(buffer, address); address += 2; --count; } } void load_u32(uint32_t* dst, const unsigned char* buffer, unsigned address, size_t count) { /* Optimization for uint32_t */ memcpy(dst, u32(buffer, address), count * sizeof(uint32_t)); } void store_u8(unsigned char* buffer, unsigned address, const uint8_t* src, size_t count) { while (count != 0) { *u8(buffer, address) = *(src++); address += 1; --count; } } void store_u16(unsigned char* buffer, unsigned address, const uint16_t* src, size_t count) { while (count != 0) { *u16(buffer, address) = *(src++); address += 2; --count; } } void store_u32(unsigned char* buffer, unsigned address, const uint32_t* src, size_t count) { /* Optimization for uint32_t */ memcpy(u32(buffer, address), src, count * sizeof(uint32_t)); } mupen64plus-rsp-hle-src-2.6.0/src/memory.h000066400000000000000000000142001464506742100203340ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - memory.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef MEMORY_H #define MEMORY_H #include #include #include #include "common.h" #include "hle_internal.h" #ifdef M64P_BIG_ENDIAN #define S 0 #define S16 0 #define S8 0 #else #define S 1 #define S16 2 #define S8 3 #endif enum { TASK_TYPE = 0xfc0, TASK_FLAGS = 0xfc4, TASK_UCODE_BOOT = 0xfc8, TASK_UCODE_BOOT_SIZE = 0xfcc, TASK_UCODE = 0xfd0, TASK_UCODE_SIZE = 0xfd4, TASK_UCODE_DATA = 0xfd8, TASK_UCODE_DATA_SIZE = 0xfdc, TASK_DRAM_STACK = 0xfe0, TASK_DRAM_STACK_SIZE = 0xfe4, TASK_OUTPUT_BUFF = 0xfe8, TASK_OUTPUT_BUFF_SIZE = 0xfec, TASK_DATA_PTR = 0xff0, TASK_DATA_SIZE = 0xff4, TASK_YIELD_DATA_PTR = 0xff8, TASK_YIELD_DATA_SIZE = 0xffc }; static inline unsigned int align(unsigned int x, unsigned amount) { --amount; return (x + amount) & ~amount; } static inline uint8_t* u8(const unsigned char* buffer, unsigned address) { return (uint8_t*)(buffer + (address ^ S8)); } static inline uint16_t* u16(const unsigned char* buffer, unsigned address) { assert((address & 1) == 0); return (uint16_t*)(buffer + (address ^ S16)); } static inline uint32_t* u32(const unsigned char* buffer, unsigned address) { assert((address & 3) == 0); return (uint32_t*)(buffer + address); } void load_u8 (uint8_t* dst, const unsigned char* buffer, unsigned address, size_t count); void load_u16(uint16_t* dst, const unsigned char* buffer, unsigned address, size_t count); void load_u32(uint32_t* dst, const unsigned char* buffer, unsigned address, size_t count); void store_u8 (unsigned char* buffer, unsigned address, const uint8_t* src, size_t count); void store_u16(unsigned char* buffer, unsigned address, const uint16_t* src, size_t count); void store_u32(unsigned char* buffer, unsigned address, const uint32_t* src, size_t count); /* convenient functions for DMEM access */ static inline uint8_t* dmem_u8(struct hle_t* hle, uint16_t address) { return u8(hle->dmem, address & 0xfff); } static inline uint16_t* dmem_u16(struct hle_t* hle, uint16_t address) { return u16(hle->dmem, address & 0xfff); } static inline uint32_t* dmem_u32(struct hle_t* hle, uint16_t address) { return u32(hle->dmem, address & 0xfff); } static inline void dmem_load_u8(struct hle_t* hle, uint8_t* dst, uint16_t address, size_t count) { load_u8(dst, hle->dmem, address & 0xfff, count); } static inline void dmem_load_u16(struct hle_t* hle, uint16_t* dst, uint16_t address, size_t count) { load_u16(dst, hle->dmem, address & 0xfff, count); } static inline void dmem_load_u32(struct hle_t* hle, uint32_t* dst, uint16_t address, size_t count) { load_u32(dst, hle->dmem, address & 0xfff, count); } static inline void dmem_store_u8(struct hle_t* hle, const uint8_t* src, uint16_t address, size_t count) { store_u8(hle->dmem, address & 0xfff, src, count); } static inline void dmem_store_u16(struct hle_t* hle, const uint16_t* src, uint16_t address, size_t count) { store_u16(hle->dmem, address & 0xfff, src, count); } static inline void dmem_store_u32(struct hle_t* hle, const uint32_t* src, uint16_t address, size_t count) { store_u32(hle->dmem, address & 0xfff, src, count); } /* convenient functions DRAM access */ static inline uint8_t* dram_u8(struct hle_t* hle, uint32_t address) { return u8(hle->dram, address & 0xffffff); } static inline uint16_t* dram_u16(struct hle_t* hle, uint32_t address) { return u16(hle->dram, address & 0xffffff); } static inline uint32_t* dram_u32(struct hle_t* hle, uint32_t address) { return u32(hle->dram, address & 0xffffff); } static inline void dram_load_u8(struct hle_t* hle, uint8_t* dst, uint32_t address, size_t count) { load_u8(dst, hle->dram, address & 0xffffff, count); } static inline void dram_load_u16(struct hle_t* hle, uint16_t* dst, uint32_t address, size_t count) { load_u16(dst, hle->dram, address & 0xffffff, count); } static inline void dram_load_u32(struct hle_t* hle, uint32_t* dst, uint32_t address, size_t count) { load_u32(dst, hle->dram, address & 0xffffff, count); } static inline void dram_store_u8(struct hle_t* hle, const uint8_t* src, uint32_t address, size_t count) { store_u8(hle->dram, address & 0xffffff, src, count); } static inline void dram_store_u16(struct hle_t* hle, const uint16_t* src, uint32_t address, size_t count) { store_u16(hle->dram, address & 0xffffff, src, count); } static inline void dram_store_u32(struct hle_t* hle, const uint32_t* src, uint32_t address, size_t count) { store_u32(hle->dram, address & 0xffffff, src, count); } #endif mupen64plus-rsp-hle-src-2.6.0/src/mp3.c000066400000000000000000000742211464506742100175270ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - mp3.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include "arithmetics.h" #include "hle_internal.h" #include "memory.h" static void InnerLoop(struct hle_t* hle, uint32_t outPtr, uint32_t inPtr, uint32_t t6, uint32_t t5, uint32_t t4); static const uint16_t DeWindowLUT [0x420] = { 0x0000, 0xFFF3, 0x005D, 0xFF38, 0x037A, 0xF736, 0x0B37, 0xC00E, 0x7FFF, 0x3FF2, 0x0B37, 0x08CA, 0x037A, 0x00C8, 0x005D, 0x000D, 0x0000, 0xFFF3, 0x005D, 0xFF38, 0x037A, 0xF736, 0x0B37, 0xC00E, 0x7FFF, 0x3FF2, 0x0B37, 0x08CA, 0x037A, 0x00C8, 0x005D, 0x000D, 0x0000, 0xFFF2, 0x005F, 0xFF1D, 0x0369, 0xF697, 0x0A2A, 0xBCE7, 0x7FEB, 0x3CCB, 0x0C2B, 0x082B, 0x0385, 0x00AF, 0x005B, 0x000B, 0x0000, 0xFFF2, 0x005F, 0xFF1D, 0x0369, 0xF697, 0x0A2A, 0xBCE7, 0x7FEB, 0x3CCB, 0x0C2B, 0x082B, 0x0385, 0x00AF, 0x005B, 0x000B, 0x0000, 0xFFF1, 0x0061, 0xFF02, 0x0354, 0xF5F9, 0x0905, 0xB9C4, 0x7FB0, 0x39A4, 0x0D08, 0x078C, 0x038C, 0x0098, 0x0058, 0x000A, 0x0000, 0xFFF1, 0x0061, 0xFF02, 0x0354, 0xF5F9, 0x0905, 0xB9C4, 0x7FB0, 0x39A4, 0x0D08, 0x078C, 0x038C, 0x0098, 0x0058, 0x000A, 0x0000, 0xFFEF, 0x0062, 0xFEE6, 0x033B, 0xF55C, 0x07C8, 0xB6A4, 0x7F4D, 0x367E, 0x0DCE, 0x06EE, 0x038F, 0x0080, 0x0056, 0x0009, 0x0000, 0xFFEF, 0x0062, 0xFEE6, 0x033B, 0xF55C, 0x07C8, 0xB6A4, 0x7F4D, 0x367E, 0x0DCE, 0x06EE, 0x038F, 0x0080, 0x0056, 0x0009, 0x0000, 0xFFEE, 0x0063, 0xFECA, 0x031C, 0xF4C3, 0x0671, 0xB38C, 0x7EC2, 0x335D, 0x0E7C, 0x0652, 0x038E, 0x006B, 0x0053, 0x0008, 0x0000, 0xFFEE, 0x0063, 0xFECA, 0x031C, 0xF4C3, 0x0671, 0xB38C, 0x7EC2, 0x335D, 0x0E7C, 0x0652, 0x038E, 0x006B, 0x0053, 0x0008, 0x0000, 0xFFEC, 0x0064, 0xFEAC, 0x02F7, 0xF42C, 0x0502, 0xB07C, 0x7E12, 0x3041, 0x0F14, 0x05B7, 0x038A, 0x0056, 0x0050, 0x0007, 0x0000, 0xFFEC, 0x0064, 0xFEAC, 0x02F7, 0xF42C, 0x0502, 0xB07C, 0x7E12, 0x3041, 0x0F14, 0x05B7, 0x038A, 0x0056, 0x0050, 0x0007, 0x0000, 0xFFEB, 0x0064, 0xFE8E, 0x02CE, 0xF399, 0x037A, 0xAD75, 0x7D3A, 0x2D2C, 0x0F97, 0x0520, 0x0382, 0x0043, 0x004D, 0x0007, 0x0000, 0xFFEB, 0x0064, 0xFE8E, 0x02CE, 0xF399, 0x037A, 0xAD75, 0x7D3A, 0x2D2C, 0x0F97, 0x0520, 0x0382, 0x0043, 0x004D, 0x0007, 0xFFFF, 0xFFE9, 0x0063, 0xFE6F, 0x029E, 0xF30B, 0x01D8, 0xAA7B, 0x7C3D, 0x2A1F, 0x1004, 0x048B, 0x0377, 0x0030, 0x004A, 0x0006, 0xFFFF, 0xFFE9, 0x0063, 0xFE6F, 0x029E, 0xF30B, 0x01D8, 0xAA7B, 0x7C3D, 0x2A1F, 0x1004, 0x048B, 0x0377, 0x0030, 0x004A, 0x0006, 0xFFFF, 0xFFE7, 0x0062, 0xFE4F, 0x0269, 0xF282, 0x001F, 0xA78D, 0x7B1A, 0x271C, 0x105D, 0x03F9, 0x036A, 0x001F, 0x0046, 0x0006, 0xFFFF, 0xFFE7, 0x0062, 0xFE4F, 0x0269, 0xF282, 0x001F, 0xA78D, 0x7B1A, 0x271C, 0x105D, 0x03F9, 0x036A, 0x001F, 0x0046, 0x0006, 0xFFFF, 0xFFE4, 0x0061, 0xFE2F, 0x022F, 0xF1FF, 0xFE4C, 0xA4AF, 0x79D3, 0x2425, 0x10A2, 0x036C, 0x0359, 0x0010, 0x0043, 0x0005, 0xFFFF, 0xFFE4, 0x0061, 0xFE2F, 0x022F, 0xF1FF, 0xFE4C, 0xA4AF, 0x79D3, 0x2425, 0x10A2, 0x036C, 0x0359, 0x0010, 0x0043, 0x0005, 0xFFFF, 0xFFE2, 0x005E, 0xFE10, 0x01EE, 0xF184, 0xFC61, 0xA1E1, 0x7869, 0x2139, 0x10D3, 0x02E3, 0x0346, 0x0001, 0x0040, 0x0004, 0xFFFF, 0xFFE2, 0x005E, 0xFE10, 0x01EE, 0xF184, 0xFC61, 0xA1E1, 0x7869, 0x2139, 0x10D3, 0x02E3, 0x0346, 0x0001, 0x0040, 0x0004, 0xFFFF, 0xFFE0, 0x005B, 0xFDF0, 0x01A8, 0xF111, 0xFA5F, 0x9F27, 0x76DB, 0x1E5C, 0x10F2, 0x025E, 0x0331, 0xFFF3, 0x003D, 0x0004, 0xFFFF, 0xFFE0, 0x005B, 0xFDF0, 0x01A8, 0xF111, 0xFA5F, 0x9F27, 0x76DB, 0x1E5C, 0x10F2, 0x025E, 0x0331, 0xFFF3, 0x003D, 0x0004, 0xFFFF, 0xFFDE, 0x0057, 0xFDD0, 0x015B, 0xF0A7, 0xF845, 0x9C80, 0x752C, 0x1B8E, 0x1100, 0x01DE, 0x0319, 0xFFE7, 0x003A, 0x0003, 0xFFFF, 0xFFDE, 0x0057, 0xFDD0, 0x015B, 0xF0A7, 0xF845, 0x9C80, 0x752C, 0x1B8E, 0x1100, 0x01DE, 0x0319, 0xFFE7, 0x003A, 0x0003, 0xFFFE, 0xFFDB, 0x0053, 0xFDB0, 0x0108, 0xF046, 0xF613, 0x99EE, 0x735C, 0x18D1, 0x10FD, 0x0163, 0x0300, 0xFFDC, 0x0037, 0x0003, 0xFFFE, 0xFFDB, 0x0053, 0xFDB0, 0x0108, 0xF046, 0xF613, 0x99EE, 0x735C, 0x18D1, 0x10FD, 0x0163, 0x0300, 0xFFDC, 0x0037, 0x0003, 0xFFFE, 0xFFD8, 0x004D, 0xFD90, 0x00B0, 0xEFF0, 0xF3CC, 0x9775, 0x716C, 0x1624, 0x10EA, 0x00EE, 0x02E5, 0xFFD2, 0x0033, 0x0003, 0xFFFE, 0xFFD8, 0x004D, 0xFD90, 0x00B0, 0xEFF0, 0xF3CC, 0x9775, 0x716C, 0x1624, 0x10EA, 0x00EE, 0x02E5, 0xFFD2, 0x0033, 0x0003, 0xFFFE, 0xFFD6, 0x0047, 0xFD72, 0x0051, 0xEFA6, 0xF16F, 0x9514, 0x6F5E, 0x138A, 0x10C8, 0x007E, 0x02CA, 0xFFC9, 0x0030, 0x0003, 0xFFFE, 0xFFD6, 0x0047, 0xFD72, 0x0051, 0xEFA6, 0xF16F, 0x9514, 0x6F5E, 0x138A, 0x10C8, 0x007E, 0x02CA, 0xFFC9, 0x0030, 0x0003, 0xFFFE, 0xFFD3, 0x0040, 0xFD54, 0xFFEC, 0xEF68, 0xEEFC, 0x92CD, 0x6D33, 0x1104, 0x1098, 0x0014, 0x02AC, 0xFFC0, 0x002D, 0x0002, 0xFFFE, 0xFFD3, 0x0040, 0xFD54, 0xFFEC, 0xEF68, 0xEEFC, 0x92CD, 0x6D33, 0x1104, 0x1098, 0x0014, 0x02AC, 0xFFC0, 0x002D, 0x0002, 0x0030, 0xFFC9, 0x02CA, 0x007E, 0x10C8, 0x138A, 0x6F5E, 0x9514, 0xF16F, 0xEFA6, 0x0051, 0xFD72, 0x0047, 0xFFD6, 0xFFFE, 0x0003, 0x0030, 0xFFC9, 0x02CA, 0x007E, 0x10C8, 0x138A, 0x6F5E, 0x9514, 0xF16F, 0xEFA6, 0x0051, 0xFD72, 0x0047, 0xFFD6, 0xFFFE, 0x0003, 0x0033, 0xFFD2, 0x02E5, 0x00EE, 0x10EA, 0x1624, 0x716C, 0x9775, 0xF3CC, 0xEFF0, 0x00B0, 0xFD90, 0x004D, 0xFFD8, 0xFFFE, 0x0003, 0x0033, 0xFFD2, 0x02E5, 0x00EE, 0x10EA, 0x1624, 0x716C, 0x9775, 0xF3CC, 0xEFF0, 0x00B0, 0xFD90, 0x004D, 0xFFD8, 0xFFFE, 0x0003, 0x0037, 0xFFDC, 0x0300, 0x0163, 0x10FD, 0x18D1, 0x735C, 0x99EE, 0xF613, 0xF046, 0x0108, 0xFDB0, 0x0053, 0xFFDB, 0xFFFE, 0x0003, 0x0037, 0xFFDC, 0x0300, 0x0163, 0x10FD, 0x18D1, 0x735C, 0x99EE, 0xF613, 0xF046, 0x0108, 0xFDB0, 0x0053, 0xFFDB, 0xFFFE, 0x0003, 0x003A, 0xFFE7, 0x0319, 0x01DE, 0x1100, 0x1B8E, 0x752C, 0x9C80, 0xF845, 0xF0A7, 0x015B, 0xFDD0, 0x0057, 0xFFDE, 0xFFFF, 0x0003, 0x003A, 0xFFE7, 0x0319, 0x01DE, 0x1100, 0x1B8E, 0x752C, 0x9C80, 0xF845, 0xF0A7, 0x015B, 0xFDD0, 0x0057, 0xFFDE, 0xFFFF, 0x0004, 0x003D, 0xFFF3, 0x0331, 0x025E, 0x10F2, 0x1E5C, 0x76DB, 0x9F27, 0xFA5F, 0xF111, 0x01A8, 0xFDF0, 0x005B, 0xFFE0, 0xFFFF, 0x0004, 0x003D, 0xFFF3, 0x0331, 0x025E, 0x10F2, 0x1E5C, 0x76DB, 0x9F27, 0xFA5F, 0xF111, 0x01A8, 0xFDF0, 0x005B, 0xFFE0, 0xFFFF, 0x0004, 0x0040, 0x0001, 0x0346, 0x02E3, 0x10D3, 0x2139, 0x7869, 0xA1E1, 0xFC61, 0xF184, 0x01EE, 0xFE10, 0x005E, 0xFFE2, 0xFFFF, 0x0004, 0x0040, 0x0001, 0x0346, 0x02E3, 0x10D3, 0x2139, 0x7869, 0xA1E1, 0xFC61, 0xF184, 0x01EE, 0xFE10, 0x005E, 0xFFE2, 0xFFFF, 0x0005, 0x0043, 0x0010, 0x0359, 0x036C, 0x10A2, 0x2425, 0x79D3, 0xA4AF, 0xFE4C, 0xF1FF, 0x022F, 0xFE2F, 0x0061, 0xFFE4, 0xFFFF, 0x0005, 0x0043, 0x0010, 0x0359, 0x036C, 0x10A2, 0x2425, 0x79D3, 0xA4AF, 0xFE4C, 0xF1FF, 0x022F, 0xFE2F, 0x0061, 0xFFE4, 0xFFFF, 0x0006, 0x0046, 0x001F, 0x036A, 0x03F9, 0x105D, 0x271C, 0x7B1A, 0xA78D, 0x001F, 0xF282, 0x0269, 0xFE4F, 0x0062, 0xFFE7, 0xFFFF, 0x0006, 0x0046, 0x001F, 0x036A, 0x03F9, 0x105D, 0x271C, 0x7B1A, 0xA78D, 0x001F, 0xF282, 0x0269, 0xFE4F, 0x0062, 0xFFE7, 0xFFFF, 0x0006, 0x004A, 0x0030, 0x0377, 0x048B, 0x1004, 0x2A1F, 0x7C3D, 0xAA7B, 0x01D8, 0xF30B, 0x029E, 0xFE6F, 0x0063, 0xFFE9, 0xFFFF, 0x0006, 0x004A, 0x0030, 0x0377, 0x048B, 0x1004, 0x2A1F, 0x7C3D, 0xAA7B, 0x01D8, 0xF30B, 0x029E, 0xFE6F, 0x0063, 0xFFE9, 0xFFFF, 0x0007, 0x004D, 0x0043, 0x0382, 0x0520, 0x0F97, 0x2D2C, 0x7D3A, 0xAD75, 0x037A, 0xF399, 0x02CE, 0xFE8E, 0x0064, 0xFFEB, 0x0000, 0x0007, 0x004D, 0x0043, 0x0382, 0x0520, 0x0F97, 0x2D2C, 0x7D3A, 0xAD75, 0x037A, 0xF399, 0x02CE, 0xFE8E, 0x0064, 0xFFEB, 0x0000, 0x0007, 0x0050, 0x0056, 0x038A, 0x05B7, 0x0F14, 0x3041, 0x7E12, 0xB07C, 0x0502, 0xF42C, 0x02F7, 0xFEAC, 0x0064, 0xFFEC, 0x0000, 0x0007, 0x0050, 0x0056, 0x038A, 0x05B7, 0x0F14, 0x3041, 0x7E12, 0xB07C, 0x0502, 0xF42C, 0x02F7, 0xFEAC, 0x0064, 0xFFEC, 0x0000, 0x0008, 0x0053, 0x006B, 0x038E, 0x0652, 0x0E7C, 0x335D, 0x7EC2, 0xB38C, 0x0671, 0xF4C3, 0x031C, 0xFECA, 0x0063, 0xFFEE, 0x0000, 0x0008, 0x0053, 0x006B, 0x038E, 0x0652, 0x0E7C, 0x335D, 0x7EC2, 0xB38C, 0x0671, 0xF4C3, 0x031C, 0xFECA, 0x0063, 0xFFEE, 0x0000, 0x0009, 0x0056, 0x0080, 0x038F, 0x06EE, 0x0DCE, 0x367E, 0x7F4D, 0xB6A4, 0x07C8, 0xF55C, 0x033B, 0xFEE6, 0x0062, 0xFFEF, 0x0000, 0x0009, 0x0056, 0x0080, 0x038F, 0x06EE, 0x0DCE, 0x367E, 0x7F4D, 0xB6A4, 0x07C8, 0xF55C, 0x033B, 0xFEE6, 0x0062, 0xFFEF, 0x0000, 0x000A, 0x0058, 0x0098, 0x038C, 0x078C, 0x0D08, 0x39A4, 0x7FB0, 0xB9C4, 0x0905, 0xF5F9, 0x0354, 0xFF02, 0x0061, 0xFFF1, 0x0000, 0x000A, 0x0058, 0x0098, 0x038C, 0x078C, 0x0D08, 0x39A4, 0x7FB0, 0xB9C4, 0x0905, 0xF5F9, 0x0354, 0xFF02, 0x0061, 0xFFF1, 0x0000, 0x000B, 0x005B, 0x00AF, 0x0385, 0x082B, 0x0C2B, 0x3CCB, 0x7FEB, 0xBCE7, 0x0A2A, 0xF697, 0x0369, 0xFF1D, 0x005F, 0xFFF2, 0x0000, 0x000B, 0x005B, 0x00AF, 0x0385, 0x082B, 0x0C2B, 0x3CCB, 0x7FEB, 0xBCE7, 0x0A2A, 0xF697, 0x0369, 0xFF1D, 0x005F, 0xFFF2, 0x0000, 0x000D, 0x005D, 0x00C8, 0x037A, 0x08CA, 0x0B37, 0x3FF2, 0x7FFF, 0xC00E, 0x0B37, 0xF736, 0x037A, 0xFF38, 0x005D, 0xFFF3, 0x0000, 0x000D, 0x005D, 0x00C8, 0x037A, 0x08CA, 0x0B37, 0x3FF2, 0x7FFF, 0xC00E, 0x0B37, 0xF736, 0x037A, 0xFF38, 0x005D, 0xFFF3, 0x0000, 0x0000 }; static void MP3AB0(int32_t* v) { /* Part 2 - 100% Accurate */ static const uint16_t LUT2[8] = { 0xFEC4, 0xF4FA, 0xC5E4, 0xE1C4, 0x1916, 0x4A50, 0xA268, 0x78AE }; static const uint16_t LUT3[4] = { 0xFB14, 0xD4DC, 0x31F2, 0x8E3A }; int i; for (i = 0; i < 8; i++) { v[16 + i] = v[0 + i] + v[8 + i]; v[24 + i] = ((v[0 + i] - v[8 + i]) * LUT2[i]) >> 0x10; } /* Part 3: 4-wide butterflies */ for (i = 0; i < 4; i++) { v[0 + i] = v[16 + i] + v[20 + i]; v[4 + i] = ((v[16 + i] - v[20 + i]) * LUT3[i]) >> 0x10; v[8 + i] = v[24 + i] + v[28 + i]; v[12 + i] = ((v[24 + i] - v[28 + i]) * LUT3[i]) >> 0x10; } /* Part 4: 2-wide butterflies - 100% Accurate */ for (i = 0; i < 16; i += 4) { v[16 + i] = v[0 + i] + v[2 + i]; v[18 + i] = ((v[0 + i] - v[2 + i]) * 0xEC84) >> 0x10; v[17 + i] = v[1 + i] + v[3 + i]; v[19 + i] = ((v[1 + i] - v[3 + i]) * 0x61F8) >> 0x10; } } void mp3_task(struct hle_t* hle, unsigned int index, uint32_t address) { uint32_t inPtr, outPtr; uint32_t t6;/* = 0x08A0; - I think these are temporary storage buffers */ uint32_t t5;/* = 0x0AC0; */ uint32_t t4;/* = (w1 & 0x1E); */ /* Initialization Code */ uint32_t readPtr; /* s5 */ uint32_t writePtr; /* s6 */ uint32_t tmp; int cnt, cnt2; /* I think these are temporary storage buffers */ t6 = 0x08A0; t5 = 0x0AC0; t4 = index; writePtr = readPtr = address; /* Just do that for efficiency... may remove and use directly later anyway */ memcpy(hle->mp3_buffer + 0xCE8, hle->dram + readPtr, 8); /* This must be a header byte or whatnot */ readPtr += 8; for (cnt = 0; cnt < 0x480; cnt += 0x180) { /* DMA: 0xCF0 <- RDRAM[s5] : 0x180 */ memcpy(hle->mp3_buffer + 0xCF0, hle->dram + readPtr, 0x180); inPtr = 0xCF0; /* s7 */ outPtr = 0xE70; /* s3 */ /* --------------- Inner Loop Start -------------------- */ for (cnt2 = 0; cnt2 < 0x180; cnt2 += 0x40) { t6 &= 0xFFE0; t5 &= 0xFFE0; t6 |= t4; t5 |= t4; InnerLoop(hle, outPtr, inPtr, t6, t5, t4); t4 = (t4 - 2) & 0x1E; tmp = t6; t6 = t5; t5 = tmp; inPtr += 0x40; outPtr += 0x40; } /* --------------- Inner Loop End -------------------- */ memcpy(hle->dram + writePtr, hle->mp3_buffer + 0xe70, 0x180); writePtr += 0x180; readPtr += 0x180; } } static void InnerLoop(struct hle_t* hle, uint32_t outPtr, uint32_t inPtr, uint32_t t6, uint32_t t5, uint32_t t4) { /* Part 1: 100% Accurate */ /* 0, 1, 3, 2, 7, 6, 4, 5, 7, 6, 4, 5, 0, 1, 3, 2 */ static const uint16_t LUT6[16] = { 0xFFB2, 0xFD3A, 0xF10A, 0xF854, 0xBDAE, 0xCDA0, 0xE76C, 0xDB94, 0x1920, 0x4B20, 0xAC7C, 0x7C68, 0xABEC, 0x9880, 0xDAE8, 0x839C }; int i; uint32_t t0; uint32_t t1; uint32_t t2; uint32_t t3; int32_t v2 = 0, v4 = 0, v6 = 0, v8 = 0; uint32_t offset; uint32_t addptr; int x; int32_t mult6; int32_t mult4; int tmp; int32_t hi0; int32_t hi1; int32_t vt; int32_t v[32]; v[0] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x00 ^ S16)); v[31] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3E ^ S16)); v[0] += v[31]; v[1] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x02 ^ S16)); v[30] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3C ^ S16)); v[1] += v[30]; v[2] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x06 ^ S16)); v[28] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x38 ^ S16)); v[2] += v[28]; v[3] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x04 ^ S16)); v[29] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3A ^ S16)); v[3] += v[29]; v[4] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0E ^ S16)); v[24] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x30 ^ S16)); v[4] += v[24]; v[5] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0C ^ S16)); v[25] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x32 ^ S16)); v[5] += v[25]; v[6] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x08 ^ S16)); v[27] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x36 ^ S16)); v[6] += v[27]; v[7] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0A ^ S16)); v[26] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x34 ^ S16)); v[7] += v[26]; v[8] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1E ^ S16)); v[16] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x20 ^ S16)); v[8] += v[16]; v[9] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1C ^ S16)); v[17] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x22 ^ S16)); v[9] += v[17]; v[10] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x18 ^ S16)); v[19] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x26 ^ S16)); v[10] += v[19]; v[11] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1A ^ S16)); v[18] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x24 ^ S16)); v[11] += v[18]; v[12] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x10 ^ S16)); v[23] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2E ^ S16)); v[12] += v[23]; v[13] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x12 ^ S16)); v[22] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2C ^ S16)); v[13] += v[22]; v[14] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x16 ^ S16)); v[20] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x28 ^ S16)); v[14] += v[20]; v[15] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x14 ^ S16)); v[21] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2A ^ S16)); v[15] += v[21]; /* Part 2-4 */ MP3AB0(v); /* Part 5 - 1-Wide Butterflies - 100% Accurate but need SSVs!!! */ t0 = t6 + 0x100; t1 = t6 + 0x200; t2 = t5 + 0x100; t3 = t5 + 0x200; /* 0x13A8 */ v[1] = 0; v[11] = ((v[16] - v[17]) * 0xB504) >> 0x10; v[16] = -v[16] - v[17]; v[2] = v[18] + v[19]; /* ** Store v[11] -> (T6 + 0)** */ *(int16_t *)(hle->mp3_buffer + ((t6 + (short)0x0))) = (short)v[11]; v[11] = -v[11]; /* ** Store v[16] -> (T3 + 0)** */ *(int16_t *)(hle->mp3_buffer + ((t3 + (short)0x0))) = (short)v[16]; /* ** Store v[11] -> (T5 + 0)** */ *(int16_t *)(hle->mp3_buffer + ((t5 + (short)0x0))) = (short)v[11]; /* 0x13E8 - Verified.... */ v[2] = -v[2]; /* ** Store v[2] -> (T2 + 0)** */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0x0))) = (short)v[2]; v[3] = (((v[18] - v[19]) * 0x16A09) >> 0x10) + v[2]; /* ** Store v[3] -> (T0 + 0)** */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0x0))) = (short)v[3]; /* 0x1400 - Verified */ v[4] = -v[20] - v[21]; v[6] = v[22] + v[23]; v[5] = ((v[20] - v[21]) * 0x16A09) >> 0x10; /* ** Store v[4] -> (T3 + 0xFF80) */ *(int16_t *)(hle->mp3_buffer + ((t3 + (short)0xFF80))) = (short)v[4]; v[7] = ((v[22] - v[23]) * 0x2D413) >> 0x10; v[5] = v[5] - v[4]; v[7] = v[7] - v[5]; v[6] = v[6] + v[6]; v[5] = v[5] - v[6]; v[4] = -v[4] - v[6]; /* *** Store v[7] -> (T1 + 0xFF80) */ *(int16_t *)(hle->mp3_buffer + ((t1 + (short)0xFF80))) = (short)v[7]; /* *** Store v[4] -> (T2 + 0xFF80) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0xFF80))) = (short)v[4]; /* *** Store v[5] -> (T0 + 0xFF80) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0xFF80))) = (short)v[5]; v[8] = v[24] + v[25]; v[9] = ((v[24] - v[25]) * 0x16A09) >> 0x10; v[2] = v[8] + v[9]; v[11] = ((v[26] - v[27]) * 0x2D413) >> 0x10; v[13] = ((v[28] - v[29]) * 0x2D413) >> 0x10; v[10] = v[26] + v[27]; v[10] = v[10] + v[10]; v[12] = v[28] + v[29]; v[12] = v[12] + v[12]; v[14] = v[30] + v[31]; v[3] = v[8] + v[10]; v[14] = v[14] + v[14]; v[13] = (v[13] - v[2]) + v[12]; v[15] = (((v[30] - v[31]) * 0x5A827) >> 0x10) - (v[11] + v[2]); v[14] = -(v[14] + v[14]) + v[3]; v[17] = v[13] - v[10]; v[9] = v[9] + v[14]; /* ** Store v[9] -> (T6 + 0x40) */ *(int16_t *)(hle->mp3_buffer + ((t6 + (short)0x40))) = (short)v[9]; v[11] = v[11] - v[13]; /* ** Store v[17] -> (T0 + 0xFFC0) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0xFFC0))) = (short)v[17]; v[12] = v[8] - v[12]; /* ** Store v[11] -> (T0 + 0x40) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0x40))) = (short)v[11]; v[8] = -v[8]; /* ** Store v[15] -> (T1 + 0xFFC0) */ *(int16_t *)(hle->mp3_buffer + ((t1 + (short)0xFFC0))) = (short)v[15]; v[10] = -v[10] - v[12]; /* ** Store v[12] -> (T2 + 0x40) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0x40))) = (short)v[12]; /* ** Store v[8] -> (T3 + 0xFFC0) */ *(int16_t *)(hle->mp3_buffer + ((t3 + (short)0xFFC0))) = (short)v[8]; /* ** Store v[14] -> (T5 + 0x40) */ *(int16_t *)(hle->mp3_buffer + ((t5 + (short)0x40))) = (short)v[14]; /* ** Store v[10] -> (T2 + 0xFFC0) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0xFFC0))) = (short)v[10]; /* 0x14FC - Verified... */ /* Part 6 - 100% Accurate */ v[0] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x00 ^ S16)); v[31] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3E ^ S16)); v[0] -= v[31]; v[1] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x02 ^ S16)); v[30] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3C ^ S16)); v[1] -= v[30]; v[2] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x06 ^ S16)); v[28] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x38 ^ S16)); v[2] -= v[28]; v[3] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x04 ^ S16)); v[29] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x3A ^ S16)); v[3] -= v[29]; v[4] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0E ^ S16)); v[24] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x30 ^ S16)); v[4] -= v[24]; v[5] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0C ^ S16)); v[25] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x32 ^ S16)); v[5] -= v[25]; v[6] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x08 ^ S16)); v[27] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x36 ^ S16)); v[6] -= v[27]; v[7] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x0A ^ S16)); v[26] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x34 ^ S16)); v[7] -= v[26]; v[8] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1E ^ S16)); v[16] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x20 ^ S16)); v[8] -= v[16]; v[9] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1C ^ S16)); v[17] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x22 ^ S16)); v[9] -= v[17]; v[10] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x18 ^ S16)); v[19] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x26 ^ S16)); v[10] -= v[19]; v[11] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x1A ^ S16)); v[18] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x24 ^ S16)); v[11] -= v[18]; v[12] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x10 ^ S16)); v[23] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2E ^ S16)); v[12] -= v[23]; v[13] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x12 ^ S16)); v[22] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2C ^ S16)); v[13] -= v[22]; v[14] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x16 ^ S16)); v[20] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x28 ^ S16)); v[14] -= v[20]; v[15] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x14 ^ S16)); v[21] = *(int16_t *)(hle->mp3_buffer + inPtr + (0x2A ^ S16)); v[15] -= v[21]; for (i = 0; i < 16; i++) v[0 + i] = (v[0 + i] * LUT6[i]) >> 0x10; v[0] = v[0] + v[0]; v[1] = v[1] + v[1]; v[2] = v[2] + v[2]; v[3] = v[3] + v[3]; v[4] = v[4] + v[4]; v[5] = v[5] + v[5]; v[6] = v[6] + v[6]; v[7] = v[7] + v[7]; v[12] = v[12] + v[12]; v[13] = v[13] + v[13]; v[15] = v[15] + v[15]; MP3AB0(v); /* Part 7: - 100% Accurate + SSV - Unoptimized */ v[0] = (v[17] + v[16]) >> 1; v[1] = ((v[17] * (int)((short)0xA57E * 2)) + (v[16] * 0xB504)) >> 0x10; v[2] = -v[18] - v[19]; v[3] = ((v[18] - v[19]) * 0x16A09) >> 0x10; v[4] = v[20] + v[21] + v[0]; v[5] = (((v[20] - v[21]) * 0x16A09) >> 0x10) + v[1]; v[6] = (((v[22] + v[23]) << 1) + v[0]) - v[2]; v[7] = (((v[22] - v[23]) * 0x2D413) >> 0x10) + v[0] + v[1] + v[3]; /* 0x16A8 */ /* Save v[0] -> (T3 + 0xFFE0) */ *(int16_t *)(hle->mp3_buffer + ((t3 + (short)0xFFE0))) = (short) - v[0]; v[8] = v[24] + v[25]; v[9] = ((v[24] - v[25]) * 0x16A09) >> 0x10; v[10] = ((v[26] + v[27]) << 1) + v[8]; v[11] = (((v[26] - v[27]) * 0x2D413) >> 0x10) + v[8] + v[9]; v[12] = v[4] - ((v[28] + v[29]) << 1); /* ** Store v12 -> (T2 + 0x20) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0x20))) = (short)v[12]; v[13] = (((v[28] - v[29]) * 0x2D413) >> 0x10) - v[12] - v[5]; v[14] = v[30] + v[31]; v[14] = v[14] + v[14]; v[14] = v[14] + v[14]; v[14] = v[6] - v[14]; v[15] = (((v[30] - v[31]) * 0x5A827) >> 0x10) - v[7]; /* Store v14 -> (T5 + 0x20) */ *(int16_t *)(hle->mp3_buffer + ((t5 + (short)0x20))) = (short)v[14]; v[14] = v[14] + v[1]; /* Store v[14] -> (T6 + 0x20) */ *(int16_t *)(hle->mp3_buffer + ((t6 + (short)0x20))) = (short)v[14]; /* Store v[15] -> (T1 + 0xFFE0) */ *(int16_t *)(hle->mp3_buffer + ((t1 + (short)0xFFE0))) = (short)v[15]; v[9] = v[9] + v[10]; v[1] = v[1] + v[6]; v[6] = v[10] - v[6]; v[1] = v[9] - v[1]; /* Store v[6] -> (T5 + 0x60) */ *(int16_t *)(hle->mp3_buffer + ((t5 + (short)0x60))) = (short)v[6]; v[10] = v[10] + v[2]; v[10] = v[4] - v[10]; /* Store v[10] -> (T2 + 0xFFA0) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0xFFA0))) = (short)v[10]; v[12] = v[2] - v[12]; /* Store v[12] -> (T2 + 0xFFE0) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0xFFE0))) = (short)v[12]; v[5] = v[4] + v[5]; v[4] = v[8] - v[4]; /* Store v[4] -> (T2 + 0x60) */ *(int16_t *)(hle->mp3_buffer + ((t2 + (short)0x60))) = (short)v[4]; v[0] = v[0] - v[8]; /* Store v[0] -> (T3 + 0xFFA0) */ *(int16_t *)(hle->mp3_buffer + ((t3 + (short)0xFFA0))) = (short)v[0]; v[7] = v[7] - v[11]; /* Store v[7] -> (T1 + 0xFFA0) */ *(int16_t *)(hle->mp3_buffer + ((t1 + (short)0xFFA0))) = (short)v[7]; v[11] = v[11] - v[3]; /* Store v[1] -> (T6 + 0x60) */ *(int16_t *)(hle->mp3_buffer + ((t6 + (short)0x60))) = (short)v[1]; v[11] = v[11] - v[5]; /* Store v[11] -> (T0 + 0x60) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0x60))) = (short)v[11]; v[3] = v[3] - v[13]; /* Store v[3] -> (T0 + 0x20) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0x20))) = (short)v[3]; v[13] = v[13] + v[2]; /* Store v[13] -> (T0 + 0xFFE0) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0xFFE0))) = (short)v[13]; v[2] = (v[5] - v[2]) - v[9]; /* Store v[2] -> (T0 + 0xFFA0) */ *(int16_t *)(hle->mp3_buffer + ((t0 + (short)0xFFA0))) = (short)v[2]; /* 0x7A8 - Verified... */ /* Step 8 - Dewindowing */ addptr = t6 & 0xFFE0; offset = 0x10 - (t4 >> 1); for (x = 0; x < 8; x++) { int32_t v0; int32_t v18; v2 = v4 = v6 = v8 = 0; for (i = 7; i >= 0; i--) { v2 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x00) * (short)DeWindowLUT[offset + 0x00] + 0x4000) >> 0xF; v4 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x10) * (short)DeWindowLUT[offset + 0x08] + 0x4000) >> 0xF; v6 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x20) * (short)DeWindowLUT[offset + 0x20] + 0x4000) >> 0xF; v8 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x30) * (short)DeWindowLUT[offset + 0x28] + 0x4000) >> 0xF; addptr += 2; offset++; } v0 = v2 + v4; v18 = v6 + v8; /* Clamp(v0); */ /* Clamp(v18); */ /* clamp??? */ *(int16_t *)(hle->mp3_buffer + (outPtr ^ S16)) = v0; *(int16_t *)(hle->mp3_buffer + ((outPtr + 2)^S16)) = v18; outPtr += 4; addptr += 0x30; offset += 0x38; } offset = 0x10 - (t4 >> 1) + 8 * 0x40; v2 = v4 = 0; for (i = 0; i < 4; i++) { v2 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x00) * (short)DeWindowLUT[offset + 0x00] + 0x4000) >> 0xF; v2 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x10) * (short)DeWindowLUT[offset + 0x08] + 0x4000) >> 0xF; addptr += 2; offset++; v4 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x00) * (short)DeWindowLUT[offset + 0x00] + 0x4000) >> 0xF; v4 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x10) * (short)DeWindowLUT[offset + 0x08] + 0x4000) >> 0xF; addptr += 2; offset++; } mult6 = *(int32_t *)(hle->mp3_buffer + 0xCE8); mult4 = *(int32_t *)(hle->mp3_buffer + 0xCEC); if (t4 & 0x2) { v2 = (v2 **(uint32_t *)(hle->mp3_buffer + 0xCE8)) >> 0x10; *(int16_t *)(hle->mp3_buffer + (outPtr ^ S16)) = v2; } else { v4 = (v4 **(uint32_t *)(hle->mp3_buffer + 0xCE8)) >> 0x10; *(int16_t *)(hle->mp3_buffer + (outPtr ^ S16)) = v4; mult4 = *(uint32_t *)(hle->mp3_buffer + 0xCE8); } addptr -= 0x50; for (x = 0; x < 8; x++) { int32_t v0; int32_t v18; v2 = v4 = v6 = v8 = 0; offset = (0x22F - (t4 >> 1) + x * 0x40); for (i = 0; i < 4; i++) { v2 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x20) * (short)DeWindowLUT[offset + 0x00] + 0x4000) >> 0xF; v2 -= ((int) * (int16_t *)(hle->mp3_buffer + ((addptr + 2)) + 0x20) * (short)DeWindowLUT[offset + 0x01] + 0x4000) >> 0xF; v4 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x30) * (short)DeWindowLUT[offset + 0x08] + 0x4000) >> 0xF; v4 -= ((int) * (int16_t *)(hle->mp3_buffer + ((addptr + 2)) + 0x30) * (short)DeWindowLUT[offset + 0x09] + 0x4000) >> 0xF; v6 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x00) * (short)DeWindowLUT[offset + 0x20] + 0x4000) >> 0xF; v6 -= ((int) * (int16_t *)(hle->mp3_buffer + ((addptr + 2)) + 0x00) * (short)DeWindowLUT[offset + 0x21] + 0x4000) >> 0xF; v8 += ((int) * (int16_t *)(hle->mp3_buffer + (addptr) + 0x10) * (short)DeWindowLUT[offset + 0x28] + 0x4000) >> 0xF; v8 -= ((int) * (int16_t *)(hle->mp3_buffer + ((addptr + 2)) + 0x10) * (short)DeWindowLUT[offset + 0x29] + 0x4000) >> 0xF; addptr += 4; offset += 2; } v0 = v2 + v4; v18 = v6 + v8; /* Clamp(v0); */ /* Clamp(v18); */ /* clamp??? */ *(int16_t *)(hle->mp3_buffer + ((outPtr + 2)^S16)) = v0; *(int16_t *)(hle->mp3_buffer + ((outPtr + 4)^S16)) = v18; outPtr += 4; addptr -= 0x50; } tmp = outPtr; hi0 = mult6; hi1 = mult4; hi0 = (int)hi0 >> 0x10; hi1 = (int)hi1 >> 0x10; for (i = 0; i < 8; i++) { /* v0 */ vt = (*(int16_t *)(hle->mp3_buffer + ((tmp - 0x40)^S16)) * hi0); *(int16_t *)((uint8_t *)hle->mp3_buffer + ((tmp - 0x40)^S16)) = clamp_s16(vt); /* v17 */ vt = (*(int16_t *)(hle->mp3_buffer + ((tmp - 0x30)^S16)) * hi0); *(int16_t *)((uint8_t *)hle->mp3_buffer + ((tmp - 0x30)^S16)) = clamp_s16(vt); /* v2 */ vt = (*(int16_t *)(hle->mp3_buffer + ((tmp - 0x1E)^S16)) * hi1); *(int16_t *)((uint8_t *)hle->mp3_buffer + ((tmp - 0x1E)^S16)) = clamp_s16(vt); /* v4 */ vt = (*(int16_t *)(hle->mp3_buffer + ((tmp - 0xE)^S16)) * hi1); *(int16_t *)((uint8_t *)hle->mp3_buffer + ((tmp - 0xE)^S16)) = clamp_s16(vt); tmp += 2; } } mupen64plus-rsp-hle-src-2.6.0/src/musyx.c000066400000000000000000001027711464506742100202170ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - musyx.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2013 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include #include "arithmetics.h" #include "audio.h" #include "common.h" #include "hle_external.h" #include "hle_internal.h" #include "memory.h" /* various constants */ enum { SUBFRAME_SIZE = 192 }; enum { MAX_VOICES = 32 }; enum { SAMPLE_BUFFER_SIZE = 0x200 }; enum { SFD_VOICE_COUNT = 0x0, SFD_SFX_INDEX = 0x2, SFD_VOICE_BITMASK = 0x4, SFD_STATE_PTR = 0x8, SFD_SFX_PTR = 0xc, SFD_VOICES = 0x10, /* v2 only */ SFD2_10_PTR = 0x10, SFD2_14_BITMASK = 0x14, SFD2_15_BITMASK = 0x15, SFD2_16_BITMASK = 0x16, SFD2_18_PTR = 0x18, SFD2_1C_PTR = 0x1c, SFD2_20_PTR = 0x20, SFD2_24_PTR = 0x24, SFD2_VOICES = 0x28 }; enum { VOICE_ENV_BEGIN = 0x00, VOICE_ENV_STEP = 0x10, VOICE_PITCH_Q16 = 0x20, VOICE_PITCH_SHIFT = 0x22, VOICE_CATSRC_0 = 0x24, VOICE_CATSRC_1 = 0x30, VOICE_ADPCM_FRAMES = 0x3c, VOICE_SKIP_SAMPLES = 0x3e, /* for PCM16 */ VOICE_U16_40 = 0x40, VOICE_U16_42 = 0x42, /* for ADPCM */ VOICE_ADPCM_TABLE_PTR = 0x40, VOICE_INTERLEAVED_PTR = 0x44, VOICE_END_POINT = 0x48, VOICE_RESTART_POINT = 0x4a, VOICE_U16_4C = 0x4c, VOICE_U16_4E = 0x4e, VOICE_SIZE = 0x50 }; enum { CATSRC_PTR1 = 0x00, CATSRC_PTR2 = 0x04, CATSRC_SIZE1 = 0x08, CATSRC_SIZE2 = 0x0a }; enum { STATE_LAST_SAMPLE = 0x0, STATE_BASE_VOL = 0x100, STATE_CC0 = 0x110, STATE_740_LAST4_V1 = 0x290, STATE_740_LAST4_V2 = 0x110 }; enum { SFX_CBUFFER_PTR = 0x00, SFX_CBUFFER_LENGTH = 0x04, SFX_TAP_COUNT = 0x08, SFX_FIR4_HGAIN = 0x0a, SFX_TAP_DELAYS = 0x0c, SFX_TAP_GAINS = 0x2c, SFX_U16_3C = 0x3c, SFX_U16_3E = 0x3e, SFX_FIR4_HCOEFFS = 0x40 }; /* struct definition */ typedef struct { /* internal subframes */ int16_t left[SUBFRAME_SIZE]; int16_t right[SUBFRAME_SIZE]; int16_t cc0[SUBFRAME_SIZE]; int16_t e50[SUBFRAME_SIZE]; /* internal subframes base volumes */ int32_t base_vol[4]; /* */ int16_t subframe_740_last4[4]; } musyx_t; typedef void (*mix_sfx_with_main_subframes_t)(musyx_t *musyx, const int16_t *subframe, const uint16_t* gains); /* helper functions prototypes */ static void load_base_vol(struct hle_t* hle, int32_t *base_vol, uint32_t address); static void save_base_vol(struct hle_t* hle, const int32_t *base_vol, uint32_t address); static void update_base_vol(struct hle_t* hle, int32_t *base_vol, uint32_t voice_mask, uint32_t last_sample_ptr, uint8_t mask_15, uint32_t ptr_24); static void init_subframes_v1(musyx_t *musyx); static void init_subframes_v2(musyx_t *musyx); static uint32_t voice_stage(struct hle_t* hle, musyx_t *musyx, uint32_t voice_ptr, uint32_t last_sample_ptr); static void dma_cat8(struct hle_t* hle, uint8_t *dst, uint32_t catsrc_ptr); static void dma_cat16(struct hle_t* hle, uint16_t *dst, uint32_t catsrc_ptr); static void load_samples_PCM16(struct hle_t* hle, uint32_t voice_ptr, int16_t *samples, unsigned *segbase, unsigned *offset); static void load_samples_ADPCM(struct hle_t* hle, uint32_t voice_ptr, int16_t *samples, unsigned *segbase, unsigned *offset); static void adpcm_decode_frames(struct hle_t* hle, int16_t *dst, const uint8_t *src, const int16_t *table, uint8_t count, uint8_t skip_samples); static void adpcm_predict_frame(int16_t *dst, const uint8_t *src, const uint8_t *nibbles, unsigned int rshift); static void mix_voice_samples(struct hle_t* hle, musyx_t *musyx, uint32_t voice_ptr, const int16_t *samples, unsigned segbase, unsigned offset, uint32_t last_sample_ptr); static void sfx_stage(struct hle_t* hle, mix_sfx_with_main_subframes_t mix_sfx_with_main_subframes, musyx_t *musyx, uint32_t sfx_ptr, uint16_t idx); static void mix_sfx_with_main_subframes_v1(musyx_t *musyx, const int16_t *subframe, const uint16_t* gains); static void mix_sfx_with_main_subframes_v2(musyx_t *musyx, const int16_t *subframe, const uint16_t* gains); static void mix_samples(int16_t *y, int16_t x, int16_t hgain); static void mix_subframes(int16_t *y, const int16_t *x, int16_t hgain); static void mix_fir4(int16_t *y, const int16_t *x, int16_t hgain, const int16_t *hcoeffs); static void interleave_stage_v1(struct hle_t* hle, musyx_t *musyx, uint32_t output_ptr); static void interleave_stage_v2(struct hle_t* hle, musyx_t *musyx, uint16_t mask_16, uint32_t ptr_18, uint32_t ptr_1c, uint32_t output_ptr); static int32_t dot4(const int16_t *x, const int16_t *y) { size_t i; int32_t accu = 0; for (i = 0; i < 4; ++i) accu = clamp_s16(accu + (((int32_t)x[i] * (int32_t)y[i]) >> 15)); return accu; } /************************************************************************** * MusyX v1 audio ucode **************************************************************************/ void musyx_v1_task(struct hle_t* hle) { uint32_t sfd_ptr = *dmem_u32(hle, TASK_DATA_PTR); uint32_t sfd_count = *dmem_u32(hle, TASK_DATA_SIZE); uint32_t state_ptr; musyx_t musyx; HleVerboseMessage(hle->user_defined, "musyx_v1_task: *data=%x, #SF=%d", sfd_ptr, sfd_count); state_ptr = *dram_u32(hle, sfd_ptr + SFD_STATE_PTR); /* load initial state */ load_base_vol(hle, musyx.base_vol, state_ptr + STATE_BASE_VOL); dram_load_u16(hle, (uint16_t *)musyx.cc0, state_ptr + STATE_CC0, SUBFRAME_SIZE); dram_load_u16(hle, (uint16_t *)musyx.subframe_740_last4, state_ptr + STATE_740_LAST4_V1, 4); for (;;) { /* parse SFD structure */ uint16_t sfx_index = *dram_u16(hle, sfd_ptr + SFD_SFX_INDEX); uint32_t voice_mask = *dram_u32(hle, sfd_ptr + SFD_VOICE_BITMASK); uint32_t sfx_ptr = *dram_u32(hle, sfd_ptr + SFD_SFX_PTR); uint32_t voice_ptr = sfd_ptr + SFD_VOICES; uint32_t last_sample_ptr = state_ptr + STATE_LAST_SAMPLE; uint32_t output_ptr; /* initialize internal subframes using updated base volumes */ update_base_vol(hle, musyx.base_vol, voice_mask, last_sample_ptr, 0, 0); init_subframes_v1(&musyx); /* active voices get mixed into L,R,cc0,e50 subframes (optional) */ output_ptr = voice_stage(hle, &musyx, voice_ptr, last_sample_ptr); /* apply delay-based effects (optional) */ sfx_stage(hle, mix_sfx_with_main_subframes_v1, &musyx, sfx_ptr, sfx_index); /* emit interleaved L,R subframes */ interleave_stage_v1(hle, &musyx, output_ptr); --sfd_count; if (sfd_count == 0) break; sfd_ptr += SFD_VOICES + MAX_VOICES * VOICE_SIZE; state_ptr = *dram_u32(hle, sfd_ptr + SFD_STATE_PTR); } /* writeback updated state */ save_base_vol(hle, musyx.base_vol, state_ptr + STATE_BASE_VOL); dram_store_u16(hle, (uint16_t *)musyx.cc0, state_ptr + STATE_CC0, SUBFRAME_SIZE); dram_store_u16(hle, (uint16_t *)musyx.subframe_740_last4, state_ptr + STATE_740_LAST4_V1, 4); rsp_break(hle, SP_STATUS_TASKDONE); } /************************************************************************** * MusyX v2 audio ucode **************************************************************************/ void musyx_v2_task(struct hle_t* hle) { uint32_t sfd_ptr = *dmem_u32(hle, TASK_DATA_PTR); uint32_t sfd_count = *dmem_u32(hle, TASK_DATA_SIZE); musyx_t musyx; HleVerboseMessage(hle->user_defined, "musyx_v2_task: *data=%x, #SF=%d", sfd_ptr, sfd_count); for (;;) { /* parse SFD structure */ uint16_t sfx_index = *dram_u16(hle, sfd_ptr + SFD_SFX_INDEX); uint32_t voice_mask = *dram_u32(hle, sfd_ptr + SFD_VOICE_BITMASK); uint32_t state_ptr = *dram_u32(hle, sfd_ptr + SFD_STATE_PTR); uint32_t sfx_ptr = *dram_u32(hle, sfd_ptr + SFD_SFX_PTR); uint32_t voice_ptr = sfd_ptr + SFD2_VOICES; uint32_t ptr_10 = *dram_u32(hle, sfd_ptr + SFD2_10_PTR); uint8_t mask_14 = *dram_u8 (hle, sfd_ptr + SFD2_14_BITMASK); uint8_t mask_15 = *dram_u8 (hle, sfd_ptr + SFD2_15_BITMASK); uint16_t mask_16 = *dram_u16(hle, sfd_ptr + SFD2_16_BITMASK); uint32_t ptr_18 = *dram_u32(hle, sfd_ptr + SFD2_18_PTR); uint32_t ptr_1c = *dram_u32(hle, sfd_ptr + SFD2_1C_PTR); uint32_t ptr_20 = *dram_u32(hle, sfd_ptr + SFD2_20_PTR); uint32_t ptr_24 = *dram_u32(hle, sfd_ptr + SFD2_24_PTR); uint32_t last_sample_ptr = state_ptr + STATE_LAST_SAMPLE; uint32_t output_ptr; /* load state */ load_base_vol(hle, musyx.base_vol, state_ptr + STATE_BASE_VOL); dram_load_u16(hle, (uint16_t *)musyx.subframe_740_last4, state_ptr + STATE_740_LAST4_V2, 4); /* initialize internal subframes using updated base volumes */ update_base_vol(hle, musyx.base_vol, voice_mask, last_sample_ptr, mask_15, ptr_24); init_subframes_v2(&musyx); if (ptr_10) { /* TODO */ HleWarnMessage(hle->user_defined, "ptr_10=%08x mask_14=%02x ptr_24=%08x", ptr_10, mask_14, ptr_24); } /* active voices get mixed into L,R,cc0,e50 subframes (optional) */ output_ptr = voice_stage(hle, &musyx, voice_ptr, last_sample_ptr); /* apply delay-based effects (optional) */ sfx_stage(hle, mix_sfx_with_main_subframes_v2, &musyx, sfx_ptr, sfx_index); dram_store_u16(hle, (uint16_t*)musyx.left, output_ptr , SUBFRAME_SIZE); dram_store_u16(hle, (uint16_t*)musyx.right, output_ptr + 2*SUBFRAME_SIZE, SUBFRAME_SIZE); dram_store_u16(hle, (uint16_t*)musyx.cc0, output_ptr + 4*SUBFRAME_SIZE, SUBFRAME_SIZE); /* store state */ save_base_vol(hle, musyx.base_vol, state_ptr + STATE_BASE_VOL); dram_store_u16(hle, (uint16_t*)musyx.subframe_740_last4, state_ptr + STATE_740_LAST4_V2, 4); if (mask_16) interleave_stage_v2(hle, &musyx, mask_16, ptr_18, ptr_1c, ptr_20); --sfd_count; if (sfd_count == 0) break; sfd_ptr += SFD2_VOICES + MAX_VOICES * VOICE_SIZE; } rsp_break(hle, SP_STATUS_TASKDONE); } static void load_base_vol(struct hle_t* hle, int32_t *base_vol, uint32_t address) { base_vol[0] = ((uint32_t)(*dram_u16(hle, address)) << 16) | (*dram_u16(hle, address + 8)); base_vol[1] = ((uint32_t)(*dram_u16(hle, address + 2)) << 16) | (*dram_u16(hle, address + 10)); base_vol[2] = ((uint32_t)(*dram_u16(hle, address + 4)) << 16) | (*dram_u16(hle, address + 12)); base_vol[3] = ((uint32_t)(*dram_u16(hle, address + 6)) << 16) | (*dram_u16(hle, address + 14)); } static void save_base_vol(struct hle_t* hle, const int32_t *base_vol, uint32_t address) { unsigned k; for (k = 0; k < 4; ++k) { *dram_u16(hle, address) = (uint16_t)(base_vol[k] >> 16); address += 2; } for (k = 0; k < 4; ++k) { *dram_u16(hle, address) = (uint16_t)(base_vol[k]); address += 2; } } static void update_base_vol(struct hle_t* hle, int32_t *base_vol, uint32_t voice_mask, uint32_t last_sample_ptr, uint8_t mask_15, uint32_t ptr_24) { unsigned i, k; uint32_t mask; HleVerboseMessage(hle->user_defined, "base_vol voice_mask = %08x", voice_mask); HleVerboseMessage(hle->user_defined, "BEFORE: base_vol = %08x %08x %08x %08x", base_vol[0], base_vol[1], base_vol[2], base_vol[3]); /* optim: skip voices contributions entirely if voice_mask is empty */ if (voice_mask != 0) { for (i = 0, mask = 1; i < MAX_VOICES; ++i, mask <<= 1, last_sample_ptr += 8) { if ((voice_mask & mask) == 0) continue; for (k = 0; k < 4; ++k) base_vol[k] += (int16_t)*dram_u16(hle, last_sample_ptr + k * 2); } } /* optim: skip contributions entirely if mask_15 is empty */ if (mask_15 != 0) { for(i = 0, mask = 1; i < 4; ++i, mask <<= 1, ptr_24 += 8) { if ((mask_15 & mask) == 0) continue; for(k = 0; k < 4; ++k) base_vol[k] += (int16_t)*dram_u16(hle, ptr_24 + k * 2); } } /* apply 3% decay */ for (k = 0; k < 4; ++k) base_vol[k] = (base_vol[k] * 0x0000f850) >> 16; HleVerboseMessage(hle->user_defined, "AFTER: base_vol = %08x %08x %08x %08x", base_vol[0], base_vol[1], base_vol[2], base_vol[3]); } static void init_subframes_v1(musyx_t *musyx) { unsigned i; int16_t base_cc0 = clamp_s16(musyx->base_vol[2]); int16_t base_e50 = clamp_s16(musyx->base_vol[3]); int16_t *left = musyx->left; int16_t *right = musyx->right; int16_t *cc0 = musyx->cc0; int16_t *e50 = musyx->e50; for (i = 0; i < SUBFRAME_SIZE; ++i) { *(e50++) = base_e50; *(left++) = clamp_s16(*cc0 + base_cc0); *(right++) = clamp_s16(-*cc0 - base_cc0); *(cc0++) = 0; } } static void init_subframes_v2(musyx_t *musyx) { unsigned i,k; int16_t values[4]; int16_t* subframes[4]; for(k = 0; k < 4; ++k) values[k] = clamp_s16(musyx->base_vol[k]); subframes[0] = musyx->left; subframes[1] = musyx->right; subframes[2] = musyx->cc0; subframes[3] = musyx->e50; for (i = 0; i < SUBFRAME_SIZE; ++i) { for(k = 0; k < 4; ++k) *(subframes[k]++) = values[k]; } } /* Process voices, and returns interleaved subframe destination address */ static uint32_t voice_stage(struct hle_t* hle, musyx_t *musyx, uint32_t voice_ptr, uint32_t last_sample_ptr) { uint32_t output_ptr; int i = 0; /* voice stage can be skipped if first voice has no samples */ if (*dram_u16(hle, voice_ptr + VOICE_CATSRC_0 + CATSRC_SIZE1) == 0) { HleVerboseMessage(hle->user_defined, "Skipping Voice stage"); output_ptr = *dram_u32(hle, voice_ptr + VOICE_INTERLEAVED_PTR); } else { /* otherwise process voices until a non null output_ptr is encountered */ for (;;) { /* load voice samples (PCM16 or APDCM) */ int16_t samples[SAMPLE_BUFFER_SIZE]; unsigned segbase; unsigned offset; HleVerboseMessage(hle->user_defined, "Processing Voice #%d", i); if (*dram_u8(hle, voice_ptr + VOICE_ADPCM_FRAMES) == 0) load_samples_PCM16(hle, voice_ptr, samples, &segbase, &offset); else load_samples_ADPCM(hle, voice_ptr, samples, &segbase, &offset); /* mix them with each internal subframes */ mix_voice_samples(hle, musyx, voice_ptr, samples, segbase, offset, last_sample_ptr + i * 8); /* check break condition */ output_ptr = *dram_u32(hle, voice_ptr + VOICE_INTERLEAVED_PTR); if (output_ptr != 0) break; /* next voice */ ++i; voice_ptr += VOICE_SIZE; } } return output_ptr; } static void dma_cat8(struct hle_t* hle, uint8_t *dst, uint32_t catsrc_ptr) { uint32_t ptr1 = *dram_u32(hle, catsrc_ptr + CATSRC_PTR1); uint32_t ptr2 = *dram_u32(hle, catsrc_ptr + CATSRC_PTR2); uint16_t size1 = *dram_u16(hle, catsrc_ptr + CATSRC_SIZE1); uint16_t size2 = *dram_u16(hle, catsrc_ptr + CATSRC_SIZE2); size_t count1 = size1; size_t count2 = size2; HleVerboseMessage(hle->user_defined, "dma_cat: %08x %08x %04x %04x", ptr1, ptr2, size1, size2); dram_load_u8(hle, dst, ptr1, count1); if (size2 == 0) return; dram_load_u8(hle, dst + count1, ptr2, count2); } static void dma_cat16(struct hle_t* hle, uint16_t *dst, uint32_t catsrc_ptr) { uint32_t ptr1 = *dram_u32(hle, catsrc_ptr + CATSRC_PTR1); uint32_t ptr2 = *dram_u32(hle, catsrc_ptr + CATSRC_PTR2); uint16_t size1 = *dram_u16(hle, catsrc_ptr + CATSRC_SIZE1); uint16_t size2 = *dram_u16(hle, catsrc_ptr + CATSRC_SIZE2); size_t count1 = size1 >> 1; size_t count2 = size2 >> 1; HleVerboseMessage(hle->user_defined, "dma_cat: %08x %08x %04x %04x", ptr1, ptr2, size1, size2); dram_load_u16(hle, dst, ptr1, count1); if (size2 == 0) return; dram_load_u16(hle, dst + count1, ptr2, count2); } static void load_samples_PCM16(struct hle_t* hle, uint32_t voice_ptr, int16_t *samples, unsigned *segbase, unsigned *offset) { uint8_t u8_3e = *dram_u8(hle, voice_ptr + VOICE_SKIP_SAMPLES); uint16_t u16_40 = *dram_u16(hle, voice_ptr + VOICE_U16_40); uint16_t u16_42 = *dram_u16(hle, voice_ptr + VOICE_U16_42); unsigned count = align(u16_40 + u8_3e, 4); HleVerboseMessage(hle->user_defined, "Format: PCM16"); *segbase = SAMPLE_BUFFER_SIZE - count; *offset = u8_3e; dma_cat16(hle, (uint16_t *)samples + *segbase, voice_ptr + VOICE_CATSRC_0); if (u16_42 != 0) dma_cat16(hle, (uint16_t *)samples, voice_ptr + VOICE_CATSRC_1); } static void load_samples_ADPCM(struct hle_t* hle, uint32_t voice_ptr, int16_t *samples, unsigned *segbase, unsigned *offset) { /* decompressed samples cannot exceed 0x400 bytes; * ADPCM has a compression ratio of 5/16 */ uint8_t buffer[SAMPLE_BUFFER_SIZE * 2 * 5 / 16]; int16_t adpcm_table[128]; uint8_t u8_3c = *dram_u8(hle, voice_ptr + VOICE_ADPCM_FRAMES ); uint8_t u8_3d = *dram_u8(hle, voice_ptr + VOICE_ADPCM_FRAMES + 1); uint8_t u8_3e = *dram_u8(hle, voice_ptr + VOICE_SKIP_SAMPLES ); uint8_t u8_3f = *dram_u8(hle, voice_ptr + VOICE_SKIP_SAMPLES + 1); uint32_t adpcm_table_ptr = *dram_u32(hle, voice_ptr + VOICE_ADPCM_TABLE_PTR); unsigned count; HleVerboseMessage(hle->user_defined, "Format: ADPCM"); HleVerboseMessage(hle->user_defined, "Loading ADPCM table: %08x", adpcm_table_ptr); dram_load_u16(hle, (uint16_t *)adpcm_table, adpcm_table_ptr, 128); count = u8_3c << 5; *segbase = SAMPLE_BUFFER_SIZE - count; *offset = u8_3e & 0x1f; dma_cat8(hle, buffer, voice_ptr + VOICE_CATSRC_0); adpcm_decode_frames(hle, samples + *segbase, buffer, adpcm_table, u8_3c, u8_3e); if (u8_3d != 0) { dma_cat8(hle, buffer, voice_ptr + VOICE_CATSRC_1); adpcm_decode_frames(hle, samples, buffer, adpcm_table, u8_3d, u8_3f); } } static void adpcm_decode_frames(struct hle_t* hle, int16_t *dst, const uint8_t *src, const int16_t *table, uint8_t count, uint8_t skip_samples) { int16_t frame[32]; const uint8_t *nibbles = src + 8; unsigned i; bool jump_gap = false; HleVerboseMessage(hle->user_defined, "ADPCM decode: count=%d, skip=%d", count, skip_samples); if (skip_samples >= 32) { jump_gap = true; nibbles += 16; src += 4; } for (i = 0; i < count; ++i) { uint8_t c2 = nibbles[0]; const int16_t *book = (c2 & 0xf0) + table; unsigned int rshift = (c2 & 0x0f); adpcm_predict_frame(frame, src, nibbles, rshift); memcpy(dst, frame, 2 * sizeof(frame[0])); adpcm_compute_residuals(dst + 2, frame + 2, book, dst , 6); adpcm_compute_residuals(dst + 8, frame + 8, book, dst + 6, 8); adpcm_compute_residuals(dst + 16, frame + 16, book, dst + 14, 8); adpcm_compute_residuals(dst + 24, frame + 24, book, dst + 22, 8); if (jump_gap) { nibbles += 8; src += 32; } jump_gap = !jump_gap; nibbles += 16; src += 4; dst += 32; } } static void adpcm_predict_frame(int16_t *dst, const uint8_t *src, const uint8_t *nibbles, unsigned int rshift) { unsigned int i; *(dst++) = (src[0] << 8) | src[1]; *(dst++) = (src[2] << 8) | src[3]; for (i = 1; i < 16; ++i) { uint8_t byte = nibbles[i]; *(dst++) = adpcm_predict_sample(byte, 0xf0, 8, rshift); *(dst++) = adpcm_predict_sample(byte, 0x0f, 12, rshift); } } static void mix_voice_samples(struct hle_t* hle, musyx_t *musyx, uint32_t voice_ptr, const int16_t *samples, unsigned segbase, unsigned offset, uint32_t last_sample_ptr) { int i, k; /* parse VOICE structure */ const uint16_t pitch_q16 = *dram_u16(hle, voice_ptr + VOICE_PITCH_Q16); const uint16_t pitch_shift = *dram_u16(hle, voice_ptr + VOICE_PITCH_SHIFT); /* Q4.12 */ const uint16_t end_point = *dram_u16(hle, voice_ptr + VOICE_END_POINT); const uint16_t restart_point = *dram_u16(hle, voice_ptr + VOICE_RESTART_POINT); const uint16_t u16_4e = *dram_u16(hle, voice_ptr + VOICE_U16_4E); /* init values and pointers */ const int16_t *sample = samples + segbase + offset + u16_4e; const int16_t *const sample_end = samples + segbase + end_point; const int16_t *const sample_restart = samples + (restart_point & 0x7fff) + (((restart_point & 0x8000) != 0) ? 0x000 : segbase); uint32_t pitch_accu = pitch_q16; uint32_t pitch_step = pitch_shift << 4; int32_t v4_env[4]; int32_t v4_env_step[4]; int16_t *v4_dst[4]; int16_t v4[4]; dram_load_u32(hle, (uint32_t *)v4_env, voice_ptr + VOICE_ENV_BEGIN, 4); dram_load_u32(hle, (uint32_t *)v4_env_step, voice_ptr + VOICE_ENV_STEP, 4); v4_dst[0] = musyx->left; v4_dst[1] = musyx->right; v4_dst[2] = musyx->cc0; v4_dst[3] = musyx->e50; HleVerboseMessage(hle->user_defined, "Voice debug: segbase=%d" "\tu16_4e=%04x\n" "\tpitch: frac0=%04x shift=%04x\n" "\tend_point=%04x restart_point=%04x\n" "\tenv = %08x %08x %08x %08x\n" "\tenv_step = %08x %08x %08x %08x\n", segbase, u16_4e, pitch_q16, pitch_shift, end_point, restart_point, v4_env[0], v4_env[1], v4_env[2], v4_env[3], v4_env_step[0], v4_env_step[1], v4_env_step[2], v4_env_step[3]); for (i = 0; i < SUBFRAME_SIZE; ++i) { /* update sample and lut pointers and then pitch_accu */ const int16_t *lut = (RESAMPLE_LUT + ((pitch_accu & 0xfc00) >> 8)); int dist; int16_t v; sample += (pitch_accu >> 16); pitch_accu &= 0xffff; pitch_accu += pitch_step; /* handle end/restart points */ dist = sample - sample_end; if (dist >= 0) sample = sample_restart + dist; /* apply resample filter */ v = clamp_s16(dot4(sample, lut)); for (k = 0; k < 4; ++k) { /* envmix */ int32_t accu = (v * (v4_env[k] >> 16)) >> 15; v4[k] = clamp_s16(accu); *(v4_dst[k]) = clamp_s16(accu + *(v4_dst[k])); /* update envelopes and dst pointers */ ++(v4_dst[k]); v4_env[k] += v4_env_step[k]; } } /* save last resampled sample */ dram_store_u16(hle, (uint16_t *)v4, last_sample_ptr, 4); HleVerboseMessage(hle->user_defined, "last_sample = %04x %04x %04x %04x", v4[0], v4[1], v4[2], v4[3]); } static void sfx_stage(struct hle_t* hle, mix_sfx_with_main_subframes_t mix_sfx_with_main_subframes, musyx_t *musyx, uint32_t sfx_ptr, uint16_t idx) { unsigned int i; int16_t buffer[SUBFRAME_SIZE + 4]; int16_t *subframe = buffer + 4; uint32_t tap_delays[8]; int16_t tap_gains[8]; int16_t fir4_hcoeffs[4]; int16_t delayed[SUBFRAME_SIZE]; int dpos, dlength; const uint32_t pos = idx * SUBFRAME_SIZE; uint32_t cbuffer_ptr; uint32_t cbuffer_length; uint16_t tap_count; int16_t fir4_hgain; uint16_t sfx_gains[2]; HleVerboseMessage(hle->user_defined, "SFX: %08x, idx=%d", sfx_ptr, idx); if (sfx_ptr == 0) return; /* load sfx parameters */ cbuffer_ptr = *dram_u32(hle, sfx_ptr + SFX_CBUFFER_PTR); cbuffer_length = *dram_u32(hle, sfx_ptr + SFX_CBUFFER_LENGTH); tap_count = *dram_u16(hle, sfx_ptr + SFX_TAP_COUNT); dram_load_u32(hle, tap_delays, sfx_ptr + SFX_TAP_DELAYS, 8); dram_load_u16(hle, (uint16_t *)tap_gains, sfx_ptr + SFX_TAP_GAINS, 8); fir4_hgain = *dram_u16(hle, sfx_ptr + SFX_FIR4_HGAIN); dram_load_u16(hle, (uint16_t *)fir4_hcoeffs, sfx_ptr + SFX_FIR4_HCOEFFS, 4); sfx_gains[0] = *dram_u16(hle, sfx_ptr + SFX_U16_3C); sfx_gains[1] = *dram_u16(hle, sfx_ptr + SFX_U16_3E); HleVerboseMessage(hle->user_defined, "cbuffer: ptr=%08x length=%x", cbuffer_ptr, cbuffer_length); HleVerboseMessage(hle->user_defined, "fir4: hgain=%04x hcoeff=%04x %04x %04x %04x", fir4_hgain, fir4_hcoeffs[0], fir4_hcoeffs[1], fir4_hcoeffs[2], fir4_hcoeffs[3]); HleVerboseMessage(hle->user_defined, "tap count=%d\n" "delays: %08x %08x %08x %08x %08x %08x %08x %08x\n" "gains: %04x %04x %04x %04x %04x %04x %04x %04x", tap_count, tap_delays[0], tap_delays[1], tap_delays[2], tap_delays[3], tap_delays[4], tap_delays[5], tap_delays[6], tap_delays[7], tap_gains[0], tap_gains[1], tap_gains[2], tap_gains[3], tap_gains[4], tap_gains[5], tap_gains[6], tap_gains[7]); HleVerboseMessage(hle->user_defined, "sfx_gains=%04x %04x", sfx_gains[0], sfx_gains[1]); /* mix up to 8 delayed subframes */ memset(subframe, 0, SUBFRAME_SIZE * sizeof(subframe[0])); for (i = 0; i < tap_count; ++i) { dpos = pos - tap_delays[i]; if (dpos <= 0) dpos += cbuffer_length; dlength = SUBFRAME_SIZE; if ((uint32_t)(dpos + SUBFRAME_SIZE) > cbuffer_length) { dlength = cbuffer_length - dpos; dram_load_u16(hle, (uint16_t *)delayed + dlength, cbuffer_ptr, SUBFRAME_SIZE - dlength); } dram_load_u16(hle, (uint16_t *)delayed, cbuffer_ptr + dpos * 2, dlength); mix_subframes(subframe, delayed, tap_gains[i]); } /* add resulting subframe to main subframes */ mix_sfx_with_main_subframes(musyx, subframe, sfx_gains); /* apply FIR4 filter and writeback filtered result */ memcpy(buffer, musyx->subframe_740_last4, 4 * sizeof(int16_t)); memcpy(musyx->subframe_740_last4, subframe + SUBFRAME_SIZE - 4, 4 * sizeof(int16_t)); mix_fir4(musyx->e50, buffer + 1, fir4_hgain, fir4_hcoeffs); dram_store_u16(hle, (uint16_t *)musyx->e50, cbuffer_ptr + pos * 2, SUBFRAME_SIZE); } static void mix_sfx_with_main_subframes_v1(musyx_t *musyx, const int16_t *subframe, const uint16_t* UNUSED(gains)) { unsigned i; for (i = 0; i < SUBFRAME_SIZE; ++i) { int16_t v = subframe[i]; musyx->left[i] = clamp_s16(musyx->left[i] + v); musyx->right[i] = clamp_s16(musyx->right[i] + v); } } static void mix_sfx_with_main_subframes_v2(musyx_t *musyx, const int16_t *subframe, const uint16_t* gains) { unsigned i; for (i = 0; i < SUBFRAME_SIZE; ++i) { int16_t v = subframe[i]; int16_t v1 = (int32_t)(v * gains[0]) >> 16; int16_t v2 = (int32_t)(v * gains[1]) >> 16; musyx->left[i] = clamp_s16(musyx->left[i] + v1); musyx->right[i] = clamp_s16(musyx->right[i] + v1); musyx->cc0[i] = clamp_s16(musyx->cc0[i] + v2); } } static void mix_samples(int16_t *y, int16_t x, int16_t hgain) { *y = clamp_s16(*y + ((x * hgain + 0x4000) >> 15)); } static void mix_subframes(int16_t *y, const int16_t *x, int16_t hgain) { unsigned int i; for (i = 0; i < SUBFRAME_SIZE; ++i) mix_samples(&y[i], x[i], hgain); } static void mix_fir4(int16_t *y, const int16_t *x, int16_t hgain, const int16_t *hcoeffs) { unsigned int i; int32_t h[4]; h[0] = (hgain * hcoeffs[0]) >> 15; h[1] = (hgain * hcoeffs[1]) >> 15; h[2] = (hgain * hcoeffs[2]) >> 15; h[3] = (hgain * hcoeffs[3]) >> 15; for (i = 0; i < SUBFRAME_SIZE; ++i) { int32_t v = (h[0] * x[i] + h[1] * x[i + 1] + h[2] * x[i + 2] + h[3] * x[i + 3]) >> 15; y[i] = clamp_s16(y[i] + v); } } static void interleave_stage_v1(struct hle_t* hle, musyx_t *musyx, uint32_t output_ptr) { size_t i; int16_t base_left; int16_t base_right; int16_t *left; int16_t *right; uint32_t *dst; HleVerboseMessage(hle->user_defined, "interleave: %08x", output_ptr); base_left = clamp_s16(musyx->base_vol[0]); base_right = clamp_s16(musyx->base_vol[1]); left = musyx->left; right = musyx->right; dst = dram_u32(hle, output_ptr); for (i = 0; i < SUBFRAME_SIZE; ++i) { uint16_t l = clamp_s16(*(left++) + base_left); uint16_t r = clamp_s16(*(right++) + base_right); *(dst++) = (l << 16) | r; } } static void interleave_stage_v2(struct hle_t* hle, musyx_t *musyx, uint16_t mask_16, uint32_t ptr_18, uint32_t ptr_1c, uint32_t output_ptr) { unsigned i, k; int16_t subframe[SUBFRAME_SIZE]; uint32_t *dst; uint16_t mask; HleVerboseMessage(hle->user_defined, "mask_16=%04x ptr_18=%08x ptr_1c=%08x output_ptr=%08x", mask_16, ptr_18, ptr_1c, output_ptr); /* compute L_total, R_total and update subframe @ptr_1c */ memset(subframe, 0, SUBFRAME_SIZE*sizeof(subframe[0])); for(i = 0; i < SUBFRAME_SIZE; ++i) { int16_t v = *dram_u16(hle, ptr_1c + i*2); musyx->left[i] = v; musyx->right[i] = clamp_s16(-v); } for (k = 0, mask = 1; k < 8; ++k, mask <<= 1, ptr_18 += 8) { int16_t hgain; uint32_t address; if ((mask_16 & mask) == 0) continue; address = *dram_u32(hle, ptr_18); hgain = *dram_u16(hle, ptr_18 + 4); for(i = 0; i < SUBFRAME_SIZE; ++i, address += 2) { mix_samples(&musyx->left[i], *dram_u16(hle, address), hgain); mix_samples(&musyx->right[i], *dram_u16(hle, address + 2*SUBFRAME_SIZE), hgain); mix_samples(&subframe[i], *dram_u16(hle, address + 4*SUBFRAME_SIZE), hgain); } } /* interleave L_total and R_total */ dst = dram_u32(hle, output_ptr); for(i = 0; i < SUBFRAME_SIZE; ++i) { uint16_t l = musyx->left[i]; uint16_t r = musyx->right[i]; *(dst++) = (l << 16) | r; } /* writeback subframe @ptr_1c */ dram_store_u16(hle, (uint16_t*)subframe, ptr_1c, SUBFRAME_SIZE); } mupen64plus-rsp-hle-src-2.6.0/src/osal_dynamiclib.h000066400000000000000000000035661464506742100221720ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-ui-console - osal_dynamiclib.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2009 Richard Goedeken * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #if !defined(OSAL_DYNAMICLIB_H) #define OSAL_DYNAMICLIB_H #include "m64p_types.h" m64p_error osal_dynlib_open(m64p_dynlib_handle *pLibHandle, const char *pccLibraryPath); void * osal_dynlib_getproc(m64p_dynlib_handle LibHandle, const char *pccProcedureName); m64p_error osal_dynlib_close(m64p_dynlib_handle LibHandle); #endif /* #define OSAL_DYNAMICLIB_H */ mupen64plus-rsp-hle-src-2.6.0/src/osal_dynamiclib_unix.c000066400000000000000000000054271464506742100232260ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-ui-console - osal_dynamiclib_unix.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2009 Richard Goedeken * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include #include "m64p_types.h" #include "hle_external.h" #include "osal_dynamiclib.h" m64p_error osal_dynlib_open(m64p_dynlib_handle *pLibHandle, const char *pccLibraryPath) { if (pLibHandle == NULL || pccLibraryPath == NULL) return M64ERR_INPUT_ASSERT; *pLibHandle = dlopen(pccLibraryPath, RTLD_NOW); if (*pLibHandle == NULL) { /* only print an error message if there is a directory separator (/) in the pathname */ /* this prevents us from throwing an error for the use case where Mupen64Plus is not installed */ if (strchr(pccLibraryPath, '/') != NULL) HleErrorMessage(NULL, "dlopen('%s') failed: %s", pccLibraryPath, dlerror()); return M64ERR_INPUT_NOT_FOUND; } return M64ERR_SUCCESS; } void * osal_dynlib_getproc(m64p_dynlib_handle LibHandle, const char *pccProcedureName) { if (pccProcedureName == NULL) return NULL; return dlsym(LibHandle, pccProcedureName); } m64p_error osal_dynlib_close(m64p_dynlib_handle LibHandle) { int rval = dlclose(LibHandle); if (rval != 0) { HleErrorMessage(NULL, "dlclose() failed: %s", dlerror()); return M64ERR_INTERNAL; } return M64ERR_SUCCESS; } mupen64plus-rsp-hle-src-2.6.0/src/osal_dynamiclib_win32.c000066400000000000000000000061151464506742100232000ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-ui-console - osal_dynamiclib_win32.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2009 Richard Goedeken * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "m64p_types.h" #include "hle_external.h" #include "osal_dynamiclib.h" m64p_error osal_dynlib_open(m64p_dynlib_handle *pLibHandle, const char *pccLibraryPath) { if (pLibHandle == NULL || pccLibraryPath == NULL) return M64ERR_INPUT_ASSERT; *pLibHandle = LoadLibrary(pccLibraryPath); if (*pLibHandle == NULL) { char *pchErrMsg; DWORD dwErr = GetLastError(); FormatMessage(FORMAT_MESSAGE_ALLOCATE_BUFFER | FORMAT_MESSAGE_FROM_SYSTEM, NULL, dwErr, MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), (LPTSTR) &pchErrMsg, 0, NULL); HleErrorMessage(NULL, "LoadLibrary('%s') error: %s", pccLibraryPath, pchErrMsg); LocalFree(pchErrMsg); return M64ERR_INPUT_NOT_FOUND; } return M64ERR_SUCCESS; } void * osal_dynlib_getproc(m64p_dynlib_handle LibHandle, const char *pccProcedureName) { if (pccProcedureName == NULL) return NULL; return GetProcAddress(LibHandle, pccProcedureName); } m64p_error osal_dynlib_close(m64p_dynlib_handle LibHandle) { int rval = FreeLibrary(LibHandle); if (rval == 0) { char *pchErrMsg; DWORD dwErr = GetLastError(); FormatMessage(FORMAT_MESSAGE_ALLOCATE_BUFFER | FORMAT_MESSAGE_FROM_SYSTEM, NULL, dwErr, MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT), (LPTSTR) &pchErrMsg, 0, NULL); HleErrorMessage(NULL, "FreeLibrary() error: %s", pchErrMsg); LocalFree(pchErrMsg); return M64ERR_INTERNAL; } return M64ERR_SUCCESS; } mupen64plus-rsp-hle-src-2.6.0/src/plugin.c000066400000000000000000000420721464506742100203250ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - plugin.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * Copyright (C) 2009 Richard Goedeken * * Copyright (C) 2002 Hacktarux * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "common.h" #include "hle.h" #include "hle_internal.h" #include "hle_external.h" #define M64P_PLUGIN_PROTOTYPES 1 #include "m64p_common.h" #include "m64p_config.h" #include "m64p_frontend.h" #include "m64p_plugin.h" #include "m64p_types.h" #include "osal_dynamiclib.h" #define CONFIG_API_VERSION 0x020100 #define CONFIG_PARAM_VERSION 1.00 #define RSP_API_VERSION 0x20000 #define RSP_HLE_VERSION 0x020600 #define RSP_PLUGIN_API_VERSION 0x020000 #define RSP_HLE_CONFIG_SECTION "Rsp-HLE" #define RSP_HLE_CONFIG_VERSION "Version" #define RSP_HLE_CONFIG_FALLBACK "RspFallback" #define RSP_HLE_CONFIG_HLE_GFX "DisplayListToGraphicsPlugin" #define RSP_HLE_CONFIG_HLE_AUD "AudioListToAudioPlugin" #define VERSION_PRINTF_SPLIT(x) (((x) >> 16) & 0xffff), (((x) >> 8) & 0xff), ((x) & 0xff) /* Handy macro to avoid code bloat when loading symbols */ #define GET_FUNC(type, field, name) \ ((field = (type)osal_dynlib_getproc(handle, name)) != NULL) /* local variables */ static struct hle_t g_hle; static void (*l_CheckInterrupts)(void) = NULL; static void (*l_ProcessDlistList)(void) = NULL; static void (*l_ProcessAlistList)(void) = NULL; static void (*l_ProcessRdpList)(void) = NULL; static void (*l_ShowCFB)(void) = NULL; static void (*l_DebugCallback)(void *, int, const char *) = NULL; static void *l_DebugCallContext = NULL; static m64p_dynlib_handle l_CoreHandle = NULL; static int l_PluginInit = 0; static m64p_handle l_ConfigRspHle; static m64p_dynlib_handle l_RspFallback; static ptr_InitiateRSP l_InitiateRSP = NULL; static ptr_DoRspCycles l_DoRspCycles = NULL; static ptr_RomClosed l_RomClosed = NULL; static ptr_PluginShutdown l_PluginShutdown = NULL; /* definitions of pointers to Core functions */ static ptr_ConfigOpenSection ConfigOpenSection = NULL; static ptr_ConfigDeleteSection ConfigDeleteSection = NULL; static ptr_ConfigSetParameter ConfigSetParameter = NULL; static ptr_ConfigGetParameter ConfigGetParameter = NULL; static ptr_ConfigSetDefaultInt ConfigSetDefaultInt = NULL; static ptr_ConfigSetDefaultFloat ConfigSetDefaultFloat = NULL; static ptr_ConfigSetDefaultBool ConfigSetDefaultBool = NULL; static ptr_ConfigSetDefaultString ConfigSetDefaultString = NULL; static ptr_ConfigGetParamInt ConfigGetParamInt = NULL; static ptr_ConfigGetParamFloat ConfigGetParamFloat = NULL; static ptr_ConfigGetParamBool ConfigGetParamBool = NULL; static ptr_ConfigGetParamString ConfigGetParamString = NULL; static ptr_CoreDoCommand CoreDoCommand = NULL; /* local function */ static void teardown_rsp_fallback() { if (l_RspFallback != NULL) { (*l_PluginShutdown)(); osal_dynlib_close(l_RspFallback); } l_RspFallback = NULL; l_DoRspCycles = NULL; l_InitiateRSP = NULL; l_RomClosed = NULL; l_PluginShutdown = NULL; } static void setup_rsp_fallback(const char* rsp_fallback_path) { m64p_dynlib_handle handle = NULL; /* reset rsp fallback */ teardown_rsp_fallback(); if (rsp_fallback_path == NULL || strlen(rsp_fallback_path) == 0) { HleInfoMessage(NULL, "RSP Fallback disabled !"); return; } /* load plugin */ if (osal_dynlib_open(&handle, rsp_fallback_path) != M64ERR_SUCCESS) { HleErrorMessage(NULL, "Can't load library: %s", rsp_fallback_path); return; } /* call the GetVersion function for the plugin and check compatibility */ ptr_PluginGetVersion PluginGetVersion = (ptr_PluginGetVersion) osal_dynlib_getproc(handle, "PluginGetVersion"); if (PluginGetVersion == NULL) { HleErrorMessage(NULL, "library '%s' is not a Mupen64Plus library.", rsp_fallback_path); goto close_handle; } m64p_plugin_type plugin_type = (m64p_plugin_type)0; int plugin_version = 0; const char *plugin_name = NULL; int api_version = 0; (*PluginGetVersion)(&plugin_type, &plugin_version, &api_version, &plugin_name, NULL); if (plugin_type != M64PLUGIN_RSP) { HleErrorMessage(NULL, "plugin %s is not an RSP plugin (%u)", plugin_name, plugin_type); goto close_handle; } if ((api_version & 0xffff0000) != (RSP_API_VERSION & 0xffff0000)) { HleErrorMessage(NULL, "plugin %s. Version mismatch: %u.%u. Expected >= %u.0", plugin_name, (uint16_t)(api_version >> 16), (uint16_t)(api_version), (uint16_t)(RSP_API_VERSION >> 16)); goto close_handle; } /* load functions */ ptr_PluginStartup PluginStartup; if (!GET_FUNC(ptr_PluginStartup, PluginStartup, "PluginStartup") || !GET_FUNC(ptr_PluginShutdown, l_PluginShutdown, "PluginShutdown") || !GET_FUNC(ptr_DoRspCycles, l_DoRspCycles, "DoRspCycles") || !GET_FUNC(ptr_InitiateRSP, l_InitiateRSP, "InitiateRSP") || !GET_FUNC(ptr_RomClosed, l_RomClosed, "RomClosed")) { HleErrorMessage(NULL, "broken RSP plugin; function(s) not found."); l_PluginShutdown = NULL; l_DoRspCycles = NULL; l_InitiateRSP = NULL; l_RomClosed = NULL; goto close_handle; } /* call the plugin's initialization function and make sure it starts okay */ if ((*PluginStartup)(l_CoreHandle, l_DebugCallContext, l_DebugCallback) != M64ERR_SUCCESS) { HleErrorMessage(NULL, "Error: %s plugin library '%s' failed to start.", plugin_name, rsp_fallback_path); goto close_handle; } /* OK we're done ! */ l_RspFallback = handle; HleInfoMessage(NULL, "RSP Fallback '%s' loaded successfully !", rsp_fallback_path); return; close_handle: osal_dynlib_close(handle); } static void DebugMessage(int level, const char *message, va_list args) { char msgbuf[1024]; if (l_DebugCallback == NULL) return; vsprintf(msgbuf, message, args); (*l_DebugCallback)(l_DebugCallContext, level, msgbuf); } /* Global functions needed by HLE core */ void HleVerboseMessage(void* UNUSED(user_defined), const char *message, ...) { va_list args; va_start(args, message); DebugMessage(M64MSG_VERBOSE, message, args); va_end(args); } void HleInfoMessage(void* UNUSED(user_defined), const char *message, ...) { va_list args; va_start(args, message); DebugMessage(M64MSG_INFO, message, args); va_end(args); } void HleErrorMessage(void* UNUSED(user_defined), const char *message, ...) { va_list args; va_start(args, message); DebugMessage(M64MSG_ERROR, message, args); va_end(args); } void HleWarnMessage(void* UNUSED(user_defined), const char *message, ...) { va_list args; va_start(args, message); DebugMessage(M64MSG_WARNING, message, args); va_end(args); } void HleCheckInterrupts(void* UNUSED(user_defined)) { if (l_CheckInterrupts == NULL) return; (*l_CheckInterrupts)(); } void HleProcessDlistList(void* UNUSED(user_defined)) { if (l_ProcessDlistList == NULL) return; (*l_ProcessDlistList)(); } void HleProcessAlistList(void* UNUSED(user_defined)) { if (l_ProcessAlistList == NULL) return; (*l_ProcessAlistList)(); } void HleProcessRdpList(void* UNUSED(user_defined)) { if (l_ProcessRdpList == NULL) return; (*l_ProcessRdpList)(); } void HleShowCFB(void* UNUSED(user_defined)) { if (l_ShowCFB == NULL) return; (*l_ShowCFB)(); } int HleForwardTask(void* user_defined) { if (l_DoRspCycles == NULL) return -1; (*l_DoRspCycles)(-1); return 0; } /* DLL-exported functions */ EXPORT m64p_error CALL PluginStartup(m64p_dynlib_handle CoreLibHandle, void *Context, void (*DebugCallback)(void *, int, const char *)) { ptr_CoreGetAPIVersions CoreAPIVersionFunc; int ConfigAPIVersion, DebugAPIVersion, VidextAPIVersion; float fConfigParamsVersion = 0.0f; if (l_PluginInit) return M64ERR_ALREADY_INIT; /* first thing is to set the callback function for debug info */ l_DebugCallback = DebugCallback; l_DebugCallContext = Context; /* attach and call the CoreGetAPIVersions function, check Config API version for compatibility */ CoreAPIVersionFunc = (ptr_CoreGetAPIVersions) osal_dynlib_getproc(CoreLibHandle, "CoreGetAPIVersions"); if (CoreAPIVersionFunc == NULL) { HleErrorMessage(NULL, "Core emulator broken; no CoreAPIVersionFunc() function found."); return M64ERR_INCOMPATIBLE; } (*CoreAPIVersionFunc)(&ConfigAPIVersion, &DebugAPIVersion, &VidextAPIVersion, NULL); if ((ConfigAPIVersion & 0xffff0000) != (CONFIG_API_VERSION & 0xffff0000)) { HleErrorMessage(NULL, "Emulator core Config API (v%i.%i.%i) incompatible with plugin (v%i.%i.%i)", VERSION_PRINTF_SPLIT(ConfigAPIVersion), VERSION_PRINTF_SPLIT(CONFIG_API_VERSION)); return M64ERR_INCOMPATIBLE; } /* Get the core config function pointers from the library handle */ ConfigOpenSection = (ptr_ConfigOpenSection) osal_dynlib_getproc(CoreLibHandle, "ConfigOpenSection"); ConfigDeleteSection = (ptr_ConfigDeleteSection) osal_dynlib_getproc(CoreLibHandle, "ConfigDeleteSection"); ConfigSetParameter = (ptr_ConfigSetParameter) osal_dynlib_getproc(CoreLibHandle, "ConfigSetParameter"); ConfigGetParameter = (ptr_ConfigGetParameter) osal_dynlib_getproc(CoreLibHandle, "ConfigGetParameter"); ConfigSetDefaultInt = (ptr_ConfigSetDefaultInt) osal_dynlib_getproc(CoreLibHandle, "ConfigSetDefaultInt"); ConfigSetDefaultFloat = (ptr_ConfigSetDefaultFloat) osal_dynlib_getproc(CoreLibHandle, "ConfigSetDefaultFloat"); ConfigSetDefaultBool = (ptr_ConfigSetDefaultBool) osal_dynlib_getproc(CoreLibHandle, "ConfigSetDefaultBool"); ConfigSetDefaultString = (ptr_ConfigSetDefaultString) osal_dynlib_getproc(CoreLibHandle, "ConfigSetDefaultString"); ConfigGetParamInt = (ptr_ConfigGetParamInt) osal_dynlib_getproc(CoreLibHandle, "ConfigGetParamInt"); ConfigGetParamFloat = (ptr_ConfigGetParamFloat) osal_dynlib_getproc(CoreLibHandle, "ConfigGetParamFloat"); ConfigGetParamBool = (ptr_ConfigGetParamBool) osal_dynlib_getproc(CoreLibHandle, "ConfigGetParamBool"); ConfigGetParamString = (ptr_ConfigGetParamString) osal_dynlib_getproc(CoreLibHandle, "ConfigGetParamString"); if (!ConfigOpenSection || !ConfigDeleteSection || !ConfigSetParameter || !ConfigGetParameter || !ConfigSetDefaultInt || !ConfigSetDefaultFloat || !ConfigSetDefaultBool || !ConfigSetDefaultString || !ConfigGetParamInt || !ConfigGetParamFloat || !ConfigGetParamBool || !ConfigGetParamString) return M64ERR_INCOMPATIBLE; /* Get core DoCommand function */ CoreDoCommand = (ptr_CoreDoCommand) osal_dynlib_getproc(CoreLibHandle, "CoreDoCommand"); if (!CoreDoCommand) { return M64ERR_INCOMPATIBLE; } /* get a configuration section handle */ if (ConfigOpenSection(RSP_HLE_CONFIG_SECTION, &l_ConfigRspHle) != M64ERR_SUCCESS) { HleErrorMessage(NULL, "Couldn't open config section '" RSP_HLE_CONFIG_SECTION "'"); return M64ERR_INPUT_NOT_FOUND; } /* check the section version number */ if (ConfigGetParameter(l_ConfigRspHle, RSP_HLE_CONFIG_VERSION, M64TYPE_FLOAT, &fConfigParamsVersion, sizeof(float)) != M64ERR_SUCCESS) { HleWarnMessage(NULL, "No version number in '" RSP_HLE_CONFIG_SECTION "' config section. Setting defaults."); ConfigDeleteSection(RSP_HLE_CONFIG_SECTION); ConfigOpenSection(RSP_HLE_CONFIG_SECTION, &l_ConfigRspHle); } else if (((int) fConfigParamsVersion) != ((int) CONFIG_PARAM_VERSION)) { HleWarnMessage(NULL, "Incompatible version %.2f in '" RSP_HLE_CONFIG_SECTION "' config section: current is %.2f. Setting defaults.", fConfigParamsVersion, (float) CONFIG_PARAM_VERSION); ConfigDeleteSection(RSP_HLE_CONFIG_SECTION); ConfigOpenSection(RSP_HLE_CONFIG_SECTION, &l_ConfigRspHle); } else if ((CONFIG_PARAM_VERSION - fConfigParamsVersion) >= 0.0001f) { /* handle upgrades */ float fVersion = CONFIG_PARAM_VERSION; ConfigSetParameter(l_ConfigRspHle, "Version", M64TYPE_FLOAT, &fVersion); HleInfoMessage(NULL, "Updating parameter set version in '" RSP_HLE_CONFIG_SECTION "' config section to %.2f", fVersion); } /* set the default values for this plugin */ ConfigSetDefaultFloat(l_ConfigRspHle, RSP_HLE_CONFIG_VERSION, CONFIG_PARAM_VERSION, "Mupen64Plus RSP HLE Plugin config parameter version number"); ConfigSetDefaultString(l_ConfigRspHle, RSP_HLE_CONFIG_FALLBACK, "", "Path to a RSP plugin which will be used when encountering an unknown ucode." "You can disable this by letting an empty string."); ConfigSetDefaultBool(l_ConfigRspHle, RSP_HLE_CONFIG_HLE_GFX, 1, "Send display lists to the graphics plugin"); ConfigSetDefaultBool(l_ConfigRspHle, RSP_HLE_CONFIG_HLE_AUD, 0, "Send audio lists to the audio plugin"); l_CoreHandle = CoreLibHandle; l_PluginInit = 1; return M64ERR_SUCCESS; } EXPORT m64p_error CALL PluginShutdown(void) { if (!l_PluginInit) return M64ERR_NOT_INIT; /* reset some local variable */ l_DebugCallback = NULL; l_DebugCallContext = NULL; l_CoreHandle = NULL; teardown_rsp_fallback(); l_PluginInit = 0; return M64ERR_SUCCESS; } EXPORT m64p_error CALL PluginGetVersion(m64p_plugin_type *PluginType, int *PluginVersion, int *APIVersion, const char **PluginNamePtr, int *Capabilities) { /* set version info */ if (PluginType != NULL) *PluginType = M64PLUGIN_RSP; if (PluginVersion != NULL) *PluginVersion = RSP_HLE_VERSION; if (APIVersion != NULL) *APIVersion = RSP_PLUGIN_API_VERSION; if (PluginNamePtr != NULL) *PluginNamePtr = "Hacktarux/Azimer High-Level Emulation RSP Plugin"; if (Capabilities != NULL) *Capabilities = 0; return M64ERR_SUCCESS; } EXPORT unsigned int CALL DoRspCycles(unsigned int Cycles) { hle_execute(&g_hle); return Cycles; } EXPORT void CALL InitiateRSP(RSP_INFO Rsp_Info, unsigned int* CycleCount) { hle_init(&g_hle, Rsp_Info.RDRAM, Rsp_Info.DMEM, Rsp_Info.IMEM, Rsp_Info.MI_INTR_REG, Rsp_Info.SP_MEM_ADDR_REG, Rsp_Info.SP_DRAM_ADDR_REG, Rsp_Info.SP_RD_LEN_REG, Rsp_Info.SP_WR_LEN_REG, Rsp_Info.SP_STATUS_REG, Rsp_Info.SP_DMA_FULL_REG, Rsp_Info.SP_DMA_BUSY_REG, Rsp_Info.SP_PC_REG, Rsp_Info.SP_SEMAPHORE_REG, Rsp_Info.DPC_START_REG, Rsp_Info.DPC_END_REG, Rsp_Info.DPC_CURRENT_REG, Rsp_Info.DPC_STATUS_REG, Rsp_Info.DPC_CLOCK_REG, Rsp_Info.DPC_BUFBUSY_REG, Rsp_Info.DPC_PIPEBUSY_REG, Rsp_Info.DPC_TMEM_REG, NULL); l_CheckInterrupts = Rsp_Info.CheckInterrupts; l_ProcessDlistList = Rsp_Info.ProcessDlistList; l_ProcessAlistList = Rsp_Info.ProcessAlistList; l_ProcessRdpList = Rsp_Info.ProcessRdpList; l_ShowCFB = Rsp_Info.ShowCFB; setup_rsp_fallback(ConfigGetParamString(l_ConfigRspHle, RSP_HLE_CONFIG_FALLBACK)); g_hle.hle_gfx = ConfigGetParamBool(l_ConfigRspHle, RSP_HLE_CONFIG_HLE_GFX); g_hle.hle_aud = ConfigGetParamBool(l_ConfigRspHle, RSP_HLE_CONFIG_HLE_AUD); /* notify fallback plugin */ if (l_InitiateRSP) { l_InitiateRSP(Rsp_Info, CycleCount); } } EXPORT void CALL RomClosed(void) { g_hle.cached_ucodes.count = 0; /* notify fallback plugin */ if (l_RomClosed) { l_RomClosed(); } } mupen64plus-rsp-hle-src-2.6.0/src/re2.c000066400000000000000000000203571464506742100175210ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - re2.c * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2016 Gilles Siberlin * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #include #include #include #include "hle_external.h" #include "hle_internal.h" #include "memory.h" #define SATURATE8(x) ((unsigned int) x <= 255 ? x : (x < 0 ? 0: 255)) /************************************************************************** * Resident evil 2 ucodes **************************************************************************/ void resize_bilinear_task(struct hle_t* hle) { int data_ptr = *dmem_u32(hle, TASK_UCODE_DATA); int src_addr = *dram_u32(hle, data_ptr); int dst_addr = *dram_u32(hle, data_ptr + 4); int dst_width = *dram_u32(hle, data_ptr + 8); int dst_height = *dram_u32(hle, data_ptr + 12); int x_ratio = *dram_u32(hle, data_ptr + 16); int y_ratio = *dram_u32(hle, data_ptr + 20); #if 0 /* unused, but keep it for documentation purpose */ int dst_stride = *dram_u32(hle, data_ptr + 24); #endif int src_offset = *dram_u32(hle, data_ptr + 36); int a, b, c ,d, index, y_index, xr, yr, blue, green, red, addr, i, j; long long x, y, x_diff, y_diff, one_min_x_diff, one_min_y_diff; unsigned short pixel; src_addr += (src_offset >> 16) * (320 * 3); x = y = 0; for(i = 0; i < dst_height; i++) { yr = (int)(y >> 16); y_diff = y - (yr << 16); one_min_y_diff = 65536 - y_diff; y_index = yr * 320; x = 0; for(j = 0; j < dst_width; j++) { xr = (int)(x >> 16); x_diff = x - (xr << 16); one_min_x_diff = 65536 - x_diff; index = y_index + xr; addr = src_addr + (index * 3); dram_load_u8(hle, (uint8_t*)&a, addr, 3); dram_load_u8(hle, (uint8_t*)&b, (addr + 3), 3); dram_load_u8(hle, (uint8_t*)&c, (addr + (320 * 3)), 3); dram_load_u8(hle, (uint8_t*)&d, (addr + (320 * 3) + 3), 3); blue = (int)(((a&0xff)*one_min_x_diff*one_min_y_diff + (b&0xff)*x_diff*one_min_y_diff + (c&0xff)*y_diff*one_min_x_diff + (d&0xff)*x_diff*y_diff) >> 32); green = (int)((((a>>8)&0xff)*one_min_x_diff*one_min_y_diff + ((b>>8)&0xff)*x_diff*one_min_y_diff + ((c>>8)&0xff)*y_diff*one_min_x_diff + ((d>>8)&0xff)*x_diff*y_diff) >> 32); red = (int)((((a>>16)&0xff)*one_min_x_diff*one_min_y_diff + ((b>>16)&0xff)*x_diff*one_min_y_diff + ((c>>16)&0xff)*y_diff*one_min_x_diff + ((d>>16)&0xff)*x_diff*y_diff) >> 32); blue = (blue >> 3) & 0x001f; green = (green >> 3) & 0x001f; red = (red >> 3) & 0x001f; pixel = (red << 11) | (green << 6) | (blue << 1) | 1; dram_store_u16(hle, &pixel, dst_addr, 1); dst_addr += 2; x += x_ratio; } y += y_ratio; } rsp_break(hle, SP_STATUS_TASKDONE); } static uint32_t YCbCr_to_RGBA(uint8_t Y, uint8_t Cb, uint8_t Cr) { int r, g, b; r = (int)(((double)Y * 0.582199097) + (0.701004028 * (double)(Cr - 128))); g = (int)(((double)Y * 0.582199097) - (0.357070923 * (double)(Cr - 128)) - (0.172073364 * (double)(Cb - 128))); b = (int)(((double)Y * 0.582199097) + (0.886001587 * (double)(Cb - 128))); r = SATURATE8(r); g = SATURATE8(g); b = SATURATE8(b); return (r << 24) | (g << 16) | (b << 8) | 0; } void decode_video_frame_task(struct hle_t* hle) { int data_ptr = *dmem_u32(hle, TASK_UCODE_DATA); int pLuminance = *dram_u32(hle, data_ptr); int pCb = *dram_u32(hle, data_ptr + 4); int pCr = *dram_u32(hle, data_ptr + 8); int pDestination = *dram_u32(hle, data_ptr + 12); int nMovieWidth = *dram_u32(hle, data_ptr + 16); int nMovieHeight = *dram_u32(hle, data_ptr + 20); #if 0 /* unused, but keep it for documentation purpose */ int nRowsPerDMEM = *dram_u32(hle, data_ptr + 24); int nDMEMPerFrame = *dram_u32(hle, data_ptr + 28); int nLengthSkipCount = *dram_u32(hle, data_ptr + 32); #endif int nScreenDMAIncrement = *dram_u32(hle, data_ptr + 36); int i, j; uint8_t Y, Cb, Cr; uint32_t pixel; int pY_1st_row, pY_2nd_row, pDest_1st_row, pDest_2nd_row; for (i = 0; i < nMovieHeight; i += 2) { pY_1st_row = pLuminance; pY_2nd_row = pLuminance + nMovieWidth; pDest_1st_row = pDestination; pDest_2nd_row = pDestination + (nScreenDMAIncrement >> 1); for (j = 0; j < nMovieWidth; j += 2) { dram_load_u8(hle, (uint8_t*)&Cb, pCb++, 1); dram_load_u8(hle, (uint8_t*)&Cr, pCr++, 1); /*1st row*/ dram_load_u8(hle, (uint8_t*)&Y, pY_1st_row++, 1); pixel = YCbCr_to_RGBA(Y, Cb, Cr); dram_store_u32(hle, &pixel, pDest_1st_row, 1); pDest_1st_row += 4; dram_load_u8(hle, (uint8_t*)&Y, pY_1st_row++, 1); pixel = YCbCr_to_RGBA(Y, Cb, Cr); dram_store_u32(hle, &pixel, pDest_1st_row, 1); pDest_1st_row += 4; /*2nd row*/ dram_load_u8(hle, (uint8_t*)&Y, pY_2nd_row++, 1); pixel = YCbCr_to_RGBA(Y, Cb, Cr); dram_store_u32(hle, &pixel, pDest_2nd_row, 1); pDest_2nd_row += 4; dram_load_u8(hle, (uint8_t*)&Y, pY_2nd_row++, 1); pixel = YCbCr_to_RGBA(Y, Cb, Cr); dram_store_u32(hle, &pixel, pDest_2nd_row, 1); pDest_2nd_row += 4; } pLuminance += (nMovieWidth << 1); pDestination += nScreenDMAIncrement; } rsp_break(hle, SP_STATUS_TASKDONE); } void fill_video_double_buffer_task(struct hle_t* hle) { int data_ptr = *dmem_u32(hle, TASK_UCODE_DATA); int pSrc = *dram_u32(hle, data_ptr); int pDest = *dram_u32(hle, data_ptr + 0x4); int width = *dram_u32(hle, data_ptr + 0x8) >> 1; int height = *dram_u32(hle, data_ptr + 0x10) << 1; int stride = *dram_u32(hle, data_ptr + 0x1c) >> 1; assert((*dram_u32(hle, data_ptr + 0x28) >> 16) == 0x8000); #if 0 /* unused, but keep it for documentation purpose */ int arg3 = *dram_u32(hle, data_ptr + 0xc); int arg5 = *dram_u32(hle, data_ptr + 0x14); int arg6 = *dram_u32(hle, data_ptr + 0x18); #endif int i, j; int r, g, b; uint32_t pixel, pixel1, pixel2; for(i = 0; i < height; i++) { for(j = 0; j < width; j=j+4) { pixel1 = *dram_u32(hle, pSrc+j); pixel2 = *dram_u32(hle, pDest+j); r = (((pixel1 >> 24) & 0xff) + ((pixel2 >> 24) & 0xff)) >> 1; g = (((pixel1 >> 16) & 0xff) + ((pixel2 >> 16) & 0xff)) >> 1; b = (((pixel1 >> 8) & 0xff) + ((pixel2 >> 8) & 0xff)) >> 1; pixel = (r << 24) | (g << 16) | (b << 8) | 0; dram_store_u32(hle, &pixel, pDest+j, 1); } pSrc += stride; pDest += stride; } rsp_break(hle, SP_STATUS_TASKDONE); } mupen64plus-rsp-hle-src-2.6.0/src/rsp_api_export.ver000066400000000000000000000001551464506742100224330ustar00rootroot00000000000000{ global: PluginStartup; PluginShutdown; PluginGetVersion; DoRspCycles; InitiateRSP; RomClosed; local: *; }; mupen64plus-rsp-hle-src-2.6.0/src/ucodes.h000066400000000000000000000117361464506742100203210ustar00rootroot00000000000000/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Mupen64plus-rsp-hle - ucodes.h * * Mupen64Plus homepage: https://mupen64plus.org/ * * Copyright (C) 2014 Bobby Smiles * * * * This program is free software; you can redistribute it and/or modify * * it under the terms of the GNU General Public License as published by * * the Free Software Foundation; either version 2 of the License, or * * (at your option) any later version. * * * * This program is distributed in the hope that it will be useful, * * but WITHOUT ANY WARRANTY; without even the implied warranty of * * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * * GNU General Public License for more details. * * * * You should have received a copy of the GNU General Public License * * along with this program; if not, write to the * * Free Software Foundation, Inc., * * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ #ifndef UCODES_H #define UCODES_H #include #define CACHED_UCODES_MAX_SIZE 16 struct hle_t; typedef void(*ucode_func_t)(struct hle_t* hle); struct ucode_info_t { uint32_t uc_start; uint32_t uc_dstart; uint16_t uc_dsize; ucode_func_t uc_pfunc; }; struct cached_ucodes_t { struct ucode_info_t infos[CACHED_UCODES_MAX_SIZE]; int count; }; /* cic_x105 ucode */ void cicx105_ucode(struct hle_t* hle); /* audio list ucodes - audio */ enum { N_SEGMENTS = 16 }; struct alist_audio_t { /* segments */ uint32_t segments[N_SEGMENTS]; /* main buffers */ uint16_t in; uint16_t out; uint16_t count; /* auxiliary buffers */ uint16_t dry_right; uint16_t wet_left; uint16_t wet_right; /* gains */ int16_t dry; int16_t wet; /* envelopes (0:left, 1:right) */ int16_t vol[2]; int16_t target[2]; int32_t rate[2]; /* ADPCM loop point address */ uint32_t loop; /* storage for ADPCM table and polef coefficients */ int16_t table[16 * 8]; }; void alist_process_audio (struct hle_t* hle); void alist_process_audio_ge(struct hle_t* hle); void alist_process_audio_bc(struct hle_t* hle); /* audio list ucodes - naudio */ struct alist_naudio_t { /* gains */ int16_t dry; int16_t wet; /* envelopes (0:left, 1:right) */ int16_t vol[2]; int16_t target[2]; int32_t rate[2]; /* ADPCM loop point address */ uint32_t loop; /* storage for ADPCM table and polef coefficients */ int16_t table[16 * 8]; }; void alist_process_naudio (struct hle_t* hle); void alist_process_naudio_bk (struct hle_t* hle); void alist_process_naudio_dk (struct hle_t* hle); void alist_process_naudio_mp3 (struct hle_t* hle); void alist_process_naudio_cbfd(struct hle_t* hle); /* audio list ucodes - nead */ struct alist_nead_t { /* main buffers */ uint16_t in; uint16_t out; uint16_t count; /* envmixer ramps */ uint16_t env_values[3]; uint16_t env_steps[3]; /* ADPCM loop point address */ uint32_t loop; /* storage for ADPCM table and polef coefficients */ int16_t table[16 * 8]; /* filter audio command state */ uint16_t filter_count; uint32_t filter_lut_address[2]; }; void alist_process_nead_mk (struct hle_t* hle); void alist_process_nead_sfj (struct hle_t* hle); void alist_process_nead_sf (struct hle_t* hle); void alist_process_nead_fz (struct hle_t* hle); void alist_process_nead_wrjb(struct hle_t* hle); void alist_process_nead_ys (struct hle_t* hle); void alist_process_nead_1080(struct hle_t* hle); void alist_process_nead_oot (struct hle_t* hle); void alist_process_nead_mm (struct hle_t* hle); void alist_process_nead_mmb (struct hle_t* hle); void alist_process_nead_ac (struct hle_t* hle); void alist_process_nead_mats(struct hle_t* hle); void alist_process_nead_efz (struct hle_t* hle); /* mp3 ucode */ void mp3_task(struct hle_t* hle, unsigned int index, uint32_t address); /* musyx ucodes */ void musyx_v1_task(struct hle_t* hle); void musyx_v2_task(struct hle_t* hle); /* jpeg ucodes */ void jpeg_decode_PS0(struct hle_t* hle); void jpeg_decode_PS(struct hle_t* hle); void jpeg_decode_OB(struct hle_t* hle); /* Resident evil 2 ucode */ void resize_bilinear_task(struct hle_t* hle); void decode_video_frame_task(struct hle_t* hle); void fill_video_double_buffer_task(struct hle_t* hle); /* hvqm2 ucode */ void hvqm2_decode_sp1_task(struct hle_t* hle); void hvqm2_decode_sp2_task(struct hle_t* hle); #endif