From c33dac5a0dee4e97416bdafdd27692363afc0d08 Mon Sep 17 00:00:00 2001 From: iximeow Date: Sat, 3 Dec 2022 14:53:40 -0800 Subject: roll up decoding loop changes for 16-bit and 32-bit decoders this applies * f338c74656f6eef8b3080fa9f249b1cb733fd1a9 * bece19e6a69b158893abbf56a6cac25eb25d9a32 * 6353f58170d28a142e3b012c2c86f684d50dea45 * 67be1c0983244645a3c762b7aa0601f0d0ba4bb3 * 091f1d66ef853d6339a96e43d71c137ee7d3907a as one unit to both the 16-bit and 32-bit decoders. --- src/protected_mode/mod.rs | 330 ++++++++++++++++++++++++++++------------------ 1 file changed, 199 insertions(+), 131 deletions(-) (limited to 'src/protected_mode') diff --git a/src/protected_mode/mod.rs b/src/protected_mode/mod.rs index e95691d..2993968 100644 --- a/src/protected_mode/mod.rs +++ b/src/protected_mode/mod.rs @@ -5546,8 +5546,8 @@ pub(self) fn read_E< T: Reader<::Address, ::Word>, S: DescriptionSink >(words: &mut T, instr: &mut Instruction, modrm: u8, width: u8, sink: &mut S) -> Result { - let bank = width_to_gp_reg_bank(width); if modrm >= 0b11000000 { + let bank = width_to_gp_reg_bank(width); read_modrm_reg(instr, modrm, bank) } else { read_M(words, instr, modrm, sink) @@ -6433,6 +6433,77 @@ impl fmt::Display for FieldDescription { } } +fn read_opc_hotpath< + T: Reader<::Address, ::Word>, + S: DescriptionSink, +>(mut b: u8, mut record: OpcodeRecord, nextb: &mut u8, next_rec: &mut OpcodeRecord, words: &mut T, instruction: &mut Instruction, sink: &mut S) -> Result, DecodeError> { + if b == 0x66 { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::Misc("operand size override (to 16 bits)"), + id: words.offset() as u32 * 8 - 8, + }); + b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + record = unsafe { + core::ptr::read_volatile(&OPCODES[b as usize]) + }; + instruction.prefixes.set_operand_size(); + } + + if let Interpretation::Instruction(opc) = record.0 { + if words.offset() > 1 { + sink.record( + words.offset() as u32 * 8 - 8 - 1, words.offset() as u32 * 8 - 8 - 1, + InnerDescription::Boundary("prefixes end") + .with_id(words.offset() as u32 * 8 - 9) + ); + } + if opc != Opcode::Invalid { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::Opcode(opc), + id: words.offset() as u32 * 8 - 8, + }); + } + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::OperandCode(OperandCodeWrapper { code: record.1 }), + id: words.offset() as u32 * 8 - 8 + 1, + }); + instruction.mem_size = 0; + instruction.operand_count = 2; + instruction.opcode = opc; + return Ok(Some(record.1)); + } else if b == 0x0f { + if words.offset() > 1 { + sink.record( + words.offset() as u32 * 8 - 8 - 1, words.offset() as u32 * 8 - 8 - 1, + InnerDescription::Boundary("prefixes end") + .with_id(words.offset() as u32 * 8 - 9) + ); + } + let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + instruction.mem_size = 0; + instruction.operand_count = 2; + let record = if b == 0x38 { + let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + read_0f38_opcode(b, &mut instruction.prefixes) + } else if b == 0x3a { + let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + read_0f3a_opcode(b, &mut instruction.prefixes) + } else { + read_0f_opcode(b, &mut instruction.prefixes) + }; + if let Interpretation::Instruction(opc) = record.0 { + instruction.opcode = opc; + } else { + unsafe { unreachable_unchecked(); } + } + return Ok(Some(record.1)); + } else { + *nextb = b; + *next_rec = record; + return Ok(None); + } +} + fn read_with_annotations< T: Reader<::Address, ::Word>, S: DescriptionSink, @@ -6440,43 +6511,21 @@ fn read_with_annotations< words.mark(); let mut nextb = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; let mut next_rec = OPCODES[nextb as usize]; - let mut prefixes = Prefixes::new(0); + instruction.prefixes = Prefixes::new(0); // default registers to `[eax; 4]` instruction.regs = unsafe { core::mem::transmute(0u64) }; - instruction.mem_size = 0; // default operands to [RegRRR, Nothing, Nothing, Nothing] instruction.operands = unsafe { core::mem::transmute(0x00_00_00_01) }; - instruction.operand_count = 2; - let record: OpcodeRecord = loop { - let record = next_rec; - if let Interpretation::Instruction(opc) = record.0 { - if words.offset() > 1 { - sink.record( - words.offset() as u32 * 8 - 8 - 1, words.offset() as u32 * 8 - 8 - 1, - InnerDescription::Boundary("prefixes end") - .with_id(words.offset() as u32 * 8 - 9) - ); - } - if opc != Opcode::Invalid { - sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::Opcode(opc), - id: words.offset() as u32 * 8 - 8, - }); - } - sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::OperandCode(OperandCodeWrapper { code: record.1 }), - id: words.offset() as u32 * 8 - 8 + 1, - }); - break record; - } else { - let b = nextb; - if words.offset() >= 15 { - return Err(DecodeError::TooLong); - } - if b == 0x0f { + let record: OperandCode = if let Some(rec) = read_opc_hotpath(nextb, next_rec, &mut nextb, &mut next_rec, words, instruction, sink)? { + rec + } else { + let prefixes = &mut instruction.prefixes; + let record = loop { + let record = next_rec; + if let Interpretation::Instruction(opc) = record.0 { if words.offset() > 1 { sink.record( words.offset() as u32 * 8 - 8 - 1, words.offset() as u32 * 8 - 8 - 1, @@ -6484,118 +6533,142 @@ fn read_with_annotations< .with_id(words.offset() as u32 * 8 - 9) ); } - let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; - if b == 0x38 { - let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; - break read_0f38_opcode(b, &mut prefixes); - } else if b == 0x3a { + if opc != Opcode::Invalid { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::Opcode(opc), + id: words.offset() as u32 * 8 - 8, + }); + } + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::OperandCode(OperandCodeWrapper { code: record.1 }), + id: words.offset() as u32 * 8 - 8 + 1, + }); + instruction.mem_size = 0; + instruction.operand_count = 2; + break record; + } else { + let b = nextb; + if b == 0x0f { + if words.offset() > 1 { + sink.record( + words.offset() as u32 * 8 - 8 - 1, words.offset() as u32 * 8 - 8 - 1, + InnerDescription::Boundary("prefixes end") + .with_id(words.offset() as u32 * 8 - 9) + ); + } let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; - break read_0f3a_opcode(b, &mut prefixes); - } else { - break read_0f_opcode(b, &mut prefixes); + instruction.mem_size = 0; + instruction.operand_count = 2; + if b == 0x38 { + let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + break read_0f38_opcode(b, prefixes); + } else if b == 0x3a { + let b = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + break read_0f3a_opcode(b, prefixes); + } else { + break read_0f_opcode(b, prefixes); + } } - } - - nextb = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; - next_rec = unsafe { - core::ptr::read_volatile(&OPCODES[nextb as usize]) - }; - match b { - 0x26 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::ES), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_es(); - }, - 0x2e => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::CS), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_cs(); - }, - 0x36 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::SS), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_ss(); - }, - 0x3e => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::DS), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_ds(); - }, - 0x64 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::FS), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_fs(); - }, - 0x65 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::SegmentPrefix(Segment::GS), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_gs(); - }, - 0x66 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { + if b == 0x66 { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { desc: InnerDescription::Misc("operand size override (to 16 bits)"), - id: words.offset() as u32 * 8 - 16, + id: words.offset() as u32 * 8 - 8, }); prefixes.set_operand_size(); - }, - 0x67 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { + } else if b == 0x67 { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { desc: InnerDescription::Misc("address size override (to 16 bits)"), - id: words.offset() as u32 * 8 - 16, + id: words.offset() as u32 * 8 - 8, }); prefixes.set_address_size(); - }, - 0xf0 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { - desc: InnerDescription::Misc("lock prefix"), - id: words.offset() as u32 * 8 - 16, - }); - prefixes.set_lock(); - }, - 0xf2 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { + } else if b == 0xf2 { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { desc: InnerDescription::Misc("repnz prefix"), - id: words.offset() as u32 * 8 - 16, + id: words.offset() as u32 * 8 - 8, }); prefixes.set_repnz(); - }, - 0xf3 => { - sink.record((words.offset() - 2) as u32 * 8, (words.offset() - 2) as u32 * 8 + 7, FieldDescription { + } else if b == 0xf3 { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { desc: InnerDescription::Misc("rep prefix"), - id: words.offset() as u32 * 8 - 16, + id: words.offset() as u32 * 8 - 8, }); prefixes.set_rep(); - }, - _ => { unsafe { unreachable_unchecked(); } } + } else { + match b { + 0x26 => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::ES), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_es(); + }, + 0x2e => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::CS), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_cs(); + }, + 0x36 => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::SS), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_ss(); + }, + 0x3e => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::DS), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_ds(); + }, + 0x64 => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::FS), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_fs(); + }, + 0x65 => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::SegmentPrefix(Segment::GS), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_gs(); + }, + 0xf0 => { + sink.record((words.offset() - 1) as u32 * 8, (words.offset() - 1) as u32 * 8 + 7, FieldDescription { + desc: InnerDescription::Misc("lock prefix"), + id: words.offset() as u32 * 8 - 8, + }); + prefixes.set_lock(); + }, + // unlike 64-bit mode, the vex/evex prefixes are not recorded as prefixes - + // they are LES/LDS/BOUND with special-case operand decoding. so we've + // actually handled all possible prefixes at this point. + _ => { unsafe { unreachable_unchecked(); } } + } + } + nextb = words.next().ok().ok_or(DecodeError::ExhaustedInput)?; + next_rec = unsafe { + core::ptr::read_volatile(&OPCODES[nextb as usize]) + }; + } + if words.offset() >= 15 { + return Err(DecodeError::TooLong); } + }; + + if let Interpretation::Instruction(opcode) = record.0 { + instruction.opcode = opcode; + } else { + unsafe { unreachable_unchecked(); } } + + record.1 }; - if record == OpcodeRecord(Interpretation::Instruction(Opcode::Invalid), OperandCode::Nothing) { - return Err(DecodeError::InvalidOpcode); - } - if let Interpretation::Instruction(opcode) = record.0 { - instruction.opcode = opcode; - } else { - unsafe { unreachable_unchecked(); } - } - instruction.prefixes = prefixes; - read_operands(decoder, words, instruction, record.1, sink)?; - instruction.length = words.offset() as u8; - if instruction.length > 15 { - return Err(DecodeError::TooLong); - } + read_operands(decoder, words, instruction, record, sink)?; if instruction.prefixes.lock() { if !LOCKABLE_INSTRUCTIONS.contains(&instruction.opcode) || !instruction.operands[0].is_memory() { @@ -6603,11 +6676,6 @@ fn read_with_annotations< } } - if decoder != &InstDecoder::default() { - // we might have to fix up or reject this instruction under whatever cpu features we need to - // pretend to have. - decoder.revise_instruction(instruction)?; - } Ok(()) } /* likely cases -- cgit v1.1