* [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode
@ 2011-10-20 21:11 Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec Stefan Weil
` (5 more replies)
0 siblings, 6 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel
This is a new version of my patch series which adds support for
QEMU on any host by using a TCG interpreter (TCI).
Some of the suggestions from the last patch version
(http://lists.nongnu.org/archive/html/qemu-devel/2011-09/msg02270.html)
are included now. Thank you to all who reviewed the code and sent
their feedback.
The new TCI is much faster by simply removing all assertion from the
interpreter code when QEMU was configured without --enable-debug-tcg
(--enable-debug also sets --enable-debug-tcg).
I still don't use computed goto. Nor did I address Stuart Brady's
remarks on possible problems with passing arguments to helper functions.
A future version might use trace events - for now I still use
macros because they support using __LINE__ and __func__.
There are still unimplemented opcodes in the code generator and in
the interpreter: I only implemented those which were needed during
my tests. Missing opcodes are reported and can be added easily when
needed.
TCI is not perfect, but it is already an exiting tool to examine
QEMU's TCG operations, to trace running emulation code or to
make statistics (for example using Valgrind).
The default QEMU builds should not be touched by my changes, therefore
the risk for the next offical QEMU release should be acceptable.
Here is a list of the patches:
[PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec
[PATCH v2 2/6] tcg: Make ARRAY_SIZE(tcg_op_defs) globally available
[PATCH v2 3/6] tcg: Add bytecode generator for tcg interpreter
[PATCH v2 4/6] tcg: Add interpreter for bytecode
[PATCH v2 5/6] tcg: Add tci disassembler
[PATCH v2 6/6] tcg: Add tcg interpreter to configure / make
The first patch cleans code - I hope it can be applied immediately.
The second patch adds a global constant used by the TCI disassembler.
The remaining 4 patches add TCI to QEMU.
Regards,
Stefan Weil
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 2/6] tcg: Make ARRAY_SIZE(tcg_op_defs) globally available Stefan Weil
` (4 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
Targets may use a non standard definition of tcg_tb_exec
by defining this macro in their tcg_target.h.
This is used here by ppc. It will be used by the TCG interpreter, too.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
tcg/ppc/tcg-target.h | 4 ++++
tcg/tcg.h | 9 ++++-----
2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/tcg/ppc/tcg-target.h b/tcg/ppc/tcg-target.h
index 5c2d612..25a6ea4 100644
--- a/tcg/ppc/tcg-target.h
+++ b/tcg/ppc/tcg-target.h
@@ -96,3 +96,7 @@ enum {
#define TCG_AREG0 TCG_REG_R27
#define TCG_TARGET_HAS_GUEST_BASE
+
+#define tcg_qemu_tb_exec(env, tb_ptr) \
+ ((long REGPARM __attribute__ ((longcall)) \
+ (*)(void *, void *))code_gen_prologue)(env, tb_ptr)
diff --git a/tcg/tcg.h b/tcg/tcg.h
index 983d59e..811e312 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -584,10 +584,9 @@ TCGv_i32 tcg_const_local_i32(int32_t val);
TCGv_i64 tcg_const_local_i64(int64_t val);
extern uint8_t code_gen_prologue[];
-#if defined(_ARCH_PPC) && !defined(_ARCH_PPC64)
-#define tcg_qemu_tb_exec(env, tb_ptr) \
- ((long REGPARM __attribute__ ((longcall)) (*)(void *, void *))code_gen_prologue)(env, tb_ptr)
-#else
-#define tcg_qemu_tb_exec(env, tb_ptr) \
+
+/* TCG targets may use a different definition of tcg_qemu_tb_exec. */
+#if !defined(tcg_qemu_tb_exec)
+# define tcg_qemu_tb_exec(env, tb_ptr) \
((long REGPARM (*)(void *, void *))code_gen_prologue)(env, tb_ptr)
#endif
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 2/6] tcg: Make ARRAY_SIZE(tcg_op_defs) globally available
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 3/6] tcg: Add bytecode generator for tcg interpreter Stefan Weil
` (3 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
tcg_op_defs was already a global array.
The tci disassembler also needs ARRAY_SIZE(tcg_op_defs),
so add a new global constant with this value.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
tcg/tcg.c | 1 +
tcg/tcg.h | 1 +
2 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/tcg/tcg.c b/tcg/tcg.c
index fefce5f..15a1f5e 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -89,6 +89,7 @@ TCGOpDef tcg_op_defs[] = {
#include "tcg-opc.h"
#undef DEF
};
+const size_t tcg_op_defs_max = ARRAY_SIZE(tcg_op_defs);
static TCGRegSet tcg_target_available_regs[2];
static TCGRegSet tcg_target_call_clobber_regs;
diff --git a/tcg/tcg.h b/tcg/tcg.h
index 811e312..79ea3cb 100644
--- a/tcg/tcg.h
+++ b/tcg/tcg.h
@@ -525,6 +525,7 @@ typedef struct TCGOpDef {
} TCGOpDef;
extern TCGOpDef tcg_op_defs[];
+extern const size_t tcg_op_defs_max;
typedef struct TCGTargetOpDef {
TCGOpcode op;
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 3/6] tcg: Add bytecode generator for tcg interpreter
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 2/6] tcg: Make ARRAY_SIZE(tcg_op_defs) globally available Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode Stefan Weil
` (2 subsequent siblings)
5 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
Unlike other tcg target code generators, this one does not generate
machine code for some cpu. It generates machine independent bytecode
which is interpreted later.
This allows running QEMU on any host.
Interpreted bytecode is slower than direct execution of generated
machine code.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
dyngen-exec.h | 12 +-
exec-all.h | 22 ++-
tcg/tci/README | 130 +++++++
tcg/tci/tcg-target.c | 919 ++++++++++++++++++++++++++++++++++++++++++++++++++
tcg/tci/tcg-target.h | 160 +++++++++
5 files changed, 1240 insertions(+), 3 deletions(-)
create mode 100644 tcg/tci/README
create mode 100644 tcg/tci/tcg-target.c
create mode 100644 tcg/tci/tcg-target.h
diff --git a/dyngen-exec.h b/dyngen-exec.h
index fbde29e..3544372 100644
--- a/dyngen-exec.h
+++ b/dyngen-exec.h
@@ -19,7 +19,12 @@
#if !defined(__DYNGEN_EXEC_H__)
#define __DYNGEN_EXEC_H__
-#if defined(__i386__)
+#if defined(CONFIG_TCG_INTERPRETER)
+/* The TCG interpreter does not need a special register AREG0,
+ * but it is possible to use one by defining AREG0.
+ * On i386, register edi seems to work. */
+/* Run without special register AREG0 or use a value defined elsewhere. */
+#elif defined(__i386__)
#define AREG0 "ebp"
#elif defined(__x86_64__)
#define AREG0 "r14"
@@ -55,6 +60,11 @@
#error unsupported CPU
#endif
+#if defined(AREG0)
register CPUState *env asm(AREG0);
+#else
+/* TODO: Try env = cpu_single_env. */
+extern CPUState *env;
+#endif
#endif /* !defined(__DYNGEN_EXEC_H__) */
diff --git a/exec-all.h b/exec-all.h
index 6e6d3c6..69fd5e4 100644
--- a/exec-all.h
+++ b/exec-all.h
@@ -122,6 +122,8 @@ void tlb_set_page(CPUState *env, target_ulong vaddr,
#if defined(_ARCH_PPC) || defined(__x86_64__) || defined(__arm__) || defined(__i386__)
#define USE_DIRECT_JUMP
+#elif defined(CONFIG_TCG_INTERPRETER)
+#define USE_DIRECT_JUMP
#endif
struct TranslationBlock {
@@ -189,7 +191,14 @@ extern TranslationBlock *tb_phys_hash[CODE_GEN_PHYS_HASH_SIZE];
#if defined(USE_DIRECT_JUMP)
-#if defined(_ARCH_PPC)
+#if defined(CONFIG_TCG_INTERPRETER)
+static inline void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr)
+{
+ /* patch the branch destination */
+ *(uint32_t *)jmp_addr = addr - (jmp_addr + 4);
+ /* no need to flush icache explicitly */
+}
+#elif defined(_ARCH_PPC)
void ppc_tb_set_jmp_target(unsigned long jmp_addr, unsigned long addr);
#define tb_set_jmp_target1 ppc_tb_set_jmp_target
#elif defined(__i386__) || defined(__x86_64__)
@@ -223,6 +232,8 @@ static inline void tb_set_jmp_target1(unsigned long jmp_addr, unsigned long addr
__asm __volatile__ ("swi 0x9f0002" : : "r" (_beg), "r" (_end), "r" (_flg));
#endif
}
+#else
+#error tb_set_jmp_target1 is missing
#endif
static inline void tb_set_jmp_target(TranslationBlock *tb,
@@ -269,7 +280,14 @@ extern int tb_invalidated_flag;
/* The return address may point to the start of the next instruction.
Subtracting one gets us the call instruction itself. */
-#if defined(__s390__) && !defined(__s390x__)
+#if defined(CONFIG_TCG_INTERPRETER)
+/* Alpha and SH4 user mode emulations and Softmmu call GETPC().
+ For all others, GETPC remains undefined (which makes TCI a little faster. */
+# if defined(CONFIG_SOFTMMU) || defined(TARGET_ALPHA) || defined(TARGET_SH4)
+extern void *tci_tb_ptr;
+# define GETPC() tci_tb_ptr
+# endif
+#elif defined(__s390__) && !defined(__s390x__)
# define GETPC() ((void*)(((unsigned long)__builtin_return_address(0) & 0x7fffffffUL) - 1))
#elif defined(__arm__)
/* Thumb return addresses have the low bit set, so we need to subtract two.
diff --git a/tcg/tci/README b/tcg/tci/README
new file mode 100644
index 0000000..6ac1ac9
--- /dev/null
+++ b/tcg/tci/README
@@ -0,0 +1,130 @@
+TCG Interpreter (TCI) - Copyright (c) 2011 Stefan Weil.
+
+This file is released under the BSD license.
+
+1) Introduction
+
+TCG (Tiny Code Generator) is a code generator which translates
+code fragments ("basic blocks") from target code (any of the
+targets supported by QEMU) to a code representation which
+can be run on a host.
+
+QEMU can create native code for some hosts (arm, hppa, i386, ia64, ppc, ppc64,
+s390, sparc, x86_64). For others, unofficial host support was written.
+
+By adding a code generator for a virtual machine and using an
+interpreter for the generated bytecode, it is possible to
+support (almost) any host.
+
+This is what TCI (Tiny Code Interpreter) does.
+
+2) Implementation
+
+Like each TCG host frontend, TCI implements the code generator in
+tcg-target.c, tcg-target.h. Both files are in directory tcg/tci.
+
+The additional file tcg/tci.c adds the interpreter.
+
+The bytecode consists of opcodes (same numeric values as those used by
+TCG), command length and arguments of variable size and number.
+
+3) Usage
+
+For hosts without native TCG, the interpreter TCI must be enabled by
+
+ configure --enable-tcg-interpreter
+
+If configure is called without --enable-tcg-interpreter, it will
+suggest using this option. Setting it automatically would need
+additional code in configure which must be fixed when new native TCG
+implementations are added.
+
+System emulation should work on any 32 or 64 bit host.
+User mode emulation might work. Maybe a new linker script (*.ld)
+is needed. Byte order might be wrong (on big endian hosts)
+and need fixes in configure.
+
+For hosts with native TCG, the interpreter TCI can be enabled by
+
+ configure --enable-tcg-interpreter
+
+The only difference from running QEMU with TCI to running without TCI
+should be speed. Especially during development of TCI, it was very
+useful to compare runs with and without TCI. Create /tmp/qemu.log by
+
+ qemu-system-i386 -d in_asm,op_opt,cpu -singlestep
+
+once with interpreter and once without interpreter and compare the resulting
+qemu.log files. This is also useful to see the effects of additional
+registers or additional opcodes (it is easy to modify the virtual machine).
+It can also be used to verify native TCGs.
+
+Hosts with native TCG can also enable TCI by claiming to be unsupported:
+
+ configure --cpu=unknown --enable-tcg-interpreter
+
+configure then no longer uses the native linker script (*.ld) for
+user mode emulation.
+
+
+4) Status
+
+TCI needs special implementation for 32 and 64 bit host, 32 and 64 bit target,
+host and target with same or different endianness.
+
+ | host (le) host (be)
+ | 32 64 32 64
+------------+------------------------------------------------------------
+target (le) | s0, u0 s1, u1 s?, u? s?, u?
+32 bit |
+ |
+target (le) | sc, uc s1, u1 s?, u? s?, u?
+64 bit |
+ |
+target (be) | sc, u0 sc, uc s?, u? s?, u?
+32 bit |
+ |
+target (be) | sc, uc sc, uc s?, u? s?, u?
+64 bit |
+ |
+
+System emulation
+s? = untested
+sc = compiles
+s0 = bios works
+s1 = grub works
+s2 = Linux boots
+
+Linux user mode emulation
+u? = untested
+uc = compiles
+u0 = static hello works
+u1 = linux-user-test works
+
+5) Todo list
+
+* TCI is not widely tested. It was written and tested on a x86_64 host
+ running i386 and x86_64 system emulation and Linux user mode.
+ A cross compiled QEMU for i386 host also works with the same basic tests.
+ A cross compiled QEMU for mipsel host works, too. It is terribly slow
+ because I run it in a mips malta emulation, so it is an interpreted
+ emulation in an emulation.
+ A cross compiled QEMU for arm host works (tested with pc bios).
+ A cross compiled QEMU for ppc host works at least partially:
+ i386-linux-user/qemu-i386 can run a simple hello-world program
+ (tested in a ppc emulation).
+
+* Some TCG opcodes are either missing in the code generator and/or
+ in the interpreter. These opcodes raise a runtime exception, so it is
+ possible to see where code must be added.
+
+* The pseudo code is not optimized and still ugly. For hosts with special
+ alignment requirements, it needs some fixes (maybe aligned bytecode
+ would also improve speed for hosts which support byte alignment).
+
+* A better disassembler for the pseudo code would be nice (a very primitive
+ disassembler is included in tcg-target.c).
+
+* It might be useful to have a runtime option which selects the native TCG
+ or TCI, so QEMU would have to include two TCGs. Today, selecting TCI
+ is a configure option, so you need two compilations of QEMU.
diff --git a/tcg/tci/tcg-target.c b/tcg/tci/tcg-target.c
new file mode 100644
index 0000000..7ad2312
--- /dev/null
+++ b/tcg/tci/tcg-target.c
@@ -0,0 +1,919 @@
+/*
+ * Tiny Code Generator for QEMU
+ *
+ * Copyright (c) 2009, 2011 Stefan Weil
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+/* TODO list:
+ * - See TODO comments in code.
+ */
+
+/* Marker for missing code. */
+#define TODO() \
+ do { \
+ fprintf(stderr, "TODO %s:%u: %s()\n", \
+ __FILE__, __LINE__, __func__); \
+ tcg_abort(); \
+ } while (0)
+
+/* Trace message to see program flow. */
+#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
+#define TRACE() \
+ loglevel \
+ ? fprintf(stderr, "TCG %s:%u: %s()\n", __FILE__, __LINE__, __func__) \
+ : (void)0
+#else
+#define TRACE() ((void)0)
+#endif
+
+/* Single bit n. */
+#define BIT(n) (1 << (n))
+
+/* Bitfield n...m (in 32 bit value). */
+#define BITS(n, m) (((0xffffffffU << (31 - n)) >> (31 - n + m)) << m)
+
+/* Used for function call generation. */
+#define TCG_REG_CALL_STACK TCG_REG_R4
+#define TCG_TARGET_STACK_ALIGN 16
+#define TCG_TARGET_CALL_STACK_OFFSET 0
+
+/* TODO: documentation. */
+static uint8_t *tb_ret_addr;
+
+/* Macros used in tcg_target_op_defs. */
+#define R "r"
+#define RI "ri"
+#if TCG_TARGET_REG_BITS == 32
+# define R64 "r", "r"
+#else
+# define R64 "r"
+#endif
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+# define L "L", "L"
+# define S "S", "S"
+#else
+# define L "L"
+# define S "S"
+#endif
+
+/* TODO: documentation. */
+static const TCGTargetOpDef tcg_target_op_defs[] = {
+ { INDEX_op_exit_tb, { } },
+ { INDEX_op_goto_tb, { } },
+ { INDEX_op_call, { RI } },
+ { INDEX_op_jmp, { RI } },
+ { INDEX_op_br, { } },
+
+ { INDEX_op_mov_i32, { R, R } },
+ { INDEX_op_movi_i32, { R } },
+
+ { INDEX_op_ld8u_i32, { R, R } },
+ { INDEX_op_ld8s_i32, { R, R } },
+ { INDEX_op_ld16u_i32, { R, R } },
+ { INDEX_op_ld16s_i32, { R, R } },
+ { INDEX_op_ld_i32, { R, R } },
+ { INDEX_op_st8_i32, { R, R } },
+ { INDEX_op_st16_i32, { R, R } },
+ { INDEX_op_st_i32, { R, R } },
+
+ { INDEX_op_add_i32, { R, RI, RI } },
+ { INDEX_op_sub_i32, { R, RI, RI } },
+ { INDEX_op_mul_i32, { R, RI, RI } },
+#if TCG_TARGET_HAS_div_i32
+ { INDEX_op_div_i32, { R, R, R } },
+ { INDEX_op_divu_i32, { R, R, R } },
+ { INDEX_op_rem_i32, { R, R, R } },
+ { INDEX_op_remu_i32, { R, R, R } },
+#elif TCG_TARGET_HAS_div2_i32
+ { INDEX_op_div2_i32, { R, R, "0", "1", R } },
+ { INDEX_op_divu2_i32, { R, R, "0", "1", R } },
+#endif
+ /* TODO: Does R, RI, RI result in faster code than R, R, RI?
+ If both operands are constants, we can optimize. */
+ { INDEX_op_and_i32, { R, RI, RI } },
+#if TCG_TARGET_HAS_andc_i32
+ { INDEX_op_andc_i32, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_eqv_i32
+ { INDEX_op_eqv_i32, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_nand_i32
+ { INDEX_op_nand_i32, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_nor_i32
+ { INDEX_op_nor_i32, { R, RI, RI } },
+#endif
+ { INDEX_op_or_i32, { R, RI, RI } },
+#if TCG_TARGET_HAS_orc_i32
+ { INDEX_op_orc_i32, { R, RI, RI } },
+#endif
+ { INDEX_op_xor_i32, { R, RI, RI } },
+ { INDEX_op_shl_i32, { R, RI, RI } },
+ { INDEX_op_shr_i32, { R, RI, RI } },
+ { INDEX_op_sar_i32, { R, RI, RI } },
+#if TCG_TARGET_HAS_rot_i32
+ { INDEX_op_rotl_i32, { R, RI, RI } },
+ { INDEX_op_rotr_i32, { R, RI, RI } },
+#endif
+
+ { INDEX_op_brcond_i32, { R, RI } },
+
+ { INDEX_op_setcond_i32, { R, R, RI } },
+#if TCG_TARGET_REG_BITS == 64
+ { INDEX_op_setcond_i64, { R, R, RI } },
+#endif /* TCG_TARGET_REG_BITS == 64 */
+
+#if TCG_TARGET_REG_BITS == 32
+ /* TODO: Support R, R, R, R, RI, RI? Will it be faster? */
+ { INDEX_op_add2_i32, { R, R, R, R, R, R } },
+ { INDEX_op_sub2_i32, { R, R, R, R, R, R } },
+ { INDEX_op_brcond2_i32, { R, R, RI, RI } },
+ { INDEX_op_mulu2_i32, { R, R, R, R } },
+ { INDEX_op_setcond2_i32, { R, R, R, RI, RI } },
+#endif
+
+#if TCG_TARGET_HAS_not_i32
+ { INDEX_op_not_i32, { R, R } },
+#endif
+#if TCG_TARGET_HAS_neg_i32
+ { INDEX_op_neg_i32, { R, R } },
+#endif
+
+#if TCG_TARGET_REG_BITS == 64
+ { INDEX_op_mov_i64, { R, R } },
+ { INDEX_op_movi_i64, { R } },
+
+ { INDEX_op_ld8u_i64, { R, R } },
+ { INDEX_op_ld8s_i64, { R, R } },
+ { INDEX_op_ld16u_i64, { R, R } },
+ { INDEX_op_ld16s_i64, { R, R } },
+ { INDEX_op_ld32u_i64, { R, R } },
+ { INDEX_op_ld32s_i64, { R, R } },
+ { INDEX_op_ld_i64, { R, R } },
+
+ { INDEX_op_st8_i64, { R, R } },
+ { INDEX_op_st16_i64, { R, R } },
+ { INDEX_op_st32_i64, { R, R } },
+ { INDEX_op_st_i64, { R, R } },
+
+ { INDEX_op_add_i64, { R, RI, RI } },
+ { INDEX_op_sub_i64, { R, RI, RI } },
+ { INDEX_op_mul_i64, { R, RI, RI } },
+#if TCG_TARGET_HAS_div_i64
+ { INDEX_op_div_i64, { R, R, R } },
+ { INDEX_op_divu_i64, { R, R, R } },
+ { INDEX_op_rem_i64, { R, R, R } },
+ { INDEX_op_remu_i64, { R, R, R } },
+#elif TCG_TARGET_HAS_div2_i64
+ { INDEX_op_div2_i64, { R, R, "0", "1", R } },
+ { INDEX_op_divu2_i64, { R, R, "0", "1", R } },
+#endif
+ { INDEX_op_and_i64, { R, RI, RI } },
+#if TCG_TARGET_HAS_andc_i64
+ { INDEX_op_andc_i64, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_eqv_i64
+ { INDEX_op_eqv_i64, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_nand_i64
+ { INDEX_op_nand_i64, { R, RI, RI } },
+#endif
+#if TCG_TARGET_HAS_nor_i64
+ { INDEX_op_nor_i64, { R, RI, RI } },
+#endif
+ { INDEX_op_or_i64, { R, RI, RI } },
+#if TCG_TARGET_HAS_orc_i64
+ { INDEX_op_orc_i64, { R, RI, RI } },
+#endif
+ { INDEX_op_xor_i64, { R, RI, RI } },
+ { INDEX_op_shl_i64, { R, RI, RI } },
+ { INDEX_op_shr_i64, { R, RI, RI } },
+ { INDEX_op_sar_i64, { R, RI, RI } },
+#if TCG_TARGET_HAS_rot_i64
+ { INDEX_op_rotl_i64, { R, RI, RI } },
+ { INDEX_op_rotr_i64, { R, RI, RI } },
+#endif
+ { INDEX_op_brcond_i64, { R, RI } },
+
+#if TCG_TARGET_HAS_ext8s_i64
+ { INDEX_op_ext8s_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext16s_i64
+ { INDEX_op_ext16s_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext32s_i64
+ { INDEX_op_ext32s_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext8u_i64
+ { INDEX_op_ext8u_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext16u_i64
+ { INDEX_op_ext16u_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext32u_i64
+ { INDEX_op_ext32u_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_bswap16_i64
+ { INDEX_op_bswap16_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_bswap32_i64
+ { INDEX_op_bswap32_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_bswap64_i64
+ { INDEX_op_bswap64_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_not_i64
+ { INDEX_op_not_i64, { R, R } },
+#endif
+#if TCG_TARGET_HAS_neg_i64
+ { INDEX_op_neg_i64, { R, R } },
+#endif
+#endif /* TCG_TARGET_REG_BITS == 64 */
+
+ { INDEX_op_qemu_ld8u, { R, L } },
+ { INDEX_op_qemu_ld8s, { R, L } },
+ { INDEX_op_qemu_ld16u, { R, L } },
+ { INDEX_op_qemu_ld16s, { R, L } },
+ { INDEX_op_qemu_ld32, { R, L } },
+#if TCG_TARGET_REG_BITS == 64
+ { INDEX_op_qemu_ld32u, { R, L } },
+ { INDEX_op_qemu_ld32s, { R, L } },
+#endif
+ { INDEX_op_qemu_ld64, { R64, L } },
+
+ { INDEX_op_qemu_st8, { R, S } },
+ { INDEX_op_qemu_st16, { R, S } },
+ { INDEX_op_qemu_st32, { R, S } },
+ { INDEX_op_qemu_st64, { R64, S } },
+
+#if TCG_TARGET_HAS_ext8s_i32
+ { INDEX_op_ext8s_i32, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext16s_i32
+ { INDEX_op_ext16s_i32, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext8u_i32
+ { INDEX_op_ext8u_i32, { R, R } },
+#endif
+#if TCG_TARGET_HAS_ext16u_i32
+ { INDEX_op_ext16u_i32, { R, R } },
+#endif
+
+#if TCG_TARGET_HAS_bswap16_i32
+ { INDEX_op_bswap16_i32, { R, R } },
+#endif
+#if TCG_TARGET_HAS_bswap32_i32
+ { INDEX_op_bswap32_i32, { R, R } },
+#endif
+
+ { -1 },
+};
+
+static const int tcg_target_reg_alloc_order[] = {
+ TCG_REG_R0,
+ TCG_REG_R1,
+ TCG_REG_R2,
+ TCG_REG_R3,
+#if 0 /* used for TCG_REG_CALL_STACK */
+ TCG_REG_R4,
+#endif
+ TCG_REG_R5,
+ TCG_REG_R6,
+ TCG_REG_R7,
+#if TCG_TARGET_NB_REGS >= 16
+ TCG_REG_R8,
+ TCG_REG_R9,
+ TCG_REG_R10,
+ TCG_REG_R11,
+ TCG_REG_R12,
+ TCG_REG_R13,
+ TCG_REG_R14,
+ TCG_REG_R15,
+#endif
+};
+
+#if MAX_OPC_PARAM_IARGS != 4
+# error Fix needed, number of supported input arguments changed!
+#endif
+
+static const int tcg_target_call_iarg_regs[] = {
+ TCG_REG_R0,
+ TCG_REG_R1,
+ TCG_REG_R2,
+ TCG_REG_R3,
+#if TCG_TARGET_REG_BITS == 32
+ /* 32 bit hosts need 2 * MAX_OPC_PARAM_IARGS registers. */
+#if 0 /* used for TCG_REG_CALL_STACK */
+ TCG_REG_R4,
+#endif
+ TCG_REG_R5,
+ TCG_REG_R6,
+ TCG_REG_R7,
+#if TCG_TARGET_NB_REGS >= 16
+ TCG_REG_R8,
+#else
+# error Too few input registers available
+#endif
+#endif
+};
+
+static const int tcg_target_call_oarg_regs[] = {
+ TCG_REG_R0,
+#if TCG_TARGET_REG_BITS == 32
+ TCG_REG_R1
+#endif
+};
+
+#ifndef NDEBUG
+static const char *const tcg_target_reg_names[TCG_TARGET_NB_REGS] = {
+ "r00",
+ "r01",
+ "r02",
+ "r03",
+ "r04",
+ "r05",
+ "r06",
+ "r07",
+#if TCG_TARGET_NB_REGS >= 16
+ "r08",
+ "r09",
+ "r10",
+ "r11",
+ "r12",
+ "r13",
+ "r14",
+ "r15",
+#if TCG_TARGET_NB_REGS >= 32
+ "r16",
+ "r17",
+ "r18",
+ "r19",
+ "r20",
+ "r21",
+ "r22",
+ "r23",
+ "r24",
+ "r25",
+ "r26",
+ "r27",
+ "r28",
+ "r29",
+ "r30",
+ "r31"
+#endif
+#endif
+};
+#endif
+
+static void flush_icache_range(unsigned long start, unsigned long stop)
+{
+ TRACE();
+}
+
+static void patch_reloc(uint8_t *code_ptr, int type,
+ tcg_target_long value, tcg_target_long addend)
+{
+ /* tcg_out_reloc always uses the same type, addend. */
+ assert(type == sizeof(tcg_target_long));
+ assert(addend == 0);
+ assert(value != 0);
+ *(tcg_target_long *)code_ptr = value;
+}
+
+/* Parse target specific constraints. */
+static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str)
+{
+ const char *ct_str = *pct_str;
+ switch (ct_str[0]) {
+ case 'r':
+ case 'L': /* qemu_ld constraint */
+ case 'S': /* qemu_st constraint */
+ ct->ct |= TCG_CT_REG;
+ tcg_regset_set32(ct->u.regs, 0, BIT(TCG_TARGET_NB_REGS) - 1);
+ break;
+ default:
+ return -1;
+ }
+ ct_str++;
+ *pct_str = ct_str;
+ return 0;
+}
+
+#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
+/* Show current bytecode. Used by tcg interpreter. */
+void tci_disas(uint8_t opc)
+{
+ const TCGOpDef *def = &tcg_op_defs[opc];
+ fprintf(stderr, "TCG %s %u, %u, %u\n",
+ def->name, def->nb_oargs, def->nb_iargs, def->nb_cargs);
+}
+#endif
+
+/* Write value (native size). */
+static void tcg_out_i(TCGContext *s, tcg_target_ulong v)
+{
+ *(tcg_target_ulong *)s->code_ptr = v;
+ s->code_ptr += sizeof(tcg_target_ulong);
+}
+
+/* Write 64 bit value. */
+static void tcg_out64(TCGContext *s, uint64_t v)
+{
+ *(uint64_t *)s->code_ptr = v;
+ s->code_ptr += sizeof(v);
+}
+
+/* Write opcode. */
+static void tcg_out_op_t(TCGContext *s, TCGOpcode op)
+{
+ tcg_out8(s, op);
+ tcg_out8(s, 0);
+}
+
+/* Write register. */
+static void tcg_out_r(TCGContext *s, TCGArg t0)
+{
+ assert(t0 < TCG_TARGET_NB_REGS);
+ tcg_out8(s, t0);
+}
+
+/* Write register or constant (native size). */
+static void tcg_out_ri(TCGContext *s, int const_arg, TCGArg arg)
+{
+ if (const_arg) {
+ assert(const_arg == 1);
+ tcg_out8(s, TCG_CONST);
+ tcg_out_i(s, arg);
+ } else {
+ tcg_out_r(s, arg);
+ }
+}
+
+/* Write register or constant (32 bit). */
+static void tcg_out_ri32(TCGContext *s, int const_arg, TCGArg arg)
+{
+ if (const_arg) {
+ assert(const_arg == 1);
+ tcg_out8(s, TCG_CONST);
+ tcg_out32(s, arg);
+ } else {
+ tcg_out_r(s, arg);
+ }
+}
+
+#if TCG_TARGET_REG_BITS == 64
+/* Write register or constant (64 bit). */
+static void tcg_out_ri64(TCGContext *s, int const_arg, TCGArg arg)
+{
+ if (const_arg) {
+ assert(const_arg == 1);
+ tcg_out8(s, TCG_CONST);
+ tcg_out64(s, arg);
+ } else {
+ tcg_out_r(s, arg);
+ }
+}
+#endif
+
+/* Write label. */
+static void tci_out_label(TCGContext *s, TCGArg arg)
+{
+ TCGLabel *label = &s->labels[arg];
+ if (label->has_value) {
+ tcg_out_i(s, label->u.value);
+ assert(label->u.value);
+ } else {
+ tcg_out_reloc(s, s->code_ptr, sizeof(tcg_target_ulong), arg, 0);
+ tcg_out_i(s, 0);
+ }
+}
+
+static void tcg_out_ld(TCGContext *s, TCGType type, int ret, int arg1,
+ tcg_target_long arg2)
+{
+ uint8_t *old_code_ptr = s->code_ptr;
+ if (type == TCG_TYPE_I32) {
+ tcg_out_op_t(s, INDEX_op_ld_i32);
+ tcg_out_r(s, ret);
+ tcg_out_r(s, arg1);
+ tcg_out32(s, arg2);
+ } else {
+ assert(type == TCG_TYPE_I64);
+#if TCG_TARGET_REG_BITS == 64
+ tcg_out_op_t(s, INDEX_op_ld_i64);
+ tcg_out_r(s, ret);
+ tcg_out_r(s, arg1);
+ assert(arg2 == (uint32_t)arg2);
+ tcg_out32(s, arg2);
+#else
+ TODO();
+#endif
+ }
+ old_code_ptr[1] = s->code_ptr - old_code_ptr;
+}
+
+static void tcg_out_mov(TCGContext *s, TCGType type, int ret, int arg)
+{
+ uint8_t *old_code_ptr = s->code_ptr;
+ assert(ret != arg);
+#if TCG_TARGET_REG_BITS == 32
+ tcg_out_op_t(s, INDEX_op_mov_i32);
+#else
+ tcg_out_op_t(s, INDEX_op_mov_i64);
+#endif
+ tcg_out_r(s, ret);
+ tcg_out_r(s, arg);
+ old_code_ptr[1] = s->code_ptr - old_code_ptr;
+}
+
+static void tcg_out_movi(TCGContext *s, TCGType type,
+ int t0, tcg_target_long arg)
+{
+ uint8_t *old_code_ptr = s->code_ptr;
+ uint32_t arg32 = arg;
+ if (type == TCG_TYPE_I32 || arg == arg32) {
+ tcg_out_op_t(s, INDEX_op_movi_i32);
+ tcg_out_r(s, t0);
+ tcg_out32(s, arg32);
+ } else {
+ assert(type == TCG_TYPE_I64);
+#if TCG_TARGET_REG_BITS == 64
+ tcg_out_op_t(s, INDEX_op_movi_i64);
+ tcg_out_r(s, t0);
+ tcg_out64(s, arg);
+#else
+ TODO();
+#endif
+ }
+ old_code_ptr[1] = s->code_ptr - old_code_ptr;
+}
+
+static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args,
+ const int *const_args)
+{
+ uint8_t *old_code_ptr = s->code_ptr;
+
+ tcg_out_op_t(s, opc);
+
+ switch (opc) {
+ case INDEX_op_exit_tb:
+ tcg_out64(s, args[0]);
+ break;
+ case INDEX_op_goto_tb:
+ if (s->tb_jmp_offset) {
+ /* Direct jump method. */
+ assert(args[0] < ARRAY_SIZE(s->tb_jmp_offset));
+ s->tb_jmp_offset[args[0]] = s->code_ptr - s->code_buf;
+ tcg_out32(s, 0);
+ } else {
+ /* Indirect jump method. */
+ TODO();
+ }
+ assert(args[0] < ARRAY_SIZE(s->tb_next_offset));
+ s->tb_next_offset[args[0]] = s->code_ptr - s->code_buf;
+ break;
+ case INDEX_op_br:
+ tci_out_label(s, args[0]);
+ break;
+ case INDEX_op_call:
+ tcg_out_ri(s, const_args[0], args[0]);
+ break;
+ case INDEX_op_jmp:
+ TODO();
+ break;
+ case INDEX_op_setcond_i32:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_ri32(s, const_args[2], args[2]);
+ tcg_out8(s, args[3]); /* condition */
+ break;
+#if TCG_TARGET_REG_BITS == 32
+ case INDEX_op_setcond2_i32:
+ /* setcond2_i32 cond, t0, t1_low, t1_high, t2_low, t2_high */
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_r(s, args[2]);
+ tcg_out_ri32(s, const_args[3], args[3]);
+ tcg_out_ri32(s, const_args[4], args[4]);
+ tcg_out8(s, args[5]); /* condition */
+ break;
+#elif TCG_TARGET_REG_BITS == 64
+ case INDEX_op_setcond_i64:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_ri64(s, const_args[2], args[2]);
+ tcg_out8(s, args[3]); /* condition */
+ break;
+#endif
+ case INDEX_op_movi_i32:
+ TODO(); /* Handled by tcg_out_movi? */
+ break;
+ case INDEX_op_ld8u_i32:
+ case INDEX_op_ld8s_i32:
+ case INDEX_op_ld16u_i32:
+ case INDEX_op_ld16s_i32:
+ case INDEX_op_ld_i32:
+ case INDEX_op_st8_i32:
+ case INDEX_op_st16_i32:
+ case INDEX_op_st_i32:
+ case INDEX_op_ld8u_i64:
+ case INDEX_op_ld8s_i64:
+ case INDEX_op_ld16u_i64:
+ case INDEX_op_ld16s_i64:
+ case INDEX_op_ld32u_i64:
+ case INDEX_op_ld32s_i64:
+ case INDEX_op_ld_i64:
+ case INDEX_op_st8_i64:
+ case INDEX_op_st16_i64:
+ case INDEX_op_st32_i64:
+ case INDEX_op_st_i64:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ assert(args[2] == (uint32_t)args[2]);
+ tcg_out32(s, args[2]);
+ break;
+ case INDEX_op_add_i32:
+ case INDEX_op_sub_i32:
+ case INDEX_op_mul_i32:
+ case INDEX_op_and_i32:
+ case INDEX_op_andc_i32: /* Optional (TCG_TARGET_HAS_andc_i32). */
+ case INDEX_op_eqv_i32: /* Optional (TCG_TARGET_HAS_eqv_i32). */
+ case INDEX_op_nand_i32: /* Optional (TCG_TARGET_HAS_nand_i32). */
+ case INDEX_op_nor_i32: /* Optional (TCG_TARGET_HAS_nor_i32). */
+ case INDEX_op_or_i32:
+ case INDEX_op_orc_i32: /* Optional (TCG_TARGET_HAS_orc_i32). */
+ case INDEX_op_xor_i32:
+ case INDEX_op_shl_i32:
+ case INDEX_op_shr_i32:
+ case INDEX_op_sar_i32:
+ case INDEX_op_rotl_i32: /* Optional (TCG_TARGET_HAS_rot_i32). */
+ case INDEX_op_rotr_i32: /* Optional (TCG_TARGET_HAS_rot_i32). */
+ tcg_out_r(s, args[0]);
+ tcg_out_ri32(s, const_args[1], args[1]);
+ tcg_out_ri32(s, const_args[2], args[2]);
+ break;
+
+#if TCG_TARGET_REG_BITS == 64
+ case INDEX_op_mov_i64:
+ case INDEX_op_movi_i64:
+ TODO();
+ break;
+ case INDEX_op_add_i64:
+ case INDEX_op_sub_i64:
+ case INDEX_op_mul_i64:
+ case INDEX_op_and_i64:
+ case INDEX_op_andc_i64: /* Optional (TCG_TARGET_HAS_andc_i64). */
+ case INDEX_op_eqv_i64: /* Optional (TCG_TARGET_HAS_eqv_i64). */
+ case INDEX_op_nand_i64: /* Optional (TCG_TARGET_HAS_nand_i64). */
+ case INDEX_op_nor_i64: /* Optional (TCG_TARGET_HAS_nor_i64). */
+ case INDEX_op_or_i64:
+ case INDEX_op_orc_i64: /* Optional (TCG_TARGET_HAS_orc_i64). */
+ case INDEX_op_xor_i64:
+ case INDEX_op_shl_i64:
+ case INDEX_op_shr_i64:
+ case INDEX_op_sar_i64:
+ /* TODO: Implementation of rotl_i64, rotr_i64 missing in tci.c. */
+ case INDEX_op_rotl_i64: /* Optional (TCG_TARGET_HAS_rot_i64). */
+ case INDEX_op_rotr_i64: /* Optional (TCG_TARGET_HAS_rot_i64). */
+ tcg_out_r(s, args[0]);
+ tcg_out_ri64(s, const_args[1], args[1]);
+ tcg_out_ri64(s, const_args[2], args[2]);
+ break;
+ case INDEX_op_div_i64: /* Optional (TCG_TARGET_HAS_div_i64). */
+ case INDEX_op_divu_i64: /* Optional (TCG_TARGET_HAS_div_i64). */
+ case INDEX_op_rem_i64: /* Optional (TCG_TARGET_HAS_div_i64). */
+ case INDEX_op_remu_i64: /* Optional (TCG_TARGET_HAS_div_i64). */
+ TODO();
+ break;
+ case INDEX_op_div2_i64: /* Optional (TCG_TARGET_HAS_div2_i64). */
+ case INDEX_op_divu2_i64: /* Optional (TCG_TARGET_HAS_div2_i64). */
+ TODO();
+ break;
+ case INDEX_op_brcond_i64:
+ tcg_out_r(s, args[0]);
+ tcg_out_ri64(s, const_args[1], args[1]);
+ tcg_out8(s, args[2]); /* condition */
+ tci_out_label(s, args[3]);
+ break;
+ case INDEX_op_bswap16_i64: /* Optional (TCG_TARGET_HAS_bswap16_i64). */
+ case INDEX_op_bswap32_i64: /* Optional (TCG_TARGET_HAS_bswap32_i64). */
+ case INDEX_op_bswap64_i64: /* Optional (TCG_TARGET_HAS_bswap64_i64). */
+ case INDEX_op_not_i64: /* Optional (TCG_TARGET_HAS_not_i64). */
+ case INDEX_op_neg_i64: /* Optional (TCG_TARGET_HAS_neg_i64). */
+ case INDEX_op_ext8s_i64: /* Optional (TCG_TARGET_HAS_ext8s_i64). */
+ case INDEX_op_ext8u_i64: /* Optional (TCG_TARGET_HAS_ext8u_i64). */
+ case INDEX_op_ext16s_i64: /* Optional (TCG_TARGET_HAS_ext16s_i64). */
+ case INDEX_op_ext16u_i64: /* Optional (TCG_TARGET_HAS_ext16u_i64). */
+ case INDEX_op_ext32s_i64: /* Optional (TCG_TARGET_HAS_ext32s_i64). */
+ case INDEX_op_ext32u_i64: /* Optional (TCG_TARGET_HAS_ext32u_i64). */
+#endif /* TCG_TARGET_REG_BITS == 64 */
+ case INDEX_op_neg_i32: /* Optional (TCG_TARGET_HAS_neg_i32). */
+ case INDEX_op_not_i32: /* Optional (TCG_TARGET_HAS_not_i32). */
+ case INDEX_op_ext8s_i32: /* Optional (TCG_TARGET_HAS_ext8s_i32). */
+ case INDEX_op_ext16s_i32: /* Optional (TCG_TARGET_HAS_ext16s_i32). */
+ case INDEX_op_ext8u_i32: /* Optional (TCG_TARGET_HAS_ext8u_i32). */
+ case INDEX_op_ext16u_i32: /* Optional (TCG_TARGET_HAS_ext16u_i32). */
+ case INDEX_op_bswap16_i32: /* Optional (TCG_TARGET_HAS_bswap16_i32). */
+ case INDEX_op_bswap32_i32: /* Optional (TCG_TARGET_HAS_bswap32_i32). */
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ break;
+ case INDEX_op_div_i32: /* Optional (TCG_TARGET_HAS_div_i32). */
+ case INDEX_op_divu_i32: /* Optional (TCG_TARGET_HAS_div_i32). */
+ case INDEX_op_rem_i32: /* Optional (TCG_TARGET_HAS_div_i32). */
+ case INDEX_op_remu_i32: /* Optional (TCG_TARGET_HAS_div_i32). */
+ tcg_out_r(s, args[0]);
+ tcg_out_ri32(s, const_args[1], args[1]);
+ tcg_out_ri32(s, const_args[2], args[2]);
+ break;
+ case INDEX_op_div2_i32: /* Optional (TCG_TARGET_HAS_div2_i32). */
+ case INDEX_op_divu2_i32: /* Optional (TCG_TARGET_HAS_div2_i32). */
+ TODO();
+ break;
+#if TCG_TARGET_REG_BITS == 32
+ case INDEX_op_add2_i32:
+ case INDEX_op_sub2_i32:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_r(s, args[2]);
+ tcg_out_r(s, args[3]);
+ tcg_out_r(s, args[4]);
+ tcg_out_r(s, args[5]);
+ break;
+ case INDEX_op_brcond2_i32:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_ri32(s, const_args[2], args[2]);
+ tcg_out_ri32(s, const_args[3], args[3]);
+ tcg_out8(s, args[4]); /* condition */
+ tci_out_label(s, args[5]);
+ break;
+ case INDEX_op_mulu2_i32:
+ tcg_out_r(s, args[0]);
+ tcg_out_r(s, args[1]);
+ tcg_out_r(s, args[2]);
+ tcg_out_r(s, args[3]);
+ break;
+#endif
+ case INDEX_op_brcond_i32:
+ tcg_out_r(s, args[0]);
+ tcg_out_ri32(s, const_args[1], args[1]);
+ tcg_out8(s, args[2]); /* condition */
+ tci_out_label(s, args[3]);
+ break;
+ case INDEX_op_qemu_ld8u:
+ case INDEX_op_qemu_ld8s:
+ case INDEX_op_qemu_ld16u:
+ case INDEX_op_qemu_ld16s:
+ case INDEX_op_qemu_ld32:
+#if TCG_TARGET_REG_BITS == 64
+ case INDEX_op_qemu_ld32s:
+ case INDEX_op_qemu_ld32u:
+#endif
+ tcg_out_r(s, *args++);
+ tcg_out_r(s, *args++);
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ tcg_out_r(s, *args++);
+#endif
+#ifdef CONFIG_SOFTMMU
+ tcg_out_i(s, *args);
+#endif
+ break;
+ case INDEX_op_qemu_ld64:
+ tcg_out_r(s, *args++);
+#if TCG_TARGET_REG_BITS == 32
+ tcg_out_r(s, *args++);
+#endif
+ tcg_out_r(s, *args++);
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ tcg_out_r(s, *args++);
+#endif
+#ifdef CONFIG_SOFTMMU
+ tcg_out_i(s, *args);
+#endif
+ break;
+ case INDEX_op_qemu_st8:
+ case INDEX_op_qemu_st16:
+ case INDEX_op_qemu_st32:
+ tcg_out_r(s, *args++);
+ tcg_out_r(s, *args++);
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ tcg_out_r(s, *args++);
+#endif
+#ifdef CONFIG_SOFTMMU
+ tcg_out_i(s, *args);
+#endif
+ break;
+ case INDEX_op_qemu_st64:
+ tcg_out_r(s, *args++);
+#if TCG_TARGET_REG_BITS == 32
+ tcg_out_r(s, *args++);
+#endif
+ tcg_out_r(s, *args++);
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ tcg_out_r(s, *args++);
+#endif
+#ifdef CONFIG_SOFTMMU
+ tcg_out_i(s, *args);
+#endif
+ break;
+ case INDEX_op_end:
+ TODO();
+ break;
+ default:
+ fprintf(stderr, "Missing: %s\n", tcg_op_defs[opc].name);
+ tcg_abort();
+ }
+ old_code_ptr[1] = s->code_ptr - old_code_ptr;
+}
+
+static void tcg_out_st(TCGContext *s, TCGType type, int arg, int arg1,
+ tcg_target_long arg2)
+{
+ uint8_t *old_code_ptr = s->code_ptr;
+ if (type == TCG_TYPE_I32) {
+ tcg_out_op_t(s, INDEX_op_st_i32);
+ tcg_out_r(s, arg);
+ tcg_out_r(s, arg1);
+ tcg_out32(s, arg2);
+ } else {
+ assert(type == TCG_TYPE_I64);
+#if TCG_TARGET_REG_BITS == 64
+ tcg_out_op_t(s, INDEX_op_st_i64);
+ tcg_out_r(s, arg);
+ tcg_out_r(s, arg1);
+ tcg_out32(s, arg2);
+#else
+ TODO();
+#endif
+ }
+ old_code_ptr[1] = s->code_ptr - old_code_ptr;
+}
+
+/* Test if a constant matches the constraint. */
+static int tcg_target_const_match(tcg_target_long val,
+ const TCGArgConstraint *arg_ct)
+{
+ /* No need to return 0 or 1, 0 or != 0 is good enough. */
+ return arg_ct->ct & TCG_CT_CONST;
+}
+
+/* Maximum number of register used for input function arguments. */
+static int tcg_target_get_call_iarg_regs_count(int flags)
+{
+ return ARRAY_SIZE(tcg_target_call_iarg_regs);
+}
+
+static void tcg_target_init(TCGContext *s)
+{
+#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
+ const char *envval = getenv("DEBUG_TCG");
+ if (envval) {
+ loglevel = strtol(envval, NULL, 0);
+ }
+#endif
+ TRACE();
+
+ /* The current code uses uint8_t for tcg operations. */
+ assert(ARRAY_SIZE(tcg_op_defs) <= UINT8_MAX);
+
+ /* Registers available for 32 bit operations. */
+ tcg_regset_set32(tcg_target_available_regs[TCG_TYPE_I32], 0,
+ BIT(TCG_TARGET_NB_REGS) - 1);
+ /* Registers available for 64 bit operations. */
+ tcg_regset_set32(tcg_target_available_regs[TCG_TYPE_I64], 0,
+ BIT(TCG_TARGET_NB_REGS) - 1);
+ /* TODO: Which registers should be set here? */
+ tcg_regset_set32(tcg_target_call_clobber_regs, 0,
+ BIT(TCG_TARGET_NB_REGS) - 1);
+ tcg_regset_clear(s->reserved_regs);
+ tcg_regset_set_reg(s->reserved_regs, TCG_REG_CALL_STACK);
+ tcg_add_target_add_op_defs(tcg_target_op_defs);
+ tcg_set_frame(s, TCG_AREG0, offsetof(CPUState, temp_buf),
+ CPU_TEMP_BUF_NLONGS * sizeof(long));
+}
+
+/* Generate global QEMU prologue and epilogue code. */
+static void tcg_target_qemu_prologue(TCGContext *s)
+{
+ TRACE();
+ tb_ret_addr = s->code_ptr;
+}
diff --git a/tcg/tci/tcg-target.h b/tcg/tci/tcg-target.h
new file mode 100644
index 0000000..81ded86
--- /dev/null
+++ b/tcg/tci/tcg-target.h
@@ -0,0 +1,160 @@
+/*
+ * Tiny Code Generator for QEMU
+ *
+ * Copyright (c) 2009, 2011 Stefan Weil
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+/*
+ * This code implements a TCG which does not generate machine code for some
+ * real target machine but which generates virtual machine code for an
+ * interpreter. Interpreted pseudo code is slow, but it works on any host.
+ *
+ * Some remarks might help in understanding the code:
+ *
+ * "target" or "TCG target" is the machine which runs the generated code.
+ * This is different to the usual meaning in QEMU where "target" is the
+ * emulated machine. So normally QEMU host is identical to TCG target.
+ * Here the TCG target is a virtual machine, but this virtual machine must
+ * use the same word size like the real machine.
+ * Therefore, we need both 32 and 64 bit virtual machines (interpreter).
+ */
+
+#if !defined(TCG_TARGET_H)
+#define TCG_TARGET_H
+
+#include "config-host.h"
+
+#define TCG_TARGET_INTERPRETER 1
+
+#ifdef CONFIG_DEBUG_TCG
+/* Enable debug output. */
+#define CONFIG_DEBUG_TCG_INTERPRETER
+#endif
+
+#if 0 /* TCI tries to emulate a little endian host. */
+#if defined(HOST_WORDS_BIGENDIAN)
+# define TCG_TARGET_WORDS_BIGENDIAN
+#endif
+#endif
+
+/* Optional instructions. */
+
+#define TCG_TARGET_HAS_bswap16_i32 1
+#define TCG_TARGET_HAS_bswap32_i32 1
+/* Not more than one of the next two defines must be 1. */
+#define TCG_TARGET_HAS_div_i32 1
+#define TCG_TARGET_HAS_div2_i32 0
+#define TCG_TARGET_HAS_ext8s_i32 1
+#define TCG_TARGET_HAS_ext16s_i32 1
+#define TCG_TARGET_HAS_ext8u_i32 1
+#define TCG_TARGET_HAS_ext16u_i32 1
+#define TCG_TARGET_HAS_andc_i32 0
+#define TCG_TARGET_HAS_deposit_i32 0
+#define TCG_TARGET_HAS_eqv_i32 0
+#define TCG_TARGET_HAS_nand_i32 0
+#define TCG_TARGET_HAS_nor_i32 0
+#define TCG_TARGET_HAS_neg_i32 1
+#define TCG_TARGET_HAS_not_i32 1
+#define TCG_TARGET_HAS_orc_i32 0
+#define TCG_TARGET_HAS_rot_i32 1
+
+#if TCG_TARGET_REG_BITS == 64
+#define TCG_TARGET_HAS_bswap16_i64 1
+#define TCG_TARGET_HAS_bswap32_i64 1
+#define TCG_TARGET_HAS_bswap64_i64 1
+#define TCG_TARGET_HAS_deposit_i64 0
+/* Not more than one of the next two defines must be 1. */
+#define TCG_TARGET_HAS_div_i64 0
+#define TCG_TARGET_HAS_div2_i64 0
+#define TCG_TARGET_HAS_ext8s_i64 1
+#define TCG_TARGET_HAS_ext16s_i64 1
+#define TCG_TARGET_HAS_ext32s_i64 1
+#define TCG_TARGET_HAS_ext8u_i64 1
+#define TCG_TARGET_HAS_ext16u_i64 1
+#define TCG_TARGET_HAS_ext32u_i64 1
+#define TCG_TARGET_HAS_andc_i64 0
+#define TCG_TARGET_HAS_eqv_i64 0
+#define TCG_TARGET_HAS_nand_i64 0
+#define TCG_TARGET_HAS_nor_i64 0
+#define TCG_TARGET_HAS_neg_i64 1
+#define TCG_TARGET_HAS_not_i64 1
+#define TCG_TARGET_HAS_orc_i64 0
+#define TCG_TARGET_HAS_rot_i64 1
+#endif /* TCG_TARGET_REG_BITS == 64 */
+
+/* Offset to user memory in user mode. */
+#define TCG_TARGET_HAS_GUEST_BASE
+
+/* Number of registers available.
+ For 32 bit hosts, we need more than 8 registers (call arguments). */
+/* #define TCG_TARGET_NB_REGS 8 */
+#define TCG_TARGET_NB_REGS 16
+/* #define TCG_TARGET_NB_REGS 32 */
+
+/* List of registers which are used by TCG. */
+typedef enum {
+ TCG_REG_R0 = 0,
+ TCG_REG_R1,
+ TCG_REG_R2,
+ TCG_REG_R3,
+ TCG_REG_R4,
+ TCG_REG_R5,
+ TCG_REG_R6,
+ TCG_REG_R7,
+ TCG_AREG0 = TCG_REG_R7,
+#if TCG_TARGET_NB_REGS >= 16
+ TCG_REG_R8,
+ TCG_REG_R9,
+ TCG_REG_R10,
+ TCG_REG_R11,
+ TCG_REG_R12,
+ TCG_REG_R13,
+ TCG_REG_R14,
+ TCG_REG_R15,
+#if TCG_TARGET_NB_REGS >= 32
+ TCG_REG_R16,
+ TCG_REG_R17,
+ TCG_REG_R18,
+ TCG_REG_R19,
+ TCG_REG_R20,
+ TCG_REG_R21,
+ TCG_REG_R22,
+ TCG_REG_R23,
+ TCG_REG_R24,
+ TCG_REG_R25,
+ TCG_REG_R26,
+ TCG_REG_R27,
+ TCG_REG_R28,
+ TCG_REG_R29,
+ TCG_REG_R30,
+ TCG_REG_R31,
+#endif
+#endif
+ /* Special value UINT8_MAX is used by TCI to encode constant values. */
+ TCG_CONST = UINT8_MAX
+} TCGRegister;
+
+void tci_disas(uint8_t opc);
+
+unsigned long tcg_qemu_tb_exec(CPUState *env, uint8_t *tb_ptr);
+#define tcg_qemu_tb_exec tcg_qemu_tb_exec
+
+#endif /* TCG_TARGET_H */
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
` (2 preceding siblings ...)
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 3/6] tcg: Add bytecode generator for tcg interpreter Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
2011-10-20 21:36 ` malc
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 5/6] tcg: Add tci disassembler Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 6/6] tcg: Add tcg interpreter to configure / make Stefan Weil
5 siblings, 1 reply; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
tci.c | 1225 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 1225 insertions(+), 0 deletions(-)
create mode 100644 tci.c
diff --git a/tci.c b/tci.c
new file mode 100644
index 0000000..470d696
--- /dev/null
+++ b/tci.c
@@ -0,0 +1,1225 @@
+/*
+ * Tiny Code Interpreter for QEMU
+ *
+ * Copyright (c) 2009, 2011 Stefan Weil
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "config.h"
+
+/* Defining NDEBUG disables assertions (which makes the code faster). */
+#if !defined(CONFIG_TCG_DEBUG) && !defined(NDEBUG)
+# define NDEBUG
+#endif
+
+#include "qemu-common.h"
+#include "dyngen-exec.h" /* env */
+#include "exec-all.h" /* MAX_OPC_PARAM_IARGS */
+#include "tcg-op.h"
+
+/* Marker for missing code. */
+#define TODO() \
+ do { \
+ fprintf(stderr, "TODO %s:%u: %s()\n", \
+ __FILE__, __LINE__, __func__); \
+ tcg_abort(); \
+ } while (0)
+
+/* Trace message to see program flow. */
+#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
+#define TRACE() \
+ loglevel \
+ ? fprintf(stderr, "TCG %s:%u: %s()\n", __FILE__, __LINE__, __func__) \
+ : (void)0
+#else
+#define TRACE() ((void)0)
+#endif
+
+#if MAX_OPC_PARAM_IARGS != 4
+# error Fix needed, number of supported input arguments changed!
+#endif
+#if TCG_TARGET_REG_BITS == 32
+typedef uint64_t (*helper_function)(tcg_target_ulong, tcg_target_ulong,
+ tcg_target_ulong, tcg_target_ulong,
+ tcg_target_ulong, tcg_target_ulong,
+ tcg_target_ulong, tcg_target_ulong);
+#else
+typedef uint64_t (*helper_function)(tcg_target_ulong, tcg_target_ulong,
+ tcg_target_ulong, tcg_target_ulong);
+#endif
+
+/* TCI can optionally use a global register variable for env. */
+#if !defined(AREG0)
+CPUState *env;
+#endif
+
+/* Targets which don't use GETPC also don't need tci_tb_ptr
+ which makes them a little faster. */
+#if defined(GETPC)
+void *tci_tb_ptr;
+#endif
+
+static tcg_target_ulong tci_reg[TCG_TARGET_NB_REGS];
+
+static tcg_target_ulong tci_read_reg(TCGRegister index)
+{
+ assert(index < ARRAY_SIZE(tci_reg));
+ return tci_reg[index];
+}
+
+#if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64
+static int8_t tci_read_reg8s(TCGRegister index)
+{
+ return (int8_t)tci_read_reg(index);
+}
+#endif
+
+#if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64
+static int16_t tci_read_reg16s(TCGRegister index)
+{
+ return (int16_t)tci_read_reg(index);
+}
+#endif
+
+#if TCG_TARGET_REG_BITS == 64
+static int32_t tci_read_reg32s(TCGRegister index)
+{
+ return (int32_t)tci_read_reg(index);
+}
+#endif
+
+static uint8_t tci_read_reg8(TCGRegister index)
+{
+ return (uint8_t)tci_read_reg(index);
+}
+
+static uint16_t tci_read_reg16(TCGRegister index)
+{
+ return (uint16_t)tci_read_reg(index);
+}
+
+static uint32_t tci_read_reg32(TCGRegister index)
+{
+ return (uint32_t)tci_read_reg(index);
+}
+
+#if TCG_TARGET_REG_BITS == 64
+static uint64_t tci_read_reg64(TCGRegister index)
+{
+ return tci_read_reg(index);
+}
+#endif
+
+static void tci_write_reg(TCGRegister index, tcg_target_ulong value)
+{
+ assert(index < ARRAY_SIZE(tci_reg));
+ assert(index != TCG_AREG0);
+ tci_reg[index] = value;
+}
+
+static void tci_write_reg8s(TCGRegister index, int8_t value)
+{
+ tci_write_reg(index, value);
+}
+
+static void tci_write_reg16s(TCGRegister index, int16_t value)
+{
+ tci_write_reg(index, value);
+}
+
+#if TCG_TARGET_REG_BITS == 64
+static void tci_write_reg32s(TCGRegister index, int32_t value)
+{
+ tci_write_reg(index, value);
+}
+#endif
+
+static void tci_write_reg8(TCGRegister index, uint8_t value)
+{
+ tci_write_reg(index, value);
+}
+
+static void tci_write_reg16(TCGRegister index, uint16_t value)
+{
+ tci_write_reg(index, value);
+}
+
+static void tci_write_reg32(TCGRegister index, uint32_t value)
+{
+ tci_write_reg(index, value);
+}
+
+#if TCG_TARGET_REG_BITS == 32
+static void tci_write_reg64(uint32_t high_index, uint32_t low_index,
+ uint64_t value)
+{
+ tci_write_reg(low_index, value);
+ tci_write_reg(high_index, value >> 32);
+}
+#elif TCG_TARGET_REG_BITS == 64
+static void tci_write_reg64(TCGRegister index, uint64_t value)
+{
+ tci_write_reg(index, value);
+}
+#endif
+
+#if TCG_TARGET_REG_BITS == 32
+/* Create a 64 bit value from two 32 bit values. */
+static uint64_t tci_uint64(uint32_t high, uint32_t low)
+{
+ return ((uint64_t)high << 32) + low;
+}
+#endif
+
+/* Read constant (native size) from bytecode. */
+static tcg_target_ulong tci_read_i(uint8_t **tb_ptr)
+{
+ tcg_target_ulong value = *(tcg_target_ulong *)(*tb_ptr);
+ *tb_ptr += sizeof(value);
+ return value;
+}
+
+/* Read constant (32 bit) from bytecode. */
+static uint32_t tci_read_i32(uint8_t **tb_ptr)
+{
+ uint32_t value = *(uint32_t *)(*tb_ptr);
+ *tb_ptr += sizeof(value);
+ return value;
+}
+
+#if TCG_TARGET_REG_BITS == 64
+/* Read constant (64 bit) from bytecode. */
+static uint64_t tci_read_i64(uint8_t **tb_ptr)
+{
+ uint64_t value = *(uint64_t *)(*tb_ptr);
+ *tb_ptr += sizeof(value);
+ return value;
+}
+#endif
+
+/* Read indexed register (native size) from bytecode. */
+static tcg_target_ulong tci_read_r(uint8_t **tb_ptr)
+{
+ tcg_target_ulong value = tci_read_reg(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+
+/* Read indexed register (8 bit) from bytecode. */
+static uint8_t tci_read_r8(uint8_t **tb_ptr)
+{
+ uint8_t value = tci_read_reg8(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+
+#if TCG_TARGET_HAS_ext8s_i32 || TCG_TARGET_HAS_ext8s_i64
+/* Read indexed register (8 bit signed) from bytecode. */
+static int8_t tci_read_r8s(uint8_t **tb_ptr)
+{
+ int8_t value = tci_read_reg8s(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+#endif
+
+/* Read indexed register (16 bit) from bytecode. */
+static uint16_t tci_read_r16(uint8_t **tb_ptr)
+{
+ uint16_t value = tci_read_reg16(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+
+#if TCG_TARGET_HAS_ext16s_i32 || TCG_TARGET_HAS_ext16s_i64
+/* Read indexed register (16 bit signed) from bytecode. */
+static int16_t tci_read_r16s(uint8_t **tb_ptr)
+{
+ int16_t value = tci_read_reg16s(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+#endif
+
+/* Read indexed register (32 bit) from bytecode. */
+static uint32_t tci_read_r32(uint8_t **tb_ptr)
+{
+ uint32_t value = tci_read_reg32(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+
+#if TCG_TARGET_REG_BITS == 32
+/* Read two indexed registers (2 * 32 bit) from bytecode. */
+static uint64_t tci_read_r64(uint8_t **tb_ptr)
+{
+ uint32_t low = tci_read_r32(tb_ptr);
+ return tci_uint64(tci_read_r32(tb_ptr), low);
+}
+#elif TCG_TARGET_REG_BITS == 64
+/* Read indexed register (32 bit signed) from bytecode. */
+static int32_t tci_read_r32s(uint8_t **tb_ptr)
+{
+ int32_t value = tci_read_reg32s(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+
+/* Read indexed register (64 bit) from bytecode. */
+static uint64_t tci_read_r64(uint8_t **tb_ptr)
+{
+ uint64_t value = tci_read_reg64(**tb_ptr);
+ *tb_ptr += 1;
+ return value;
+}
+#endif
+
+/* Read indexed register(s) with target address from bytecode. */
+static target_ulong tci_read_ulong(uint8_t **tb_ptr)
+{
+ target_ulong taddr = tci_read_r(tb_ptr);
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ taddr += (uint64_t)tci_read_r(tb_ptr) << 32;
+#endif
+ return taddr;
+}
+
+/* Read indexed register or constant (native size) from bytecode. */
+static tcg_target_ulong tci_read_ri(uint8_t **tb_ptr)
+{
+ tcg_target_ulong value;
+ TCGRegister r = **tb_ptr;
+ *tb_ptr += 1;
+ if (r == TCG_CONST) {
+ value = tci_read_i(tb_ptr);
+ } else {
+ value = tci_read_reg(r);
+ }
+ return value;
+}
+
+/* Read indexed register or constant (32 bit) from bytecode. */
+static uint32_t tci_read_ri32(uint8_t **tb_ptr)
+{
+ uint32_t value;
+ TCGRegister r = **tb_ptr;
+ *tb_ptr += 1;
+ if (r == TCG_CONST) {
+ value = tci_read_i32(tb_ptr);
+ } else {
+ value = tci_read_reg32(r);
+ }
+ return value;
+}
+
+#if TCG_TARGET_REG_BITS == 32
+/* Read two indexed registers or constants (2 * 32 bit) from bytecode. */
+static uint64_t tci_read_ri64(uint8_t **tb_ptr)
+{
+ uint32_t low = tci_read_ri32(tb_ptr);
+ return tci_uint64(tci_read_ri32(tb_ptr), low);
+}
+#elif TCG_TARGET_REG_BITS == 64
+/* Read indexed register or constant (64 bit) from bytecode. */
+static uint64_t tci_read_ri64(uint8_t **tb_ptr)
+{
+ uint64_t value;
+ TCGRegister r = **tb_ptr;
+ *tb_ptr += 1;
+ if (r == TCG_CONST) {
+ value = tci_read_i64(tb_ptr);
+ } else {
+ value = tci_read_reg64(r);
+ }
+ return value;
+}
+#endif
+
+static target_ulong tci_read_label(uint8_t **tb_ptr)
+{
+ target_ulong label = tci_read_i(tb_ptr);
+ assert(label != 0);
+ return label;
+}
+
+static bool tci_compare32(uint32_t u0, uint32_t u1, TCGCond condition)
+{
+ bool result = false;
+ int32_t i0 = u0;
+ int32_t i1 = u1;
+ switch (condition) {
+ case TCG_COND_EQ:
+ result = (u0 == u1);
+ break;
+ case TCG_COND_NE:
+ result = (u0 != u1);
+ break;
+ case TCG_COND_LT:
+ result = (i0 < i1);
+ break;
+ case TCG_COND_GE:
+ result = (i0 >= i1);
+ break;
+ case TCG_COND_LE:
+ result = (i0 <= i1);
+ break;
+ case TCG_COND_GT:
+ result = (i0 > i1);
+ break;
+ case TCG_COND_LTU:
+ result = (u0 < u1);
+ break;
+ case TCG_COND_GEU:
+ result = (u0 >= u1);
+ break;
+ case TCG_COND_LEU:
+ result = (u0 <= u1);
+ break;
+ case TCG_COND_GTU:
+ result = (u0 > u1);
+ break;
+ default:
+ TODO();
+ }
+ return result;
+}
+
+static bool tci_compare64(uint64_t u0, uint64_t u1, TCGCond condition)
+{
+ bool result = false;
+ int64_t i0 = u0;
+ int64_t i1 = u1;
+ switch (condition) {
+ case TCG_COND_EQ:
+ result = (u0 == u1);
+ break;
+ case TCG_COND_NE:
+ result = (u0 != u1);
+ break;
+ case TCG_COND_LT:
+ result = (i0 < i1);
+ break;
+ case TCG_COND_GE:
+ result = (i0 >= i1);
+ break;
+ case TCG_COND_LE:
+ result = (i0 <= i1);
+ break;
+ case TCG_COND_GT:
+ result = (i0 > i1);
+ break;
+ case TCG_COND_LTU:
+ result = (u0 < u1);
+ break;
+ case TCG_COND_GEU:
+ result = (u0 >= u1);
+ break;
+ case TCG_COND_LEU:
+ result = (u0 <= u1);
+ break;
+ case TCG_COND_GTU:
+ result = (u0 > u1);
+ break;
+ default:
+ TODO();
+ }
+ return result;
+}
+
+/* Interpret pseudo code in tb. */
+unsigned long tcg_qemu_tb_exec(CPUState *cpustate, uint8_t *tb_ptr)
+{
+ unsigned long next_tb = 0;
+
+ TRACE();
+
+ env = cpustate;
+ tci_reg[TCG_AREG0] = (tcg_target_ulong)env;
+ assert(tb_ptr);
+
+ for (;;) {
+#if defined(GETPC)
+ tci_tb_ptr = tb_ptr;
+#endif
+#if !defined(NDEBUG)
+ uint8_t *old_code_ptr = tb_ptr;
+#endif
+ TCGOpcode opc = *tb_ptr++;
+ uint8_t op_size = *tb_ptr++;
+ tcg_target_ulong t0;
+ tcg_target_ulong t1;
+ tcg_target_ulong t2;
+ tcg_target_ulong label;
+ TCGCond condition;
+ target_ulong taddr;
+#ifndef CONFIG_SOFTMMU
+ tcg_target_ulong host_addr;
+#endif
+ uint8_t tmp8;
+ uint16_t tmp16;
+ uint32_t tmp32;
+ uint64_t tmp64;
+#if TCG_TARGET_REG_BITS == 32
+ uint64_t v64;
+#endif
+
+ (void)op_size;
+
+#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
+ if (loglevel) {
+ tci_disas(opc);
+ }
+#endif
+
+ switch (opc) {
+ case INDEX_op_end:
+ case INDEX_op_nop:
+ break;
+ case INDEX_op_nop1:
+ case INDEX_op_nop2:
+ case INDEX_op_nop3:
+ case INDEX_op_nopn:
+ case INDEX_op_discard:
+ TODO();
+ break;
+ case INDEX_op_set_label:
+ TODO();
+ break;
+ case INDEX_op_call:
+ t0 = tci_read_ri(&tb_ptr);
+#if TCG_TARGET_REG_BITS == 32
+ tmp64 = ((helper_function)t0)(tci_read_reg(TCG_REG_R0),
+ tci_read_reg(TCG_REG_R1),
+ tci_read_reg(TCG_REG_R2),
+ tci_read_reg(TCG_REG_R3),
+ tci_read_reg(TCG_REG_R5),
+ tci_read_reg(TCG_REG_R6),
+ tci_read_reg(TCG_REG_R7),
+ tci_read_reg(TCG_REG_R8));
+ tci_write_reg(TCG_REG_R0, tmp64);
+ tci_write_reg(TCG_REG_R1, tmp64 >> 32);
+#else
+ tmp64 = ((helper_function)t0)(tci_read_reg(TCG_REG_R0),
+ tci_read_reg(TCG_REG_R1),
+ tci_read_reg(TCG_REG_R2),
+ tci_read_reg(TCG_REG_R3));
+ tci_write_reg(TCG_REG_R0, tmp64);
+#endif
+ break;
+ case INDEX_op_jmp:
+ case INDEX_op_br:
+ label = tci_read_label(&tb_ptr);
+ assert(tb_ptr == old_code_ptr + op_size);
+ tb_ptr = (uint8_t *)label;
+ continue;
+ case INDEX_op_setcond_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ condition = *tb_ptr++;
+ tci_write_reg32(t0, tci_compare32(t1, t2, condition));
+ break;
+#if TCG_TARGET_REG_BITS == 32
+ case INDEX_op_setcond2_i32:
+ t0 = *tb_ptr++;
+ tmp64 = tci_read_r64(&tb_ptr);
+ v64 = tci_read_ri64(&tb_ptr);
+ condition = *tb_ptr++;
+ tci_write_reg32(t0, tci_compare64(tmp64, v64, condition));
+ break;
+#elif TCG_TARGET_REG_BITS == 64
+ case INDEX_op_setcond_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ condition = *tb_ptr++;
+ tci_write_reg64(t0, tci_compare64(t1, t2, condition));
+ break;
+#endif
+ case INDEX_op_mov_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+ case INDEX_op_movi_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_i32(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+
+ /* Load/store operations (32 bit). */
+
+ case INDEX_op_ld8u_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg8(t0, *(uint8_t *)(t1 + t2));
+ break;
+ case INDEX_op_ld8s_i32:
+ case INDEX_op_ld16u_i32:
+ TODO();
+ break;
+ case INDEX_op_ld16s_i32:
+ TODO();
+ break;
+ case INDEX_op_ld_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg32(t0, *(uint32_t *)(t1 + t2));
+ break;
+ case INDEX_op_st8_i32:
+ t0 = tci_read_r8(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint8_t *)(t1 + t2) = t0;
+ break;
+ case INDEX_op_st16_i32:
+ t0 = tci_read_r16(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint16_t *)(t1 + t2) = t0;
+ break;
+ case INDEX_op_st_i32:
+ t0 = tci_read_r32(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint32_t *)(t1 + t2) = t0;
+ break;
+
+ /* Arithmetic operations (32 bit). */
+
+ case INDEX_op_add_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 + t2);
+ break;
+ case INDEX_op_sub_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 - t2);
+ break;
+ case INDEX_op_mul_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 * t2);
+ break;
+#if TCG_TARGET_HAS_div_i32
+ case INDEX_op_div_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, (int32_t)t1 / (int32_t)t2);
+ break;
+ case INDEX_op_divu_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 / t2);
+ break;
+ case INDEX_op_rem_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, (int32_t)t1 % (int32_t)t2);
+ break;
+ case INDEX_op_remu_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 % t2);
+ break;
+#elif TCG_TARGET_HAS_div2_i32
+ case INDEX_op_div2_i32:
+ case INDEX_op_divu2_i32:
+ TODO();
+ break;
+#endif
+ case INDEX_op_and_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 & t2);
+ break;
+ case INDEX_op_or_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 | t2);
+ break;
+ case INDEX_op_xor_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 ^ t2);
+ break;
+
+ /* Shift/rotate operations (32 bit). */
+
+ case INDEX_op_shl_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 << t2);
+ break;
+ case INDEX_op_shr_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, t1 >> t2);
+ break;
+ case INDEX_op_sar_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, ((int32_t)t1 >> t2));
+ break;
+#if TCG_TARGET_HAS_rot_i32
+ case INDEX_op_rotl_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, (t1 << t2) | (t1 >> (32 - t2)));
+ break;
+ case INDEX_op_rotr_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri32(&tb_ptr);
+ t2 = tci_read_ri32(&tb_ptr);
+ tci_write_reg32(t0, (t1 >> t2) | (t1 << (32 - t2)));
+ break;
+#endif
+ case INDEX_op_brcond_i32:
+ t0 = tci_read_r32(&tb_ptr);
+ t1 = tci_read_ri32(&tb_ptr);
+ condition = *tb_ptr++;
+ label = tci_read_label(&tb_ptr);
+ if (tci_compare32(t0, t1, condition)) {
+ assert(tb_ptr == old_code_ptr + op_size);
+ tb_ptr = (uint8_t *)label;
+ continue;
+ }
+ break;
+#if TCG_TARGET_REG_BITS == 32
+ case INDEX_op_add2_i32:
+ t0 = *tb_ptr++;
+ t1 = *tb_ptr++;
+ tmp64 = tci_read_r64(&tb_ptr);
+ tmp64 += tci_read_r64(&tb_ptr);
+ tci_write_reg64(t1, t0, tmp64);
+ break;
+ case INDEX_op_sub2_i32:
+ t0 = *tb_ptr++;
+ t1 = *tb_ptr++;
+ tmp64 = tci_read_r64(&tb_ptr);
+ tmp64 -= tci_read_r64(&tb_ptr);
+ tci_write_reg64(t1, t0, tmp64);
+ break;
+ case INDEX_op_brcond2_i32:
+ tmp64 = tci_read_r64(&tb_ptr);
+ v64 = tci_read_ri64(&tb_ptr);
+ condition = *tb_ptr++;
+ label = tci_read_label(&tb_ptr);
+ if (tci_compare64(tmp64, v64, condition)) {
+ assert(tb_ptr == old_code_ptr + op_size);
+ tb_ptr = (uint8_t *)label;
+ continue;
+ }
+ break;
+ case INDEX_op_mulu2_i32:
+ t0 = *tb_ptr++;
+ t1 = *tb_ptr++;
+ t2 = tci_read_r32(&tb_ptr);
+ tmp64 = tci_read_r32(&tb_ptr);
+ tci_write_reg64(t1, t0, t2 * tmp64);
+ break;
+#endif /* TCG_TARGET_REG_BITS == 32 */
+#if TCG_TARGET_HAS_ext8s_i32
+ case INDEX_op_ext8s_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r8s(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext16s_i32
+ case INDEX_op_ext16s_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16s(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext8u_i32
+ case INDEX_op_ext8u_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r8(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext16u_i32
+ case INDEX_op_ext16u_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16(&tb_ptr);
+ tci_write_reg32(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_bswap16_i32
+ case INDEX_op_bswap16_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16(&tb_ptr);
+ tci_write_reg32(t0, bswap16(t1));
+ break;
+#endif
+#if TCG_TARGET_HAS_bswap32_i32
+ case INDEX_op_bswap32_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg32(t0, bswap32(t1));
+ break;
+#endif
+#if TCG_TARGET_HAS_not_i32
+ case INDEX_op_not_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg32(t0, ~t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_neg_i32
+ case INDEX_op_neg_i32:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg32(t0, -t1);
+ break;
+#endif
+#if TCG_TARGET_REG_BITS == 64
+ case INDEX_op_mov_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r64(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+ case INDEX_op_movi_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_i64(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+
+ /* Load/store operations (64 bit). */
+
+ case INDEX_op_ld8u_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg8(t0, *(uint8_t *)(t1 + t2));
+ break;
+ case INDEX_op_ld8s_i64:
+ case INDEX_op_ld16u_i64:
+ case INDEX_op_ld16s_i64:
+ TODO();
+ break;
+ case INDEX_op_ld32u_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg32(t0, *(uint32_t *)(t1 + t2));
+ break;
+ case INDEX_op_ld32s_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg32s(t0, *(int32_t *)(t1 + t2));
+ break;
+ case INDEX_op_ld_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ tci_write_reg64(t0, *(uint64_t *)(t1 + t2));
+ break;
+ case INDEX_op_st8_i64:
+ t0 = tci_read_r8(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint8_t *)(t1 + t2) = t0;
+ break;
+ case INDEX_op_st16_i64:
+ t0 = tci_read_r16(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint16_t *)(t1 + t2) = t0;
+ break;
+ case INDEX_op_st32_i64:
+ t0 = tci_read_r32(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint32_t *)(t1 + t2) = t0;
+ break;
+ case INDEX_op_st_i64:
+ t0 = tci_read_r64(&tb_ptr);
+ t1 = tci_read_r(&tb_ptr);
+ t2 = tci_read_i32(&tb_ptr);
+ *(uint64_t *)(t1 + t2) = t0;
+ break;
+
+ /* Arithmetic operations (64 bit). */
+
+ case INDEX_op_add_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 + t2);
+ break;
+ case INDEX_op_sub_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 - t2);
+ break;
+ case INDEX_op_mul_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 * t2);
+ break;
+#if TCG_TARGET_HAS_div_i64
+ case INDEX_op_div_i64:
+ case INDEX_op_divu_i64:
+ case INDEX_op_rem_i64:
+ case INDEX_op_remu_i64:
+ TODO();
+ break;
+#elif TCG_TARGET_HAS_div2_i64
+ case INDEX_op_div2_i64:
+ case INDEX_op_divu2_i64:
+ TODO();
+ break;
+#endif
+ case INDEX_op_and_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 & t2);
+ break;
+ case INDEX_op_or_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 | t2);
+ break;
+ case INDEX_op_xor_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 ^ t2);
+ break;
+
+ /* Shift/rotate operations (64 bit). */
+
+ case INDEX_op_shl_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 << t2);
+ break;
+ case INDEX_op_shr_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, t1 >> t2);
+ break;
+ case INDEX_op_sar_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_ri64(&tb_ptr);
+ t2 = tci_read_ri64(&tb_ptr);
+ tci_write_reg64(t0, ((int64_t)t1 >> t2));
+ break;
+#if TCG_TARGET_HAS_rot_i64
+ case INDEX_op_rotl_i64:
+ case INDEX_op_rotr_i64:
+ TODO();
+ break;
+#endif
+ case INDEX_op_brcond_i64:
+ t0 = tci_read_r64(&tb_ptr);
+ t1 = tci_read_ri64(&tb_ptr);
+ condition = *tb_ptr++;
+ label = tci_read_label(&tb_ptr);
+ if (tci_compare64(t0, t1, condition)) {
+ assert(tb_ptr == old_code_ptr + op_size);
+ tb_ptr = (uint8_t *)label;
+ continue;
+ }
+ break;
+#if TCG_TARGET_HAS_ext8u_i64
+ case INDEX_op_ext8u_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r8(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext8s_i64
+ case INDEX_op_ext8s_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r8s(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext16s_i64
+ case INDEX_op_ext16s_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16s(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext16u_i64
+ case INDEX_op_ext16u_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext32s_i64
+ case INDEX_op_ext32s_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32s(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_ext32u_i64
+ case INDEX_op_ext32u_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg64(t0, t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_bswap16_i64
+ case INDEX_op_bswap16_i64:
+ TODO();
+ t0 = *tb_ptr++;
+ t1 = tci_read_r16(&tb_ptr);
+ tci_write_reg64(t0, bswap16(t1));
+ break;
+#endif
+#if TCG_TARGET_HAS_bswap32_i64
+ case INDEX_op_bswap32_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r32(&tb_ptr);
+ tci_write_reg64(t0, bswap32(t1));
+ break;
+#endif
+#if TCG_TARGET_HAS_bswap64_i64
+ case INDEX_op_bswap64_i64:
+ TODO();
+ t0 = *tb_ptr++;
+ t1 = tci_read_r64(&tb_ptr);
+ tci_write_reg64(t0, bswap64(t1));
+ break;
+#endif
+#if TCG_TARGET_HAS_not_i64
+ case INDEX_op_not_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r64(&tb_ptr);
+ tci_write_reg64(t0, ~t1);
+ break;
+#endif
+#if TCG_TARGET_HAS_neg_i64
+ case INDEX_op_neg_i64:
+ t0 = *tb_ptr++;
+ t1 = tci_read_r64(&tb_ptr);
+ tci_write_reg64(t0, -t1);
+ break;
+#endif
+#endif /* TCG_TARGET_REG_BITS == 64 */
+
+ /* QEMU specific operations. */
+
+#if TARGET_LONG_BITS > TCG_TARGET_REG_BITS
+ case INDEX_op_debug_insn_start:
+ TODO();
+ break;
+#else
+ case INDEX_op_debug_insn_start:
+ TODO();
+ break;
+#endif
+ case INDEX_op_exit_tb:
+ next_tb = *(uint64_t *)tb_ptr;
+ goto exit;
+ break;
+ case INDEX_op_goto_tb:
+ t0 = tci_read_i32(&tb_ptr);
+ assert(tb_ptr == old_code_ptr + op_size);
+ tb_ptr += (int32_t)t0;
+ continue;
+ case INDEX_op_qemu_ld8u:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp8 = __ldb_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp8 = *(uint8_t *)(host_addr + GUEST_BASE);
+#endif
+ tci_write_reg8(t0, tmp8);
+ break;
+ case INDEX_op_qemu_ld8s:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp8 = __ldb_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp8 = *(uint8_t *)(host_addr + GUEST_BASE);
+#endif
+ tci_write_reg8s(t0, tmp8);
+ break;
+ case INDEX_op_qemu_ld16u:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp16 = __ldw_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp16 = tswap16(*(uint16_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg16(t0, tmp16);
+ break;
+ case INDEX_op_qemu_ld16s:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp16 = __ldw_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp16 = tswap16(*(uint16_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg16s(t0, tmp16);
+ break;
+#if TCG_TARGET_REG_BITS == 64
+ case INDEX_op_qemu_ld32u:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp32 = __ldl_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp32 = tswap32(*(uint32_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg32(t0, tmp32);
+ break;
+ case INDEX_op_qemu_ld32s:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp32 = __ldl_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp32 = tswap32(*(uint32_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg32s(t0, tmp32);
+ break;
+#endif /* TCG_TARGET_REG_BITS == 64 */
+ case INDEX_op_qemu_ld32:
+ t0 = *tb_ptr++;
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp32 = __ldl_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp32 = tswap32(*(uint32_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg32(t0, tmp32);
+ break;
+ case INDEX_op_qemu_ld64:
+ t0 = *tb_ptr++;
+#if TCG_TARGET_REG_BITS == 32
+ t1 = *tb_ptr++;
+#endif
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ tmp64 = __ldq_mmu(taddr, tci_read_i(&tb_ptr));
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ tmp64 = tswap64(*(uint64_t *)(host_addr + GUEST_BASE));
+#endif
+ tci_write_reg(t0, tmp64);
+#if TCG_TARGET_REG_BITS == 32
+ tci_write_reg(t1, tmp64 >> 32);
+#endif
+ break;
+ case INDEX_op_qemu_st8:
+ t0 = tci_read_r8(&tb_ptr);
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ t2 = tci_read_i(&tb_ptr);
+ __stb_mmu(taddr, t0, t2);
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ *(uint8_t *)(host_addr + GUEST_BASE) = t0;
+#endif
+ break;
+ case INDEX_op_qemu_st16:
+ t0 = tci_read_r16(&tb_ptr);
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ t2 = tci_read_i(&tb_ptr);
+ __stw_mmu(taddr, t0, t2);
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ *(uint16_t *)(host_addr + GUEST_BASE) = tswap16(t0);
+#endif
+ break;
+ case INDEX_op_qemu_st32:
+ t0 = tci_read_r32(&tb_ptr);
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ t2 = tci_read_i(&tb_ptr);
+ __stl_mmu(taddr, t0, t2);
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ *(uint32_t *)(host_addr + GUEST_BASE) = tswap32(t0);
+#endif
+ break;
+ case INDEX_op_qemu_st64:
+ tmp64 = tci_read_r64(&tb_ptr);
+ taddr = tci_read_ulong(&tb_ptr);
+#ifdef CONFIG_SOFTMMU
+ t2 = tci_read_i(&tb_ptr);
+ __stq_mmu(taddr, tmp64, t2);
+#else
+ host_addr = (tcg_target_ulong)taddr;
+ assert(taddr == host_addr);
+ *(uint64_t *)(host_addr + GUEST_BASE) = tswap64(tmp64);
+#endif
+ break;
+ default:
+ TODO();
+ break;
+ }
+ assert(tb_ptr == old_code_ptr + op_size);
+ }
+exit:
+ return next_tb;
+}
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 5/6] tcg: Add tci disassembler
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
` (3 preceding siblings ...)
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 6/6] tcg: Add tcg interpreter to configure / make Stefan Weil
5 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
dis-asm.h | 1 +
disas.c | 4 +++-
tci-dis.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 63 insertions(+), 1 deletions(-)
create mode 100644 tci-dis.c
diff --git a/dis-asm.h b/dis-asm.h
index 5b07d7f..4f15fad 100644
--- a/dis-asm.h
+++ b/dis-asm.h
@@ -365,6 +365,7 @@ typedef struct disassemble_info {
target address. Return number of bytes processed. */
typedef int (*disassembler_ftype) (bfd_vma, disassemble_info *);
+int print_insn_tci(bfd_vma, disassemble_info*);
int print_insn_big_mips (bfd_vma, disassemble_info*);
int print_insn_little_mips (bfd_vma, disassemble_info*);
int print_insn_i386 (bfd_vma, disassemble_info*);
diff --git a/disas.c b/disas.c
index 611b30b..3b1fd97 100644
--- a/disas.c
+++ b/disas.c
@@ -273,7 +273,9 @@ void disas(FILE *out, void *code, unsigned long size)
#else
disasm_info.endian = BFD_ENDIAN_LITTLE;
#endif
-#if defined(__i386__)
+#if defined(CONFIG_TCG_INTERPRETER)
+ print_insn = print_insn_tci;
+#elif defined(__i386__)
disasm_info.mach = bfd_mach_i386_i386;
print_insn = print_insn_i386;
#elif defined(__x86_64__)
diff --git a/tci-dis.c b/tci-dis.c
new file mode 100644
index 0000000..10c411b
--- /dev/null
+++ b/tci-dis.c
@@ -0,0 +1,59 @@
+/*
+ * Tiny Code Interpreter for QEMU - disassembler
+ *
+ * Copyright (c) 2011 Stefan Weil
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "dis-asm.h"
+#include "tcg/tcg.h"
+
+/* Disassemble TCI bytecode. */
+int print_insn_tci(bfd_vma addr, disassemble_info *info)
+{
+ int length;
+ uint8_t byte;
+ int status;
+ TCGOpcode op;
+
+ status = info->read_memory_func(addr, &byte, 1, info);
+ if (status != 0) {
+ info->memory_error_func(status, addr, info);
+ return -1;
+ }
+ op = byte;
+
+ addr++;
+ status = info->read_memory_func(addr, &byte, 1, info);
+ if (status != 0) {
+ info->memory_error_func(status, addr, info);
+ return -1;
+ }
+ length = byte;
+
+ if (op >= tcg_op_defs_max) {
+ info->fprintf_func(info->stream, "illegal opcode %d", op);
+ } else {
+ const TCGOpDef *def = &tcg_op_defs[op];
+ int nb_oargs = def->nb_oargs;
+ int nb_iargs = def->nb_iargs;
+ int nb_cargs = def->nb_cargs;
+ /* TODO: Improve disassembler output. */
+ info->fprintf_func(info->stream, "%s\to=%d i=%d c=%d",
+ def->name, nb_oargs, nb_iargs, nb_cargs);
+ }
+
+ return length;
+}
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [Qemu-devel] [PATCH v2 6/6] tcg: Add tcg interpreter to configure / make
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
` (4 preceding siblings ...)
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 5/6] tcg: Add tci disassembler Stefan Weil
@ 2011-10-20 21:11 ` Stefan Weil
5 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-20 21:11 UTC (permalink / raw)
To: qemu-devel; +Cc: Stefan Weil
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
Makefile.target | 4 ++++
configure | 34 ++++++++++++++++++++++++++++++++--
2 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/Makefile.target b/Makefile.target
index 417f23e..924f733 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -69,6 +69,7 @@ all: $(PROGS) stap
# cpu emulator library
libobj-y = exec.o translate-all.o cpu-exec.o translate.o
libobj-y += tcg/tcg.o tcg/optimize.o
+libobj-$(CONFIG_TCG_INTERPRETER) += tci.o
libobj-y += fpu/softfloat.o
libobj-y += op_helper.o helper.o
ifeq ($(TARGET_BASE_ARCH), i386)
@@ -78,6 +79,9 @@ libobj-$(CONFIG_NEED_MMU) += mmu.o
libobj-$(TARGET_ARM) += neon_helper.o iwmmxt_helper.o
libobj-y += disas.o
+libobj-$(CONFIG_TCI_DIS) += tci-dis.o
+
+tci-dis.o: QEMU_CFLAGS += -I$(SRC_PATH)/tcg -I$(SRC_PATH)/tcg/tci
$(libobj-y): $(GENERATED_HEADERS)
diff --git a/configure b/configure
index 4f87e0a..b634cc8 100755
--- a/configure
+++ b/configure
@@ -138,6 +138,7 @@ debug_tcg="no"
debug_mon="no"
debug="no"
strip_opt="yes"
+tcg_interpreter="no"
bigendian="no"
mingw32="no"
EXESUF=""
@@ -653,6 +654,10 @@ for opt do
;;
--enable-kvm) kvm="yes"
;;
+ --disable-tcg-interpreter) tcg_interpreter="no"
+ ;;
+ --enable-tcg-interpreter) tcg_interpreter="yes"
+ ;;
--disable-spice) spice="no"
;;
--enable-spice) spice="yes"
@@ -1004,6 +1009,7 @@ echo " --enable-bluez enable bluez stack connectivity"
echo " --disable-slirp disable SLIRP userspace network connectivity"
echo " --disable-kvm disable KVM acceleration support"
echo " --enable-kvm enable KVM acceleration support"
+echo " --enable-tcg-interpreter enable TCG with bytecode interpreter (TCI)"
echo " --disable-nptl disable usermode NPTL support"
echo " --enable-nptl enable usermode NPTL support"
echo " --enable-system enable all system emulation targets"
@@ -2729,6 +2735,7 @@ echo "Linux AIO support $linux_aio"
echo "ATTR/XATTR support $attr"
echo "Install blobs $blobs"
echo "KVM support $kvm"
+echo "TCG interpreter $tcg_interpreter"
echo "fdt support $fdt"
echo "preadv support $preadv"
echo "fdatasync $fdatasync"
@@ -2776,6 +2783,15 @@ case "$cpu" in
armv4b|armv4l)
ARCH=arm
;;
+ *)
+ if test "$tcg_interpreter" = "yes" ; then
+ echo "Unsupported CPU = $cpu, will use TCG with TCI (experimental)"
+ ARCH=tci
+ else
+ echo "Unsupported CPU = $cpu, try --enable-tcg-interpreter"
+ exit 1
+ fi
+ ;;
esac
echo "ARCH=$ARCH" >> $config_host_mak
if test "$debug_tcg" = "yes" ; then
@@ -3009,6 +3025,9 @@ fi
if test "$signalfd" = "yes" ; then
echo "CONFIG_SIGNALFD=y" >> $config_host_mak
fi
+if test "$tcg_interpreter" = "yes" ; then
+ echo "CONFIG_TCG_INTERPRETER=y" >> $config_host_mak
+fi
if test "$need_offsetof" = "yes" ; then
echo "CONFIG_NEED_OFFSETOF=y" >> $config_host_mak
fi
@@ -3477,7 +3496,9 @@ cflags=""
includes=""
ldflags=""
-if test "$ARCH" = "sparc64" ; then
+if test "$tcg_interpreter" = "yes"; then
+ includes="-I\$(SRC_PATH)/tcg/tci $includes"
+elif test "$ARCH" = "sparc64" ; then
includes="-I\$(SRC_PATH)/tcg/sparc $includes"
elif test "$ARCH" = "s390x" ; then
includes="-I\$(SRC_PATH)/tcg/s390 $includes"
@@ -3554,6 +3575,10 @@ for i in $ARCH $TARGET_BASE_ARCH ; do
;;
esac
done
+if test "$tcg_interpreter" = "yes" ; then
+ echo "CONFIG_TCI_DIS=y" >> $config_target_mak
+ echo "CONFIG_TCI_DIS=y" >> $libdis_config_mak
+fi
case "$ARCH" in
alpha)
@@ -3600,7 +3625,12 @@ if test "$gprof" = "yes" ; then
fi
fi
-linker_script="-Wl,-T../config-host.ld -Wl,-T,\$(SRC_PATH)/\$(ARCH).ld"
+if test "$ARCH" = "tci"; then
+ linker_script=""
+else
+ linker_script="-Wl,-T../config-host.ld -Wl,-T,\$(SRC_PATH)/\$(ARCH).ld"
+fi
+
if test "$target_linux_user" = "yes" -o "$target_bsd_user" = "yes" ; then
case "$ARCH" in
sparc)
--
1.7.2.5
^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode Stefan Weil
@ 2011-10-20 21:36 ` malc
2011-10-21 6:03 ` Stefan Weil
0 siblings, 1 reply; 9+ messages in thread
From: malc @ 2011-10-20 21:36 UTC (permalink / raw)
To: Stefan Weil; +Cc: qemu-devel
On Thu, 20 Oct 2011, Stefan Weil wrote:
[..snip..]
> +/* Trace message to see program flow. */
> +#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
> +#define TRACE() \
> + loglevel \
> + ? fprintf(stderr, "TCG %s:%u: %s()\n", __FILE__, __LINE__, __func__) \
> + : (void)0
This is wrong, fprintf's return value is int and not void. Similar issue
was present today on comp.std.c
http://groups.google.com/group/comp.std.c/browse_thread/thread/4e01fece59a80572#
[..snip..]
--
mailto:av1474@comtv.ru
^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode
2011-10-20 21:36 ` malc
@ 2011-10-21 6:03 ` Stefan Weil
0 siblings, 0 replies; 9+ messages in thread
From: Stefan Weil @ 2011-10-21 6:03 UTC (permalink / raw)
To: malc; +Cc: qemu-devel
Am 20.10.2011 23:36, schrieb malc:
> On Thu, 20 Oct 2011, Stefan Weil wrote:
>
> [..snip..]
>
>> +/* Trace message to see program flow. */
>> +#if defined(CONFIG_DEBUG_TCG_INTERPRETER)
>> +#define TRACE() \
>> + loglevel \
>> + ? fprintf(stderr, "TCG %s:%u: %s()\n", __FILE__, __LINE__, __func__) \
>> + : (void)0
> This is wrong, fprintf's return value is int and not void. Similar issue
> was present today on comp.std.c
> http://groups.google.com/group/comp.std.c/browse_thread/thread/4e01fece59a80572#
>
> [..snip..]
You are right, but it does not matter. TRACE() is never used in
assignments, and gcc accepts the statement as it is.
Nevertheless, I' ll fix it in the next release.
Thanks,
Stefan
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2011-10-21 6:04 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-10-20 21:11 [Qemu-devel] [PATCH v2 0/6] tcg/interpreter: Add TCG + interpreter for bytecode Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 1/6] tcg: TCG targets may define tcg_qemu_tb_exec Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 2/6] tcg: Make ARRAY_SIZE(tcg_op_defs) globally available Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 3/6] tcg: Add bytecode generator for tcg interpreter Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 4/6] tcg: Add interpreter for bytecode Stefan Weil
2011-10-20 21:36 ` malc
2011-10-21 6:03 ` Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 5/6] tcg: Add tci disassembler Stefan Weil
2011-10-20 21:11 ` [Qemu-devel] [PATCH v2 6/6] tcg: Add tcg interpreter to configure / make Stefan Weil
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.