-
Notifications
You must be signed in to change notification settings - Fork 17
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add LoongArch support #47
Conversation
The MCInstDesc table changed. Bsides this only minor changes were done and some additional code is emitted now for LLVM. This commit is the combination of all previous Auto-Sync commits. The list of commit messages follows: ----------- Combination of all commits of the refactored tablegen backends. These are the changes made for LLVM 16. Refactor Capstone relevant TableGen Emitter backends. This commit extracts the code which emits generated tables into two printer classes. The Printer is called whenever actual code is written to a file. There is the PrinterLLVM which emits tht code as before and PrinterCapstone which is tailored to or needs (emitting C and generates more info). Additionally missing memory access properties were added to ARMs td files. Emit a single header for all files. Captialize Target name for enums. Add lay metric to emit enum value for Banked and system regs. Malloc substr Sort instructions in ascending order. Free substr after use Add vanished constrainsts Fix `regInfoEmitEnums()` and indent Fix `GenDisassemblerTables.inc#checkDecoderPredicate()` Fix `TriCoreGenRegisterInfo.inc` | `PrinterCapstone::regInfoEmitRegClasses` revert changes to NEON instructions Add instructions with duplicate operands as Matchables. Add memory load and store info Correct memory access and out operand info Set register lists again as read ops due to llvm/llvm-project#62455 Make printAliasInstr and getMnemonic static. Generate CS instruction enums from actual mnemonic. Not via the flawed AsmMatcher. Fix typo in InstrInfoEmitter.cpp Add deprecated QPX feature Replace + and - with p and m Add AssemblerPredicates to PPC Generate RegEncodingTable Define functions which are called by the Mapper as static. Necessary because these functions are present in each arch' Remove set_mem_access(). The cases where this is used to mark access to actual memory operands are either very rare, or those are neon lane indicies. Generate correct op type for absolute addresses. Check for RegisterPointer operands first to prevent mis-categorization. Add missing Operand types Generate Instruction formats for PPC. Add Paired Single instructions. Partly revert 94e41ce (introduces accidentially removed code.) Set correct operand types for PS operands Add memory read/write attributes Add missing operand types Add mayLoad and mayStore information. Add documentation. Handle special AArch64 operand Replace C++ with C code. Check for duplicate enum instr. names Check for duplicate defintions of system registers. Add note about missing target names. Resolve templates in a single static method and add docs about it. Revert printing target name in upper case. Revert partially C++ syntax fixes in .td files. They break the TemplateCOllector since it searches for exactly those references but can't find any' Add all SubtargetFeatures to feature enum. Not just the one used by CGIs. Pass Decoder Enable to check specific table fields to determine if reg enum must be emitted. Allow to add namespace to type name/ Formatting Rework emitting of tables. The system operands are now emitted in reg, imm and aliass groups. Also a bug was fixed which emitted incorrect code.. Check for rename IMPLICIT_IMM operand types Pass DecodeComplete as pointer not as reference Print undef when it needs to be printed. Add namespace ids to all types and functions. Rework C translation. Pass MCOp as pointer not as ref Add missing SysImm type Fix syntax mistakes Generate additonal sys immediates and op groups. Handle edge case for printSVERegOp Handle default arguments of template functions. Add two missing op groups Generate a static RecEncodingTable Set enum values to encodings of the sys ops Generate a single Enum value file for system operands. Replace System operand groups with their operand types Fix missing braces warning Emit MCOperand validator. Emit lookupByName functions for sys operands Add namespaces for ARM. Check for Target if default arguments of template functions are resolved. auto-sync opcode & operand encoding info generation (capstone-engine#14) * Added operand and opcode info generation * Wrapped deprecated macro under an intellisense check Basically intellisense fails, causing multiple errors in other files, so when intellisense parses the code it will use the different version of the macro * Fixed a small bug Used double braces to prevent an old bug Removed extra new line and fixed a bug regarding move semantics
…thub action for branch checkout.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please, before you continue with this rebase it on top of the LLVM 18 branch from #45. There are many bug fixes and it is based on LLVM-18.
Now to the changes:
- Before you proceed, please make sure you have read the architecture about the
auto-sync
updater. I would advice you to read this blog entry first and afterward check out the docs about: the general architecture, the auto-sync updater and lastly the cpp-translator. Last but not least I highly recommend to step with a debugger through a decoding of an ARM, PPC or AArch64 instruction. This way you can see in practice how it is done. - Please do not make any changes to the LLVM files here. Exceptions are
TableGen
related files. But any target related source code should stay the same. The reason is, that it is a lot and very very annoying work to rebase our patched LLVM code onto the newest release. We need to have as little changes in the original LLVM source code as possible. This is why all our editing of theLoongArchDisassembler
etc files, all happen in the auto-sync scripts and not here. Because the scritps automate a lot of them and keep the changes in this repo small.
You are one of the first users of the auto-sync updater. So I would love to improve the first steps for people like you. If you could tell me where you got confused I would like to know it, so I can make the "new-user" steps easier.
llvm/lib/Target/LoongArch/Disassembler/LoongArchDisassembler.cpp
Outdated
Show resolved
Hide resolved
llvm/lib/Target/LoongArch/Disassembler/LoongArchDisassembler.cpp
Outdated
Show resolved
Hide resolved
llvm/lib/Target/LoongArch/Disassembler/LoongArchDisassembler.cpp
Outdated
Show resolved
Hide resolved
llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchInstPrinter.cpp
Outdated
Show resolved
Hide resolved
llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchInstPrinter.cpp
Outdated
Show resolved
Hide resolved
llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchInstPrinter.cpp
Outdated
Show resolved
Hide resolved
I have rebased on top of auto-sync-18 branch and dropped LoongArch-specific changes under |
Using capstone
|
Update: oh, I need to use auto-sync-18 from your repo, not capstone-engine. |
- Determine memory operands by there membership of patterns. Operands can be defined via patterns. The pattern has a type itself. Patterns are commonly used to define memory operands (as a collection of multiple other operands). Here we check, if the operand is part of a pattern with type iPTR. Type iPTR indicates a memory operand. - Search for unspecified instruction patterns and extract memory op info from them. AArch64 has the case of patterns which are not set in the CGI class. Instead those patterns are free floating in the record keeper. With this commit we first collect all such unassigned patterns and save them in a map (indexed by instruction name they belong to). When instruction operand details are generated, the operand type is searched in the pattern of the instruction. If there is an operand with the same type in the pattern, which is also of a iPTR pattern, we treat it as memory operand. Note that we can not compare the operands of the instruction and the pattern operands by name. Because they simply are not named the same way. This is a short coming in the td files and should be addressed in LLVM. Because it is enormiously hard to get a 1 to 1 mapping from a pattern to an instruction and its operands.
Merged with #45 |
See capstone-engine/capstone#2349