Skip to contentSkip to conversation

QEMU AVR 8 bit cores

PATCH Superseded Michael Rolnik 1 participant 14 messages Oct 25, 2019

v34 of 41 Superseded Latest is v41 →

lore
Share & tools
lore
#3 lore raw

PATCH v34 3/13 target/avr: Add instruction decoding

In reply to Michael Rolnik

This includes:
- encoding of all 16 bit instructions
- encoding of all 32 bit instructions

---
 target/avr/insn.decode | 175 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 175 insertions(+)
 create mode 100644 target/avr/insn.decode
target/avr/insn.decode new file +175 -0
new file mode 100644
index 0000000000..6b387762c6
@@ -0,0 +1,175 @@
+#
+#   A = [16 .. 31]
+#   B = [16 .. 23]
+#   C = [24, 26, 28, 30]
+#   D = [0, 2, 4, 6, 8, .. 30]
+
+%rd             4:5
+%rr             9:1 0:4
+
+&rd_rr          rd rr
+&rd_imm         rd imm
+
+@op_rd_rr       .... .. . ..... ....        &rd_rr      rd=%rd rr=%rr
+ADD             0000 11 . ..... ....        @op_rd_rr
+ADC             0001 11 . ..... ....        @op_rd_rr
+AND             0010 00 . ..... ....        @op_rd_rr
+CP              0001 01 . ..... ....        @op_rd_rr
+CPC             0000 01 . ..... ....        @op_rd_rr
+CPSE            0001 00 . ..... ....        @op_rd_rr
+EOR             0010 01 . ..... ....        @op_rd_rr
+MOV             0010 11 . ..... ....        @op_rd_rr
+MUL             1001 11 . ..... ....        @op_rd_rr
+OR              0010 10 . ..... ....        @op_rd_rr
+SBC             0000 10 . ..... ....        @op_rd_rr
+SUB             0001 10 . ..... ....        @op_rd_rr
+
+
+%rd_c           4:2                         !function=to_C
+%imm6           6:2 0:4
+
+@op_rd_imm6     .... .... .. .. ....        &rd_imm     rd=%rd_c imm=%imm6
+ADIW            1001 0110 .. .. ....        @op_rd_imm6
+SBIW            1001 0111 .. .. ....        @op_rd_imm6
+
+
+%rd_a           4:4                         !function=to_A
+%rr_a           0:4                         !function=to_A
+%rd_d           4:4                         !function=to_D
+%rr_d           0:4                         !function=to_D
+%imm8           8:4 0:4
+
+@op_rd_imm8     .... .... .... ....         &rd_imm     rd=%rd_a imm=%imm8
+ANDI            0111 .... .... ....         @op_rd_imm8
+CPI             0011 .... .... ....         @op_rd_imm8
+LDI             1110 .... .... ....         @op_rd_imm8
+ORI             0110 .... .... ....         @op_rd_imm8
+SBCI            0100 .... .... ....         @op_rd_imm8
+SUBI            0101 .... .... ....         @op_rd_imm8
+
+
+@op_rd          .... ... rd:5 ....
+ASR             1001 010 ..... 0101         @op_rd
+COM             1001 010 ..... 0000         @op_rd
+DEC             1001 010 ..... 1010         @op_rd
+ELPM2           1001 000 ..... 0110         @op_rd
+ELPMX           1001 000 ..... 0111         @op_rd
+INC             1001 010 ..... 0011         @op_rd
+LDX1            1001 000 ..... 1100         @op_rd
+LDX2            1001 000 ..... 1101         @op_rd
+LDX3            1001 000 ..... 1110         @op_rd
+LDY2            1001 000 ..... 1001         @op_rd
+LDY3            1001 000 ..... 1010         @op_rd
+LDZ2            1001 000 ..... 0001         @op_rd
+LDZ3            1001 000 ..... 0010         @op_rd
+LPM2            1001 000 ..... 0100         @op_rd
+LPMX            1001 000 ..... 0101         @op_rd
+LSR             1001 010 ..... 0110         @op_rd
+NEG             1001 010 ..... 0001         @op_rd
+POP             1001 000 ..... 1111         @op_rd
+PUSH            1001 001 ..... 1111         @op_rd
+ROR             1001 010 ..... 0111         @op_rd
+STY2            1001 001 ..... 1001         @op_rd
+STY3            1001 001 ..... 1010         @op_rd
+STZ2            1001 001 ..... 0001         @op_rd
+STZ3            1001 001 ..... 0010         @op_rd
+SWAP            1001 010 ..... 0010         @op_rd
+
+
+@op_bit         .... .... . bit:3 ....
+BCLR            1001 0100 1 ... 1000        @op_bit
+BSET            1001 0100 0 ... 1000        @op_bit
+
+
+@op_rd_bit      .... ... rd:5 . bit:3
+BLD             1111 100 ..... 0 ...        @op_rd_bit
+BST             1111 101 ..... 0 ...        @op_rd_bit
+
+
+@op_bit_imm     .... .. imm:s7 bit:3
+BRBC            1111 01 ....... ...         @op_bit_imm
+BRBS            1111 00 ....... ...         @op_bit_imm
+
+
+BREAK           1001 0101 1001 1000
+EICALL          1001 0101 0001 1001
+EIJMP           1001 0100 0001 1001
+ELPM1           1001 0101 1101 1000
+ICALL           1001 0101 0000 1001
+IJMP            1001 0100 0000 1001
+LPM1            1001 0101 1100 1000
+NOP             0000 0000 0000 0000
+RET             1001 0101 0000 1000
+RETI            1001 0101 0001 1000
+SLEEP           1001 0101 1000 1000
+SPM             1001 0101 1110 1000
+SPMX            1001 0101 1111 1000
+WDR             1001 0101 1010 1000
+
+
+@op_reg_bit     .... .... reg:5 bit:3
+CBI             1001 1000 ..... ...         @op_reg_bit
+SBI             1001 1010 ..... ...         @op_reg_bit
+SBIC            1001 1001 ..... ...         @op_reg_bit
+SBIS            1001 1011 ..... ...         @op_reg_bit
+
+
+DES             1001 0100 imm:4 1011
+
+
+%rd_b           4:3                         !function=to_B
+%rr_b           0:3                         !function=to_B
+@fmul           .... .... . ... . ...       &rd_rr      rd=%rd_b rr=%rr_b
+FMUL            0000 0011 0 ... 1 ...       @fmul
+FMULS           0000 0011 1 ... 0 ...       @fmul
+FMULSU          0000 0011 1 ... 1 ...       @fmul
+MULSU           0000 0011 0 ... 0 ...       @fmul
+
+
+%io_imm         9:2 0:4
+@io_rd_imm      .... . .. ..... ....        &rd_imm     rd=%rd imm=%io_imm
+IN              1011 0 .. ..... ....        @io_rd_imm
+OUT             1011 1 .. ..... ....        @io_rd_imm
+
+
+XCH             1001 001 rd:5 0100
+LAC             1001 001 rd:5 0110
+LAS             1001 001 rd:5 0101
+LAT             1001 001 rd:5 0111
+STX1            1001 001 rr:5 1100
+STX2            1001 001 rr:5 1101
+STX3            1001 001 rr:5 1110
+
+
+%ldst_d_imm     13:1 10:2 0:3
+@ldst_d         .. . . .. . rd:5  . ...     &rd_imm     imm=%ldst_d_imm
+LDDY            10 . 0 .. 0 ..... 1 ...     @ldst_d
+LDDZ            10 . 0 .. 0 ..... 0 ...     @ldst_d
+STDY            10 . 0 .. 1 ..... 1 ...     @ldst_d
+STDZ            10 . 0 .. 1 ..... 0 ...     @ldst_d
+
+
+MOVW            0000 0001 .... ....         &rd_rr      rd=%rd_d rr=%rr_d
+MULS            0000 0010 .... ....         &rd_rr      rd=%rd_a rr=%rr_a
+
+RCALL           1101 imm:s12
+RJMP            1100 imm:s12
+
+SBRC            1111 110 rr:5 0 bit:3
+SBRS            1111 111 rr:5 0 bit:3
+
+# The 22-bit immediate is partially in the opcode word,
+# and partially in the next.  Use append_16 to build the
+# complete 22-bit value.
+%imm_call       4:5 0:1                     !function=append_16
+CALL            1001 010 ..... 111 .        imm=%imm_call
+JMP             1001 010 ..... 110 .        imm=%imm_call
+
+
+# The 16-bit immediate is completely in the next word.
+# Fields cannot be defined with no bits, so we cannot play
+# the same trick and append to a zero-bit value.
+# Defer reading the immediate until trans_{LDS,STS}.
+@ldst_s         .... ... rd:5 ....          imm=0
+LDS             1001 000 ..... 0000         @ldst_s
+STS             1001 001 ..... 0000         @ldst_s
Signature · 1 line
2.17.2 (Apple Git-113)
#0 lore raw

v34 cover QEMU AVR 8 bit cores

This series of patches adds 8bit AVR cores to QEMU.
All instruction, except BREAK/DES/SPM/SPMX, are implemented. Not fully tested yet. However I was able to execute simple code with functions. e.g fibonacci calculation. This series of patches include a non real, sample board.
No fuses support yet. PC is set to 0 at reset.

the patches include the following
1. just a basic 8bit AVR CPU, without instruction decoding or translation
2. CPU features which allow define the following 8bit AVR cores

     avr1
     avr2 avr25
     avr3 avr31 avr35
     avr4
     avr5 avr51
     avr6
     xmega2 xmega4 xmega5 xmega6 xmega7

3. a definition of sample machine with SRAM, FLASH and CPU which allows to execute simple code
4. encoding for all AVR instructions
5. interrupt handling
6. helpers for IN, OUT, SLEEP, WBR & unsupported instructions
7. a decoder which given an opcode decides what istruction it is
8. translation of AVR instruction into TCG
9. all features together

changes since v3
1. rampD/X/Y/Z registers are encoded as 0x00ff0000 (instead of 0x000000ff) for faster address manipulaton
2. ffs changed to ctz32
3. duplicate code removed at avr_cpu_do_interrupt
4. using andc instead of not + and
5. fixing V flag calculation in varios instructions
6. freeing local variables in PUSH
7. tcg_const_local_i32 -> tcg_const_i32
8. using sextract32 instead of my implementation
9. fixing BLD instruction
10.xor(r) instead of 0xff - r at COM
11.fixing MULS/MULSU not to modify inputs' content
12.using SUB for NEG
13.fixing tcg_gen_qemu_ld/st call in XCH

changes since v4
1. target is now defined as big endian in order to optimize push_ret/pop_ret
2. all style warnings are fixed
3. adding cpu_set/get_sreg functions
4. simplifying gen_goto_tb as there is no real paging
5. env->pc -> env->pc_w
6. making flag dump more compact
7. more spacing
8. renaming CODE/DATA_INDEX -> MMU_CODE/DATA_IDX
9. removing avr_set_feature
10. SPL/SPH set bug fix
11. switching stb_phys to cpu_stb_data
12. cleaning up avr_decode
13. saving sreg, rampD/X/Y/Z, eind in HW format (savevm)
14. saving CPU features (savevm)

changes since v5
1. BLD bug fix
2. decoder generator is added

chages since v6
1. using cpu_get_sreg/cpu_set_sreg in avr_cpu_gdb_read_register/avr_cpu_gdb_write_register
2. configure the target as little endian because otherwise GDB does not work
3. fixing and testing gen_push_ret/gen_pop_ret

changes since v7
1. folding back v6
2. logging at helper_outb and helper_inb are done for non supported yet registers only
3. MAINTAINERS updated

changes since v8
1. removing hw/avr from hw/Makefile.obj as it should not be built for all
2. making linux compilable
3. testing on

    a. Mac, Apple LLVM version 7.0.0
    b. Ubuntu 12.04, gcc 4.9.2
    c. Fedora 23, gcc 5.3.1

4. folding back some patches
5. translation bug fixes for ORI, CPI, XOR instructions
6. propper handling of cpu register writes though memory

changes since v9
1. removing forward declarations of static functions
2. disabling debug prints
3. switching to case range instead of if else if ...
4. LD/ST IN/OUT accessing CPU maintainder registers are not routed to any device
5. commenst about sample board and sample IO device added
6. sample board description is more descriptive now
7. memory_region_allocate_system_memory is used to create RAM
8. now there are helper_fullrd & helper_fullwr when LD/ST try to access registers

changes since v10
1. movig back fullwr & fullrd into the commit where outb and inb were introduced
2. changing tlb_fill function signature
3. adding empty line between functions
4. adding newline on the last line of the file
5. using tb->flags to generae full access ST/LD instructions
6. fixing SBRC bug
7. folding back 10th commit
8. whenever a new file is introduced it's added to Makefile.objs

changes since v11
1. updating to v2.7.0-rc
2. removing assignment to env->fullacc from gen_intermediate_code

changes since v12
1. fixing spacing
2. fixing get/put_segment functions
3. removing target-avr/machine.h file
4. VMSTATE_SINGLE_TEST -> VMSTATE_SINGLE
5. comment spelling
6. removing hw/avr/sample_io.c
7. char const* -> const char*
8. proper ram allocation
9. fixing breakpoint functionality.
10.env1 -> env
11.fixing avr_cpu_gdb_write_register & avr_cpu_gdb_read_register functions 12.any cpu is removed
12.feature bits are not saved into vm state

changes since v13
1. rebasing to v2.7.0-rc1

changes since v14
1. I made self review with git gui tool. (I did not know such a thing exists)
2. removing all double/tripple spaces
3. removing comment reference to SampleIO
4. folding back some changes, so there is not deleted lines in my code
5. moving avr configuration, within configure file, before chris

changes since v15
1. removing IO registers cache from CPU
2. implementing CBI/SBI as read(helper_inb), modify, write(helper_outb)
3. implementing CBIC/SBIC as read(helper_inb), check, branch
4. adding missing tcg_temp_free_i32 for tcg_const_i32

changes since v16
1. removing EXT IO registers knoledge from CPU. These registers are accessible by LD/ST only. CPU has no interest in them

changes since v17 (by Richard Henderson)
This is Michael's v17, with some adjustments of my own:

1. Fix the whitespace errors reported by "git am",
2. Replace the utf-8 characters with normal ascii,
3. Ditch the separate compilation of translate.c.

I retained the two separate files that could be regenerated from the included cpugen program, but merged in translate-insn.c. Not that it matters, but the code generated is about 3k smaller.

changes since v18

1.  moving target-avr into target/avr
2.  do not call cpu_exec_initfn function from avr_cpu_initfn
3.  call cpu_exec_realizefn avr_cpu_realizefn
4.  do not fail sample machine creation if no rom is suplied
5.  add tcg_gen_exit_tb(0) for BS_BRANCH in gen_intermediate_code
6.  fix a register getters/setters in machine.c
7.  changing QEMU_ARCH_AVR from 1<<17 to 1<<18

changes since v19

1.  use decodetree.py tool to decode instructions
2.  adding USART
3.  adding 16 bit timer peripherals
4.  changing QEMU_ARCH_AVR from 1<<18 to 1<<20
5.  renaming tlb_fill to avr_cpu_tlb_fill

changes since v20

1.  use one CPU naming convention
2.  merging insn16.decode & insn32.decode files
3.  modifying skip next instruction mechanizm
4.  translate BREAK as NOP for now

changes since v21
1. Reorganize bstate.
This will make transition to <exec/translator.h> easier, and fixes a couple of bugs wrt single stepping by [email protected]
2. Drop cpc and fix page cross condition.
by [email protected]
3. Refactor checking supported/unsupported instructions
4. Add gdb-xml/avr-cpu.xml

changes since v22
1. Rebase
2. Split long comment

changes since v23

1.  remove avr_cpu_list_compare function
2.  modify avr_cpu_class_by_name function
3.  modify avr_cpu_list_entry function
4.  modify avr_cpu_list function

changes since v24
1. remove AVR_CPU_TYPE_NAME macro

changes since v25

1.  fix patches. every file belong to one patch only
2.  change copyright year from 2016 to 2019
3.  create mask device to emulate prr0/prr1

changes since v26
1. add avocado acceptence test
2. add boot serial test

changes since v27
1. list atmel2560 devices as unimplemented
2. fix sram base/size

changes since v28
1. rebase
2. fix includes & build

changes since v29
1. fixing ownership
2. using 'since' instread of 'added in'

changes since v30
1. rebase

changes since v31

1.  splitting 'Add instruction translation' commit
2.  fixing typo in qapi/machine.json sicne -> since
3.  removing unintended changes in configure file
4.  adding Richard Henderson as a co developer to 'Add instruction translation - CPU main translation funcions' commit

changes since v32
1. modify cpu_get_sreg to treat sreg C as other flags, except sreg Z

changes since v33
1. ensure flag C is always calculated as one bit
2. calculate flag Z as one bit, without using inverse logic

Michael Rolnik (12):
target/avr: Add outward facing interfaces and core CPU logic target/avr: Add instruction helpers
target/avr: Add instruction decoding
target/avr: Add instruction translation - Registers definition target/avr: Add instruction translation - Arithmetic and Logic
Instructions
target/avr: Add instruction translation - Branch Instructions target/avr: Add instruction translation - Bit and Bit-test
Instructions
target/avr: Add instruction translation - MCU Control Instructions target/avr: Add instruction translation - CPU main translation
function
target/avr: Add example board configuration
target/avr: Register AVR support with the rest of QEMU, the build
system, and the WMAINTAINERS file
target/avr: Add tests

Sarah Harris (1):
target/avr: Add limited support for USART and 16 bit timer peripherals

 MAINTAINERS                      |    9 +
 arch_init.c                      |    2 +
 configure                        |    7 +
 default-configs/avr-softmmu.mak  |    5 +
 gdb-xml/avr-cpu.xml              |   49 +
 hw/Kconfig                       |    1 +
 hw/avr/Kconfig                   |    6 +
 hw/avr/Makefile.objs             |    1 +
 hw/avr/sample.c                  |  282 +++
 hw/char/Kconfig                  |    3 +
 hw/char/Makefile.objs            |    1 +
 hw/char/avr_usart.c              |  324 ++++
 hw/misc/Kconfig                  |    3 +
 hw/misc/Makefile.objs            |    2 +
 hw/misc/avr_mask.c               |  112 ++
 hw/timer/Kconfig                 |    3 +
 hw/timer/Makefile.objs           |    1 +
 hw/timer/avr_timer16.c           |  605 ++++++
 include/disas/dis-asm.h          |    6 +
 include/hw/char/avr_usart.h      |   97 +
 include/hw/misc/avr_mask.h       |   47 +
 include/hw/timer/avr_timer16.h   |   97 +
 include/sysemu/arch_init.h       |    1 +
 qapi/machine.json                |    3 +-
 target/avr/Makefile.objs         |   33 +
 target/avr/cpu-param.h           |   37 +
 target/avr/cpu-qom.h             |   54 +
 target/avr/cpu.c                 |  576 ++++++
 target/avr/cpu.h                 |  253 +++
 target/avr/gdbstub.c             |   85 +
 target/avr/helper.c              |  354 ++++
 target/avr/helper.h              |   29 +
 target/avr/insn.decode           |  175 ++
 target/avr/machine.c             |  121 ++
 target/avr/translate.c           | 3028 ++++++++++++++++++++++++++++++
 tests/Makefile.include           |    2 +
 tests/acceptance/machine_avr6.py |   36 +
 tests/boot-serial-test.c         |   10 +
 tests/machine-none-test.c        |    1 +
 39 files changed, 6460 insertions(+), 1 deletion(-)
 create mode 100644 default-configs/avr-softmmu.mak
 create mode 100644 gdb-xml/avr-cpu.xml
 create mode 100644 hw/avr/Kconfig
 create mode 100644 hw/avr/Makefile.objs
 create mode 100644 hw/avr/sample.c
 create mode 100644 hw/char/avr_usart.c
 create mode 100644 hw/misc/avr_mask.c
 create mode 100644 hw/timer/avr_timer16.c
 create mode 100644 include/hw/char/avr_usart.h
 create mode 100644 include/hw/misc/avr_mask.h
 create mode 100644 include/hw/timer/avr_timer16.h
 create mode 100644 target/avr/Makefile.objs
 create mode 100644 target/avr/cpu-param.h
 create mode 100644 target/avr/cpu-qom.h
 create mode 100644 target/avr/cpu.c
 create mode 100644 target/avr/cpu.h
 create mode 100644 target/avr/gdbstub.c
 create mode 100644 target/avr/helper.c
 create mode 100644 target/avr/helper.h
 create mode 100644 target/avr/insn.decode
 create mode 100644 target/avr/machine.c
 create mode 100644 target/avr/translate.c
 create mode 100644 tests/acceptance/machine_avr6.py
Signature · 1 line
2.17.2 (Apple Git-113)
#12 lore raw

PATCH v34 12/13 target/avr: Register AVR support with the rest of QEMU, the build system, and the WMAINTAINERS file

---
 MAINTAINERS                     |  9 +++++++++
 arch_init.c                     |  2 ++
 configure                       |  7 +++++++
 default-configs/avr-softmmu.mak |  5 +++++
 include/disas/dis-asm.h         |  6 ++++++
 include/sysemu/arch_init.h      |  1 +
 qapi/machine.json               |  3 ++-
 target/avr/Makefile.objs        | 33 +++++++++++++++++++++++++++++++++
 tests/machine-none-test.c       |  1 +
 9 files changed, 66 insertions(+), 1 deletion(-)
 create mode 100644 default-configs/avr-softmmu.mak
 create mode 100644 target/avr/Makefile.objs
MAINTAINERS +9 -0
index ed41d7d1b6..931bbd17dd 100644
@@ -163,6 +163,15 @@ S: Maintained
 F: hw/arm/smmu*
 F: include/hw/arm/smmu*
 
+AVR TCG CPUs
+M: Michael Rolnik <[email protected]>
+S: Maintained
+F: target/avr/
+F: hw/misc/avr_mask.c
+F: hw/char/avr_usart.c
+F: hw/timer/avr_timer16.c
+F: hw/avr/
+
 CRIS TCG CPUs
 M: Edgar E. Iglesias <[email protected]>
 S: Maintained
arch_init.c +2 -0
index 0a1531124c..fb308aa802 100644
@@ -85,6 +85,8 @@ int graphic_depth = 32;
 #define QEMU_ARCH QEMU_ARCH_UNICORE32
 #elif defined(TARGET_XTENSA)
 #define QEMU_ARCH QEMU_ARCH_XTENSA
+#elif defined(TARGET_AVR)
+#define QEMU_ARCH QEMU_ARCH_AVR
 #endif
 
 const uint32_t arch_type = QEMU_ARCH;
configure +7 -0
index 145fcabbb3..b2a5ad4225 100755
@@ -7457,6 +7457,10 @@ case "$target_name" in
     mttcg="yes"
     gdb_xml_files="aarch64-core.xml aarch64-fpu.xml arm-core.xml arm-vfp.xml arm-vfp3.xml arm-neon.xml"
   ;;
+  avr)
+    gdb_xml_files="avr-cpu.xml"
+    target_compiler=$cross_cc_avr
+  ;;
   cris)
   ;;
   hppa)
@@ -7676,6 +7680,9 @@ for i in $ARCH $TARGET_BASE_ARCH ; do
       disas_config "ARM_A64"
     fi
   ;;
+  avr)
+    disas_config "AVR"
+  ;;
   cris)
     disas_config "CRIS"
   ;;
default-configs/avr-softmmu.mak new file +5 -0
new file mode 100644
index 0000000000..d1e1c28118
@@ -0,0 +1,5 @@
+# Default configuration for avr-softmmu
+
+# Boards:
+#
+CONFIG_AVR_SAMPLE=y
include/disas/dis-asm.h +6 -0
index e9c7dd8eb4..8bedce17ac 100644
@@ -211,6 +211,12 @@ enum bfd_architecture
 #define bfd_mach_m32r          0  /* backwards compatibility */
   bfd_arch_mn10200,    /* Matsushita MN10200 */
   bfd_arch_mn10300,    /* Matsushita MN10300 */
+  bfd_arch_avr,       /* Atmel AVR microcontrollers.  */
+#define bfd_mach_avr1          1
+#define bfd_mach_avr2          2
+#define bfd_mach_avr3          3
+#define bfd_mach_avr4          4
+#define bfd_mach_avr5          5
   bfd_arch_cris,       /* Axis CRIS */
 #define bfd_mach_cris_v0_v10   255
 #define bfd_mach_cris_v32      32
include/sysemu/arch_init.h +1 -0
index 62c6fe4cf1..893df26ce2 100644
@@ -24,6 +24,7 @@ enum {
     QEMU_ARCH_NIOS2 = (1 << 17),
     QEMU_ARCH_HPPA = (1 << 18),
     QEMU_ARCH_RISCV = (1 << 19),
+    QEMU_ARCH_AVR = (1 << 20),
 };
 
 extern const uint32_t arch_type;
qapi/machine.json +2 -1
index ca26779f1a..1fa2917ba9 100644
@@ -21,11 +21,12 @@
 #        is true even for "qemu-system-x86_64".
 #
 # ppcemb: dropped in 3.1
+# avr: since 4.2
 #
 # Since: 3.0
 ##
 { 'enum' : 'SysEmuTarget',
-  'data' : [ 'aarch64', 'alpha', 'arm', 'cris', 'hppa', 'i386', 'lm32',
+  'data' : [ 'aarch64', 'alpha', 'arm', 'avr', 'cris', 'hppa', 'i386', 'lm32',
              'm68k', 'microblaze', 'microblazeel', 'mips', 'mips64',
              'mips64el', 'mipsel', 'moxie', 'nios2', 'or1k', 'ppc',
              'ppc64', 'riscv32', 'riscv64', 's390x', 'sh4',
target/avr/Makefile.objs new file +33 -0
new file mode 100644
index 0000000000..2976affd95
@@ -0,0 +1,33 @@
+#
+#  QEMU AVR CPU
+#
+#  Copyright (c) 2019 Michael Rolnik
+#
+#  This library is free software; you can redistribute it and/or
+#  modify it under the terms of the GNU Lesser General Public
+#  License as published by the Free Software Foundation; either
+#  version 2.1 of the License, or (at your option) any later version.
+#
+#  This library 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
+#  Lesser General Public License for more details.
+#
+#  You should have received a copy of the GNU Lesser General Public
+#  License along with this library; if not, see
+#  <http://www.gnu.org/licenses/lgpl-2.1.html>
+#
+
+DECODETREE = $(SRC_PATH)/scripts/decodetree.py
+decode-y = $(SRC_PATH)/target/avr/insn.decode
+
+target/avr/decode_insn.inc.c: $(decode-y) $(DECODETREE)
+	$(call quiet-command, \
+	  $(PYTHON) $(DECODETREE) -o $@ --decode decode_insn --insnwidth 16 $<, \
+	  "GEN", $(TARGET_DIR)$@)
+
+target/avr/translate.o: target/avr/decode_insn.inc.c
+
+obj-y += translate.o cpu.o helper.o
+obj-y += gdbstub.o
+obj-$(CONFIG_SOFTMMU) += machine.o
tests/machine-none-test.c +1 -0
index 5953d31755..3e5c74e73e 100644
@@ -27,6 +27,7 @@ static struct arch2cpu cpus_map[] = {
     /* tested targets list */
     { "arm", "cortex-a15" },
     { "aarch64", "cortex-a57" },
+    { "avr", "avr6-avr-cpu" },
     { "x86_64", "qemu64,apic-id=0" },
     { "i386", "qemu32,apic-id=0" },
     { "alpha", "ev67" },
Signature · 1 line
2.17.2 (Apple Git-113)
#6 lore raw

PATCH v34 6/13 target/avr: Add instruction translation - Branch Instructions

In reply to Michael Rolnik

This includes:
- RJMP, IJMP, EIJMP, JMP
- RCALL, ICALL, EICALL, CALL
- RET, RETI
- CPSE, CP, CPC, CPI
- SBRC, SBRS, SBIC, SBIS
- BRBC, BRBS

---
 target/avr/translate.c | 549 ++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 546 insertions(+), 3 deletions(-)
target/avr/translate.c +546 -3
index 573c9988b5..4ef3294a40 100644
@@ -224,6 +224,86 @@ static void gen_ZNSf(TCGv R)
 }
 
 
+static void gen_push_ret(DisasContext *ctx, int ret)
+{
+    if (avr_feature(ctx->env, AVR_FEATURE_1_BYTE_PC)) {
+
+        TCGv t0 = tcg_const_i32((ret & 0x0000ff));
+
+        tcg_gen_qemu_st_tl(t0, cpu_sp, MMU_DATA_IDX, MO_UB);
+        tcg_gen_subi_tl(cpu_sp, cpu_sp, 1);
+
+        tcg_temp_free_i32(t0);
+    } else if (avr_feature(ctx->env, AVR_FEATURE_2_BYTE_PC)) {
+
+        TCGv t0 = tcg_const_i32((ret & 0x00ffff));
+
+        tcg_gen_subi_tl(cpu_sp, cpu_sp, 1);
+        tcg_gen_qemu_st_tl(t0, cpu_sp, MMU_DATA_IDX, MO_BEUW);
+        tcg_gen_subi_tl(cpu_sp, cpu_sp, 1);
+
+        tcg_temp_free_i32(t0);
+
+    } else if (avr_feature(ctx->env, AVR_FEATURE_3_BYTE_PC)) {
+
+        TCGv lo = tcg_const_i32((ret & 0x0000ff));
+        TCGv hi = tcg_const_i32((ret & 0xffff00) >> 8);
+
+        tcg_gen_qemu_st_tl(lo, cpu_sp, MMU_DATA_IDX, MO_UB);
+        tcg_gen_subi_tl(cpu_sp, cpu_sp, 2);
+        tcg_gen_qemu_st_tl(hi, cpu_sp, MMU_DATA_IDX, MO_BEUW);
+        tcg_gen_subi_tl(cpu_sp, cpu_sp, 1);
+
+        tcg_temp_free_i32(lo);
+        tcg_temp_free_i32(hi);
+    }
+}
+
+
+static void gen_pop_ret(DisasContext *ctx, TCGv ret)
+{
+    if (avr_feature(ctx->env, AVR_FEATURE_1_BYTE_PC)) {
+        tcg_gen_addi_tl(cpu_sp, cpu_sp, 1);
+        tcg_gen_qemu_ld_tl(ret, cpu_sp, MMU_DATA_IDX, MO_UB);
+    } else if (avr_feature(ctx->env, AVR_FEATURE_2_BYTE_PC)) {
+        tcg_gen_addi_tl(cpu_sp, cpu_sp, 1);
+        tcg_gen_qemu_ld_tl(ret, cpu_sp, MMU_DATA_IDX, MO_BEUW);
+        tcg_gen_addi_tl(cpu_sp, cpu_sp, 1);
+    } else if (avr_feature(ctx->env, AVR_FEATURE_3_BYTE_PC)) {
+        TCGv lo = tcg_temp_new_i32();
+        TCGv hi = tcg_temp_new_i32();
+
+        tcg_gen_addi_tl(cpu_sp, cpu_sp, 1);
+        tcg_gen_qemu_ld_tl(hi, cpu_sp, MMU_DATA_IDX, MO_BEUW);
+
+        tcg_gen_addi_tl(cpu_sp, cpu_sp, 2);
+        tcg_gen_qemu_ld_tl(lo, cpu_sp, MMU_DATA_IDX, MO_UB);
+
+        tcg_gen_deposit_tl(ret, lo, hi, 8, 16);
+
+        tcg_temp_free_i32(lo);
+        tcg_temp_free_i32(hi);
+    }
+}
+
+
+static void gen_goto_tb(DisasContext *ctx, int n, target_ulong dest)
+{
+    TranslationBlock *tb = ctx->tb;
+
+    if (ctx->singlestep == 0) {
+        tcg_gen_goto_tb(n);
+        tcg_gen_movi_i32(cpu_pc, dest);
+        tcg_gen_exit_tb(tb, n);
+    } else {
+        tcg_gen_movi_i32(cpu_pc, dest);
+        gen_helper_debug(cpu_env);
+        tcg_gen_exit_tb(NULL, 0);
+    }
+    ctx->bstate = DISAS_NORETURN;
+}
+
+
 /*
  *  Adds two registers without the C Flag and places the result in the
  *  destination register Rd.
@@ -797,13 +877,14 @@ static bool trans_MULSU(DisasContext *ctx, arg_MULSU *a)
     TCGv t0 = tcg_temp_new_i32();
 
     tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */
-    tcg_gen_mul_tl(R, t0, Rr); /* R = Rd *Rr */
+    tcg_gen_mul_tl(R, t0, Rr); /* R = Rd * Rr */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make R 16 bits */
 
     tcg_gen_andi_tl(R0, R, 0xff);
     tcg_gen_shri_tl(R1, R, 8);
 
-    tcg_gen_shri_tl(cpu_Cf, R, 16); /* Cf = R(16) */
-    tcg_gen_mov_tl(cpu_Zf, R);
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
 
     tcg_temp_free_i32(t0);
     tcg_temp_free_i32(R);
@@ -952,3 +1033,465 @@ static bool trans_DES(DisasContext *ctx, arg_DES *a)
 
     return true;
 }
+
+
+/*
+ *  Relative jump to an address within PC - 2K +1 and PC + 2K (words). For
+ *  AVR microcontrollers with Program memory not exceeding 4K words (8KB) this
+ *  instruction can address the entire memory from every address location. See
+ *  also JMP.
+ */
+static bool trans_RJMP(DisasContext *ctx, arg_RJMP *a)
+{
+    int dst = ctx->npc + a->imm;
+
+    gen_goto_tb(ctx, 0, dst);
+
+    return true;
+}
+
+
+/*
+ *  Indirect jump to the address pointed to by the Z (16 bits) Pointer
+ *  Register in the Register File. The Z-pointer Register is 16 bits wide and
+ *  allows jump within the lowest 64K words (128KB) section of Program memory.
+ *  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_IJMP(DisasContext *ctx, arg_IJMP *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_IJMP_ICALL)) {
+        return true;
+    }
+
+    gen_jmp_z(ctx);
+
+    return true;
+}
+
+
+/*
+ *  Indirect jump to the address pointed to by the Z (16 bits) Pointer
+ *  Register in the Register File and the EIND Register in the I/O space. This
+ *  instruction allows for indirect jumps to the entire 4M (words) Program
+ *  memory space. See also IJMP.  This instruction is not available in all
+ *  devices. Refer to the device specific instruction set summary.
+ */
+static bool trans_EIJMP(DisasContext *ctx, arg_EIJMP *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_EIJMP_EICALL)) {
+        return true;
+    }
+
+    gen_jmp_ez(ctx);
+    return true;
+}
+
+
+/*
+ *  Jump to an address within the entire 4M (words) Program memory. See also
+ *  RJMP.  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.0
+ */
+static bool trans_JMP(DisasContext *ctx, arg_JMP *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_JMP_CALL)) {
+        return true;
+    }
+
+    gen_goto_tb(ctx, 0, a->imm);
+
+    return true;
+}
+
+
+/*
+ *  Relative call to an address within PC - 2K + 1 and PC + 2K (words). The
+ *  return address (the instruction after the RCALL) is stored onto the Stack.
+ *  See also CALL. For AVR microcontrollers with Program memory not exceeding 4K
+ *  words (8KB) this instruction can address the entire memory from every
+ *  address location. The Stack Pointer uses a post-decrement scheme during
+ *  RCALL.
+ */
+static bool trans_RCALL(DisasContext *ctx, arg_RCALL *a)
+{
+    int ret = ctx->npc;
+    int dst = ctx->npc + a->imm;
+
+    gen_push_ret(ctx, ret);
+    gen_goto_tb(ctx, 0, dst);
+
+    return true;
+}
+
+
+/*
+ *  Calls to a subroutine within the entire 4M (words) Program memory. The
+ *  return address (to the instruction after the CALL) will be stored onto the
+ *  Stack. See also RCALL. The Stack Pointer uses a post-decrement scheme during
+ *  CALL.  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_ICALL(DisasContext *ctx, arg_ICALL *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_IJMP_ICALL)) {
+        return true;
+    }
+
+    int ret = ctx->npc;
+
+    gen_push_ret(ctx, ret);
+    gen_jmp_z(ctx);
+
+    return true;
+}
+
+
+/*
+ *  Indirect call of a subroutine pointed to by the Z (16 bits) Pointer
+ *  Register in the Register File and the EIND Register in the I/O space. This
+ *  instruction allows for indirect calls to the entire 4M (words) Program
+ *  memory space. See also ICALL. The Stack Pointer uses a post-decrement scheme
+ *  during EICALL.  This instruction is not available in all devices. Refer to
+ *  the device specific instruction set summary.
+ */
+static bool trans_EICALL(DisasContext *ctx, arg_EICALL *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_EIJMP_EICALL)) {
+        return true;
+    }
+
+    int ret = ctx->npc;
+
+    gen_push_ret(ctx, ret);
+    gen_jmp_ez(ctx);
+    return true;
+}
+
+
+/*
+ *  Calls to a subroutine within the entire Program memory. The return
+ *  address (to the instruction after the CALL) will be stored onto the Stack.
+ *  (See also RCALL). The Stack Pointer uses a post-decrement scheme during
+ *  CALL.  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_CALL(DisasContext *ctx, arg_CALL *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_JMP_CALL)) {
+        return true;
+    }
+
+    int Imm = a->imm;
+    int ret = ctx->npc;
+
+    gen_push_ret(ctx, ret);
+    gen_goto_tb(ctx, 0, Imm);
+
+    return true;
+}
+
+
+/*
+ *  Returns from subroutine. The return address is loaded from the STACK.
+ *  The Stack Pointer uses a preincrement scheme during RET.
+ */
+static bool trans_RET(DisasContext *ctx, arg_RET *a)
+{
+    gen_pop_ret(ctx, cpu_pc);
+
+    ctx->bstate = DISAS_LOOKUP;
+    return true;
+}
+
+
+/*
+ *  Returns from interrupt. The return address is loaded from the STACK and
+ *  the Global Interrupt Flag is set.  Note that the Status Register is not
+ *  automatically stored when entering an interrupt routine, and it is not
+ *  restored when returning from an interrupt routine. This must be handled by
+ *  the application program. The Stack Pointer uses a pre-increment scheme
+ *  during RETI.
+ */
+static bool trans_RETI(DisasContext *ctx, arg_RETI *a)
+{
+    gen_pop_ret(ctx, cpu_pc);
+    tcg_gen_movi_tl(cpu_If, 1);
+
+    /* Need to return to main loop to re-evaluate interrupts.  */
+    ctx->bstate = DISAS_EXIT;
+    return true;
+}
+
+
+/*
+ *  This instruction performs a compare between two registers Rd and Rr, and
+ *  skips the next instruction if Rd = Rr.
+ */
+static bool trans_CPSE(DisasContext *ctx, arg_CPSE *a)
+{
+    ctx->skip_cond = TCG_COND_EQ;
+    ctx->skip_var0 = cpu_r[a->rd];
+    ctx->skip_var1 = cpu_r[a->rr];
+    return true;
+}
+
+
+/*
+ *  This instruction performs a compare between two registers Rd and Rr.
+ *  None of the registers are changed. All conditional branches can be used
+ *  after this instruction.
+ */
+static bool trans_CP(DisasContext *ctx, arg_CP *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs a compare between two registers Rd and Rr and
+ *  also takes into account the previous carry. None of the registers are
+ *  changed. All conditional branches can be used after this instruction.
+ */
+static bool trans_CPC(DisasContext *ctx, arg_CPC *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv zero = tcg_const_i32(0);
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr - Cf */
+    tcg_gen_sub_tl(R, R, cpu_Cf);
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_NSf(R);
+
+    /*
+     * Previous value remains unchanged when the result is zero;
+     * cleared otherwise.
+     */
+    tcg_gen_movcond_tl(TCG_COND_EQ, cpu_Zf, R, zero, cpu_Zf, zero);
+
+    tcg_temp_free_i32(zero);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs a compare between register Rd and a constant.
+ *  The register is not changed. All conditional branches can be used after this
+ *  instruction.
+ */
+static bool trans_CPI(DisasContext *ctx, arg_CPI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    int Imm = a->imm;
+    TCGv Rr = tcg_const_i32(Imm);
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    tcg_temp_free_i32(R);
+    tcg_temp_free_i32(Rr);
+
+    return true;
+}
+
+
+/*
+ *  This instruction tests a single bit in a register and skips the next
+ *  instruction if the bit is cleared.
+ */
+static bool trans_SBRC(DisasContext *ctx, arg_SBRC *a)
+{
+    TCGv Rr = cpu_r[a->rr];
+
+    ctx->skip_cond = TCG_COND_EQ;
+    ctx->skip_var0 = tcg_temp_new();
+    ctx->free_skip_var0 = true;
+
+    tcg_gen_andi_tl(ctx->skip_var0, Rr, 1 << a->bit);
+    return true;
+}
+
+
+/*
+ *  This instruction tests a single bit in a register and skips the next
+ *  instruction if the bit is set.
+ */
+static bool trans_SBRS(DisasContext *ctx, arg_SBRS *a)
+{
+    TCGv Rr = cpu_r[a->rr];
+
+    ctx->skip_cond = TCG_COND_NE;
+    ctx->skip_var0 = tcg_temp_new();
+    ctx->free_skip_var0 = true;
+
+    tcg_gen_andi_tl(ctx->skip_var0, Rr, 1 << a->bit);
+    return true;
+}
+
+
+/*
+ *  This instruction tests a single bit in an I/O Register and skips the
+ *  next instruction if the bit is cleared. This instruction operates on the
+ *  lower 32 I/O Registers -- addresses 0-31.
+ */
+static bool trans_SBIC(DisasContext *ctx, arg_SBIC *a)
+{
+    TCGv temp = tcg_const_i32(a->reg);
+
+    gen_helper_inb(temp, cpu_env, temp);
+    tcg_gen_andi_tl(temp, temp, 1 << a->bit);
+    ctx->skip_cond = TCG_COND_EQ;
+    ctx->skip_var0 = temp;
+    ctx->free_skip_var0 = true;
+
+    return true;
+}
+
+
+/*
+ *  This instruction tests a single bit in an I/O Register and skips the
+ *  next instruction if the bit is set. This instruction operates on the lower
+ *  32 I/O Registers -- addresses 0-31.
+ */
+static bool trans_SBIS(DisasContext *ctx, arg_SBIS *a)
+{
+    TCGv temp = tcg_const_i32(a->reg);
+
+    gen_helper_inb(temp, cpu_env, temp);
+    tcg_gen_andi_tl(temp, temp, 1 << a->bit);
+    ctx->skip_cond = TCG_COND_NE;
+    ctx->skip_var0 = temp;
+    ctx->free_skip_var0 = true;
+
+    return true;
+}
+
+
+/*
+ *  Conditional relative branch. Tests a single bit in SREG and branches
+ *  relatively to PC if the bit is cleared. This instruction branches relatively
+ *  to PC in either direction (PC - 63 < = destination <= PC + 64). The
+ *  parameter k is the offset from PC and is represented in two's complement
+ *  form.
+ */
+static bool trans_BRBC(DisasContext *ctx, arg_BRBC *a)
+{
+    TCGLabel *not_taken = gen_new_label();
+
+    TCGv var;
+
+    switch (a->bit) {
+    case 0x00:
+        var = cpu_Cf;
+        break;
+    case 0x01:
+        var = cpu_Zf;
+        break;
+    case 0x02:
+        var = cpu_Nf;
+        break;
+    case 0x03:
+        var = cpu_Vf;
+        break;
+    case 0x04:
+        var = cpu_Sf;
+        break;
+    case 0x05:
+        var = cpu_Hf;
+        break;
+    case 0x06:
+        var = cpu_Tf;
+        break;
+    case 0x07:
+        var = cpu_If;
+        break;
+    default:
+        g_assert_not_reached();
+    }
+
+    tcg_gen_brcondi_i32(TCG_COND_NE, var, 0, not_taken);
+    gen_goto_tb(ctx, 0, ctx->npc + a->imm);
+    gen_set_label(not_taken);
+
+    ctx->bstate = DISAS_CHAIN;
+    return true;
+}
+
+
+/*
+ *  Conditional relative branch. Tests a single bit in SREG and branches
+ *  relatively to PC if the bit is set. This instruction branches relatively to
+ *  PC in either direction (PC - 63 < = destination <= PC + 64). The parameter k
+ *  is the offset from PC and is represented in two's complement form.
+ */
+static bool trans_BRBS(DisasContext *ctx, arg_BRBS *a)
+{
+    TCGLabel *not_taken = gen_new_label();
+
+    TCGv var;
+
+    switch (a->bit) {
+    case 0x00:
+        var = cpu_Cf;
+        break;
+    case 0x01:
+        var = cpu_Zf;
+        break;
+    case 0x02:
+        var = cpu_Nf;
+        break;
+    case 0x03:
+        var = cpu_Vf;
+        break;
+    case 0x04:
+        var = cpu_Sf;
+        break;
+    case 0x05:
+        var = cpu_Hf;
+        break;
+    case 0x06:
+        var = cpu_Tf;
+        break;
+    case 0x07:
+        var = cpu_If;
+        break;
+    default:
+        g_assert_not_reached();
+    }
+
+    tcg_gen_brcondi_i32(TCG_COND_EQ, var, 0, not_taken);
+    gen_goto_tb(ctx, 0, ctx->npc + a->imm);
+    gen_set_label(not_taken);
+
+    ctx->bstate = DISAS_CHAIN;
+    return true;
+}
+
Signature · 1 line
2.17.2 (Apple Git-113)
#2 lore raw

PATCH v34 2/13 target/avr: Add instruction helpers

In reply to Michael Rolnik

Stubs for unimplemented instructions and helpers for instructions that need to interact with QEMU. SPM and WDR are unimplemented because they require emulation of complex peripherals.
The implementation of SLEEP is very limited due to the lack of peripherals to generate wake interrupts. Memory access instructions are implemented here because some address ranges actually refer to CPU registers.

---
 target/avr/helper.c | 354 ++++++++++++++++++++++++++++++++++++++++++++
 target/avr/helper.h |  29 ++++
 2 files changed, 383 insertions(+)
 create mode 100644 target/avr/helper.c
 create mode 100644 target/avr/helper.h
target/avr/helper.c new file +354 -0
new file mode 100644
index 0000000000..f0f0d4f15a
@@ -0,0 +1,354 @@
+/*
+ * QEMU AVR CPU
+ *
+ * Copyright (c) 2019 Michael Rolnik
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see
+ * <http://www.gnu.org/licenses/lgpl-2.1.html>
+ */
+
+#include "qemu/osdep.h"
+
+#include "cpu.h"
+#include "hw/irq.h"
+#include "hw/sysbus.h"
+#include "sysemu/sysemu.h"
+#include "exec/exec-all.h"
+#include "exec/cpu_ldst.h"
+#include "exec/helper-proto.h"
+#include "exec/ioport.h"
+#include "qemu/host-utils.h"
+#include "qemu/error-report.h"
+
+bool avr_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
+{
+    bool ret = false;
+    CPUClass *cc = CPU_GET_CLASS(cs);
+    AVRCPU *cpu = AVR_CPU(cs);
+    CPUAVRState *env = &cpu->env;
+
+    if (interrupt_request & CPU_INTERRUPT_RESET) {
+        if (cpu_interrupts_enabled(env)) {
+            cs->exception_index = EXCP_RESET;
+            cc->do_interrupt(cs);
+
+            cs->interrupt_request &= ~CPU_INTERRUPT_RESET;
+
+            ret = true;
+        }
+    }
+    if (interrupt_request & CPU_INTERRUPT_HARD) {
+        if (cpu_interrupts_enabled(env) && env->intsrc != 0) {
+            int index = ctz32(env->intsrc);
+            cs->exception_index = EXCP_INT(index);
+            cc->do_interrupt(cs);
+
+            env->intsrc &= env->intsrc - 1; /* clear the interrupt */
+            cs->interrupt_request &= ~CPU_INTERRUPT_HARD;
+
+            ret = true;
+        }
+    }
+    return ret;
+}
+
+void avr_cpu_do_interrupt(CPUState *cs)
+{
+    AVRCPU *cpu = AVR_CPU(cs);
+    CPUAVRState *env = &cpu->env;
+
+    uint32_t ret = env->pc_w;
+    int vector = 0;
+    int size = avr_feature(env, AVR_FEATURE_JMP_CALL) ? 2 : 1;
+    int base = 0;
+
+    if (cs->exception_index == EXCP_RESET) {
+        vector = 0;
+    } else if (env->intsrc != 0) {
+        vector = ctz32(env->intsrc) + 1;
+    }
+
+    if (avr_feature(env, AVR_FEATURE_3_BYTE_PC)) {
+        cpu_stb_data(env, env->sp--, (ret & 0x0000ff));
+        cpu_stb_data(env, env->sp--, (ret & 0x00ff00) >> 8);
+        cpu_stb_data(env, env->sp--, (ret & 0xff0000) >> 16);
+    } else if (avr_feature(env, AVR_FEATURE_2_BYTE_PC)) {
+        cpu_stb_data(env, env->sp--, (ret & 0x0000ff));
+        cpu_stb_data(env, env->sp--, (ret & 0x00ff00) >> 8);
+    } else {
+        cpu_stb_data(env, env->sp--, (ret & 0x0000ff));
+    }
+
+    env->pc_w = base + vector * size;
+    env->sregI = 0; /* clear Global Interrupt Flag */
+
+    cs->exception_index = -1;
+}
+
+int avr_cpu_memory_rw_debug(CPUState *cs, vaddr addr, uint8_t *buf,
+                                int len, bool is_write)
+{
+    return cpu_memory_rw_debug(cs, addr, buf, len, is_write);
+}
+
+hwaddr avr_cpu_get_phys_page_debug(CPUState *cs, vaddr addr)
+{
+    return addr; /* I assume 1:1 address correspondance */
+}
+
+int avr_cpu_handle_mmu_fault(
+    CPUState *cs, vaddr address, int size, int rw, int mmu_idx)
+{
+    /* currently it's assumed that this will never happen */
+    cs->exception_index = EXCP_DEBUG;
+    cpu_dump_state(cs, stderr, 0);
+    return 1;
+}
+
+bool avr_cpu_tlb_fill(CPUState *cs, vaddr address, int size,
+                        MMUAccessType access_type, int mmu_idx,
+                        bool probe, uintptr_t retaddr)
+{
+    int prot = 0;
+    MemTxAttrs attrs = {};
+    uint32_t paddr;
+
+    address &= TARGET_PAGE_MASK;
+
+    if (mmu_idx == MMU_CODE_IDX) {
+        /* access to code in flash */
+        paddr = OFFSET_CODE + address;
+        prot = PAGE_READ | PAGE_EXEC;
+        if (paddr + TARGET_PAGE_SIZE > OFFSET_DATA) {
+            error_report("execution left flash memory");
+            exit(1);
+        }
+    } else if (address < NO_CPU_REGISTERS + NO_IO_REGISTERS) {
+        /*
+         * access to CPU registers, exit and rebuilt this TB to use full access
+         * incase it touches specially handled registers like SREG or SP
+         */
+        AVRCPU *cpu = AVR_CPU(cs);
+        CPUAVRState *env = &cpu->env;
+        env->fullacc = 1;
+        cpu_loop_exit_restore(cs, retaddr);
+    } else {
+        /* access to memory. nothing special */
+        paddr = OFFSET_DATA + address;
+        prot = PAGE_READ | PAGE_WRITE;
+    }
+
+    tlb_set_page_with_attrs(
+        cs, address, paddr, attrs, prot, mmu_idx, TARGET_PAGE_SIZE);
+
+    return true;
+}
+
+void helper_sleep(CPUAVRState *env)
+{
+    CPUState *cs = env_cpu(env);
+
+    cs->exception_index = EXCP_HLT;
+    cpu_loop_exit(cs);
+}
+
+void helper_unsupported(CPUAVRState *env)
+{
+    CPUState *cs = env_cpu(env);
+
+    /*
+     *  I count not find what happens on the real platform, so
+     *  it's EXCP_DEBUG for meanwhile
+     */
+    cs->exception_index = EXCP_DEBUG;
+    if (qemu_loglevel_mask(LOG_UNIMP)) {
+        qemu_log("UNSUPPORTED\n");
+        cpu_dump_state(cs, qemu_logfile, 0);
+    }
+    cpu_loop_exit(cs);
+}
+
+void helper_debug(CPUAVRState *env)
+{
+    CPUState *cs = env_cpu(env);
+
+    cs->exception_index = EXCP_DEBUG;
+    cpu_loop_exit(cs);
+}
+
+void helper_break(CPUAVRState *env)
+{
+    CPUState *cs = env_cpu(env);
+
+    cs->exception_index = EXCP_DEBUG;
+    cpu_loop_exit(cs);
+}
+
+void helper_wdr(CPUAVRState *env)
+{
+    CPUState *cs = env_cpu(env);
+
+    /* WD is not implemented yet, placeholder */
+    cs->exception_index = EXCP_DEBUG;
+    cpu_loop_exit(cs);
+}
+
+/*
+ * This function implements IN instruction
+ *
+ * It does the following
+ * a.  if an IO register belongs to CPU, its value is read and returned
+ * b.  otherwise io address is translated to mem address and physical memory
+ *     is read.
+ * c.  it caches the value for sake of SBI, SBIC, SBIS & CBI implementation
+ *
+ */
+target_ulong helper_inb(CPUAVRState *env, uint32_t port)
+{
+    target_ulong data = 0;
+
+    switch (port) {
+    case 0x38: /* RAMPD */
+        data = 0xff & (env->rampD >> 16);
+        break;
+    case 0x39: /* RAMPX */
+        data = 0xff & (env->rampX >> 16);
+        break;
+    case 0x3a: /* RAMPY */
+        data = 0xff & (env->rampY >> 16);
+        break;
+    case 0x3b: /* RAMPZ */
+        data = 0xff & (env->rampZ >> 16);
+        break;
+    case 0x3c: /* EIND */
+        data = 0xff & (env->eind >> 16);
+        break;
+    case 0x3d: /* SPL */
+        data = env->sp & 0x00ff;
+        break;
+    case 0x3e: /* SPH */
+        data = env->sp >> 8;
+        break;
+    case 0x3f: /* SREG */
+        data = cpu_get_sreg(env);
+        break;
+    default:
+        /* not a special register, pass to normal memory access */
+        cpu_physical_memory_read(OFFSET_IO_REGISTERS + port, &data, 1);
+    }
+
+    return data;
+}
+
+/*
+ *  This function implements OUT instruction
+ *
+ *  It does the following
+ *  a.  if an IO register belongs to CPU, its value is written into the register
+ *  b.  otherwise io address is translated to mem address and physical memory
+ *      is written.
+ *  c.  it caches the value for sake of SBI, SBIC, SBIS & CBI implementation
+ *
+ */
+void helper_outb(CPUAVRState *env, uint32_t port, uint32_t data)
+{
+    data &= 0x000000ff;
+
+    switch (port) {
+    case 0x38: /* RAMPD */
+        if (avr_feature(env, AVR_FEATURE_RAMPD)) {
+            env->rampD = (data & 0xff) << 16;
+        }
+        break;
+    case 0x39: /* RAMPX */
+        if (avr_feature(env, AVR_FEATURE_RAMPX)) {
+            env->rampX = (data & 0xff) << 16;
+        }
+        break;
+    case 0x3a: /* RAMPY */
+        if (avr_feature(env, AVR_FEATURE_RAMPY)) {
+            env->rampY = (data & 0xff) << 16;
+        }
+        break;
+    case 0x3b: /* RAMPZ */
+        if (avr_feature(env, AVR_FEATURE_RAMPZ)) {
+            env->rampZ = (data & 0xff) << 16;
+        }
+        break;
+    case 0x3c: /* EIDN */
+        env->eind = (data & 0xff) << 16;
+        break;
+    case 0x3d: /* SPL */
+        env->sp = (env->sp & 0xff00) | (data);
+        break;
+    case 0x3e: /* SPH */
+        if (avr_feature(env, AVR_FEATURE_2_BYTE_SP)) {
+            env->sp = (env->sp & 0x00ff) | (data << 8);
+        }
+        break;
+    case 0x3f: /* SREG */
+        cpu_set_sreg(env, data);
+        break;
+    default:
+        /* not a special register, pass to normal memory access */
+        cpu_physical_memory_write(OFFSET_IO_REGISTERS + port, &data, 1);
+    }
+}
+
+/*
+ *  this function implements LD instruction when there is a posibility to read
+ *  from a CPU register
+ */
+target_ulong helper_fullrd(CPUAVRState *env, uint32_t addr)
+{
+    uint8_t data;
+
+    env->fullacc = false;
+
+    if (addr < NO_CPU_REGISTERS) {
+        /* CPU registers */
+        data = env->r[addr];
+    } else if (addr < NO_CPU_REGISTERS + NO_IO_REGISTERS) {
+        /* IO registers */
+        data = helper_inb(env, addr - NO_CPU_REGISTERS);
+    } else {
+        /* memory */
+        cpu_physical_memory_read(OFFSET_DATA + addr, &data, 1);
+    }
+    return data;
+}
+
+/*
+ *  this function implements ST instruction when there is a posibility to write
+ *  into a CPU register
+ */
+void helper_fullwr(CPUAVRState *env, uint32_t data, uint32_t addr)
+{
+    env->fullacc = false;
+
+    /* Following logic assumes this: */
+    assert(OFFSET_CPU_REGISTERS == OFFSET_DATA);
+    assert(OFFSET_IO_REGISTERS == OFFSET_CPU_REGISTERS + NO_CPU_REGISTERS);
+
+    if (addr < NO_CPU_REGISTERS) {
+        /* CPU registers */
+        env->r[addr] = data;
+    } else if (addr < NO_CPU_REGISTERS + NO_IO_REGISTERS) {
+        /* IO registers */
+        helper_outb(env, addr - NO_CPU_REGISTERS, data);
+    } else {
+        /* memory */
+        cpu_physical_memory_write(OFFSET_DATA + addr, &data, 1);
+    }
+}
target/avr/helper.h new file +29 -0
new file mode 100644
index 0000000000..bf087504a8
@@ -0,0 +1,29 @@
+/*
+ * QEMU AVR CPU
+ *
+ * Copyright (c) 2019 Michael Rolnik
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see
+ * <http://www.gnu.org/licenses/lgpl-2.1.html>
+ */
+
+DEF_HELPER_1(wdr, void, env)
+DEF_HELPER_1(debug, void, env)
+DEF_HELPER_1(break, void, env)
+DEF_HELPER_1(sleep, void, env)
+DEF_HELPER_1(unsupported, void, env)
+DEF_HELPER_3(outb, void, env, i32, i32)
+DEF_HELPER_2(inb, tl, env, i32)
+DEF_HELPER_3(fullwr, void, env, i32, i32)
+DEF_HELPER_2(fullrd, tl, env, i32)
Signature · 1 line
2.17.2 (Apple Git-113)
#1 lore raw

PATCH v34 1/13 target/avr: Add outward facing interfaces and core CPU logic

In reply to Michael Rolnik

This includes:
- CPU data structures
- object model classes and functions
- migration functions
- GDB hooks

---
 gdb-xml/avr-cpu.xml    |  49 ++++
 target/avr/cpu-param.h |  37 +++
 target/avr/cpu-qom.h   |  54 ++++
 target/avr/cpu.c       | 576 +++++++++++++++++++++++++++++++++++++++++
 target/avr/cpu.h       | 253 ++++++++++++++++++
 target/avr/gdbstub.c   |  85 ++++++
 target/avr/machine.c   | 121 +++++++++
 7 files changed, 1175 insertions(+)
 create mode 100644 gdb-xml/avr-cpu.xml
 create mode 100644 target/avr/cpu-param.h
 create mode 100644 target/avr/cpu-qom.h
 create mode 100644 target/avr/cpu.c
 create mode 100644 target/avr/cpu.h
 create mode 100644 target/avr/gdbstub.c
 create mode 100644 target/avr/machine.c
gdb-xml/avr-cpu.xml new file +49 -0
new file mode 100644
index 0000000000..c4747f5b40
@@ -0,0 +1,49 @@
+<?xml version="1.0"?>
+<!-- Copyright (C) 2018-2019 Free Software Foundation, Inc.
+
+     Copying and distribution of this file, with or without modification,
+     are permitted in any medium without royalty provided the copyright
+     notice and this notice are preserved.  -->
+
+<!-- Register numbers are hard-coded in order to maintain backward
+     compatibility with older versions of tools that didn't use xml
+     register descriptions.  -->
+
+<!DOCTYPE feature SYSTEM "gdb-target.dtd">
+<feature name="org.gnu.gdb.riscv.cpu">
+  <reg name="r0" bitsize="8" type="int" regnum="0"/>
+  <reg name="r1" bitsize="8" type="int"/>
+  <reg name="r2" bitsize="8" type="int"/>
+  <reg name="r3" bitsize="8" type="int"/>
+  <reg name="r4" bitsize="8" type="int"/>
+  <reg name="r5" bitsize="8" type="int"/>
+  <reg name="r6" bitsize="8" type="int"/>
+  <reg name="r7" bitsize="8" type="int"/>
+  <reg name="r8" bitsize="8" type="int"/>
+  <reg name="r9" bitsize="8" type="int"/>
+  <reg name="r10" bitsize="8" type="int"/>
+  <reg name="r11" bitsize="8" type="int"/>
+  <reg name="r12" bitsize="8" type="int"/>
+  <reg name="r13" bitsize="8" type="int"/>
+  <reg name="r14" bitsize="8" type="int"/>
+  <reg name="r15" bitsize="8" type="int"/>
+  <reg name="r16" bitsize="8" type="int"/>
+  <reg name="r17" bitsize="8" type="int"/>
+  <reg name="r18" bitsize="8" type="int"/>
+  <reg name="r19" bitsize="8" type="int"/>
+  <reg name="r20" bitsize="8" type="int"/>
+  <reg name="r21" bitsize="8" type="int"/>
+  <reg name="r22" bitsize="8" type="int"/>
+  <reg name="r23" bitsize="8" type="int"/>
+  <reg name="r24" bitsize="8" type="int"/>
+  <reg name="r25" bitsize="8" type="int"/>
+  <reg name="r26" bitsize="8" type="int"/>
+  <reg name="r27" bitsize="8" type="int"/>
+  <reg name="r28" bitsize="8" type="int"/>
+  <reg name="r29" bitsize="8" type="int"/>
+  <reg name="r30" bitsize="8" type="int"/>
+  <reg name="r31" bitsize="8" type="int"/>
+  <reg name="sreg" bitsize="8" type="int"/>
+  <reg name="sp" bitsize="8" type="int"/>
+  <reg name="pc" bitsize="8" type="int"/>
+</feature>
target/avr/cpu-param.h new file +37 -0
new file mode 100644
index 0000000000..ccd1ea3429
@@ -0,0 +1,37 @@
+/*
+ * QEMU AVR CPU
+ *
+ * Copyright (c) 2019 Michael Rolnik
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see
+ * <http://www.gnu.org/licenses/lgpl-2.1.html>
+ */
+
+#ifndef AVR_CPU_PARAM_H
+#define AVR_CPU_PARAM_H 1
+
+#define TARGET_LONG_BITS 32
+/*
+ * TARGET_PAGE_BITS cannot be more than 8 bits because
+ * 1.  all IO registers occupy [0x0000 .. 0x00ff] address range, and they
+ *     should be implemented as a device and not memory
+ * 2.  SRAM starts at the address 0x0100
+ */
+#define TARGET_PAGE_BITS 8
+#define TARGET_PHYS_ADDR_SPACE_BITS 24
+#define TARGET_VIRT_ADDR_SPACE_BITS 24
+#define NB_MMU_MODES 2
+
+
+#endif
target/avr/cpu-qom.h new file +54 -0

54 lines not on this page · raw message

target/avr/cpu.c new file +576 -0

576 lines not on this page · raw message

target/avr/cpu.h new file +253 -0

253 lines not on this page · raw message

target/avr/gdbstub.c new file +85 -0

85 lines not on this page · raw message

target/avr/machine.c new file +121 -0

121 lines not on this page · raw message

Signature · 1 line
2.17.2 (Apple Git-113)
#7 lore raw

PATCH v34 7/13 target/avr: Add instruction translation - Bit and Bit-test Instructions

In reply to Michael Rolnik

This includes:
- LSR, ROR
- ASR
- SWAP
- SBI, CBI
- BST, BLD
- BSET, BCLR

---
 target/avr/translate.c | 1123 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 1123 insertions(+)
target/avr/translate.c +1123 -0
index 4ef3294a40..f2ec2e2d2f 100644
@@ -304,6 +304,15 @@ static void gen_goto_tb(DisasContext *ctx, int n, target_ulong dest)
 }
 
 
+static void gen_rshift_ZNVSf(TCGv R)
+{
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+    tcg_gen_shri_tl(cpu_Nf, R, 7); /* Nf = R(7) */
+    tcg_gen_xor_tl(cpu_Vf, cpu_Nf, cpu_Cf);
+    tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */
+}
+
+
 /*
  *  Adds two registers without the C Flag and places the result in the
  *  destination register Rd.
@@ -1495,3 +1504,1117 @@ static bool trans_BRBS(DisasContext *ctx, arg_BRBS *a)
     return true;
 }
 
+
+/*
+ *  This instruction makes a copy of one register into another. The source
+ *  register Rr is left unchanged, while the destination register Rd is loaded
+ *  with a copy of Rr.
+ */
+static bool trans_MOV(DisasContext *ctx, arg_MOV *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+
+    tcg_gen_mov_tl(Rd, Rr);
+
+    return true;
+}
+
+
+/*
+ *  This instruction makes a copy of one register pair into another register
+ *  pair. The source register pair Rr+1:Rr is left unchanged, while the
+ *  destination register pair Rd+1:Rd is loaded with a copy of Rr + 1:Rr.  This
+ *  instruction is not available in all devices. Refer to the device specific
+ *  instruction set summary.
+ */
+static bool trans_MOVW(DisasContext *ctx, arg_MOVW *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MOVW)) {
+        return true;
+    }
+
+    TCGv RdL = cpu_r[a->rd];
+    TCGv RdH = cpu_r[a->rd + 1];
+    TCGv RrL = cpu_r[a->rr];
+    TCGv RrH = cpu_r[a->rr + 1];
+
+    tcg_gen_mov_tl(RdH, RrH);
+    tcg_gen_mov_tl(RdL, RrL);
+
+    return true;
+}
+
+
+/*
+ * Loads an 8 bit constant directly to register 16 to 31.
+ */
+static bool trans_LDI(DisasContext *ctx, arg_LDI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    int imm = a->imm;
+
+    tcg_gen_movi_tl(Rd, imm);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte from the data space to a register. For parts with SRAM,
+ *  the data space consists of the Register File, I/O memory and internal SRAM
+ *  (and external SRAM if applicable). For parts without SRAM, the data space
+ *  consists of the register file only. The EEPROM has a separate address space.
+ *  A 16-bit address must be supplied. Memory access is limited to the current
+ *  data segment of 64KB. The LDS instruction uses the RAMPD Register to access
+ *  memory above 64KB. To access another data segment in devices with more than
+ *  64KB data space, the RAMPD in register in the I/O area has to be changed.
+ *  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_LDS(DisasContext *ctx, arg_LDS *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = tcg_temp_new_i32();
+    TCGv H = cpu_rampD;
+    a->imm = next_word(ctx);
+
+    tcg_gen_mov_tl(addr, H); /* addr = H:M:L */
+    tcg_gen_shli_tl(addr, addr, 16);
+    tcg_gen_ori_tl(addr, addr, a->imm);
+
+    gen_data_load(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte indirect from the data space to a register. For parts
+ *  with SRAM, the data space consists of the Register File, I/O memory and
+ *  internal SRAM (and external SRAM if applicable). For parts without SRAM, the
+ *  data space consists of the Register File only. In some parts the Flash
+ *  Memory has been mapped to the data space and can be read using this command.
+ *  The EEPROM has a separate address space.  The data location is pointed to by
+ *  the X (16 bits) Pointer Register in the Register File. Memory access is
+ *  limited to the current data segment of 64KB. To access another data segment
+ *  in devices with more than 64KB data space, the RAMPX in register in the I/O
+ *  area has to be changed.  The X-pointer Register can either be left unchanged
+ *  by the operation, or it can be post-incremented or predecremented.  These
+ *  features are especially suited for accessing arrays, tables, and Stack
+ *  Pointer usage of the X-pointer Register. Note that only the low byte of the
+ *  X-pointer is updated in devices with no more than 256 bytes data space. For
+ *  such devices, the high byte of the pointer is not used by this instruction
+ *  and can be used for other purposes. The RAMPX Register in the I/O area is
+ *  updated in parts with more than 64KB data space or more than 64KB Program
+ *  memory, and the increment/decrement is added to the entire 24-bit address on
+ *  such devices.  Not all variants of this instruction is available in all
+ *  devices. Refer to the device specific instruction set summary.  In the
+ *  Reduced Core tinyAVR the LD instruction can be used to achieve the same
+ *  operation as LPM since the program memory is mapped to the data memory
+ *  space.
+ */
+static bool trans_LDX1(DisasContext *ctx, arg_LDX1 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_xaddr();
+
+    gen_data_load(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDX2(DisasContext *ctx, arg_LDX2 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_xaddr();
+
+    gen_data_load(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    gen_set_xaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDX3(DisasContext *ctx, arg_LDX3 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_xaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_load(ctx, Rd, addr);
+    gen_set_xaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STX1(DisasContext *ctx, arg_STX1 *a)
+{
+    TCGv Rd = cpu_r[a->rr];
+    TCGv addr = gen_get_xaddr();
+
+    gen_data_store(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+static bool trans_STX2(DisasContext *ctx, arg_STX2 *a)
+{
+    TCGv Rd = cpu_r[a->rr];
+    TCGv addr = gen_get_xaddr();
+
+    gen_data_store(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+    gen_set_xaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+static bool trans_STX3(DisasContext *ctx, arg_STX3 *a)
+{
+    TCGv Rd = cpu_r[a->rr];
+    TCGv addr = gen_get_xaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_store(ctx, Rd, addr);
+    gen_set_xaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte indirect with or without displacement from the data space
+ *  to a register. For parts with SRAM, the data space consists of the Register
+ *  File, I/O memory and internal SRAM (and external SRAM if applicable). For
+ *  parts without SRAM, the data space consists of the Register File only. In
+ *  some parts the Flash Memory has been mapped to the data space and can be
+ *  read using this command. The EEPROM has a separate address space.  The data
+ *  location is pointed to by the Y (16 bits) Pointer Register in the Register
+ *  File. Memory access is limited to the current data segment of 64KB. To
+ *  access another data segment in devices with more than 64KB data space, the
+ *  RAMPY in register in the I/O area has to be changed.  The Y-pointer Register
+ *  can either be left unchanged by the operation, or it can be post-incremented
+ *  or predecremented.  These features are especially suited for accessing
+ *  arrays, tables, and Stack Pointer usage of the Y-pointer Register. Note that
+ *  only the low byte of the Y-pointer is updated in devices with no more than
+ *  256 bytes data space. For such devices, the high byte of the pointer is not
+ *  used by this instruction and can be used for other purposes. The RAMPY
+ *  Register in the I/O area is updated in parts with more than 64KB data space
+ *  or more than 64KB Program memory, and the increment/decrement/displacement
+ *  is added to the entire 24-bit address on such devices.  Not all variants of
+ *  this instruction is available in all devices. Refer to the device specific
+ *  instruction set summary.  In the Reduced Core tinyAVR the LD instruction can
+ *  be used to achieve the same operation as LPM since the program memory is
+ *  mapped to the data memory space.
+ */
+static bool trans_LDY2(DisasContext *ctx, arg_LDY2 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    gen_data_load(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    gen_set_yaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDY3(DisasContext *ctx, arg_LDY3 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_load(ctx, Rd, addr);
+    gen_set_yaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDDY(DisasContext *ctx, arg_LDDY *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    tcg_gen_addi_tl(addr, addr, a->imm); /* addr = addr + q */
+    gen_data_load(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STY2(DisasContext *ctx, arg_STY2 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    gen_data_store(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+    gen_set_yaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STY3(DisasContext *ctx, arg_STY3 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_store(ctx, Rd, addr);
+    gen_set_yaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STDY(DisasContext *ctx, arg_STDY *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_yaddr();
+
+    tcg_gen_addi_tl(addr, addr, a->imm); /* addr = addr + q */
+    gen_data_store(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte indirect with or without displacement from the data space
+ *  to a register. For parts with SRAM, the data space consists of the Register
+ *  File, I/O memory and internal SRAM (and external SRAM if applicable). For
+ *  parts without SRAM, the data space consists of the Register File only. In
+ *  some parts the Flash Memory has been mapped to the data space and can be
+ *  read using this command. The EEPROM has a separate address space.  The data
+ *  location is pointed to by the Z (16 bits) Pointer Register in the Register
+ *  File. Memory access is limited to the current data segment of 64KB. To
+ *  access another data segment in devices with more than 64KB data space, the
+ *  RAMPZ in register in the I/O area has to be changed.  The Z-pointer Register
+ *  can either be left unchanged by the operation, or it can be post-incremented
+ *  or predecremented.  These features are especially suited for Stack Pointer
+ *  usage of the Z-pointer Register, however because the Z-pointer Register can
+ *  be used for indirect subroutine calls, indirect jumps and table lookup, it
+ *  is often more convenient to use the X or Y-pointer as a dedicated Stack
+ *  Pointer. Note that only the low byte of the Z-pointer is updated in devices
+ *  with no more than 256 bytes data space. For such devices, the high byte of
+ *  the pointer is not used by this instruction and can be used for other
+ *  purposes. The RAMPZ Register in the I/O area is updated in parts with more
+ *  than 64KB data space or more than 64KB Program memory, and the
+ *  increment/decrement/displacement is added to the entire 24-bit address on
+ *  such devices.  Not all variants of this instruction is available in all
+ *  devices. Refer to the device specific instruction set summary.  In the
+ *  Reduced Core tinyAVR the LD instruction can be used to achieve the same
+ *  operation as LPM since the program memory is mapped to the data memory
+ *  space.  For using the Z-pointer for table lookup in Program memory see the
+ *  LPM and ELPM instructions.
+ */
+static bool trans_LDZ2(DisasContext *ctx, arg_LDZ2 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    gen_data_load(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    gen_set_zaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDZ3(DisasContext *ctx, arg_LDZ3 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_load(ctx, Rd, addr);
+
+    gen_set_zaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LDDZ(DisasContext *ctx, arg_LDDZ *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_addi_tl(addr, addr, a->imm); /* addr = addr + q */
+    gen_data_load(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STZ2(DisasContext *ctx, arg_STZ2 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    gen_data_store(ctx, Rd, addr);
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    gen_set_zaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STZ3(DisasContext *ctx, arg_STZ3 *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_subi_tl(addr, addr, 1); /* addr = addr - 1 */
+    gen_data_store(ctx, Rd, addr);
+
+    gen_set_zaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_STDZ(DisasContext *ctx, arg_STDZ *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_addi_tl(addr, addr, a->imm); /* addr = addr + q */
+    gen_data_store(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+
+/*
+ *  Stores one byte from a Register to the data space. For parts with SRAM,
+ *  the data space consists of the Register File, I/O memory and internal SRAM
+ *  (and external SRAM if applicable). For parts without SRAM, the data space
+ *  consists of the Register File only. The EEPROM has a separate address space.
+ *  A 16-bit address must be supplied. Memory access is limited to the current
+ *  data segment of 64KB. The STS instruction uses the RAMPD Register to access
+ *  memory above 64KB. To access another data segment in devices with more than
+ *  64KB data space, the RAMPD in register in the I/O area has to be changed.
+ *  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_STS(DisasContext *ctx, arg_STS *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = tcg_temp_new_i32();
+    TCGv H = cpu_rampD;
+    a->imm = next_word(ctx);
+
+    tcg_gen_mov_tl(addr, H); /* addr = H:M:L */
+    tcg_gen_shli_tl(addr, addr, 16);
+    tcg_gen_ori_tl(addr, addr, a->imm);
+
+    gen_data_store(ctx, Rd, addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte pointed to by the Z-register into the destination
+ *  register Rd. This instruction features a 100% space effective constant
+ *  initialization or constant data fetch. The Program memory is organized in
+ *  16-bit words while the Z-pointer is a byte address. Thus, the least
+ *  significant bit of the Z-pointer selects either low byte (ZLSB = 0) or high
+ *  byte (ZLSB = 1). This instruction can address the first 64KB (32K words) of
+ *  Program memory. The Zpointer Register can either be left unchanged by the
+ *  operation, or it can be incremented. The incrementation does not apply to
+ *  the RAMPZ Register.  Devices with Self-Programming capability can use the
+ *  LPM instruction to read the Fuse and Lock bit values.  Refer to the device
+ *  documentation for a detailed description.  The LPM instruction is not
+ *  available in all devices. Refer to the device specific instruction set
+ *  summary
+ */
+static bool trans_LPM1(DisasContext *ctx, arg_LPM1 *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_LPM)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[0];
+    TCGv addr = tcg_temp_new_i32();
+    TCGv H = cpu_r[31];
+    TCGv L = cpu_r[30];
+
+    tcg_gen_shli_tl(addr, H, 8); /* addr = H:L */
+    tcg_gen_or_tl(addr, addr, L);
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LPM2(DisasContext *ctx, arg_LPM2 *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_LPM)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = tcg_temp_new_i32();
+    TCGv H = cpu_r[31];
+    TCGv L = cpu_r[30];
+
+    tcg_gen_shli_tl(addr, H, 8); /* addr = H:L */
+    tcg_gen_or_tl(addr, addr, L);
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_LPMX(DisasContext *ctx, arg_LPMX *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_LPMX)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = tcg_temp_new_i32();
+    TCGv H = cpu_r[31];
+    TCGv L = cpu_r[30];
+
+    tcg_gen_shli_tl(addr, H, 8); /* addr = H:L */
+    tcg_gen_or_tl(addr, addr, L);
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    tcg_gen_andi_tl(L, addr, 0xff);
+
+    tcg_gen_shri_tl(addr, addr, 8);
+    tcg_gen_andi_tl(H, addr, 0xff);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Loads one byte pointed to by the Z-register and the RAMPZ Register in
+ *  the I/O space, and places this byte in the destination register Rd. This
+ *  instruction features a 100% space effective constant initialization or
+ *  constant data fetch. The Program memory is organized in 16-bit words while
+ *  the Z-pointer is a byte address. Thus, the least significant bit of the
+ *  Z-pointer selects either low byte (ZLSB = 0) or high byte (ZLSB = 1). This
+ *  instruction can address the entire Program memory space. The Z-pointer
+ *  Register can either be left unchanged by the operation, or it can be
+ *  incremented. The incrementation applies to the entire 24-bit concatenation
+ *  of the RAMPZ and Z-pointer Registers.  Devices with Self-Programming
+ *  capability can use the ELPM instruction to read the Fuse and Lock bit value.
+ *  Refer to the device documentation for a detailed description.  This
+ *  instruction is not available in all devices. Refer to the device specific
+ *  instruction set summary.
+ */
+static bool trans_ELPM1(DisasContext *ctx, arg_ELPM1 *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_ELPM)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[0];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_ELPM2(DisasContext *ctx, arg_ELPM2 *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_ELPM)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+static bool trans_ELPMX(DisasContext *ctx, arg_ELPMX *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_ELPMX)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+
+    tcg_gen_qemu_ld8u(Rd, addr, MMU_CODE_IDX); /* Rd = mem[addr] */
+
+    tcg_gen_addi_tl(addr, addr, 1); /* addr = addr + 1 */
+
+    gen_set_zaddr(addr);
+
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  SPM can be used to erase a page in the Program memory, to write a page
+ *  in the Program memory (that is already erased), and to set Boot Loader Lock
+ *  bits. In some devices, the Program memory can be written one word at a time,
+ *  in other devices an entire page can be programmed simultaneously after first
+ *  filling a temporary page buffer. In all cases, the Program memory must be
+ *  erased one page at a time. When erasing the Program memory, the RAMPZ and
+ *  Z-register are used as page address. When writing the Program memory, the
+ *  RAMPZ and Z-register are used as page or word address, and the R1:R0
+ *  register pair is used as data(1). When setting the Boot Loader Lock bits,
+ *  the R1:R0 register pair is used as data. Refer to the device documentation
+ *  for detailed description of SPM usage. This instruction can address the
+ *  entire Program memory.  The SPM instruction is not available in all devices.
+ *  Refer to the device specific instruction set summary.  Note: 1. R1
+ *  determines the instruction high byte, and R0 determines the instruction low
+ *  byte.
+ */
+static bool trans_SPM(DisasContext *ctx, arg_SPM *a)
+{
+    /* TODO */
+    if (!avr_have_feature(ctx, AVR_FEATURE_SPM)) {
+        return true;
+    }
+
+    return true;
+}
+
+
+static bool trans_SPMX(DisasContext *ctx, arg_SPMX *a)
+{
+    /* TODO */
+    if (!avr_have_feature(ctx, AVR_FEATURE_SPMX)) {
+        return true;
+    }
+
+    return true;
+}
+
+
+/*
+ *  Loads data from the I/O Space (Ports, Timers, Configuration Registers,
+ *  etc.) into register Rd in the Register File.
+ */
+static bool trans_IN(DisasContext *ctx, arg_IN *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv port = tcg_const_i32(a->imm);
+
+    gen_helper_inb(Rd, cpu_env, port);
+
+    tcg_temp_free_i32(port);
+
+    return true;
+}
+
+
+/*
+ *  Stores data from register Rr in the Register File to I/O Space (Ports,
+ *  Timers, Configuration Registers, etc.).
+ */
+static bool trans_OUT(DisasContext *ctx, arg_OUT *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv port = tcg_const_i32(a->imm);
+
+    gen_helper_outb(cpu_env, port, Rd);
+
+    tcg_temp_free_i32(port);
+
+    return true;
+}
+
+
+/*
+ *  This instruction stores the contents of register Rr on the STACK. The
+ *  Stack Pointer is post-decremented by 1 after the PUSH.  This instruction is
+ *  not available in all devices. Refer to the device specific instruction set
+ *  summary.
+ */
+static bool trans_PUSH(DisasContext *ctx, arg_PUSH *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+
+    gen_data_store(ctx, Rd, cpu_sp);
+    tcg_gen_subi_tl(cpu_sp, cpu_sp, 1);
+
+    return true;
+}
+
+
+/*
+ *  This instruction loads register Rd with a byte from the STACK. The Stack
+ *  Pointer is pre-incremented by 1 before the POP.  This instruction is not
+ *  available in all devices. Refer to the device specific instruction set
+ *  summary.
+ */
+static bool trans_POP(DisasContext *ctx, arg_POP *a)
+{
+    /*
+     * Using a temp to work around some strange behaviour:
+     * tcg_gen_addi_tl(cpu_sp, cpu_sp, 1);
+     * gen_data_load(ctx, Rd, cpu_sp);
+     * seems to cause the add to happen twice.
+     * This doesn't happen if either the add or the load is removed.
+     */
+    TCGv t1 = tcg_temp_new_i32();
+    TCGv Rd = cpu_r[a->rd];
+
+    tcg_gen_addi_tl(t1, cpu_sp, 1);
+    gen_data_load(ctx, Rd, t1);
+    tcg_gen_mov_tl(cpu_sp, t1);
+
+    return true;
+}
+
+
+/*
+ *  Exchanges one byte indirect between register and data space.  The data
+ *  location is pointed to by the Z (16 bits) Pointer Register in the Register
+ *  File. Memory access is limited to the current data segment of 64KB. To
+ *  access another data segment in devices with more than 64KB data space, the
+ *  RAMPZ in register in the I/O area has to be changed.  The Z-pointer Register
+ *  is left unchanged by the operation. This instruction is especially suited
+ *  for writing/reading status bits stored in SRAM.
+ */
+static bool trans_XCH(DisasContext *ctx, arg_XCH *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_RMW)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv addr = gen_get_zaddr();
+
+    gen_data_load(ctx, t0, addr);
+    gen_data_store(ctx, Rd, addr);
+    tcg_gen_mov_tl(Rd, t0);
+
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Load one byte indirect from data space to register and set bits in data
+ *  space specified by the register. The instruction can only be used towards
+ *  internal SRAM.  The data location is pointed to by the Z (16 bits) Pointer
+ *  Register in the Register File. Memory access is limited to the current data
+ *  segment of 64KB. To access another data segment in devices with more than
+ *  64KB data space, the RAMPZ in register in the I/O area has to be changed.
+ *  The Z-pointer Register is left unchanged by the operation. This instruction
+ *  is especially suited for setting status bits stored in SRAM.
+ */
+static bool trans_LAS(DisasContext *ctx, arg_LAS *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_RMW)) {
+        return true;
+    }
+
+    TCGv Rr = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    gen_data_load(ctx, t0, addr); /* t0 = mem[addr] */
+    tcg_gen_or_tl(t1, t0, Rr);
+
+    tcg_gen_mov_tl(Rr, t0); /* Rr = t0 */
+    gen_data_store(ctx, t1, addr); /* mem[addr] = t1 */
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Load one byte indirect from data space to register and stores and clear
+ *  the bits in data space specified by the register. The instruction can
+ *  only be used towards internal SRAM.  The data location is pointed to by
+ *  the Z (16 bits) Pointer Register in the Register File. Memory access is
+ *  limited to the current data segment of 64KB. To access another data
+ *  segment in devices with more than 64KB data space, the RAMPZ in register
+ *  in the I/O area has to be changed.  The Z-pointer Register is left
+ *  unchanged by the operation. This instruction is especially suited for
+ *  clearing status bits stored in SRAM.
+ */
+static bool trans_LAC(DisasContext *ctx, arg_LAC *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_RMW)) {
+        return true;
+    }
+
+    TCGv Rr = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    gen_data_load(ctx, t0, addr); /* t0 = mem[addr] */
+        /* t1 = t0 & (0xff - Rr) = t0 and ~Rr */
+    tcg_gen_andc_tl(t1, t0, Rr);
+
+    tcg_gen_mov_tl(Rr, t0); /* Rr = t0 */
+    gen_data_store(ctx, t1, addr); /* mem[addr] = t1 */
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Load one byte indirect from data space to register and toggles bits in
+ *  the data space specified by the register.  The instruction can only be used
+ *  towards SRAM.  The data location is pointed to by the Z (16 bits) Pointer
+ *  Register in the Register File. Memory access is limited to the current data
+ *  segment of 64KB. To access another data segment in devices with more than
+ *  64KB data space, the RAMPZ in register in the I/O area has to be changed.
+ *  The Z-pointer Register is left unchanged by the operation. This instruction
+ *  is especially suited for changing status bits stored in SRAM.
+ */
+static bool trans_LAT(DisasContext *ctx, arg_LAT *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_RMW)) {
+        return true;
+    }
+
+    TCGv Rd = cpu_r[a->rd];
+    TCGv addr = gen_get_zaddr();
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    gen_data_load(ctx, t0, addr); /* t0 = mem[addr] */
+    tcg_gen_xor_tl(t1, t0, Rd);
+
+    tcg_gen_mov_tl(Rd, t0); /* Rd = t0 */
+    gen_data_store(ctx, t1, addr); /* mem[addr] = t1 */
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(addr);
+
+    return true;
+}
+
+
+/*
+ *  Shifts all bits in Rd one place to the right. Bit 7 is cleared. Bit 0 is
+ *  loaded into the C Flag of the SREG. This operation effectively divides an
+ *  unsigned value by two. The C Flag can be used to round the result.
+ */
+static bool trans_LSR(DisasContext *ctx, arg_LSR *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+
+    tcg_gen_andi_tl(cpu_Cf, Rd, 1);
+
+    tcg_gen_shri_tl(Rd, Rd, 1);
+
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, Rd, 0); /* Zf = Rd == 0 */
+    tcg_gen_movi_tl(cpu_Nf, 0);
+    tcg_gen_mov_tl(cpu_Vf, cpu_Cf);
+    tcg_gen_mov_tl(cpu_Sf, cpu_Vf);
+
+    return true;
+}
+
+
+/*
+ *  Shifts all bits in Rd one place to the right. The C Flag is shifted into
+ *  bit 7 of Rd. Bit 0 is shifted into the C Flag.  This operation, combined
+ *  with ASR, effectively divides multi-byte signed values by two. Combined with
+ *  LSR it effectively divides multi-byte unsigned values by two. The Carry Flag
+ *  can be used to round the result.
+ */
+static bool trans_ROR(DisasContext *ctx, arg_ROR *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t0 = tcg_temp_new_i32();
+
+    tcg_gen_shli_tl(t0, cpu_Cf, 7);
+    tcg_gen_andi_tl(cpu_Cf, Rd, 1);
+    tcg_gen_shri_tl(Rd, Rd, 1);
+    tcg_gen_or_tl(Rd, Rd, t0);
+
+    gen_rshift_ZNVSf(Rd);
+
+    tcg_temp_free_i32(t0);
+
+    return true;
+}
+
+
+/*
+ *  Shifts all bits in Rd one place to the right. Bit 7 is held constant. Bit 0
+ *  is loaded into the C Flag of the SREG. This operation effectively divides a
+ *  signed value by two without changing its sign. The Carry Flag can be used to
+ *  round the result.
+ */
+static bool trans_ASR(DisasContext *ctx, arg_ASR *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t0 = tcg_temp_new_i32();
+
+    /* Cf */
+    tcg_gen_andi_tl(cpu_Cf, Rd, 1); /* Cf = Rd(0) */
+
+    /* op */
+    tcg_gen_andi_tl(t0, Rd, 0x80); /* Rd = (Rd & 0x80) | (Rd >> 1) */
+    tcg_gen_shri_tl(Rd, Rd, 1);
+    tcg_gen_or_tl(Rd, Rd, t0);
+
+    gen_rshift_ZNVSf(Rd);
+
+    tcg_temp_free_i32(t0);
+
+    return true;
+}
+
+
+/*
+ *  Swaps high and low nibbles in a register.
+ */
+static bool trans_SWAP(DisasContext *ctx, arg_SWAP *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    tcg_gen_andi_tl(t0, Rd, 0x0f);
+    tcg_gen_shli_tl(t0, t0, 4);
+    tcg_gen_andi_tl(t1, Rd, 0xf0);
+    tcg_gen_shri_tl(t1, t1, 4);
+    tcg_gen_or_tl(Rd, t0, t1);
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+
+    return true;
+}
+
+
+/*
+ *  Sets a specified bit in an I/O Register. This instruction operates on
+ *  the lower 32 I/O Registers -- addresses 0-31.
+ */
+static bool trans_SBI(DisasContext *ctx, arg_SBI *a)
+{
+    TCGv data = tcg_temp_new_i32();
+    TCGv port = tcg_const_i32(a->reg);
+
+    gen_helper_inb(data, cpu_env, port);
+    tcg_gen_ori_tl(data, data, 1 << a->bit);
+    gen_helper_outb(cpu_env, port, data);
+
+    tcg_temp_free_i32(port);
+    tcg_temp_free_i32(data);
+
+    return true;
+}
+
+
+/*
+ *  Clears a specified bit in an I/O Register. This instruction operates on
+ *  the lower 32 I/O Registers -- addresses 0-31.
+ */
+static bool trans_CBI(DisasContext *ctx, arg_CBI *a)
+{
+    TCGv data = tcg_temp_new_i32();
+    TCGv port = tcg_const_i32(a->reg);
+
+    gen_helper_inb(data, cpu_env, port);
+    tcg_gen_andi_tl(data, data, ~(1 << a->bit));
+    gen_helper_outb(cpu_env, port, data);
+
+    tcg_temp_free_i32(data);
+    tcg_temp_free_i32(port);
+
+    return true;
+}
+
+
+/*
+ *  Stores bit b from Rd to the T Flag in SREG (Status Register).
+ */
+static bool trans_BST(DisasContext *ctx, arg_BST *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+
+    tcg_gen_andi_tl(cpu_Tf, Rd, 1 << a->bit);
+    tcg_gen_shri_tl(cpu_Tf, cpu_Tf, a->bit);
+
+    return true;
+}
+
+
+/*
+ *  Copies the T Flag in the SREG (Status Register) to bit b in register Rd.
+ */
+static bool trans_BLD(DisasContext *ctx, arg_BLD *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t1 = tcg_temp_new_i32();
+
+    tcg_gen_andi_tl(Rd, Rd, ~(1u << a->bit)); /* clear bit */
+    tcg_gen_shli_tl(t1, cpu_Tf, a->bit); /* create mask */
+    tcg_gen_or_tl(Rd, Rd, t1);
+
+    tcg_temp_free_i32(t1);
+
+    return true;
+}
+
+
+/*
+ *  Sets a single Flag or bit in SREG.
+ */
+static bool trans_BSET(DisasContext *ctx, arg_BSET *a)
+{
+    switch (a->bit) {
+    case 0x00:
+        tcg_gen_movi_tl(cpu_Cf, 0x01);
+        break;
+    case 0x01:
+        tcg_gen_movi_tl(cpu_Zf, 0x01);
+        break;
+    case 0x02:
+        tcg_gen_movi_tl(cpu_Nf, 0x01);
+        break;
+    case 0x03:
+        tcg_gen_movi_tl(cpu_Vf, 0x01);
+        break;
+    case 0x04:
+        tcg_gen_movi_tl(cpu_Sf, 0x01);
+        break;
+    case 0x05:
+        tcg_gen_movi_tl(cpu_Hf, 0x01);
+        break;
+    case 0x06:
+        tcg_gen_movi_tl(cpu_Tf, 0x01);
+        break;
+    case 0x07:
+        tcg_gen_movi_tl(cpu_If, 0x01);
+        break;
+    }
+
+    return true;
+}
+
+
+/*
+ *  Clears a single Flag in SREG.
+ */
+static bool trans_BCLR(DisasContext *ctx, arg_BCLR *a)
+{
+    switch (a->bit) {
+    case 0x00:
+        tcg_gen_movi_tl(cpu_Cf, 0x00);
+        break;
+    case 0x01:
+        tcg_gen_movi_tl(cpu_Zf, 0x00);
+        break;
+    case 0x02:
+        tcg_gen_movi_tl(cpu_Nf, 0x00);
+        break;
+    case 0x03:
+        tcg_gen_movi_tl(cpu_Vf, 0x00);
+        break;
+    case 0x04:
+        tcg_gen_movi_tl(cpu_Sf, 0x00);
+        break;
+    case 0x05:
+        tcg_gen_movi_tl(cpu_Hf, 0x00);
+        break;
+    case 0x06:
+        tcg_gen_movi_tl(cpu_Tf, 0x00);
+        break;
+    case 0x07:
+        tcg_gen_movi_tl(cpu_If, 0x00);
+        break;
+    }
+
+    return true;
+}
Signature · 1 line
2.17.2 (Apple Git-113)
#4 lore raw

PATCH v34 4/13 target/avr: Add instruction translation - Registers definition

In reply to Michael Rolnik
---
 target/avr/translate.c | 132 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 132 insertions(+)
 create mode 100644 target/avr/translate.c
target/avr/translate.c new file +132 -0
new file mode 100644
index 0000000000..53c9892a60
@@ -0,0 +1,132 @@
+/*
+ * QEMU AVR CPU
+ *
+ * Copyright (c) 2019 Michael Rolnik
+ *
+ * This library is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * This library 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
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with this library; if not, see
+ * <http://www.gnu.org/licenses/lgpl-2.1.html>
+ */
+
+#include "qemu/osdep.h"
+#include "qemu/qemu-print.h"
+#include "tcg/tcg.h"
+#include "cpu.h"
+#include "exec/exec-all.h"
+#include "disas/disas.h"
+#include "tcg-op.h"
+#include "exec/cpu_ldst.h"
+#include "exec/helper-proto.h"
+#include "exec/helper-gen.h"
+#include "exec/log.h"
+#include "exec/gdbstub.h"
+#include "exec/translator.h"
+#include "exec/gen-icount.h"
+
+/*
+ *  Define if you want a BREAK instruction translated to a breakpoint
+ *  Active debugging connection is assumed
+ *  This is for
+ *  https://github.com/seharris/qemu-avr-tests/tree/master/instruction-tests
+ *  tests
+ */
+#undef BREAKPOINT_ON_BREAK
+
+static TCGv cpu_pc;
+
+static TCGv cpu_Cf;
+static TCGv cpu_Zf;
+static TCGv cpu_Nf;
+static TCGv cpu_Vf;
+static TCGv cpu_Sf;
+static TCGv cpu_Hf;
+static TCGv cpu_Tf;
+static TCGv cpu_If;
+
+static TCGv cpu_rampD;
+static TCGv cpu_rampX;
+static TCGv cpu_rampY;
+static TCGv cpu_rampZ;
+
+static TCGv cpu_r[NO_CPU_REGISTERS];
+static TCGv cpu_eind;
+static TCGv cpu_sp;
+
+static TCGv cpu_skip;
+
+static const char reg_names[NO_CPU_REGISTERS][8] = {
+    "r0",  "r1",  "r2",  "r3",  "r4",  "r5",  "r6",  "r7",
+    "r8",  "r9",  "r10", "r11", "r12", "r13", "r14", "r15",
+    "r16", "r17", "r18", "r19", "r20", "r21", "r22", "r23",
+    "r24", "r25", "r26", "r27", "r28", "r29", "r30", "r31",
+};
+#define REG(x) (cpu_r[x])
+
+enum {
+    DISAS_EXIT   = DISAS_TARGET_0,  /* We want return to the cpu main loop.  */
+    DISAS_LOOKUP = DISAS_TARGET_1,  /* We have a variable condition exit.  */
+    DISAS_CHAIN  = DISAS_TARGET_2,  /* We have a single condition exit.  */
+};
+
+typedef struct DisasContext DisasContext;
+
+/* This is the state at translation time. */
+struct DisasContext {
+    TranslationBlock *tb;
+
+    CPUAVRState *env;
+    CPUState *cs;
+
+    target_long npc;
+    uint32_t opcode;
+
+    /* Routine used to access memory */
+    int memidx;
+    int bstate;
+    int singlestep;
+
+    TCGv skip_var0;
+    TCGv skip_var1;
+    TCGCond skip_cond;
+    bool free_skip_var0;
+};
+
+static int to_A(DisasContext *ctx, int indx) { return 16 + (indx % 16); }
+static int to_B(DisasContext *ctx, int indx) { return 16 + (indx % 8); }
+static int to_C(DisasContext *ctx, int indx) { return 24 + (indx % 4) * 2; }
+static int to_D(DisasContext *ctx, int indx) { return (indx % 16) * 2; }
+
+static uint16_t next_word(DisasContext *ctx)
+{
+    return cpu_lduw_code(ctx->env, ctx->npc++ * 2);
+}
+
+static int append_16(DisasContext *ctx, int x)
+{
+    return x << 16 | next_word(ctx);
+}
+
+
+static bool avr_have_feature(DisasContext *ctx, int feature)
+{
+    if (!avr_feature(ctx->env, feature)) {
+        gen_helper_unsupported(cpu_env);
+        ctx->bstate = DISAS_NORETURN;
+        return false;
+    }
+    return true;
+}
+
+static bool decode_insn(DisasContext *ctx, uint16_t insn);
+#include "decode_insn.inc.c"
+
Signature · 1 line
2.17.2 (Apple Git-113)
#11 lore raw

PATCH v34 11/13 target/avr: Add example board configuration

In reply to Michael Rolnik

A simple board setup that configures an AVR CPU to run a given firmware image.
This is all that's useful to implement without peripheral emulation as AVR CPUs include a lot of on-board peripherals.

NOTE: this is not a real board !!!!
NOTE: it's used for CPU testing!!!!

---
 hw/Kconfig           |   1 +
 hw/avr/Kconfig       |   6 +
 hw/avr/Makefile.objs |   1 +
 hw/avr/sample.c      | 282 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 290 insertions(+)
 create mode 100644 hw/avr/Kconfig
 create mode 100644 hw/avr/Makefile.objs
 create mode 100644 hw/avr/sample.c
hw/Kconfig +1 -0
index b45db3c813..f185fcb19e 100644
@@ -42,6 +42,7 @@ source watchdog/Kconfig
 # arch Kconfig
 source arm/Kconfig
 source alpha/Kconfig
+source avr/Kconfig
 source cris/Kconfig
 source hppa/Kconfig
 source i386/Kconfig
hw/avr/Kconfig new file +6 -0
new file mode 100644
index 0000000000..92aa1e6afb
@@ -0,0 +1,6 @@
+config AVR_SAMPLE
+    bool
+    select AVR_TIMER16
+    select AVR_USART
+    select AVR_MASK
+    select UNIMP
hw/avr/Makefile.objs new file +1 -0
new file mode 100644
index 0000000000..626b7064b3
@@ -0,0 +1 @@
+obj-y += sample.o
hw/avr/sample.c new file +282 -0

282 lines not on this page · raw message

Signature · 1 line
2.17.2 (Apple Git-113)
#5 lore raw

PATCH v34 5/13 target/avr: Add instruction translation - Arithmetic and Logic Instructions

In reply to Michael Rolnik

This includes:
- ADD, ADC, ADIW
- SBIW, SUB, SUBI, SBC, SBCI
- AND, ANDI
- OR, ORI, EOR
- COM, NEG
- INC, DEC
- MUL, MULS, MULSU
- FMUL, FMULS, FMULSU
- DES

---
 target/avr/translate.c | 822 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 822 insertions(+)
target/avr/translate.c +822 -0
index 53c9892a60..573c9988b5 100644
@@ -130,3 +130,825 @@ static bool avr_have_feature(DisasContext *ctx, int feature)
 static bool decode_insn(DisasContext *ctx, uint16_t insn);
 #include "decode_insn.inc.c"
 
+
+static void gen_add_CHf(TCGv R, TCGv Rd, TCGv Rr)
+{
+    TCGv t1 = tcg_temp_new_i32();
+    TCGv t2 = tcg_temp_new_i32();
+    TCGv t3 = tcg_temp_new_i32();
+
+    tcg_gen_and_tl(t1, Rd, Rr); /* t1 = Rd & Rr */
+    tcg_gen_andc_tl(t2, Rd, R); /* t2 = Rd & ~R */
+    tcg_gen_andc_tl(t3, Rr, R); /* t3 = Rr & ~R */
+    tcg_gen_or_tl(t1, t1, t2); /* t1 = t1 | t2 | t3 */
+    tcg_gen_or_tl(t1, t1, t3);
+
+    tcg_gen_shri_tl(cpu_Cf, t1, 7); /* Cf = t1(7) */
+    tcg_gen_shri_tl(cpu_Hf, t1, 3); /* Hf = t1(3) */
+    tcg_gen_andi_tl(cpu_Hf, cpu_Hf, 1);
+
+    tcg_temp_free_i32(t3);
+    tcg_temp_free_i32(t2);
+    tcg_temp_free_i32(t1);
+}
+
+
+static void gen_add_Vf(TCGv R, TCGv Rd, TCGv Rr)
+{
+    TCGv t1 = tcg_temp_new_i32();
+    TCGv t2 = tcg_temp_new_i32();
+
+    /* t1 = Rd & Rr & ~R | ~Rd & ~Rr & R = (Rd ^ R) & ~(Rd ^ Rr) */
+    tcg_gen_xor_tl(t1, Rd, R);
+    tcg_gen_xor_tl(t2, Rd, Rr);
+    tcg_gen_andc_tl(t1, t1, t2);
+
+    tcg_gen_shri_tl(cpu_Vf, t1, 7); /* Vf = t1(7) */
+
+    tcg_temp_free_i32(t2);
+    tcg_temp_free_i32(t1);
+}
+
+
+static void gen_sub_CHf(TCGv R, TCGv Rd, TCGv Rr)
+{
+    TCGv t1 = tcg_temp_new_i32();
+    TCGv t2 = tcg_temp_new_i32();
+    TCGv t3 = tcg_temp_new_i32();
+
+    /* Cf & Hf */
+    tcg_gen_not_tl(t1, Rd); /* t1 = ~Rd */
+    tcg_gen_and_tl(t2, t1, Rr); /* t2 = ~Rd & Rr */
+    tcg_gen_or_tl(t3, t1, Rr); /* t3 = (~Rd | Rr) & R */
+    tcg_gen_and_tl(t3, t3, R);
+    tcg_gen_or_tl(t2, t2, t3); /* t2 = ~Rd & Rr | ~Rd & R | R & Rr */
+    tcg_gen_shri_tl(cpu_Cf, t2, 7); /* Cf = t2(7) */
+    tcg_gen_shri_tl(cpu_Hf, t2, 3); /* Hf = t2(3) */
+    tcg_gen_andi_tl(cpu_Hf, cpu_Hf, 1);
+
+    tcg_temp_free_i32(t3);
+    tcg_temp_free_i32(t2);
+    tcg_temp_free_i32(t1);
+}
+
+
+static void gen_sub_Vf(TCGv R, TCGv Rd, TCGv Rr)
+{
+    TCGv t1 = tcg_temp_new_i32();
+    TCGv t2 = tcg_temp_new_i32();
+
+    /* Vf */
+    /* t1 = Rd & ~Rr & ~R | ~Rd & Rr & R = (Rd ^ R) & (Rd ^ R) */
+    tcg_gen_xor_tl(t1, Rd, R);
+    tcg_gen_xor_tl(t2, Rd, Rr);
+    tcg_gen_and_tl(t1, t1, t2);
+    tcg_gen_shri_tl(cpu_Vf, t1, 7); /* Vf = t1(7) */
+
+    tcg_temp_free_i32(t2);
+    tcg_temp_free_i32(t1);
+}
+
+
+static void gen_NSf(TCGv R)
+{
+    tcg_gen_shri_tl(cpu_Nf, R, 7); /* Nf = R(7) */
+    tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */
+}
+
+
+static void gen_ZNSf(TCGv R)
+{
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+    tcg_gen_shri_tl(cpu_Nf, R, 7); /* Nf = R(7) */
+    tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */
+}
+
+
+/*
+ *  Adds two registers without the C Flag and places the result in the
+ *  destination register Rd.
+ */
+static bool trans_ADD(DisasContext *ctx, arg_ADD *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_add_tl(R, Rd, Rr); /* Rd = Rd + Rr */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_add_CHf(R, Rd, Rr);
+    gen_add_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Adds two registers and the contents of the C Flag and places the result in
+ *  the destination register Rd.
+ */
+static bool trans_ADC(DisasContext *ctx, arg_ADC *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_add_tl(R, Rd, Rr); /* R = Rd + Rr + Cf */
+    tcg_gen_add_tl(R, R, cpu_Cf);
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_add_CHf(R, Rd, Rr);
+    gen_add_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Subtracts an immediate value (0-63) from a register pair and places the
+ *  result in the register pair. This instruction operates on the upper four
+ *  register pairs, and is well suited for operations on the Pointer Registers.
+ *  This instruction is not available in all devices. Refer to the device
+ *  specific instruction set summary.
+ */
+static bool trans_SBIW(DisasContext *ctx, arg_SBIW *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_ADIW_SBIW)) {
+        return true;
+    }
+
+    TCGv RdL = cpu_r[a->rd];
+    TCGv RdH = cpu_r[a->rd + 1];
+    int Imm = (a->imm);
+    TCGv R = tcg_temp_new_i32();
+    TCGv Rd = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_deposit_tl(Rd, RdL, RdH, 8, 8); /* Rd = RdH:RdL */
+    tcg_gen_subi_tl(R, Rd, Imm); /* R = Rd - Imm */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */
+
+    /* Cf */
+    tcg_gen_andc_tl(cpu_Cf, R, Rd);
+    tcg_gen_shri_tl(cpu_Cf, cpu_Cf, 15); /* Cf = R & ~Rd */
+
+    /* Vf */
+    tcg_gen_andc_tl(cpu_Vf, Rd, R);
+    tcg_gen_shri_tl(cpu_Vf, cpu_Vf, 15); /* Vf = Rd & ~R */
+
+    /* Zf */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    /* Nf */
+    tcg_gen_shri_tl(cpu_Nf, R, 15); /* Nf = R(15) */
+
+    /* Sf */
+    tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */
+
+    /* R */
+    tcg_gen_andi_tl(RdL, R, 0xff);
+    tcg_gen_shri_tl(RdH, R, 8);
+
+    tcg_temp_free_i32(Rd);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Adds an immediate value (0 - 63) to a register pair and places the result
+ *  in the register pair. This instruction operates on the upper four register
+ *  pairs, and is well suited for operations on the pointer registers.  This
+ *  instruction is not available in all devices. Refer to the device specific
+ *  instruction set summary.
+ */
+static bool trans_ADIW(DisasContext *ctx, arg_ADIW *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_ADIW_SBIW)) {
+        return true;
+    }
+
+    TCGv RdL = cpu_r[a->rd];
+    TCGv RdH = cpu_r[a->rd + 1];
+    int Imm = (a->imm);
+    TCGv R = tcg_temp_new_i32();
+    TCGv Rd = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_deposit_tl(Rd, RdL, RdH, 8, 8); /* Rd = RdH:RdL */
+    tcg_gen_addi_tl(R, Rd, Imm); /* R = Rd + Imm */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */
+
+    /* Cf */
+    tcg_gen_andc_tl(cpu_Cf, Rd, R); /* Cf = Rd & ~R */
+    tcg_gen_shri_tl(cpu_Cf, cpu_Cf, 15);
+
+    /* Vf */
+    tcg_gen_andc_tl(cpu_Vf, R, Rd); /* Vf = R & ~Rd */
+    tcg_gen_shri_tl(cpu_Vf, cpu_Vf, 15);
+
+    /* Zf */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    /* Nf */
+    tcg_gen_shri_tl(cpu_Nf, R, 15); /* Nf = R(15) */
+
+    /* Sf */
+    tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf);/* Sf = Nf ^ Vf */
+
+    /* R */
+    tcg_gen_andi_tl(RdL, R, 0xff);
+    tcg_gen_shri_tl(RdH, R, 8);
+
+    tcg_temp_free_i32(Rd);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Subtracts two registers and places the result in the destination
+ *  register Rd.
+ */
+static bool trans_SUB(DisasContext *ctx, arg_SUB *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+/*
+ *  Subtracts a register and a constant and places the result in the
+ *  destination register Rd. This instruction is working on Register R16 to R31
+ *  and is very well suited for operations on the X, Y, and Z-pointers.
+ */
+static bool trans_SUBI(DisasContext *ctx, arg_SUBI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = tcg_const_i32(a->imm);
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Imm */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+    tcg_temp_free_i32(Rr);
+
+    return true;
+}
+
+
+/*
+ *  Subtracts two registers and subtracts with the C Flag and places the
+ *  result in the destination register Rd.
+ */
+static bool trans_SBC(DisasContext *ctx, arg_SBC *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv zero = tcg_const_i32(0);
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr - Cf */
+    tcg_gen_sub_tl(R, R, cpu_Cf);
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_NSf(R);
+
+    /*
+     * Previous value remains unchanged when the result is zero;
+     * cleared otherwise.
+     */
+    tcg_gen_movcond_tl(TCG_COND_EQ, cpu_Zf, R, zero, cpu_Zf, zero);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(zero);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  SBCI -- Subtract Immediate with Carry
+ */
+static bool trans_SBCI(DisasContext *ctx, arg_SBCI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = tcg_const_i32(a->imm);
+    TCGv R = tcg_temp_new_i32();
+    TCGv zero = tcg_const_i32(0);
+
+    /* op */
+    tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr - Cf */
+    tcg_gen_sub_tl(R, R, cpu_Cf);
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, Rd, Rr);
+    gen_sub_Vf(R, Rd, Rr);
+    gen_NSf(R);
+
+    /*
+     * Previous value remains unchanged when the result is zero;
+     * cleared otherwise.
+     */
+    tcg_gen_movcond_tl(TCG_COND_EQ, cpu_Zf, R, zero, cpu_Zf, zero);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(zero);
+    tcg_temp_free_i32(R);
+    tcg_temp_free_i32(Rr);
+
+    return true;
+}
+
+
+/*
+ *  Performs the logical AND between the contents of register Rd and register
+ *  Rr and places the result in the destination register Rd.
+ */
+static bool trans_AND(DisasContext *ctx, arg_AND *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_and_tl(R, Rd, Rr); /* Rd = Rd and Rr */
+
+    /* Vf */
+    tcg_gen_movi_tl(cpu_Vf, 0); /* Vf = 0 */
+
+    /* Zf */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Performs the logical AND between the contents of register Rd and a constant
+ *  and places the result in the destination register Rd.
+ */
+static bool trans_ANDI(DisasContext *ctx, arg_ANDI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    int Imm = (a->imm);
+
+    /* op */
+    tcg_gen_andi_tl(Rd, Rd, Imm); /* Rd = Rd & Imm */
+
+    tcg_gen_movi_tl(cpu_Vf, 0x00); /* Vf = 0 */
+    gen_ZNSf(Rd);
+
+    return true;
+}
+
+
+/*
+ *  Performs the logical OR between the contents of register Rd and register
+ *  Rr and places the result in the destination register Rd.
+ */
+static bool trans_OR(DisasContext *ctx, arg_OR *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    tcg_gen_or_tl(R, Rd, Rr);
+
+    tcg_gen_movi_tl(cpu_Vf, 0);
+    gen_ZNSf(R);
+
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Performs the logical OR between the contents of register Rd and a
+ *  constant and places the result in the destination register Rd.
+ */
+static bool trans_ORI(DisasContext *ctx, arg_ORI *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    int Imm = (a->imm);
+
+    tcg_gen_ori_tl(Rd, Rd, Imm); /* Rd = Rd | Imm */
+
+    tcg_gen_movi_tl(cpu_Vf, 0x00); /* Vf = 0 */
+    gen_ZNSf(Rd);
+
+    return true;
+}
+
+
+/*
+ *  Performs the logical EOR between the contents of register Rd and
+ *  register Rr and places the result in the destination register Rd.
+ */
+static bool trans_EOR(DisasContext *ctx, arg_EOR *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+
+    tcg_gen_xor_tl(Rd, Rd, Rr);
+
+    tcg_gen_movi_tl(cpu_Vf, 0);
+    gen_ZNSf(Rd);
+
+    return true;
+}
+
+
+/*
+ *  Clears the specified bits in register Rd. Performs the logical AND
+ *  between the contents of register Rd and the complement of the constant mask
+ *  K. The result will be placed in register Rd.
+ */
+static bool trans_COM(DisasContext *ctx, arg_COM *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv R = tcg_temp_new_i32();
+
+    tcg_gen_xori_tl(Rd, Rd, 0xff);
+
+    tcg_gen_movi_tl(cpu_Cf, 1); /* Cf = 1 */
+    tcg_gen_movi_tl(cpu_Vf, 0); /* Vf = 0 */
+    gen_ZNSf(Rd);
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Replaces the contents of register Rd with its two's complement; the
+ *  value $80 is left unchanged.
+ */
+static bool trans_NEG(DisasContext *ctx, arg_NEG *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+    TCGv t0 = tcg_const_i32(0);
+    TCGv R = tcg_temp_new_i32();
+
+    /* op */
+    tcg_gen_sub_tl(R, t0, Rd); /* R = 0 - Rd */
+    tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */
+
+    gen_sub_CHf(R, t0, Rd);
+    gen_sub_Vf(R, t0, Rd);
+    gen_ZNSf(R);
+
+    /* R */
+    tcg_gen_mov_tl(Rd, R);
+
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  Adds one -1- to the contents of register Rd and places the result in the
+ *  destination register Rd.  The C Flag in SREG is not affected by the
+ *  operation, thus allowing the INC instruction to be used on a loop counter in
+ *  multiple-precision computations.  When operating on unsigned numbers, only
+ *  BREQ and BRNE branches can be expected to perform consistently. When
+ *  operating on two's complement values, all signed branches are available.
+ */
+static bool trans_INC(DisasContext *ctx, arg_INC *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+
+    tcg_gen_addi_tl(Rd, Rd, 1);
+    tcg_gen_andi_tl(Rd, Rd, 0xff);
+
+    /* cpu_Vf = Rd == 0x80 */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Vf, Rd, 0x80);
+    gen_ZNSf(Rd);
+
+    return true;
+}
+
+
+/*
+ *  Subtracts one -1- from the contents of register Rd and places the result
+ *  in the destination register Rd.  The C Flag in SREG is not affected by the
+ *  operation, thus allowing the DEC instruction to be used on a loop counter in
+ *  multiple-precision computations.  When operating on unsigned values, only
+ *  BREQ and BRNE branches can be expected to perform consistently.  When
+ *  operating on two's complement values, all signed branches are available.
+ */
+static bool trans_DEC(DisasContext *ctx, arg_DEC *a)
+{
+    TCGv Rd = cpu_r[a->rd];
+
+    tcg_gen_subi_tl(Rd, Rd, 1); /* Rd = Rd - 1 */
+    tcg_gen_andi_tl(Rd, Rd, 0xff); /* make it 8 bits */
+
+    /* cpu_Vf = Rd == 0x7f */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Vf, Rd, 0x7f);
+    gen_ZNSf(Rd);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit unsigned multiplication.
+ */
+static bool trans_MUL(DisasContext *ctx, arg_MUL *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    tcg_gen_mul_tl(R, Rd, Rr); /* R = Rd * Rr */
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication.
+ */
+static bool trans_MULS(DisasContext *ctx, arg_MULS *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */
+    tcg_gen_ext8s_tl(t1, Rr); /* make Rr full 32 bit signed */
+    tcg_gen_mul_tl(R, t0, t1); /* R = Rd * Rr */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit multiplication of a
+ *  signed and an unsigned number.
+ */
+static bool trans_MULSU(DisasContext *ctx, arg_MULSU *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv t0 = tcg_temp_new_i32();
+
+    tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */
+    tcg_gen_mul_tl(R, t0, Rr); /* R = Rd *Rr */
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+
+    tcg_gen_shri_tl(cpu_Cf, R, 16); /* Cf = R(16) */
+    tcg_gen_mov_tl(cpu_Zf, R);
+
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit unsigned
+ *  multiplication and shifts the result one bit left.
+ */
+static bool trans_FMUL(DisasContext *ctx, arg_FMUL *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+
+    tcg_gen_mul_tl(R, Rd, Rr); /* R = Rd * Rr */
+
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    tcg_gen_shli_tl(R, R, 1);
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+    tcg_gen_andi_tl(R1, R1, 0xff);
+
+
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication
+ *  and shifts the result one bit left.
+ */
+static bool trans_FMULS(DisasContext *ctx, arg_FMULS *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv t0 = tcg_temp_new_i32();
+    TCGv t1 = tcg_temp_new_i32();
+
+    tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */
+    tcg_gen_ext8s_tl(t1, Rr); /* make Rr full 32 bit signed */
+    tcg_gen_mul_tl(R, t0, t1); /* R = Rd * Rr */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */
+
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    tcg_gen_shli_tl(R, R, 1);
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+    tcg_gen_andi_tl(R1, R1, 0xff);
+
+    tcg_temp_free_i32(t1);
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication
+ *  and shifts the result one bit left.
+ */
+static bool trans_FMULSU(DisasContext *ctx, arg_FMULSU *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) {
+        return true;
+    }
+
+    TCGv R0 = cpu_r[0];
+    TCGv R1 = cpu_r[1];
+    TCGv Rd = cpu_r[a->rd];
+    TCGv Rr = cpu_r[a->rr];
+    TCGv R = tcg_temp_new_i32();
+    TCGv t0 = tcg_temp_new_i32();
+
+    tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */
+    tcg_gen_mul_tl(R, t0, Rr); /* R = Rd * Rr */
+    tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */
+
+    tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */
+    tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */
+
+    tcg_gen_shli_tl(R, R, 1);
+
+    tcg_gen_andi_tl(R0, R, 0xff);
+    tcg_gen_shri_tl(R1, R, 8);
+    tcg_gen_andi_tl(R1, R1, 0xff);
+
+    tcg_temp_free_i32(t0);
+    tcg_temp_free_i32(R);
+
+    return true;
+}
+
+
+/*
+ *  The module is an instruction set extension to the AVR CPU, performing
+ *  DES iterations. The 64-bit data block (plaintext or ciphertext) is placed in
+ *  the CPU register file, registers R0-R7, where LSB of data is placed in LSB
+ *  of R0 and MSB of data is placed in MSB of R7. The full 64-bit key (including
+ *  parity bits) is placed in registers R8- R15, organized in the register file
+ *  with LSB of key in LSB of R8 and MSB of key in MSB of R15. Executing one DES
+ *  instruction performs one round in the DES algorithm. Sixteen rounds must be
+ *  executed in increasing order to form the correct DES ciphertext or
+ *  plaintext. Intermediate results are stored in the register file (R0-R15)
+ *  after each DES instruction. The instruction's operand (K) determines which
+ *  round is executed, and the half carry flag (H) determines whether encryption
+ *  or decryption is performed.  The DES algorithm is described in
+ *  "Specifications for the Data Encryption Standard" (Federal Information
+ *  Processing Standards Publication 46). Intermediate results in this
+ *  implementation differ from the standard because the initial permutation and
+ *  the inverse initial permutation are performed each iteration. This does not
+ *  affect the result in the final ciphertext or plaintext, but reduces
+ *  execution time.
+ */
+static bool trans_DES(DisasContext *ctx, arg_DES *a)
+{
+    /* TODO */
+    if (!avr_have_feature(ctx, AVR_FEATURE_DES)) {
+        return true;
+    }
+
+    return true;
+}
Signature · 1 line
2.17.2 (Apple Git-113)
#13 lore raw

PATCH v34 13/13 target/avr: Add tests

In reply to Michael Rolnik

1. Avocado test
The test is based on
https://github.com/seharris/qemu-avr-tests/tree/master/free-rtos/Demo
demo which. If working correctly, prints 'ABCDEFGHIJKLMNOPQRSTUVWX' out. it also demostrates that timer and IRQ are working

2. Boot serial test
Print out 'T' through serial port

---
 tests/Makefile.include           |  2 ++
 tests/acceptance/machine_avr6.py | 36 ++++++++++++++++++++++++++++++++
 tests/boot-serial-test.c         | 10 +++++++++
 3 files changed, 48 insertions(+)
 create mode 100644 tests/acceptance/machine_avr6.py
tests/Makefile.include +2 -0
index 09e5b410dc..7f3f971191 100644
@@ -206,6 +206,8 @@ check-qtest-i386-y += tests/test-x86-cpuid-compat$(EXESUF)
 check-qtest-i386-y += tests/numa-test$(EXESUF)
 check-qtest-x86_64-y += $(check-qtest-i386-y)
 
+check-qtest-avr-y += tests/boot-serial-test$(EXESUF)
+
 check-qtest-alpha-y += tests/boot-serial-test$(EXESUF)
 check-qtest-alpha-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF)
 
tests/acceptance/machine_avr6.py new file +36 -0
new file mode 100644
index 0000000000..0601080b01
@@ -0,0 +1,36 @@
+import logging
+import time
+import distutils.spawn
+
+from avocado import skipUnless
+from avocado_qemu import Test
+from avocado.utils import process
+
+class AVR6Machine(Test):
+    timeout = 5
+
+    def test_freertos(self):
+        """
+        :avocado: tags=arch:avr
+        :avocado: tags=machine:sample
+        """
+        """
+        https://github.com/seharris/qemu-avr-tests/raw/master/free-rtos/Demo/AVR_ATMega2560_GCC/demo.elf
+        constantly prints out 'ABCDEFGHIJKLMNOPQRSTUVWXABCDEFGHIJKLMNOPQRSTUVWX'
+        """
+        rom_url = 'https://github.com/seharris/qemu-avr-tests'
+        rom_url += '/raw/master/free-rtos/Demo/AVR_ATMega2560_GCC/demo.elf'
+        rom_hash = '7eb521f511ca8f2622e0a3c5e8dd686efbb911d4'
+        rom_path = self.fetch_asset(rom_url, asset_hash=rom_hash)
+
+        self.vm.set_machine('sample')
+        self.vm.add_args('-bios', rom_path)
+        self.vm.add_args('-nographic')
+        self.vm.launch()
+
+        time.sleep(2)
+        self.vm.shutdown()
+
+        match = 'ABCDEFGHIJKLMNOPQRSTUVWXABCDEFGHIJKLMNOPQRSTUVWX'
+
+        self.assertIn(match, self.vm.get_log())
tests/boot-serial-test.c +10 -0
index d3a54a0ba5..1121ed0db2 100644
@@ -16,6 +16,15 @@
 #include "qemu/osdep.h"
 #include "libqtest.h"
 
+static const uint8_t bios_avr[] = {
+    0x88, 0xe0,             /* ldi r24, 0x08   */
+    0x80, 0x93, 0xc1, 0x00, /* sts 0x00C1, r24 ; Enable tx */
+    0x86, 0xe0,             /* ldi r24, 0x06   */
+    0x80, 0x93, 0xc2, 0x00, /* sts 0x00C2, r24 ; Set the data bits to 8 */
+    0x84, 0xe5,             /* ldi r24, 0x54   */
+    0x80, 0x93, 0xc6, 0x00, /* sts 0x00C6, r24 ; Output 'T' */
+};
+
 static const uint8_t kernel_mcf5208[] = {
     0x41, 0xf9, 0xfc, 0x06, 0x00, 0x00,     /* lea 0xfc060000,%a0 */
     0x10, 0x3c, 0x00, 0x54,                 /* move.b #'T',%d0 */
@@ -103,6 +112,7 @@ typedef struct testdef {
 
 static testdef_t tests[] = {
     { "alpha", "clipper", "", "PCI:" },
+    { "avr", "sample", "", "T", sizeof(bios_avr), NULL, bios_avr },
     { "ppc", "ppce500", "", "U-Boot" },
     { "ppc", "40p", "-vga none -boot d", "Trying cd:," },
     { "ppc", "g3beige", "", "PowerPC,750" },
Signature · 1 line
2.17.2 (Apple Git-113)
#8 lore raw

PATCH v34 8/13 target/avr: Add instruction translation - MCU Control Instructions

In reply to Michael Rolnik

This includes:
- BREAK
- NOP
- SLEEP
- WDR

---
 target/avr/translate.c | 174 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 174 insertions(+)
target/avr/translate.c +174 -0
index f2ec2e2d2f..30ba13bdd7 100644
@@ -368,6 +368,115 @@ static bool trans_ADC(DisasContext *ctx, arg_ADC *a)
 }
 
 
+static void gen_jmp_ez(DisasContext *ctx)
+{
+    tcg_gen_deposit_tl(cpu_pc, cpu_r[30], cpu_r[31], 8, 8);
+    tcg_gen_or_tl(cpu_pc, cpu_pc, cpu_eind);
+    ctx->bstate = DISAS_LOOKUP;
+}
+
+
+static void gen_jmp_z(DisasContext *ctx)
+{
+    tcg_gen_deposit_tl(cpu_pc, cpu_r[30], cpu_r[31], 8, 8);
+    ctx->bstate = DISAS_LOOKUP;
+}
+
+
+/*
+ *  in the gen_set_addr & gen_get_addr functions
+ *  H assumed to be in 0x00ff0000 format
+ *  M assumed to be in 0x000000ff format
+ *  L assumed to be in 0x000000ff format
+ */
+static void gen_set_addr(TCGv addr, TCGv H, TCGv M, TCGv L)
+{
+
+    tcg_gen_andi_tl(L, addr, 0x000000ff);
+
+    tcg_gen_andi_tl(M, addr, 0x0000ff00);
+    tcg_gen_shri_tl(M, M, 8);
+
+    tcg_gen_andi_tl(H, addr, 0x00ff0000);
+}
+
+
+static void gen_set_xaddr(TCGv addr)
+{
+    gen_set_addr(addr, cpu_rampX, cpu_r[27], cpu_r[26]);
+}
+
+
+static void gen_set_yaddr(TCGv addr)
+{
+    gen_set_addr(addr, cpu_rampY, cpu_r[29], cpu_r[28]);
+}
+
+
+static void gen_set_zaddr(TCGv addr)
+{
+    gen_set_addr(addr, cpu_rampZ, cpu_r[31], cpu_r[30]);
+}
+
+
+static TCGv gen_get_addr(TCGv H, TCGv M, TCGv L)
+{
+    TCGv addr = tcg_temp_new_i32();
+
+    tcg_gen_deposit_tl(addr, M, H, 8, 8);
+    tcg_gen_deposit_tl(addr, L, addr, 8, 16);
+
+    return addr;
+}
+
+
+static TCGv gen_get_xaddr(void)
+{
+    return gen_get_addr(cpu_rampX, cpu_r[27], cpu_r[26]);
+}
+
+
+static TCGv gen_get_yaddr(void)
+{
+    return gen_get_addr(cpu_rampY, cpu_r[29], cpu_r[28]);
+}
+
+
+static TCGv gen_get_zaddr(void)
+{
+    return gen_get_addr(cpu_rampZ, cpu_r[31], cpu_r[30]);
+}
+
+
+/*
+ *  Load one byte indirect from data space to register and stores an clear
+ *  the bits in data space specified by the register. The instruction can only
+ *  be used towards internal SRAM.  The data location is pointed to by the Z (16
+ *  bits) Pointer Register in the Register File. Memory access is limited to the
+ *  current data segment of 64KB. To access another data segment in devices with
+ *  more than 64KB data space, the RAMPZ in register in the I/O area has to be
+ *  changed.  The Z-pointer Register is left unchanged by the operation. This
+ *  instruction is especially suited for clearing status bits stored in SRAM.
+ */
+static void gen_data_store(DisasContext *ctx, TCGv data, TCGv addr)
+{
+    if (ctx->tb->flags & TB_FLAGS_FULL_ACCESS) {
+        gen_helper_fullwr(cpu_env, data, addr);
+    } else {
+        tcg_gen_qemu_st8(data, addr, MMU_DATA_IDX); /* mem[addr] = data */
+    }
+}
+
+static void gen_data_load(DisasContext *ctx, TCGv data, TCGv addr)
+{
+    if (ctx->tb->flags & TB_FLAGS_FULL_ACCESS) {
+        gen_helper_fullrd(data, cpu_env, addr);
+    } else {
+        tcg_gen_qemu_ld8u(data, addr, MMU_DATA_IDX); /* data = mem[addr] */
+    }
+}
+
+
 /*
  *  Subtracts an immediate value (0-63) from a register pair and places the
  *  result in the register pair. This instruction operates on the upper four
@@ -2618,3 +2727,68 @@ static bool trans_BCLR(DisasContext *ctx, arg_BCLR *a)
 
     return true;
 }
+
+
+/*
+ *  The BREAK instruction is used by the On-chip Debug system, and is
+ *  normally not used in the application software. When the BREAK instruction is
+ *  executed, the AVR CPU is set in the Stopped Mode. This gives the On-chip
+ *  Debugger access to internal resources.  If any Lock bits are set, or either
+ *  the JTAGEN or OCDEN Fuses are unprogrammed, the CPU will treat the BREAK
+ *  instruction as a NOP and will not enter the Stopped mode.  This instruction
+ *  is not available in all devices. Refer to the device specific instruction
+ *  set summary.
+ */
+static bool trans_BREAK(DisasContext *ctx, arg_BREAK *a)
+{
+    if (!avr_have_feature(ctx, AVR_FEATURE_BREAK)) {
+        return true;
+    }
+
+#ifdef BREAKPOINT_ON_BREAK
+    tcg_gen_movi_tl(cpu_pc, ctx->npc - 1);
+    gen_helper_debug(cpu_env);
+    ctx->bstate = DISAS_EXIT;
+#else
+    /* NOP */
+#endif
+
+    return true;
+}
+
+
+/*
+ *  This instruction performs a single cycle No Operation.
+ */
+static bool trans_NOP(DisasContext *ctx, arg_NOP *a)
+{
+
+    /* NOP */
+
+    return true;
+}
+
+
+/*
+ *  This instruction sets the circuit in sleep mode defined by the MCU
+ *  Control Register.
+ */
+static bool trans_SLEEP(DisasContext *ctx, arg_SLEEP *a)
+{
+    gen_helper_sleep(cpu_env);
+    ctx->bstate = DISAS_NORETURN;
+    return true;
+}
+
+
+/*
+ *  This instruction resets the Watchdog Timer. This instruction must be
+ *  executed within a limited time given by the WD prescaler. See the Watchdog
+ *  Timer hardware specification.
+ */
+static bool trans_WDR(DisasContext *ctx, arg_WDR *a)
+{
+    gen_helper_wdr(cpu_env);
+
+    return true;
+}
Signature · 1 line
2.17.2 (Apple Git-113)
#10 lore raw

PATCH v34 10/13 target/avr: Add limited support for USART and 16 bit timer peripherals

In reply to Michael Rolnik

From: Sarah Harris <[email protected]>

These were designed to facilitate testing but should provide enough function to be useful in other contexts.
Only a subset of the functions of each peripheral is implemented, mainly due to the lack of a standard way to handle electrical connections (like GPIO pins).

---
 hw/char/Kconfig                |   3 +
 hw/char/Makefile.objs          |   1 +
 hw/char/avr_usart.c            | 324 ++++++++++++++++++
 hw/misc/Kconfig                |   3 +
 hw/misc/Makefile.objs          |   2 +
 hw/misc/avr_mask.c             | 112 ++++++
 hw/timer/Kconfig               |   3 +
 hw/timer/Makefile.objs         |   1 +
 hw/timer/avr_timer16.c         | 605 +++++++++++++++++++++++++++++++++
 include/hw/char/avr_usart.h    |  97 ++++++
 include/hw/misc/avr_mask.h     |  47 +++
 include/hw/timer/avr_timer16.h |  97 ++++++
 12 files changed, 1295 insertions(+)
 create mode 100644 hw/char/avr_usart.c
 create mode 100644 hw/misc/avr_mask.c
 create mode 100644 hw/timer/avr_timer16.c
 create mode 100644 include/hw/char/avr_usart.h
 create mode 100644 include/hw/misc/avr_mask.h
 create mode 100644 include/hw/timer/avr_timer16.h
hw/char/Kconfig +3 -0
index 40e7a8b8bb..331b20983f 100644
@@ -46,3 +46,6 @@ config SCLPCONSOLE
 
 config TERMINAL3270
     bool
+
+config AVR_USART
+    bool
hw/char/Makefile.objs +1 -0
index 02d8a66925..09ed50f1d0 100644
@@ -21,6 +21,7 @@ obj-$(CONFIG_PSERIES) += spapr_vty.o
 obj-$(CONFIG_DIGIC) += digic-uart.o
 obj-$(CONFIG_STM32F2XX_USART) += stm32f2xx_usart.o
 obj-$(CONFIG_RASPI) += bcm2835_aux.o
+obj-$(CONFIG_AVR_USART) += avr_usart.o
 
 common-obj-$(CONFIG_CMSDK_APB_UART) += cmsdk-apb-uart.o
 common-obj-$(CONFIG_ETRAXFS) += etraxfs_ser.o
hw/char/avr_usart.c new file +324 -0

324 lines not on this page · raw message

hw/misc/Kconfig +3 -0
index 51754bb47c..eb9e3210f4 100644
@@ -120,4 +120,7 @@ config AUX
 config UNIMP
     bool
 
+config AVR_MASK
+    bool
+
 source macio/Kconfig
hw/misc/Makefile.objs +2 -0
index a150680966..946bfc396d 100644
@@ -80,3 +80,5 @@ common-obj-$(CONFIG_MSF2) += msf2-sysreg.o
 common-obj-$(CONFIG_NRF51_SOC) += nrf51_rng.o
 
 common-obj-$(CONFIG_GRLIB) += grlib_ahb_apb_pnp.o
+
+obj-$(CONFIG_AVR_MASK) += avr_mask.o
hw/misc/avr_mask.c new file +112 -0

112 lines not on this page · raw message

hw/timer/Kconfig +3 -0
index eefc95f35e..3bea322184 100644
@@ -58,3 +58,6 @@ config CMSDK_APB_TIMER
 config CMSDK_APB_DUALTIMER
     bool
     select PTIMER
+
+config AVR_TIMER16
+    bool
hw/timer/Makefile.objs +1 -0
index 123d92c969..4f853e265d 100644
@@ -35,6 +35,7 @@ obj-$(CONFIG_PXA2XX) += pxa2xx_timer.o
 obj-$(CONFIG_SH4) += sh_timer.o
 obj-$(CONFIG_DIGIC) += digic-timer.o
 obj-$(CONFIG_MIPS_CPS) += mips_gictimer.o
+obj-$(CONFIG_AVR_TIMER16) += avr_timer16.o
 
 obj-$(CONFIG_MC146818RTC) += mc146818rtc.o
 
hw/timer/avr_timer16.c new file +605 -0

605 lines not on this page · raw message

include/hw/char/avr_usart.h new file +97 -0

97 lines not on this page · raw message

include/hw/misc/avr_mask.h new file +47 -0

47 lines not on this page · raw message

include/hw/timer/avr_timer16.h new file +97 -0

97 lines not on this page · raw message

Signature · 1 line
2.17.2 (Apple Git-113)
#9 lore raw

PATCH v34 9/13 target/avr: Add instruction translation - CPU main translation function

In reply to Michael Rolnik
---
 target/avr/translate.c | 234 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 234 insertions(+)
target/avr/translate.c +234 -0
index 30ba13bdd7..fdf4e11f58 100644
@@ -2792,3 +2792,237 @@ static bool trans_WDR(DisasContext *ctx, arg_WDR *a)
 
     return true;
 }
+
+
+void avr_cpu_tcg_init(void)
+{
+    int i;
+
+#define AVR_REG_OFFS(x) offsetof(CPUAVRState, x)
+    cpu_pc = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(pc_w), "pc");
+    cpu_Cf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregC), "Cf");
+    cpu_Zf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregZ), "Zf");
+    cpu_Nf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregN), "Nf");
+    cpu_Vf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregV), "Vf");
+    cpu_Sf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregS), "Sf");
+    cpu_Hf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregH), "Hf");
+    cpu_Tf = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregT), "Tf");
+    cpu_If = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sregI), "If");
+    cpu_rampD = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(rampD), "rampD");
+    cpu_rampX = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(rampX), "rampX");
+    cpu_rampY = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(rampY), "rampY");
+    cpu_rampZ = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(rampZ), "rampZ");
+    cpu_eind = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(eind), "eind");
+    cpu_sp = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(sp), "sp");
+    cpu_skip = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(skip), "skip");
+
+    for (i = 0; i < NO_CPU_REGISTERS; i++) {
+        cpu_r[i] = tcg_global_mem_new_i32(cpu_env, AVR_REG_OFFS(r[i]),
+                                          reg_names[i]);
+    }
+#undef AVR_REG_OFFS
+}
+
+static void translate(DisasContext *ctx)
+{
+    uint32_t opcode = next_word(ctx);
+
+    if (!decode_insn(ctx, opcode)) {
+        gen_helper_unsupported(cpu_env);
+        ctx->bstate = DISAS_NORETURN;
+    }
+}
+
+/* Standardize the cpu_skip condition to NE.  */
+static bool canonicalize_skip(DisasContext *ctx)
+{
+    switch (ctx->skip_cond) {
+    case TCG_COND_NEVER:
+        /* Normal case: cpu_skip is known to be false.  */
+        return false;
+
+    case TCG_COND_ALWAYS:
+        /*
+         * Breakpoint case: cpu_skip is known to be true, via TB_FLAGS_SKIP.
+         * The breakpoint is on the instruction being skipped, at the start
+         * of the TranslationBlock.  No need to update.
+         */
+        return false;
+
+    case TCG_COND_NE:
+        if (ctx->skip_var1 == NULL) {
+            tcg_gen_mov_tl(cpu_skip, ctx->skip_var0);
+        } else {
+            tcg_gen_xor_tl(cpu_skip, ctx->skip_var0, ctx->skip_var1);
+            ctx->skip_var1 = NULL;
+        }
+        break;
+
+    default:
+        /* Convert to a NE condition vs 0. */
+        if (ctx->skip_var1 == NULL) {
+            tcg_gen_setcondi_tl(ctx->skip_cond, cpu_skip, ctx->skip_var0, 0);
+        } else {
+            tcg_gen_setcond_tl(ctx->skip_cond, cpu_skip,
+                               ctx->skip_var0, ctx->skip_var1);
+            ctx->skip_var1 = NULL;
+        }
+        ctx->skip_cond = TCG_COND_NE;
+        break;
+    }
+    if (ctx->free_skip_var0) {
+        tcg_temp_free(ctx->skip_var0);
+        ctx->free_skip_var0 = false;
+    }
+    ctx->skip_var0 = cpu_skip;
+    return true;
+}
+
+void gen_intermediate_code(CPUState *cs, TranslationBlock *tb, int max_insns)
+{
+    CPUAVRState *env = cs->env_ptr;
+    DisasContext ctx = {
+        .tb = tb,
+        .cs = cs,
+        .env = env,
+        .memidx = 0,
+        .bstate = DISAS_NEXT,
+        .skip_cond = TCG_COND_NEVER,
+        .singlestep = cs->singlestep_enabled,
+    };
+    target_ulong pc_start = tb->pc / 2;
+    int num_insns = 0;
+
+    if (tb->flags & TB_FLAGS_FULL_ACCESS) {
+        /*
+         * This flag is set by ST/LD instruction we will regenerate it ONLY
+         * with mem/cpu memory access instead of mem access
+         */
+        max_insns = 1;
+    }
+    if (ctx.singlestep) {
+        max_insns = 1;
+    }
+
+    gen_tb_start(tb);
+
+    ctx.npc = pc_start;
+    if (tb->flags & TB_FLAGS_SKIP) {
+        ctx.skip_cond = TCG_COND_ALWAYS;
+        ctx.skip_var0 = cpu_skip;
+    }
+
+    do {
+        TCGLabel *skip_label = NULL;
+
+        /* translate current instruction */
+        tcg_gen_insn_start(ctx.npc);
+        num_insns++;
+
+        /*
+         * this is due to some strange GDB behavior
+         * let's assume main has address 0x100
+         * b main   - sets breakpoint at address 0x00000100 (code)
+         * b *0x100 - sets breakpoint at address 0x00800100 (data)
+         */
+        if (unlikely(!ctx.singlestep &&
+                (cpu_breakpoint_test(cs, OFFSET_CODE + ctx.npc * 2, BP_ANY) ||
+                 cpu_breakpoint_test(cs, OFFSET_DATA + ctx.npc * 2, BP_ANY)))) {
+            canonicalize_skip(&ctx);
+            tcg_gen_movi_tl(cpu_pc, ctx.npc);
+            gen_helper_debug(cpu_env);
+            goto done_generating;
+        }
+
+        /* Conditionally skip the next instruction, if indicated.  */
+        if (ctx.skip_cond != TCG_COND_NEVER) {
+            skip_label = gen_new_label();
+            if (ctx.skip_var0 == cpu_skip) {
+                /*
+                 * Copy cpu_skip so that we may zero it before the branch.
+                 * This ensures that cpu_skip is non-zero after the label
+                 * if and only if the skipped insn itself sets a skip.
+                 */
+                ctx.free_skip_var0 = true;
+                ctx.skip_var0 = tcg_temp_new();
+                tcg_gen_mov_tl(ctx.skip_var0, cpu_skip);
+                tcg_gen_movi_tl(cpu_skip, 0);
+            }
+            if (ctx.skip_var1 == NULL) {
+                tcg_gen_brcondi_tl(ctx.skip_cond, ctx.skip_var0, 0, skip_label);
+            } else {
+                tcg_gen_brcond_tl(ctx.skip_cond, ctx.skip_var0,
+                                  ctx.skip_var1, skip_label);
+                ctx.skip_var1 = NULL;
+            }
+            if (ctx.free_skip_var0) {
+                tcg_temp_free(ctx.skip_var0);
+                ctx.free_skip_var0 = false;
+            }
+            ctx.skip_cond = TCG_COND_NEVER;
+            ctx.skip_var0 = NULL;
+        }
+
+        translate(&ctx);
+
+        if (skip_label) {
+            canonicalize_skip(&ctx);
+            gen_set_label(skip_label);
+            if (ctx.bstate == DISAS_NORETURN) {
+                ctx.bstate = DISAS_CHAIN;
+            }
+        }
+    } while (ctx.bstate == DISAS_NEXT
+             && num_insns < max_insns
+             && (ctx.npc - pc_start) * 2 < TARGET_PAGE_SIZE - 4
+             && !tcg_op_buf_full());
+
+    if (tb->cflags & CF_LAST_IO) {
+        gen_io_end();
+    }
+
+    bool nonconst_skip = canonicalize_skip(&ctx);
+
+    switch (ctx.bstate) {
+    case DISAS_NORETURN:
+        assert(!nonconst_skip);
+        break;
+    case DISAS_NEXT:
+    case DISAS_TOO_MANY:
+    case DISAS_CHAIN:
+        if (!nonconst_skip) {
+            /* Note gen_goto_tb checks singlestep.  */
+            gen_goto_tb(&ctx, 1, ctx.npc);
+            break;
+        }
+        tcg_gen_movi_tl(cpu_pc, ctx.npc);
+        /* fall through */
+    case DISAS_LOOKUP:
+        if (!ctx.singlestep) {
+            tcg_gen_lookup_and_goto_ptr();
+            break;
+        }
+        /* fall through */
+    case DISAS_EXIT:
+        if (ctx.singlestep) {
+            gen_helper_debug(cpu_env);
+        } else {
+            tcg_gen_exit_tb(NULL, 0);
+        }
+        break;
+    default:
+        g_assert_not_reached();
+    }
+
+done_generating:
+    gen_tb_end(tb, num_insns);
+
+    tb->size = (ctx.npc - pc_start) * 2;
+    tb->icount = num_insns;
+}
+
+void restore_state_to_opc(CPUAVRState *env, TranslationBlock *tb,
+                            target_ulong *data)
+{
+    env->pc_w = data[0];
+}
Signature · 1 line
2.17.2 (Apple Git-113)