diff options
59 files changed, 3351 insertions, 1614 deletions
diff --git a/audio/audio_template.h b/audio/audio_template.h index 519432a7c5..16f7880770 100644 --- a/audio/audio_template.h +++ b/audio/audio_template.h @@ -410,15 +410,15 @@ SW *glue (AUD_open_, TYPE) ( SW *old_sw = NULL; #endif - ldebug ("open %s, freq %d, nchannels %d, fmt %d\n", - name, as->freq, as->nchannels, as->fmt); - if (audio_bug (AUDIO_FUNC, !card || !name || !callback_fn || !as)) { dolog ("card=%p name=%p callback_fn=%p as=%p\n", card, name, callback_fn, as); goto fail; } + ldebug ("open %s, freq %d, nchannels %d, fmt %d\n", + name, as->freq, as->nchannels, as->fmt); + if (audio_bug (AUDIO_FUNC, audio_validate_settings (as))) { audio_print_settings (as); goto fail; diff --git a/block/iscsi.c b/block/iscsi.c index 0b96165ec2..fb001b955c 100644 --- a/block/iscsi.c +++ b/block/iscsi.c @@ -167,12 +167,6 @@ iscsi_set_events(IscsiLun *iscsilun) } - /* If we just added an event, the callback might be delayed - * unless we call qemu_notify_event(). - */ - if (ev & ~iscsilun->events) { - qemu_notify_event(); - } iscsilun->events = ev; } @@ -628,9 +622,17 @@ static BlockDriverAIOCB *iscsi_aio_ioctl(BlockDriverState *bs, return &acb->common; } + +static void ioctl_cb(void *opaque, int status) +{ + int *p_status = opaque; + *p_status = status; +} + static int iscsi_ioctl(BlockDriverState *bs, unsigned long int req, void *buf) { IscsiLun *iscsilun = bs->opaque; + int status; switch (req) { case SG_GET_VERSION_NUM: @@ -639,6 +641,15 @@ static int iscsi_ioctl(BlockDriverState *bs, unsigned long int req, void *buf) case SG_GET_SCSI_ID: ((struct sg_scsi_id *)buf)->scsi_type = iscsilun->type; break; + case SG_IO: + status = -EINPROGRESS; + iscsi_aio_ioctl(bs, req, buf, ioctl_cb, &status); + + while (status == -EINPROGRESS) { + qemu_aio_wait(); + } + + return 0; default: return -1; } @@ -111,7 +111,6 @@ source_path=`dirname "$0"` cpu="" interp_prefix="/usr/gnemul/qemu-%M" static="no" -sparc_cpu="" cross_prefix="" audio_drv_list="" audio_card_list="ac97 es1370 sb16 hda" @@ -241,21 +240,6 @@ for opt do ;; --disable-debug-info) debug_info="no" ;; - --sparc_cpu=*) - sparc_cpu="$optarg" - case $sparc_cpu in - v7|v8|v8plus|v8plusa) - cpu="sparc" - ;; - v9) - cpu="sparc64" - ;; - *) - echo "undefined SPARC architecture. Exiting"; - exit 1 - ;; - esac - ;; esac done # OS specific @@ -343,8 +327,6 @@ elif check_define __i386__ ; then elif check_define __x86_64__ ; then cpu="x86_64" elif check_define __sparc__ ; then - # We can't check for 64 bit (when gcc is biarch) or V8PLUSA - # They must be specified using --sparc_cpu if check_define __arch64__ ; then cpu="sparc64" else @@ -792,8 +774,6 @@ for opt do ;; --enable-uname-release=*) uname_release="$optarg" ;; - --sparc_cpu=*) - ;; --enable-werror) werror="yes" ;; --disable-werror) werror="no" @@ -881,35 +861,17 @@ for opt do esac done -# -# If cpu ~= sparc and sparc_cpu hasn't been defined, plug in the right -# QEMU_CFLAGS/LDFLAGS (assume sparc_v8plus for 32-bit and sparc_v9 for 64-bit) -# host_guest_base="no" case "$cpu" in - sparc) case $sparc_cpu in - v7|v8) - QEMU_CFLAGS="-mcpu=${sparc_cpu} -D__sparc_${sparc_cpu}__ $QEMU_CFLAGS" - ;; - v8plus|v8plusa) - QEMU_CFLAGS="-mcpu=ultrasparc -D__sparc_${sparc_cpu}__ $QEMU_CFLAGS" - ;; - *) # sparc_cpu not defined in the command line - QEMU_CFLAGS="-mcpu=ultrasparc -D__sparc_v8plus__ $QEMU_CFLAGS" - esac + sparc) LDFLAGS="-m32 $LDFLAGS" - QEMU_CFLAGS="-m32 -ffixed-g2 -ffixed-g3 $QEMU_CFLAGS" - if test "$solaris" = "no" ; then - QEMU_CFLAGS="-ffixed-g1 -ffixed-g6 $QEMU_CFLAGS" - fi + QEMU_CFLAGS="-m32 -mcpu=ultrasparc $QEMU_CFLAGS" + host_guest_base="yes" ;; sparc64) - QEMU_CFLAGS="-m64 -mcpu=ultrasparc -D__sparc_v9__ $QEMU_CFLAGS" LDFLAGS="-m64 $LDFLAGS" - QEMU_CFLAGS="-ffixed-g5 -ffixed-g6 -ffixed-g7 $QEMU_CFLAGS" - if test "$solaris" != "no" ; then - QEMU_CFLAGS="-ffixed-g1 $QEMU_CFLAGS" - fi + QEMU_CFLAGS="-m64 -mcpu=ultrasparc $QEMU_CFLAGS" + host_guest_base="yes" ;; s390) QEMU_CFLAGS="-m31 -march=z990 $QEMU_CFLAGS" @@ -2795,7 +2757,7 @@ if test "$usb_redir" != "no" ; then usb_redir_cflags=$($pkg_config --cflags libusbredirparser 2>/dev/null) usb_redir_libs=$($pkg_config --libs libusbredirparser 2>/dev/null) QEMU_CFLAGS="$QEMU_CFLAGS $usb_redir_cflags" - LIBS="$LIBS $usb_redir_libs" + libs_softmmu="$libs_softmmu $usb_redir_libs" else if test "$usb_redir" = "yes"; then feature_not_found "usb-redir" @@ -4131,10 +4093,6 @@ fi if test "$target_linux_user" = "yes" -o "$target_bsd_user" = "yes" ; then case "$ARCH" in - sparc) - # -static is used to avoid g1/g3 usage by the dynamic linker - ldflags="$linker_script -static $ldflags" - ;; alpha | s390x) # The default placement of the application is fine. ;; @@ -316,9 +316,7 @@ void disas(FILE *out, void *code, unsigned long size) print_insn = print_insn_alpha; #elif defined(__sparc__) print_insn = print_insn_sparc; -#if defined(__sparc_v8plus__) || defined(__sparc_v8plusa__) || defined(__sparc_v9__) disasm_info.mach = bfd_mach_sparc_v9b; -#endif #elif defined(__arm__) print_insn = print_insn_arm; #elif defined(__MIPSEB__) diff --git a/exec-all.h b/exec-all.h index dba96098b8..6516da071d 100644 --- a/exec-all.h +++ b/exec-all.h @@ -132,9 +132,10 @@ static inline void tlb_flush(CPUArchState *env, int flush_global) #define CODE_GEN_AVG_BLOCK_SIZE 64 #endif -#if defined(_ARCH_PPC) || defined(__x86_64__) || defined(__arm__) || defined(__i386__) -#define USE_DIRECT_JUMP -#elif defined(CONFIG_TCG_INTERPRETER) +#if defined(__arm__) || defined(_ARCH_PPC) \ + || defined(__x86_64__) || defined(__i386__) \ + || defined(__sparc__) \ + || defined(CONFIG_TCG_INTERPRETER) #define USE_DIRECT_JUMP #endif @@ -244,6 +245,8 @@ static inline void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr) __asm __volatile__ ("swi 0x9f0002" : : "r" (_beg), "r" (_end), "r" (_flg)); #endif } +#elif defined(__sparc__) +void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr); #else #error tb_set_jmp_target1 is missing #endif @@ -86,7 +86,7 @@ static int nb_tbs; /* any access to the tbs or the page table must use this lock */ spinlock_t tb_lock = SPIN_LOCK_UNLOCKED; -#if defined(__arm__) || defined(__sparc_v9__) +#if defined(__arm__) || defined(__sparc__) /* The prologue must be reachable with a direct jump. ARM and Sparc64 have limited branch ranges (possibly also PPC) so place it in a section close to code segment. */ @@ -541,10 +541,9 @@ static void code_gen_alloc(unsigned long tb_size) /* Cannot map more than that */ if (code_gen_buffer_size > (800 * 1024 * 1024)) code_gen_buffer_size = (800 * 1024 * 1024); -#elif defined(__sparc_v9__) +#elif defined(__sparc__) && HOST_LONG_BITS == 64 // Map the buffer below 2G, so we can use direct calls and branches - flags |= MAP_FIXED; - start = (void *) 0x60000000UL; + start = (void *) 0x40000000UL; if (code_gen_buffer_size > (512 * 1024 * 1024)) code_gen_buffer_size = (512 * 1024 * 1024); #elif defined(__arm__) @@ -582,10 +581,9 @@ static void code_gen_alloc(unsigned long tb_size) /* Cannot map more than that */ if (code_gen_buffer_size > (800 * 1024 * 1024)) code_gen_buffer_size = (800 * 1024 * 1024); -#elif defined(__sparc_v9__) +#elif defined(__sparc__) && HOST_LONG_BITS == 64 // Map the buffer below 2G, so we can use direct calls and branches - flags |= MAP_FIXED; - addr = (void *) 0x60000000UL; + addr = (void *) 0x40000000UL; if (code_gen_buffer_size > (512 * 1024 * 1024)) { code_gen_buffer_size = (512 * 1024 * 1024); } diff --git a/fpu/softfloat-specialize.h b/fpu/softfloat-specialize.h index 490245004f..a1d489e425 100644 --- a/fpu/softfloat-specialize.h +++ b/fpu/softfloat-specialize.h @@ -41,6 +41,13 @@ these four paragraphs for those parts of this code that are retained. #define SNAN_BIT_IS_ONE 0 #endif +#if defined(TARGET_XTENSA) +/* Define for architectures which deviate from IEEE in not supporting + * signaling NaNs (so all NaNs are treated as quiet). + */ +#define NO_SIGNALING_NANS 1 +#endif + /*---------------------------------------------------------------------------- | The pattern for a default generated half-precision NaN. *----------------------------------------------------------------------------*/ @@ -57,7 +64,8 @@ const float16 float16_default_nan = const_float16(0xFE00); *----------------------------------------------------------------------------*/ #if defined(TARGET_SPARC) const float32 float32_default_nan = const_float32(0x7FFFFFFF); -#elif defined(TARGET_PPC) || defined(TARGET_ARM) || defined(TARGET_ALPHA) +#elif defined(TARGET_PPC) || defined(TARGET_ARM) || defined(TARGET_ALPHA) || \ + defined(TARGET_XTENSA) const float32 float32_default_nan = const_float32(0x7FC00000); #elif SNAN_BIT_IS_ONE const float32 float32_default_nan = const_float32(0x7FBFFFFF); @@ -127,6 +135,17 @@ typedef struct { uint64_t high, low; } commonNaNT; +#ifdef NO_SIGNALING_NANS +int float16_is_quiet_nan(float16 a_) +{ + return float16_is_any_nan(a_); +} + +int float16_is_signaling_nan(float16 a_) +{ + return 0; +} +#else /*---------------------------------------------------------------------------- | Returns 1 if the half-precision floating-point value `a' is a quiet | NaN; otherwise returns 0. @@ -156,6 +175,7 @@ int float16_is_signaling_nan(float16 a_) return (((a >> 9) & 0x3F) == 0x3E) && (a & 0x1FF); #endif } +#endif /*---------------------------------------------------------------------------- | Returns a quiet NaN if the half-precision floating point value `a' is a @@ -217,6 +237,17 @@ static float16 commonNaNToFloat16(commonNaNT a STATUS_PARAM) } } +#ifdef NO_SIGNALING_NANS +int float32_is_quiet_nan(float32 a_) +{ + return float32_is_any_nan(a_); +} + +int float32_is_signaling_nan(float32 a_) +{ + return 0; +} +#else /*---------------------------------------------------------------------------- | Returns 1 if the single-precision floating-point value `a' is a quiet | NaN; otherwise returns 0. @@ -246,6 +277,7 @@ int float32_is_signaling_nan( float32 a_ ) return ( ( ( a>>22 ) & 0x1FF ) == 0x1FE ) && ( a & 0x003FFFFF ); #endif } +#endif /*---------------------------------------------------------------------------- | Returns a quiet NaN if the single-precision floating point value `a' is a @@ -372,7 +404,7 @@ static int pickNaN(flag aIsQNaN, flag aIsSNaN, flag bIsQNaN, flag bIsSNaN, return 1; } } -#elif defined(TARGET_PPC) +#elif defined(TARGET_PPC) || defined(TARGET_XTENSA) static int pickNaN(flag aIsQNaN, flag aIsSNaN, flag bIsQNaN, flag bIsSNaN, flag aIsLargerSignificand) { @@ -586,6 +618,17 @@ static float32 propagateFloat32MulAddNaN(float32 a, float32 b, } } +#ifdef NO_SIGNALING_NANS +int float64_is_quiet_nan(float64 a_) +{ + return float64_is_any_nan(a_); +} + +int float64_is_signaling_nan(float64 a_) +{ + return 0; +} +#else /*---------------------------------------------------------------------------- | Returns 1 if the double-precision floating-point value `a' is a quiet | NaN; otherwise returns 0. @@ -619,6 +662,7 @@ int float64_is_signaling_nan( float64 a_ ) && ( a & LIT64( 0x0007FFFFFFFFFFFF ) ); #endif } +#endif /*---------------------------------------------------------------------------- | Returns a quiet NaN if the double-precision floating point value `a' is a @@ -773,6 +817,17 @@ static float64 propagateFloat64MulAddNaN(float64 a, float64 b, } } +#ifdef NO_SIGNALING_NANS +int floatx80_is_quiet_nan(floatx80 a_) +{ + return floatx80_is_any_nan(a_); +} + +int floatx80_is_signaling_nan(floatx80 a_) +{ + return 0; +} +#else /*---------------------------------------------------------------------------- | Returns 1 if the extended double-precision floating-point value `a' is a | quiet NaN; otherwise returns 0. This slightly differs from the same @@ -816,6 +871,7 @@ int floatx80_is_signaling_nan( floatx80 a ) && ( a.low == aLow ); #endif } +#endif /*---------------------------------------------------------------------------- | Returns a quiet NaN if the extended double-precision floating point value @@ -929,6 +985,17 @@ static floatx80 propagateFloatx80NaN( floatx80 a, floatx80 b STATUS_PARAM) } } +#ifdef NO_SIGNALING_NANS +int float128_is_quiet_nan(float128 a_) +{ + return float128_is_any_nan(a_); +} + +int float128_is_signaling_nan(float128 a_) +{ + return 0; +} +#else /*---------------------------------------------------------------------------- | Returns 1 if the quadruple-precision floating-point value `a' is a quiet | NaN; otherwise returns 0. @@ -964,6 +1031,7 @@ int float128_is_signaling_nan( float128 a ) && ( a.low || ( a.high & LIT64( 0x00007FFFFFFFFFFF ) ) ); #endif } +#endif /*---------------------------------------------------------------------------- | Returns a quiet NaN if the quadruple-precision floating point value `a' is diff --git a/fpu/softfloat.h b/fpu/softfloat.h index feec3a180e..d8999b311a 100644 --- a/fpu/softfloat.h +++ b/fpu/softfloat.h @@ -219,7 +219,7 @@ void float_raise( int8 flags STATUS_PARAM); enum { float_muladd_negate_c = 1, float_muladd_negate_product = 2, - float_muladd_negate_result = 3, + float_muladd_negate_result = 4, }; /*---------------------------------------------------------------------------- @@ -251,6 +251,11 @@ int float16_is_quiet_nan( float16 ); int float16_is_signaling_nan( float16 ); float16 float16_maybe_silence_nan( float16 ); +INLINE int float16_is_any_nan(float16 a) +{ + return ((float16_val(a) & ~0x8000) > 0x7c00); +} + /*---------------------------------------------------------------------------- | The pattern for a default generated half-precision NaN. *----------------------------------------------------------------------------*/ @@ -1226,33 +1226,48 @@ static int cpu_gdb_write_register(CPUOpenRISCState *env, static int cpu_gdb_read_register(CPUSH4State *env, uint8_t *mem_buf, int n) { - if (n < 8) { + switch (n) { + case 0 ... 7: if ((env->sr & (SR_MD | SR_RB)) == (SR_MD | SR_RB)) { GET_REGL(env->gregs[n + 16]); } else { GET_REGL(env->gregs[n]); } - } else if (n < 16) { + case 8 ... 15: GET_REGL(env->gregs[n]); - } else if (n >= 25 && n < 41) { - GET_REGL(env->fregs[(n - 25) + ((env->fpscr & FPSCR_FR) ? 16 : 0)]); - } else if (n >= 43 && n < 51) { - GET_REGL(env->gregs[n - 43]); - } else if (n >= 51 && n < 59) { - GET_REGL(env->gregs[n - (51 - 16)]); - } - switch (n) { - case 16: GET_REGL(env->pc); - case 17: GET_REGL(env->pr); - case 18: GET_REGL(env->gbr); - case 19: GET_REGL(env->vbr); - case 20: GET_REGL(env->mach); - case 21: GET_REGL(env->macl); - case 22: GET_REGL(env->sr); - case 23: GET_REGL(env->fpul); - case 24: GET_REGL(env->fpscr); - case 41: GET_REGL(env->ssr); - case 42: GET_REGL(env->spc); + case 16: + GET_REGL(env->pc); + case 17: + GET_REGL(env->pr); + case 18: + GET_REGL(env->gbr); + case 19: + GET_REGL(env->vbr); + case 20: + GET_REGL(env->mach); + case 21: + GET_REGL(env->macl); + case 22: + GET_REGL(env->sr); + case 23: + GET_REGL(env->fpul); + case 24: + GET_REGL(env->fpscr); + case 25 ... 40: + if (env->fpscr & FPSCR_FR) { + stfl_p(mem_buf, env->fregs[n - 9]); + } else { + stfl_p(mem_buf, env->fregs[n - 25]); + } + return 4; + case 41: + GET_REGL(env->ssr); + case 42: + GET_REGL(env->spc); + case 43 ... 50: + GET_REGL(env->gregs[n - 43]); + case 51 ... 58: + GET_REGL(env->gregs[n - (51 - 16)]); } return 0; @@ -1260,42 +1275,63 @@ static int cpu_gdb_read_register(CPUSH4State *env, uint8_t *mem_buf, int n) static int cpu_gdb_write_register(CPUSH4State *env, uint8_t *mem_buf, int n) { - uint32_t tmp; - - tmp = ldl_p(mem_buf); - - if (n < 8) { + switch (n) { + case 0 ... 7: if ((env->sr & (SR_MD | SR_RB)) == (SR_MD | SR_RB)) { - env->gregs[n + 16] = tmp; + env->gregs[n + 16] = ldl_p(mem_buf); } else { - env->gregs[n] = tmp; + env->gregs[n] = ldl_p(mem_buf); } - return 4; - } else if (n < 16) { - env->gregs[n] = tmp; - return 4; - } else if (n >= 25 && n < 41) { - env->fregs[(n - 25) + ((env->fpscr & FPSCR_FR) ? 16 : 0)] = tmp; - return 4; - } else if (n >= 43 && n < 51) { - env->gregs[n - 43] = tmp; - return 4; - } else if (n >= 51 && n < 59) { - env->gregs[n - (51 - 16)] = tmp; - return 4; - } - switch (n) { - case 16: env->pc = tmp; break; - case 17: env->pr = tmp; break; - case 18: env->gbr = tmp; break; - case 19: env->vbr = tmp; break; - case 20: env->mach = tmp; break; - case 21: env->macl = tmp; break; - case 22: env->sr = tmp; break; - case 23: env->fpul = tmp; break; - case 24: env->fpscr = tmp; break; - case 41: env->ssr = tmp; break; - case 42: env->spc = tmp; break; + break; + case 8 ... 15: + env->gregs[n] = ldl_p(mem_buf); + break; + case 16: + env->pc = ldl_p(mem_buf); + break; + case 17: + env->pr = ldl_p(mem_buf); + break; + case 18: + env->gbr = ldl_p(mem_buf); + break; + case 19: + env->vbr = ldl_p(mem_buf); + break; + case 20: + env->mach = ldl_p(mem_buf); + break; + case 21: + env->macl = ldl_p(mem_buf); + break; + case 22: + env->sr = ldl_p(mem_buf); + break; + case 23: + env->fpul = ldl_p(mem_buf); + break; + case 24: + env->fpscr = ldl_p(mem_buf); + break; + case 25 ... 40: + if (env->fpscr & FPSCR_FR) { + env->fregs[n - 9] = ldfl_p(mem_buf); + } else { + env->fregs[n - 25] = ldfl_p(mem_buf); + } + break; + case 41: + env->ssr = ldl_p(mem_buf); + break; + case 42: + env->spc = ldl_p(mem_buf); + break; + case 43 ... 50: + env->gregs[n - 43] = ldl_p(mem_buf); + break; + case 51 ... 58: + env->gregs[n - (51 - 16)] = ldl_p(mem_buf); + break; default: return 0; } @@ -1660,6 +1696,10 @@ static int cpu_gdb_read_register(CPUXtensaState *env, uint8_t *mem_buf, int n) GET_REG32(env->uregs[reg->targno & 0xff]); break; + case 4: /*f*/ + GET_REG32(float32_val(env->fregs[reg->targno & 0x0f])); + break; + case 8: /*a*/ GET_REG32(env->regs[reg->targno & 0x0f]); break; @@ -1700,6 +1740,10 @@ static int cpu_gdb_write_register(CPUXtensaState *env, uint8_t *mem_buf, int n) env->uregs[reg->targno & 0xff] = tmp; break; + case 4: /*f*/ + env->fregs[reg->targno & 0x0f] = make_float32(tmp); + break; + case 8: /*a*/ env->regs[reg->targno & 0x0f] = tmp; break; diff --git a/hw/scsi-bus.c b/hw/scsi-bus.c index 4981a02436..058d3b237f 100644 --- a/hw/scsi-bus.c +++ b/hw/scsi-bus.c @@ -801,26 +801,39 @@ static int ata_passthrough_16_xfer_size(SCSIDevice *dev, uint8_t *buf) return xfer * unit; } -static int scsi_req_length(SCSICommand *cmd, SCSIDevice *dev, uint8_t *buf) +uint32_t scsi_data_cdb_length(uint8_t *buf) +{ + if ((buf[0] >> 5) == 0 && buf[4] == 0) { + return 256; + } else { + return scsi_cdb_length(buf); + } +} + +uint32_t scsi_cdb_length(uint8_t *buf) { switch (buf[0] >> 5) { case 0: - cmd->xfer = buf[4]; + return buf[4]; break; case 1: case 2: - cmd->xfer = lduw_be_p(&buf[7]); + return lduw_be_p(&buf[7]); break; case 4: - cmd->xfer = ldl_be_p(&buf[10]) & 0xffffffffULL; + return ldl_be_p(&buf[10]) & 0xffffffffULL; break; case 5: - cmd->xfer = ldl_be_p(&buf[6]) & 0xffffffffULL; + return ldl_be_p(&buf[6]) & 0xffffffffULL; break; default: return -1; } +} +static int scsi_req_length(SCSICommand *cmd, SCSIDevice *dev, uint8_t *buf) +{ + cmd->xfer = scsi_cdb_length(buf); switch (buf[0]) { case TEST_UNIT_READY: case REWIND: diff --git a/hw/scsi-disk.c b/hw/scsi-disk.c index 1585683bce..95e91585e6 100644 --- a/hw/scsi-disk.c +++ b/hw/scsi-disk.c @@ -678,7 +678,7 @@ static int scsi_disk_emulate_inquiry(SCSIRequest *req, uint8_t *outbuf) * is actually implemented, but we're good enough. */ outbuf[2] = 5; - outbuf[3] = 2; /* Format 2 */ + outbuf[3] = 2 | 0x10; /* Format 2, HiSup */ if (buflen > 36) { outbuf[4] = buflen - 5; /* Additional Length = (Len - 1) - 4 */ @@ -1449,6 +1449,22 @@ invalid_field: return; } +static inline bool check_lba_range(SCSIDiskState *s, + uint64_t sector_num, uint32_t nb_sectors) +{ + /* + * The first line tests that no overflow happens when computing the last + * sector. The second line tests that the last accessed sector is in + * range. + * + * Careful, the computations should not underflow for nb_sectors == 0, + * and a 0-block read to the first LBA beyond the end of device is + * valid. + */ + return (sector_num <= sector_num + nb_sectors && + sector_num + nb_sectors <= s->qdev.max_lba + 1); +} + typedef struct UnmapCBData { SCSIDiskReq *r; uint8_t *inbuf; @@ -1473,8 +1489,7 @@ static void scsi_unmap_complete(void *opaque, int ret) if (data->count > 0 && !r->req.io_canceled) { sector_num = ldq_be_p(&data->inbuf[0]); nb_sectors = ldl_be_p(&data->inbuf[8]) & 0xffffffffULL; - if (sector_num > sector_num + nb_sectors || - sector_num + nb_sectors - 1 > s->qdev.max_lba) { + if (!check_lba_range(s, sector_num, nb_sectors)) { scsi_check_condition(r, SENSE_CODE(LBA_OUT_OF_RANGE)); goto done; } @@ -1793,17 +1808,13 @@ static int32_t scsi_disk_emulate_command(SCSIRequest *req, uint8_t *buf) DPRINTF("Unmap (len %lu)\n", (long)r->req.cmd.xfer); break; case WRITE_SAME_10: - nb_sectors = lduw_be_p(&req->cmd.buf[7]); - goto write_same; case WRITE_SAME_16: - nb_sectors = ldl_be_p(&req->cmd.buf[10]) & 0xffffffffULL; - write_same: + nb_sectors = scsi_data_cdb_length(r->req.cmd.buf); if (bdrv_is_read_only(s->qdev.conf.bs)) { scsi_check_condition(r, SENSE_CODE(WRITE_PROTECTED)); return 0; } - if (r->req.cmd.lba > r->req.cmd.lba + nb_sectors || - r->req.cmd.lba + nb_sectors - 1 > s->qdev.max_lba) { + if (!check_lba_range(s, r->req.cmd.lba, nb_sectors)) { goto illegal_lba; } @@ -1858,7 +1869,7 @@ static int32_t scsi_disk_dma_command(SCSIRequest *req, uint8_t *buf) { SCSIDiskReq *r = DO_UPCAST(SCSIDiskReq, req, req); SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); - int32_t len; + uint32_t len; uint8_t command; command = buf[0]; @@ -1868,18 +1879,17 @@ static int32_t scsi_disk_dma_command(SCSIRequest *req, uint8_t *buf) return 0; } + len = scsi_data_cdb_length(r->req.cmd.buf); switch (command) { case READ_6: case READ_10: case READ_12: case READ_16: - len = r->req.cmd.xfer / s->qdev.blocksize; - DPRINTF("Read (sector %" PRId64 ", count %d)\n", r->req.cmd.lba, len); + DPRINTF("Read (sector %" PRId64 ", count %u)\n", r->req.cmd.lba, len); if (r->req.cmd.buf[1] & 0xe0) { goto illegal_request; } - if (r->req.cmd.lba > r->req.cmd.lba + len || - r->req.cmd.lba + len - 1 > s->qdev.max_lba) { + if (!check_lba_range(s, r->req.cmd.lba, len)) { goto illegal_lba; } r->sector = r->req.cmd.lba * (s->qdev.blocksize / 512); @@ -1900,15 +1910,13 @@ static int32_t scsi_disk_dma_command(SCSIRequest *req, uint8_t *buf) case VERIFY_10: case VERIFY_12: case VERIFY_16: - len = r->req.cmd.xfer / s->qdev.blocksize; - DPRINTF("Write %s(sector %" PRId64 ", count %d)\n", + DPRINTF("Write %s(sector %" PRId64 ", count %u)\n", (command & 0xe) == 0xe ? "And Verify " : "", r->req.cmd.lba, len); if (r->req.cmd.buf[1] & 0xe0) { goto illegal_request; } - if (r->req.cmd.lba > r->req.cmd.lba + len || - r->req.cmd.lba + len - 1 > s->qdev.max_lba) { + if (!check_lba_range(s, r->req.cmd.lba, len)) { goto illegal_lba; } r->sector = r->req.cmd.lba * (s->qdev.blocksize / 512); @@ -218,6 +218,8 @@ extern const struct SCSISense sense_code_WRITE_PROTECTED; #define SENSE_CODE(x) sense_code_ ## x +uint32_t scsi_data_cdb_length(uint8_t *buf); +uint32_t scsi_cdb_length(uint8_t *buf); int scsi_sense_valid(SCSISense sense); int scsi_build_sense(uint8_t *in_buf, int in_len, uint8_t *buf, int len, bool fixed); diff --git a/hw/usb/hcd-ehci.c b/hw/usb/hcd-ehci.c index 2f3e9c03a1..6a5da8413f 100644 --- a/hw/usb/hcd-ehci.c +++ b/hw/usb/hcd-ehci.c @@ -34,6 +34,7 @@ #include "monitor.h" #include "trace.h" #include "dma.h" +#include "sysemu.h" #define EHCI_DEBUG 0 @@ -139,6 +140,7 @@ #define NB_PORTS 6 // Number of downstream ports #define BUFF_SIZE 5*4096 // Max bytes to transfer per transaction #define MAX_QH 100 // Max allowable queue heads in a chain +#define MIN_FR_PER_TICK 3 // Min frames to process when catching up /* Internal periodic / asynchronous schedule state machine states */ @@ -389,6 +391,9 @@ struct EHCIState { USBBus bus; qemu_irq irq; MemoryRegion mem; + MemoryRegion mem_caps; + MemoryRegion mem_opreg; + MemoryRegion mem_ports; int companion_count; /* properties */ @@ -398,10 +403,10 @@ struct EHCIState { * EHCI spec version 1.0 Section 2.3 * Host Controller Operational Registers */ + uint8_t caps[OPREGBASE]; union { - uint8_t mmio[MMIO_SIZE]; + uint32_t opreg[(PORTSC_BEGIN-OPREGBASE)/sizeof(uint32_t)]; struct { - uint8_t cap[OPREGBASE]; uint32_t usbcmd; uint32_t usbsts; uint32_t usbintr; @@ -411,9 +416,9 @@ struct EHCIState { uint32_t asynclistaddr; uint32_t notused[9]; uint32_t configflag; - uint32_t portsc[NB_PORTS]; }; }; + uint32_t portsc[NB_PORTS]; /* * Internal states, shadow registers, etc @@ -471,22 +476,12 @@ static const char *ehci_state_names[] = { }; static const char *ehci_mmio_names[] = { - [CAPLENGTH] = "CAPLENGTH", - [HCIVERSION] = "HCIVERSION", - [HCSPARAMS] = "HCSPARAMS", - [HCCPARAMS] = "HCCPARAMS", [USBCMD] = "USBCMD", [USBSTS] = "USBSTS", [USBINTR] = "USBINTR", [FRINDEX] = "FRINDEX", [PERIODICLISTBASE] = "P-LIST BASE", [ASYNCLISTADDR] = "A-LIST ADDR", - [PORTSC_BEGIN] = "PORTSC #0", - [PORTSC_BEGIN + 4] = "PORTSC #1", - [PORTSC_BEGIN + 8] = "PORTSC #2", - [PORTSC_BEGIN + 12] = "PORTSC #3", - [PORTSC_BEGIN + 16] = "PORTSC #4", - [PORTSC_BEGIN + 20] = "PORTSC #5", [CONFIGFLAG] = "CONFIGFLAG", }; @@ -509,7 +504,8 @@ static const char *state2str(uint32_t state) static const char *addr2str(target_phys_addr_t addr) { - return nr2str(ehci_mmio_names, ARRAY_SIZE(ehci_mmio_names), addr); + return nr2str(ehci_mmio_names, ARRAY_SIZE(ehci_mmio_names), + addr + OPREGBASE); } static void ehci_trace_usbsts(uint32_t mask, int state) @@ -853,10 +849,10 @@ static EHCIQueue *ehci_find_queue_by_qh(EHCIState *ehci, uint32_t addr, return NULL; } -static void ehci_queues_rip_unused(EHCIState *ehci, int async, int flush) +static void ehci_queues_rip_unused(EHCIState *ehci, int async) { EHCIQueueHead *head = async ? &ehci->aqueues : &ehci->pqueues; - const char *warn = (async && !flush) ? "guest unlinked busy QH" : NULL; + const char *warn = async ? "guest unlinked busy QH" : NULL; uint64_t maxage = FRAME_TIMER_NS * ehci->maxframes * 4; EHCIQueue *q, *tmp; @@ -866,13 +862,25 @@ static void ehci_queues_rip_unused(EHCIState *ehci, int async, int flush) q->ts = ehci->last_run_ns; continue; } - if (!flush && ehci->last_run_ns < q->ts + maxage) { + if (ehci->last_run_ns < q->ts + maxage) { continue; } ehci_free_queue(q, warn); } } +static void ehci_queues_rip_unseen(EHCIState *ehci, int async) +{ + EHCIQueueHead *head = async ? &ehci->aqueues : &ehci->pqueues; + EHCIQueue *q, *tmp; + + QTAILQ_FOREACH_SAFE(q, head, next, tmp) { + if (!q->seen) { + ehci_free_queue(q, NULL); + } + } +} + static void ehci_queues_rip_device(EHCIState *ehci, USBDevice *dev, int async) { EHCIQueueHead *head = async ? &ehci->aqueues : &ehci->pqueues; @@ -1018,7 +1026,7 @@ static int ehci_register_companion(USBBus *bus, USBPort *ports[], } s->companion_count++; - s->mmio[0x05] = (s->companion_count << 4) | portcount; + s->caps[0x05] = (s->companion_count << 4) | portcount; return 0; } @@ -1063,7 +1071,8 @@ static void ehci_reset(void *opaque) } } - memset(&s->mmio[OPREGBASE], 0x00, MMIO_SIZE - OPREGBASE); + memset(&s->opreg, 0x00, sizeof(s->opreg)); + memset(&s->portsc, 0x00, sizeof(s->portsc)); s->usbcmd = NB_MAXINTRATE << USBCMD_ITC_SH; s->usbsts = USBSTS_HALT; @@ -1090,50 +1099,35 @@ static void ehci_reset(void *opaque) qemu_bh_cancel(s->async_bh); } -static uint32_t ehci_mem_readb(void *ptr, target_phys_addr_t addr) +static uint64_t ehci_caps_read(void *ptr, target_phys_addr_t addr, + unsigned size) { EHCIState *s = ptr; - uint32_t val; - - val = s->mmio[addr]; - - return val; + return s->caps[addr]; } -static uint32_t ehci_mem_readw(void *ptr, target_phys_addr_t addr) +static uint64_t ehci_opreg_read(void *ptr, target_phys_addr_t addr, + unsigned size) { EHCIState *s = ptr; uint32_t val; - val = s->mmio[addr] | (s->mmio[addr+1] << 8); - + val = s->opreg[addr >> 2]; + trace_usb_ehci_opreg_read(addr + OPREGBASE, addr2str(addr), val); return val; } -static uint32_t ehci_mem_readl(void *ptr, target_phys_addr_t addr) +static uint64_t ehci_port_read(void *ptr, target_phys_addr_t addr, + unsigned size) { EHCIState *s = ptr; uint32_t val; - val = s->mmio[addr] | (s->mmio[addr+1] << 8) | - (s->mmio[addr+2] << 16) | (s->mmio[addr+3] << 24); - - trace_usb_ehci_mmio_readl(addr, addr2str(addr), val); + val = s->portsc[addr >> 2]; + trace_usb_ehci_portsc_read(addr + PORTSC_BEGIN, addr >> 2, val); return val; } -static void ehci_mem_writeb(void *ptr, target_phys_addr_t addr, uint32_t val) -{ - fprintf(stderr, "EHCI doesn't handle byte writes to MMIO\n"); - exit(1); -} - -static void ehci_mem_writew(void *ptr, target_phys_addr_t addr, uint32_t val) -{ - fprintf(stderr, "EHCI doesn't handle 16-bit writes to MMIO\n"); - exit(1); -} - static void handle_port_owner_write(EHCIState *s, int port, uint32_t owner) { USBDevice *dev = s->ports[port].dev; @@ -1162,11 +1156,17 @@ static void handle_port_owner_write(EHCIState *s, int port, uint32_t owner) } } -static void handle_port_status_write(EHCIState *s, int port, uint32_t val) +static void ehci_port_write(void *ptr, target_phys_addr_t addr, + uint64_t val, unsigned size) { + EHCIState *s = ptr; + int port = addr >> 2; uint32_t *portsc = &s->portsc[port]; + uint32_t old = *portsc; USBDevice *dev = s->ports[port].dev; + trace_usb_ehci_portsc_write(addr + PORTSC_BEGIN, addr >> 2, val); + /* Clear rwc bits */ *portsc &= ~(val & PORTSC_RWC_MASK); /* The guest may clear, but not set the PED bit */ @@ -1198,39 +1198,20 @@ static void handle_port_status_write(EHCIState *s, int port, uint32_t val) *portsc &= ~PORTSC_RO_MASK; *portsc |= val; + trace_usb_ehci_portsc_change(addr + PORTSC_BEGIN, addr >> 2, *portsc, old); } -static void ehci_mem_writel(void *ptr, target_phys_addr_t addr, uint32_t val) +static void ehci_opreg_write(void *ptr, target_phys_addr_t addr, + uint64_t val, unsigned size) { EHCIState *s = ptr; - uint32_t *mmio = (uint32_t *)(&s->mmio[addr]); + uint32_t *mmio = s->opreg + (addr >> 2); uint32_t old = *mmio; int i; - trace_usb_ehci_mmio_writel(addr, addr2str(addr), val); - - /* Only aligned reads are allowed on OHCI */ - if (addr & 3) { - fprintf(stderr, "usb-ehci: Mis-aligned write to addr 0x" - TARGET_FMT_plx "\n", addr); - return; - } - - if (addr >= PORTSC && addr < PORTSC + 4 * NB_PORTS) { - handle_port_status_write(s, (addr-PORTSC)/4, val); - trace_usb_ehci_mmio_change(addr, addr2str(addr), *mmio, old); - return; - } - - if (addr < OPREGBASE) { - fprintf(stderr, "usb-ehci: write attempt to read-only register" - TARGET_FMT_plx "\n", addr); - return; - } + trace_usb_ehci_opreg_write(addr + OPREGBASE, addr2str(addr), val); - - /* Do any register specific pre-write processing here. */ - switch(addr) { + switch (addr + OPREGBASE) { case USBCMD: if (val & USBCMD_HCRESET) { ehci_reset(s); @@ -1241,7 +1222,7 @@ static void ehci_mem_writel(void *ptr, target_phys_addr_t addr, uint32_t val) /* not supporting dynamic frame list size at the moment */ if ((val & USBCMD_FLS) && !(s->usbcmd & USBCMD_FLS)) { fprintf(stderr, "attempt to set frame list size -- value %d\n", - val & USBCMD_FLS); + (int)val & USBCMD_FLS); val &= ~USBCMD_FLS; } @@ -1308,7 +1289,7 @@ static void ehci_mem_writel(void *ptr, target_phys_addr_t addr, uint32_t val) } *mmio = val; - trace_usb_ehci_mmio_change(addr, addr2str(addr), *mmio, old); + trace_usb_ehci_opreg_change(addr + OPREGBASE, addr2str(addr), *mmio, old); } @@ -1732,7 +1713,7 @@ static int ehci_state_waitlisthead(EHCIState *ehci, int async) ehci_set_usbsts(ehci, USBSTS_REC); } - ehci_queues_rip_unused(ehci, async, 0); + ehci_queues_rip_unused(ehci, async); /* Find the head of the list (4.9.1.1) */ for(i = 0; i < MAX_QH; i++) { @@ -2364,7 +2345,7 @@ static void ehci_advance_async_state(EHCIState *ehci) */ if (ehci->usbcmd & USBCMD_IAAD) { /* Remove all unseen qhs from the async qhs queue */ - ehci_queues_rip_unused(ehci, async, 1); + ehci_queues_rip_unseen(ehci, async); trace_usb_ehci_doorbell_ack(); ehci->usbcmd &= ~USBCMD_IAAD; ehci_raise_irq(ehci, USBSTS_IAA); @@ -2417,7 +2398,7 @@ static void ehci_advance_periodic_state(EHCIState *ehci) ehci_set_fetch_addr(ehci, async,entry); ehci_set_state(ehci, async, EST_FETCHENTRY); ehci_advance_state(ehci, async); - ehci_queues_rip_unused(ehci, async, 0); + ehci_queues_rip_unused(ehci, async); break; default: @@ -2446,7 +2427,7 @@ static void ehci_update_frindex(EHCIState *ehci, int frames) if (ehci->frindex == 0x00004000) { ehci_raise_irq(ehci, USBSTS_FLR); ehci->frindex = 0; - if (ehci->usbsts_frindex > 0x00004000) { + if (ehci->usbsts_frindex >= 0x00004000) { ehci->usbsts_frindex -= 0x00004000; } else { ehci->usbsts_frindex = 0; @@ -2481,6 +2462,19 @@ static void ehci_frame_timer(void *opaque) } for (i = 0; i < frames; i++) { + /* + * If we're running behind schedule, we should not catch up + * too fast, as that will make some guests unhappy: + * 1) We must process a minimum of MIN_FR_PER_TICK frames, + * otherwise we will never catch up + * 2) Process frames until the guest has requested an irq (IOC) + */ + if (i >= MIN_FR_PER_TICK) { + ehci_commit_irq(ehci); + if ((ehci->usbsts & USBINTR_MASK) & ehci->usbintr) { + break; + } + } ehci_update_frindex(ehci, 1); ehci_advance_periodic_state(ehci); ehci->last_run_ns += FRAME_TIMER_NS; @@ -2520,11 +2514,28 @@ static void ehci_async_bh(void *opaque) ehci_advance_async_state(ehci); } -static const MemoryRegionOps ehci_mem_ops = { - .old_mmio = { - .read = { ehci_mem_readb, ehci_mem_readw, ehci_mem_readl }, - .write = { ehci_mem_writeb, ehci_mem_writew, ehci_mem_writel }, - }, +static const MemoryRegionOps ehci_mmio_caps_ops = { + .read = ehci_caps_read, + .valid.min_access_size = 1, + .valid.max_access_size = 4, + .impl.min_access_size = 1, + .impl.max_access_size = 1, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static const MemoryRegionOps ehci_mmio_opreg_ops = { + .read = ehci_opreg_read, + .write = ehci_opreg_write, + .valid.min_access_size = 4, + .valid.max_access_size = 4, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static const MemoryRegionOps ehci_mmio_port_ops = { + .read = ehci_port_read, + .write = ehci_port_write, + .valid.min_access_size = 4, + .valid.max_access_size = 4, .endianness = DEVICE_LITTLE_ENDIAN, }; @@ -2562,6 +2573,32 @@ static int usb_ehci_post_load(void *opaque, int version_id) return 0; } +static void usb_ehci_vm_state_change(void *opaque, int running, RunState state) +{ + EHCIState *ehci = opaque; + + /* + * We don't migrate the EHCIQueue-s, instead we rebuild them for the + * schedule in guest memory. We must do the rebuilt ASAP, so that + * USB-devices which have async handled packages have a packet in the + * ep queue to match the completion with. + */ + if (state == RUN_STATE_RUNNING) { + ehci_advance_async_state(ehci); + } + + /* + * The schedule rebuilt from guest memory could cause the migration dest + * to miss a QH unlink, and fail to cancel packets, since the unlinked QH + * will never have existed on the destination. Therefor we must flush the + * async schedule on savevm to catch any not yet noticed unlinks. + */ + if (state == RUN_STATE_SAVE_VM) { + ehci_advance_async_state(ehci); + ehci_queues_rip_unseen(ehci, 1); + } +} + static const VMStateDescription vmstate_ehci = { .name = "ehci", .version_id = 2, @@ -2681,19 +2718,19 @@ static int usb_ehci_initfn(PCIDevice *dev) pci_conf[0x6e] = 0x00; pci_conf[0x6f] = 0xc0; // USBLEFCTLSTS - // 2.2 host controller interface version - s->mmio[0x00] = (uint8_t) OPREGBASE; - s->mmio[0x01] = 0x00; - s->mmio[0x02] = 0x00; - s->mmio[0x03] = 0x01; // HC version - s->mmio[0x04] = NB_PORTS; // Number of downstream ports - s->mmio[0x05] = 0x00; // No companion ports at present - s->mmio[0x06] = 0x00; - s->mmio[0x07] = 0x00; - s->mmio[0x08] = 0x80; // We can cache whole frame, not 64-bit capable - s->mmio[0x09] = 0x68; // EECP - s->mmio[0x0a] = 0x00; - s->mmio[0x0b] = 0x00; + /* 2.2 host controller interface version */ + s->caps[0x00] = (uint8_t) OPREGBASE; + s->caps[0x01] = 0x00; + s->caps[0x02] = 0x00; + s->caps[0x03] = 0x01; /* HC version */ + s->caps[0x04] = NB_PORTS; /* Number of downstream ports */ + s->caps[0x05] = 0x00; /* No companion ports at present */ + s->caps[0x06] = 0x00; + s->caps[0x07] = 0x00; + s->caps[0x08] = 0x80; /* We can cache whole frame, no 64-bit */ + s->caps[0x09] = 0x68; /* EECP */ + s->caps[0x0a] = 0x00; + s->caps[0x0b] = 0x00; s->irq = s->dev.irq[3]; @@ -2711,8 +2748,20 @@ static int usb_ehci_initfn(PCIDevice *dev) usb_packet_init(&s->ipacket); qemu_register_reset(ehci_reset, s); + qemu_add_vm_change_state_handler(usb_ehci_vm_state_change, s); + + memory_region_init(&s->mem, "ehci", MMIO_SIZE); + memory_region_init_io(&s->mem_caps, &ehci_mmio_caps_ops, s, + "capabilities", OPREGBASE); + memory_region_init_io(&s->mem_opreg, &ehci_mmio_opreg_ops, s, + "operational", PORTSC_BEGIN - OPREGBASE); + memory_region_init_io(&s->mem_ports, &ehci_mmio_port_ops, s, + "ports", PORTSC_END - PORTSC_BEGIN); + + memory_region_add_subregion(&s->mem, 0, &s->mem_caps); + memory_region_add_subregion(&s->mem, OPREGBASE, &s->mem_opreg); + memory_region_add_subregion(&s->mem, PORTSC_BEGIN, &s->mem_ports); - memory_region_init_io(&s->mem, &ehci_mem_ops, s, "ehci", MMIO_SIZE); pci_register_bar(&s->dev, 0, PCI_BASE_ADDRESS_SPACE_MEMORY, &s->mem); return 0; diff --git a/hw/usb/hcd-uhci.c b/hw/usb/hcd-uhci.c index c7c878635a..cdc8bc3fba 100644 --- a/hw/usb/hcd-uhci.c +++ b/hw/usb/hcd-uhci.c @@ -1000,6 +1000,9 @@ static void uhci_fill_queue(UHCIState *s, UHCI_TD *td) } assert(ret == TD_RESULT_ASYNC_START); assert(int_mask == 0); + if (ptd.ctrl & TD_CTRL_SPD) { + break; + } plink = ptd.link; } } @@ -1097,7 +1100,7 @@ static void uhci_process_frame(UHCIState *s) case TD_RESULT_ASYNC_START: trace_usb_uhci_td_async(curr_qh & ~0xf, link & ~0xf); - if (is_valid(td.link)) { + if (is_valid(td.link) && !(td.ctrl & TD_CTRL_SPD)) { uhci_fill_queue(s, &td); } link = curr_qh ? qh.link : td.link; diff --git a/hw/usb/host-linux.c b/hw/usb/host-linux.c index 8df92074d3..44f1a64b35 100644 --- a/hw/usb/host-linux.c +++ b/hw/usb/host-linux.c @@ -1045,7 +1045,6 @@ static int usb_host_handle_control(USBDevice *dev, USBPacket *p, /* Note request is (bRequestType << 8) | bRequest */ trace_usb_host_req_control(s->bus_num, s->addr, p, request, value, index); - assert(p->result == 0); switch (request) { case DeviceOutRequest | USB_REQ_SET_ADDRESS: @@ -1074,6 +1073,7 @@ static int usb_host_handle_control(USBDevice *dev, USBPacket *p, } /* The rest are asynchronous */ + assert(p && p->result == 0); if (length > sizeof(dev->data_buf)) { fprintf(stderr, "husb: ctrl buffer too small (%d > %zu)\n", diff --git a/hw/usb/redirect.c b/hw/usb/redirect.c index 5301a69c4f..b10241a139 100644 --- a/hw/usb/redirect.c +++ b/hw/usb/redirect.c @@ -43,7 +43,6 @@ #define EP2I(ep_address) (((ep_address & 0x80) >> 3) | (ep_address & 0x0f)) #define I2EP(i) (((i & 0x10) << 3) | (i & 0x0f)) -typedef struct Cancelled Cancelled; typedef struct USBRedirDevice USBRedirDevice; /* Struct to hold buffered packets (iso or int input packets) */ @@ -58,6 +57,7 @@ struct endp_data { uint8_t type; uint8_t interval; uint8_t interface; /* bInterfaceNumber this ep belongs to */ + uint16_t max_packet_size; /* In bytes, not wMaxPacketSize format !! */ uint8_t iso_started; uint8_t iso_error; /* For reporting iso errors to the HC */ uint8_t interrupt_started; @@ -65,8 +65,20 @@ struct endp_data { uint8_t bufpq_prefilled; uint8_t bufpq_dropping_packets; QTAILQ_HEAD(, buf_packet) bufpq; - int bufpq_size; - int bufpq_target_size; + int32_t bufpq_size; + int32_t bufpq_target_size; +}; + +struct PacketIdQueueEntry { + uint64_t id; + QTAILQ_ENTRY(PacketIdQueueEntry)next; +}; + +struct PacketIdQueue { + USBRedirDevice *dev; + const char *name; + QTAILQ_HEAD(, PacketIdQueueEntry) head; + int size; }; struct USBRedirDevice { @@ -86,7 +98,8 @@ struct USBRedirDevice { int64_t next_attach_time; struct usbredirparser *parser; struct endp_data endpoint[MAX_ENDPOINTS]; - QTAILQ_HEAD(, Cancelled) cancelled; + struct PacketIdQueue cancelled; + struct PacketIdQueue already_in_flight; /* Data for device filtering */ struct usb_redir_device_connect_header device_info; struct usb_redir_interface_info_header interface_info; @@ -94,11 +107,6 @@ struct USBRedirDevice { int filter_rules_count; }; -struct Cancelled { - uint64_t id; - QTAILQ_ENTRY(Cancelled)next; -}; - static void usbredir_hello(void *priv, struct usb_redir_hello_header *h); static void usbredir_device_connect(void *priv, struct usb_redir_device_connect_header *device_connect); @@ -134,6 +142,8 @@ static void usbredir_interrupt_packet(void *priv, uint64_t id, static int usbredir_handle_status(USBRedirDevice *dev, int status, int actual_len); +#define VERSION "qemu usb-redir guest " QEMU_VERSION + /* * Logging stuff */ @@ -232,6 +242,11 @@ static int usbredir_write(void *priv, uint8_t *data, int count) return 0; } + /* Don't send new data to the chardev until our state is fully synced */ + if (!runstate_check(RUN_STATE_RUNNING)) { + return 0; + } + return qemu_chr_fe_write(dev->cs, data, count); } @@ -239,37 +254,103 @@ static int usbredir_write(void *priv, uint8_t *data, int count) * Cancelled and buffered packets helpers */ -static void usbredir_cancel_packet(USBDevice *udev, USBPacket *p) +static void packet_id_queue_init(struct PacketIdQueue *q, + USBRedirDevice *dev, const char *name) { - USBRedirDevice *dev = DO_UPCAST(USBRedirDevice, dev, udev); - Cancelled *c; + q->dev = dev; + q->name = name; + QTAILQ_INIT(&q->head); + q->size = 0; +} + +static void packet_id_queue_add(struct PacketIdQueue *q, uint64_t id) +{ + USBRedirDevice *dev = q->dev; + struct PacketIdQueueEntry *e; + + DPRINTF("adding packet id %"PRIu64" to %s queue\n", id, q->name); - DPRINTF("cancel packet id %"PRIu64"\n", p->id); + e = g_malloc0(sizeof(struct PacketIdQueueEntry)); + e->id = id; + QTAILQ_INSERT_TAIL(&q->head, e, next); + q->size++; +} - c = g_malloc0(sizeof(Cancelled)); - c->id = p->id; - QTAILQ_INSERT_TAIL(&dev->cancelled, c, next); +static int packet_id_queue_remove(struct PacketIdQueue *q, uint64_t id) +{ + USBRedirDevice *dev = q->dev; + struct PacketIdQueueEntry *e; + + QTAILQ_FOREACH(e, &q->head, next) { + if (e->id == id) { + DPRINTF("removing packet id %"PRIu64" from %s queue\n", + id, q->name); + QTAILQ_REMOVE(&q->head, e, next); + q->size--; + g_free(e); + return 1; + } + } + return 0; +} +static void packet_id_queue_empty(struct PacketIdQueue *q) +{ + USBRedirDevice *dev = q->dev; + struct PacketIdQueueEntry *e, *next_e; + + DPRINTF("removing %d packet-ids from %s queue\n", q->size, q->name); + + QTAILQ_FOREACH_SAFE(e, &q->head, next, next_e) { + QTAILQ_REMOVE(&q->head, e, next); + g_free(e); + } + q->size = 0; +} + +static void usbredir_cancel_packet(USBDevice *udev, USBPacket *p) +{ + USBRedirDevice *dev = DO_UPCAST(USBRedirDevice, dev, udev); + + packet_id_queue_add(&dev->cancelled, p->id); usbredirparser_send_cancel_data_packet(dev->parser, p->id); usbredirparser_do_write(dev->parser); } static int usbredir_is_cancelled(USBRedirDevice *dev, uint64_t id) { - Cancelled *c; - if (!dev->dev.attached) { return 1; /* Treat everything as cancelled after a disconnect */ } + return packet_id_queue_remove(&dev->cancelled, id); +} + +static void usbredir_fill_already_in_flight_from_ep(USBRedirDevice *dev, + struct USBEndpoint *ep) +{ + static USBPacket *p; + + QTAILQ_FOREACH(p, &ep->queue, queue) { + packet_id_queue_add(&dev->already_in_flight, p->id); + } +} - QTAILQ_FOREACH(c, &dev->cancelled, next) { - if (c->id == id) { - QTAILQ_REMOVE(&dev->cancelled, c, next); - g_free(c); - return 1; - } +static void usbredir_fill_already_in_flight(USBRedirDevice *dev) +{ + int ep; + struct USBDevice *udev = &dev->dev; + + usbredir_fill_already_in_flight_from_ep(dev, &udev->ep_ctl); + + for (ep = 0; ep < USB_MAX_ENDPOINTS; ep++) { + usbredir_fill_already_in_flight_from_ep(dev, &udev->ep_in[ep]); + usbredir_fill_already_in_flight_from_ep(dev, &udev->ep_out[ep]); } - return 0; +} + +static int usbredir_already_in_flight(USBRedirDevice *dev, uint64_t id) +{ + return packet_id_queue_remove(&dev->already_in_flight, id); } static USBPacket *usbredir_find_packet_by_id(USBRedirDevice *dev, @@ -487,6 +568,10 @@ static int usbredir_handle_bulk_data(USBRedirDevice *dev, USBPacket *p, DPRINTF("bulk-out ep %02X len %zd id %"PRIu64"\n", ep, p->iov.size, p->id); + if (usbredir_already_in_flight(dev, p->id)) { + return USB_RET_ASYNC; + } + bulk_packet.endpoint = ep; bulk_packet.length = p->iov.size; bulk_packet.stream_id = 0; @@ -567,6 +652,10 @@ static int usbredir_handle_interrupt_data(USBRedirDevice *dev, DPRINTF("interrupt-out ep %02X len %zd id %"PRIu64"\n", ep, p->iov.size, p->id); + if (usbredir_already_in_flight(dev, p->id)) { + return USB_RET_ASYNC; + } + interrupt_packet.endpoint = ep; interrupt_packet.length = p->iov.size; @@ -709,6 +798,10 @@ static int usbredir_handle_control(USBDevice *udev, USBPacket *p, USBRedirDevice *dev = DO_UPCAST(USBRedirDevice, dev, udev); struct usb_redir_control_packet_header control_packet; + if (usbredir_already_in_flight(dev, p->id)) { + return USB_RET_ASYNC; + } + /* Special cases for certain standard device requests */ switch (request) { case DeviceOutRequest | USB_REQ_SET_ADDRESS: @@ -763,6 +856,7 @@ static void usbredir_chardev_close_bh(void *opaque) usbredir_device_disconnect(dev); if (dev->parser) { + DPRINTF("destroying usbredirparser\n"); usbredirparser_destroy(dev->parser); dev->parser = NULL; } @@ -771,14 +865,13 @@ static void usbredir_chardev_close_bh(void *opaque) static void usbredir_chardev_open(USBRedirDevice *dev) { uint32_t caps[USB_REDIR_CAPS_SIZE] = { 0, }; - char version[32]; + int flags = 0; /* Make sure any pending closes are handled (no-op if none pending) */ usbredir_chardev_close_bh(dev); qemu_bh_cancel(dev->chardev_close_bh); - strcpy(version, "qemu usb-redir guest "); - pstrcat(version, sizeof(version), qemu_get_version()); + DPRINTF("creating usbredirparser\n"); dev->parser = qemu_oom_check(usbredirparser_create()); dev->parser->priv = dev; @@ -807,7 +900,12 @@ static void usbredir_chardev_open(USBRedirDevice *dev) usbredirparser_caps_set_cap(caps, usb_redir_cap_filter); usbredirparser_caps_set_cap(caps, usb_redir_cap_ep_info_max_packet_size); usbredirparser_caps_set_cap(caps, usb_redir_cap_64bits_ids); - usbredirparser_init(dev->parser, version, caps, USB_REDIR_CAPS_SIZE, 0); + + if (runstate_check(RUN_STATE_INMIGRATE)) { + flags |= usbredirparser_fl_no_hello; + } + usbredirparser_init(dev->parser, VERSION, caps, USB_REDIR_CAPS_SIZE, + flags); usbredirparser_do_write(dev->parser); } @@ -853,6 +951,11 @@ static int usbredir_chardev_can_read(void *opaque) return 0; } + /* Don't read new data from the chardev until our state is fully synced */ + if (!runstate_check(RUN_STATE_RUNNING)) { + return 0; + } + /* usbredir_parser_do_read will consume *all* data we give it */ return 1024 * 1024; } @@ -878,9 +981,11 @@ static void usbredir_chardev_event(void *opaque, int event) switch (event) { case CHR_EVENT_OPENED: + DPRINTF("chardev open\n"); usbredir_chardev_open(dev); break; case CHR_EVENT_CLOSED: + DPRINTF("chardev close\n"); qemu_bh_schedule(dev->chardev_close_bh); break; } @@ -890,6 +995,15 @@ static void usbredir_chardev_event(void *opaque, int event) * init + destroy */ +static void usbredir_vm_state_change(void *priv, int running, RunState state) +{ + USBRedirDevice *dev = priv; + + if (state == RUN_STATE_RUNNING && dev->parser != NULL) { + usbredirparser_do_write(dev->parser); /* Flush any pending writes */ + } +} + static int usbredir_initfn(USBDevice *udev) { USBRedirDevice *dev = DO_UPCAST(USBRedirDevice, dev, udev); @@ -914,7 +1028,8 @@ static int usbredir_initfn(USBDevice *udev) dev->chardev_close_bh = qemu_bh_new(usbredir_chardev_close_bh, dev); dev->attach_timer = qemu_new_timer_ms(vm_clock, usbredir_do_attach, dev); - QTAILQ_INIT(&dev->cancelled); + packet_id_queue_init(&dev->cancelled, dev, "cancelled"); + packet_id_queue_init(&dev->already_in_flight, dev, "already-in-flight"); for (i = 0; i < MAX_ENDPOINTS; i++) { QTAILQ_INIT(&dev->endpoint[i].bufpq); } @@ -927,19 +1042,17 @@ static int usbredir_initfn(USBDevice *udev) qemu_chr_add_handlers(dev->cs, usbredir_chardev_can_read, usbredir_chardev_read, usbredir_chardev_event, dev); + qemu_add_vm_change_state_handler(usbredir_vm_state_change, dev); add_boot_device_path(dev->bootindex, &udev->qdev, NULL); return 0; } static void usbredir_cleanup_device_queues(USBRedirDevice *dev) { - Cancelled *c, *next_c; int i; - QTAILQ_FOREACH_SAFE(c, &dev->cancelled, next, next_c) { - QTAILQ_REMOVE(&dev->cancelled, c, next); - g_free(c); - } + packet_id_queue_empty(&dev->cancelled); + packet_id_queue_empty(&dev->already_in_flight); for (i = 0; i < MAX_ENDPOINTS; i++) { usbredir_free_bufpq(dev, I2EP(i)); } @@ -1118,6 +1231,7 @@ static void usbredir_device_disconnect(void *priv) qemu_del_timer(dev->attach_timer); if (dev->dev.attached) { + DPRINTF("detaching device\n"); usb_device_detach(&dev->dev); /* * Delay next usb device attach to give the guest a chance to see @@ -1195,7 +1309,8 @@ static void usbredir_ep_info(void *priv, usb_ep->ifnum = dev->endpoint[i].interface; if (usbredirparser_peer_has_cap(dev->parser, usb_redir_cap_ep_info_max_packet_size)) { - usb_ep->max_packet_size = ep_info->max_packet_size[i]; + dev->endpoint[i].max_packet_size = + usb_ep->max_packet_size = ep_info->max_packet_size[i]; } if (ep_info->type[i] == usb_redir_type_bulk) { usb_ep->pipeline = true; @@ -1418,6 +1533,322 @@ static void usbredir_interrupt_packet(void *priv, uint64_t id, } } +/* + * Migration code + */ + +static void usbredir_pre_save(void *priv) +{ + USBRedirDevice *dev = priv; + + usbredir_fill_already_in_flight(dev); +} + +static int usbredir_post_load(void *priv, int version_id) +{ + USBRedirDevice *dev = priv; + struct USBEndpoint *usb_ep; + int i; + + switch (dev->device_info.speed) { + case usb_redir_speed_low: + dev->dev.speed = USB_SPEED_LOW; + break; + case usb_redir_speed_full: + dev->dev.speed = USB_SPEED_FULL; + break; + case usb_redir_speed_high: + dev->dev.speed = USB_SPEED_HIGH; + break; + case usb_redir_speed_super: + dev->dev.speed = USB_SPEED_SUPER; + break; + default: + dev->dev.speed = USB_SPEED_FULL; + } + dev->dev.speedmask = (1 << dev->dev.speed); + + for (i = 0; i < MAX_ENDPOINTS; i++) { + usb_ep = usb_ep_get(&dev->dev, + (i & 0x10) ? USB_TOKEN_IN : USB_TOKEN_OUT, + i & 0x0f); + usb_ep->type = dev->endpoint[i].type; + usb_ep->ifnum = dev->endpoint[i].interface; + usb_ep->max_packet_size = dev->endpoint[i].max_packet_size; + if (dev->endpoint[i].type == usb_redir_type_bulk) { + usb_ep->pipeline = true; + } + } + return 0; +} + +/* For usbredirparser migration */ +static void usbredir_put_parser(QEMUFile *f, void *priv, size_t unused) +{ + USBRedirDevice *dev = priv; + uint8_t *data; + int len; + + if (dev->parser == NULL) { + qemu_put_be32(f, 0); + return; + } + + usbredirparser_serialize(dev->parser, &data, &len); + qemu_oom_check(data); + + qemu_put_be32(f, len); + qemu_put_buffer(f, data, len); + + free(data); +} + +static int usbredir_get_parser(QEMUFile *f, void *priv, size_t unused) +{ + USBRedirDevice *dev = priv; + uint8_t *data; + int len, ret; + + len = qemu_get_be32(f); + if (len == 0) { + return 0; + } + + /* + * Our chardev should be open already at this point, otherwise + * the usbredir channel will be broken (ie spice without seamless) + */ + if (dev->parser == NULL) { + ERROR("get_parser called with closed chardev, failing migration\n"); + return -1; + } + + data = g_malloc(len); + qemu_get_buffer(f, data, len); + + ret = usbredirparser_unserialize(dev->parser, data, len); + + g_free(data); + + return ret; +} + +static const VMStateInfo usbredir_parser_vmstate_info = { + .name = "usb-redir-parser", + .put = usbredir_put_parser, + .get = usbredir_get_parser, +}; + + +/* For buffered packets (iso/irq) queue migration */ +static void usbredir_put_bufpq(QEMUFile *f, void *priv, size_t unused) +{ + struct endp_data *endp = priv; + struct buf_packet *bufp; + int remain = endp->bufpq_size; + + qemu_put_be32(f, endp->bufpq_size); + QTAILQ_FOREACH(bufp, &endp->bufpq, next) { + qemu_put_be32(f, bufp->len); + qemu_put_be32(f, bufp->status); + qemu_put_buffer(f, bufp->data, bufp->len); + remain--; + } + assert(remain == 0); +} + +static int usbredir_get_bufpq(QEMUFile *f, void *priv, size_t unused) +{ + struct endp_data *endp = priv; + struct buf_packet *bufp; + int i; + + endp->bufpq_size = qemu_get_be32(f); + for (i = 0; i < endp->bufpq_size; i++) { + bufp = g_malloc(sizeof(struct buf_packet)); + bufp->len = qemu_get_be32(f); + bufp->status = qemu_get_be32(f); + bufp->data = qemu_oom_check(malloc(bufp->len)); /* regular malloc! */ + qemu_get_buffer(f, bufp->data, bufp->len); + QTAILQ_INSERT_TAIL(&endp->bufpq, bufp, next); + } + return 0; +} + +static const VMStateInfo usbredir_ep_bufpq_vmstate_info = { + .name = "usb-redir-bufpq", + .put = usbredir_put_bufpq, + .get = usbredir_get_bufpq, +}; + + +/* For endp_data migration */ +static const VMStateDescription usbredir_ep_vmstate = { + .name = "usb-redir-ep", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT8(type, struct endp_data), + VMSTATE_UINT8(interval, struct endp_data), + VMSTATE_UINT8(interface, struct endp_data), + VMSTATE_UINT16(max_packet_size, struct endp_data), + VMSTATE_UINT8(iso_started, struct endp_data), + VMSTATE_UINT8(iso_error, struct endp_data), + VMSTATE_UINT8(interrupt_started, struct endp_data), + VMSTATE_UINT8(interrupt_error, struct endp_data), + VMSTATE_UINT8(bufpq_prefilled, struct endp_data), + VMSTATE_UINT8(bufpq_dropping_packets, struct endp_data), + { + .name = "bufpq", + .version_id = 0, + .field_exists = NULL, + .size = 0, + .info = &usbredir_ep_bufpq_vmstate_info, + .flags = VMS_SINGLE, + .offset = 0, + }, + VMSTATE_INT32(bufpq_target_size, struct endp_data), + VMSTATE_END_OF_LIST() + } +}; + + +/* For PacketIdQueue migration */ +static void usbredir_put_packet_id_q(QEMUFile *f, void *priv, size_t unused) +{ + struct PacketIdQueue *q = priv; + USBRedirDevice *dev = q->dev; + struct PacketIdQueueEntry *e; + int remain = q->size; + + DPRINTF("put_packet_id_q %s size %d\n", q->name, q->size); + qemu_put_be32(f, q->size); + QTAILQ_FOREACH(e, &q->head, next) { + qemu_put_be64(f, e->id); + remain--; + } + assert(remain == 0); +} + +static int usbredir_get_packet_id_q(QEMUFile *f, void *priv, size_t unused) +{ + struct PacketIdQueue *q = priv; + USBRedirDevice *dev = q->dev; + int i, size; + uint64_t id; + + size = qemu_get_be32(f); + DPRINTF("get_packet_id_q %s size %d\n", q->name, size); + for (i = 0; i < size; i++) { + id = qemu_get_be64(f); + packet_id_queue_add(q, id); + } + assert(q->size == size); + return 0; +} + +static const VMStateInfo usbredir_ep_packet_id_q_vmstate_info = { + .name = "usb-redir-packet-id-q", + .put = usbredir_put_packet_id_q, + .get = usbredir_get_packet_id_q, +}; + +static const VMStateDescription usbredir_ep_packet_id_queue_vmstate = { + .name = "usb-redir-packet-id-queue", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + { + .name = "queue", + .version_id = 0, + .field_exists = NULL, + .size = 0, + .info = &usbredir_ep_packet_id_q_vmstate_info, + .flags = VMS_SINGLE, + .offset = 0, + }, + VMSTATE_END_OF_LIST() + } +}; + + +/* For usb_redir_device_connect_header migration */ +static const VMStateDescription usbredir_device_info_vmstate = { + .name = "usb-redir-device-info", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT8(speed, struct usb_redir_device_connect_header), + VMSTATE_UINT8(device_class, struct usb_redir_device_connect_header), + VMSTATE_UINT8(device_subclass, struct usb_redir_device_connect_header), + VMSTATE_UINT8(device_protocol, struct usb_redir_device_connect_header), + VMSTATE_UINT16(vendor_id, struct usb_redir_device_connect_header), + VMSTATE_UINT16(product_id, struct usb_redir_device_connect_header), + VMSTATE_UINT16(device_version_bcd, + struct usb_redir_device_connect_header), + VMSTATE_END_OF_LIST() + } +}; + + +/* For usb_redir_interface_info_header migration */ +static const VMStateDescription usbredir_interface_info_vmstate = { + .name = "usb-redir-interface-info", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(interface_count, + struct usb_redir_interface_info_header), + VMSTATE_UINT8_ARRAY(interface, + struct usb_redir_interface_info_header, 32), + VMSTATE_UINT8_ARRAY(interface_class, + struct usb_redir_interface_info_header, 32), + VMSTATE_UINT8_ARRAY(interface_subclass, + struct usb_redir_interface_info_header, 32), + VMSTATE_UINT8_ARRAY(interface_protocol, + struct usb_redir_interface_info_header, 32), + VMSTATE_END_OF_LIST() + } +}; + + +/* And finally the USBRedirDevice vmstate itself */ +static const VMStateDescription usbredir_vmstate = { + .name = "usb-redir", + .version_id = 1, + .minimum_version_id = 1, + .pre_save = usbredir_pre_save, + .post_load = usbredir_post_load, + .fields = (VMStateField[]) { + VMSTATE_USB_DEVICE(dev, USBRedirDevice), + VMSTATE_TIMER(attach_timer, USBRedirDevice), + { + .name = "parser", + .version_id = 0, + .field_exists = NULL, + .size = 0, + .info = &usbredir_parser_vmstate_info, + .flags = VMS_SINGLE, + .offset = 0, + }, + VMSTATE_STRUCT_ARRAY(endpoint, USBRedirDevice, MAX_ENDPOINTS, 1, + usbredir_ep_vmstate, struct endp_data), + VMSTATE_STRUCT(cancelled, USBRedirDevice, 1, + usbredir_ep_packet_id_queue_vmstate, + struct PacketIdQueue), + VMSTATE_STRUCT(already_in_flight, USBRedirDevice, 1, + usbredir_ep_packet_id_queue_vmstate, + struct PacketIdQueue), + VMSTATE_STRUCT(device_info, USBRedirDevice, 1, + usbredir_device_info_vmstate, + struct usb_redir_device_connect_header), + VMSTATE_STRUCT(interface_info, USBRedirDevice, 1, + usbredir_interface_info_vmstate, + struct usb_redir_interface_info_header), + VMSTATE_END_OF_LIST() + } +}; + static Property usbredir_properties[] = { DEFINE_PROP_CHR("chardev", USBRedirDevice, cs), DEFINE_PROP_UINT8("debug", USBRedirDevice, debug, 0), @@ -1438,6 +1869,7 @@ static void usbredir_class_initfn(ObjectClass *klass, void *data) uc->handle_reset = usbredir_handle_reset; uc->handle_data = usbredir_handle_data; uc->handle_control = usbredir_handle_control; + dc->vmsd = &usbredir_vmstate; dc->props = usbredir_properties; } diff --git a/linux-user/qemu.h b/linux-user/qemu.h index 69b27d7146..fc4cc00b9f 100644 --- a/linux-user/qemu.h +++ b/linux-user/qemu.h @@ -289,46 +289,29 @@ static inline int access_ok(int type, abi_ulong addr, abi_ulong size) * struct has been locked - usually with lock_user_struct(). */ #define __put_user(x, hptr)\ -({\ +({ __typeof(*hptr) pu_ = (x);\ switch(sizeof(*hptr)) {\ - case 1:\ - *(uint8_t *)(hptr) = (uint8_t)(typeof(*hptr))(x);\ - break;\ - case 2:\ - *(uint16_t *)(hptr) = tswap16((uint16_t)(typeof(*hptr))(x));\ - break;\ - case 4:\ - *(uint32_t *)(hptr) = tswap32((uint32_t)(typeof(*hptr))(x));\ - break;\ - case 8:\ - *(uint64_t *)(hptr) = tswap64((typeof(*hptr))(x));\ - break;\ - default:\ - abort();\ + case 1: break;\ + case 2: pu_ = tswap16(pu_); break; \ + case 4: pu_ = tswap32(pu_); break; \ + case 8: pu_ = tswap64(pu_); break; \ + default: abort();\ }\ + memcpy(hptr, &pu_, sizeof(pu_)); \ 0;\ }) #define __get_user(x, hptr) \ -({\ +({ __typeof(*hptr) gu_; \ + memcpy(&gu_, hptr, sizeof(gu_)); \ switch(sizeof(*hptr)) {\ - case 1:\ - x = (typeof(*hptr))*(uint8_t *)(hptr);\ - break;\ - case 2:\ - x = (typeof(*hptr))tswap16(*(uint16_t *)(hptr));\ - break;\ - case 4:\ - x = (typeof(*hptr))tswap32(*(uint32_t *)(hptr));\ - break;\ - case 8:\ - x = (typeof(*hptr))tswap64(*(uint64_t *)(hptr));\ - break;\ - default:\ - /* avoid warning */\ - x = 0;\ - abort();\ + case 1: break; \ + case 2: gu_ = tswap16(gu_); break; \ + case 4: gu_ = tswap32(gu_); break; \ + case 8: gu_ = tswap64(gu_); break; \ + default: abort();\ }\ + (x) = gu_; \ 0;\ }) @@ -57,9 +57,12 @@ /* This is all part of the "official" NBD API */ +#define NBD_REQUEST_SIZE (4 + 4 + 8 + 8 + 4) #define NBD_REPLY_SIZE (4 + 4 + 8) #define NBD_REQUEST_MAGIC 0x25609513 #define NBD_REPLY_MAGIC 0x67446698 +#define NBD_OPTS_MAGIC 0x49484156454F5054LL +#define NBD_CLIENT_MAGIC 0x0000420281861253LL #define NBD_SET_SOCK _IO(0xab, 0) #define NBD_SET_BLKSIZE _IO(0xab, 1) @@ -75,6 +78,49 @@ #define NBD_OPT_EXPORT_NAME (1 << 0) +/* Definitions for opaque data types */ + +typedef struct NBDRequest NBDRequest; + +struct NBDRequest { + QSIMPLEQ_ENTRY(NBDRequest) entry; + NBDClient *client; + uint8_t *data; +}; + +struct NBDExport { + int refcount; + void (*close)(NBDExport *exp); + + BlockDriverState *bs; + char *name; + off_t dev_offset; + off_t size; + uint32_t nbdflags; + QTAILQ_HEAD(, NBDClient) clients; + QSIMPLEQ_HEAD(, NBDRequest) requests; + QTAILQ_ENTRY(NBDExport) next; +}; + +static QTAILQ_HEAD(, NBDExport) exports = QTAILQ_HEAD_INITIALIZER(exports); + +struct NBDClient { + int refcount; + void (*close)(NBDClient *client); + + NBDExport *exp; + int sock; + + Coroutine *recv_coroutine; + + CoMutex send_lock; + Coroutine *send_coroutine; + + QTAILQ_ENTRY(NBDClient) next; + int nb_requests; + bool closing; +}; + /* That's all folks */ ssize_t nbd_wr_sync(int fd, void *buffer, size_t size, bool do_read) @@ -192,11 +238,23 @@ int unix_socket_outgoing(const char *path) return unix_connect(path); } -/* Basic flow +/* Basic flow for negotiation Server Client - Negotiate + + or + + Server Client + Negotiate #1 + Option + Negotiate #2 + + ---- + + followed by + + Server Client Request Response Request @@ -204,19 +262,112 @@ int unix_socket_outgoing(const char *path) ... ... Request (type == 2) + */ -static int nbd_send_negotiate(int csock, off_t size, uint32_t flags) +static int nbd_receive_options(NBDClient *client) +{ + int csock = client->sock; + char name[256]; + uint32_t tmp, length; + uint64_t magic; + int rc; + + /* Client sends: + [ 0 .. 3] reserved (0) + [ 4 .. 11] NBD_OPTS_MAGIC + [12 .. 15] NBD_OPT_EXPORT_NAME + [16 .. 19] length + [20 .. xx] export name (length bytes) + */ + + rc = -EINVAL; + if (read_sync(csock, &tmp, sizeof(tmp)) != sizeof(tmp)) { + LOG("read failed"); + goto fail; + } + TRACE("Checking reserved"); + if (tmp != 0) { + LOG("Bad reserved received"); + goto fail; + } + + if (read_sync(csock, &magic, sizeof(magic)) != sizeof(magic)) { + LOG("read failed"); + goto fail; + } + TRACE("Checking reserved"); + if (magic != be64_to_cpu(NBD_OPTS_MAGIC)) { + LOG("Bad magic received"); + goto fail; + } + + if (read_sync(csock, &tmp, sizeof(tmp)) != sizeof(tmp)) { + LOG("read failed"); + goto fail; + } + TRACE("Checking option"); + if (tmp != be32_to_cpu(NBD_OPT_EXPORT_NAME)) { + LOG("Bad option received"); + goto fail; + } + + if (read_sync(csock, &length, sizeof(length)) != sizeof(length)) { + LOG("read failed"); + goto fail; + } + TRACE("Checking length"); + length = be32_to_cpu(length); + if (length > 255) { + LOG("Bad length received"); + goto fail; + } + if (read_sync(csock, name, length) != length) { + LOG("read failed"); + goto fail; + } + name[length] = '\0'; + + client->exp = nbd_export_find(name); + if (!client->exp) { + LOG("export not found"); + goto fail; + } + + QTAILQ_INSERT_TAIL(&client->exp->clients, client, next); + nbd_export_get(client->exp); + + TRACE("Option negotiation succeeded."); + rc = 0; +fail: + return rc; +} + +static int nbd_send_negotiate(NBDClient *client) { + int csock = client->sock; char buf[8 + 8 + 8 + 128]; int rc; + const int myflags = (NBD_FLAG_HAS_FLAGS | NBD_FLAG_SEND_TRIM | + NBD_FLAG_SEND_FLUSH | NBD_FLAG_SEND_FUA); - /* Negotiate - [ 0 .. 7] passwd ("NBDMAGIC") - [ 8 .. 15] magic (0x00420281861253) + /* Negotiation header without options: + [ 0 .. 7] passwd ("NBDMAGIC") + [ 8 .. 15] magic (NBD_CLIENT_MAGIC) [16 .. 23] size - [24 .. 27] flags - [28 .. 151] reserved (0) + [24 .. 25] server flags (0) + [24 .. 27] export flags + [28 .. 151] reserved (0) + + Negotiation header with options, part 1: + [ 0 .. 7] passwd ("NBDMAGIC") + [ 8 .. 15] magic (NBD_OPTS_MAGIC) + [16 .. 17] server flags (0) + + part 2 (after options are sent): + [18 .. 25] size + [26 .. 27] export flags + [28 .. 151] reserved (0) */ socket_set_block(csock); @@ -224,16 +375,39 @@ static int nbd_send_negotiate(int csock, off_t size, uint32_t flags) TRACE("Beginning negotiation."); memcpy(buf, "NBDMAGIC", 8); - cpu_to_be64w((uint64_t*)(buf + 8), 0x00420281861253LL); - cpu_to_be64w((uint64_t*)(buf + 16), size); - cpu_to_be32w((uint32_t*)(buf + 24), - flags | NBD_FLAG_HAS_FLAGS | NBD_FLAG_SEND_TRIM | - NBD_FLAG_SEND_FLUSH | NBD_FLAG_SEND_FUA); + if (client->exp) { + assert ((client->exp->nbdflags & ~65535) == 0); + cpu_to_be64w((uint64_t*)(buf + 8), NBD_CLIENT_MAGIC); + cpu_to_be64w((uint64_t*)(buf + 16), client->exp->size); + cpu_to_be16w((uint16_t*)(buf + 26), client->exp->nbdflags | myflags); + } else { + cpu_to_be64w((uint64_t*)(buf + 8), NBD_OPTS_MAGIC); + } memset(buf + 28, 0, 124); - if (write_sync(csock, buf, sizeof(buf)) != sizeof(buf)) { - LOG("write failed"); - goto fail; + if (client->exp) { + if (write_sync(csock, buf, sizeof(buf)) != sizeof(buf)) { + LOG("write failed"); + goto fail; + } + } else { + if (write_sync(csock, buf, 18) != 18) { + LOG("write failed"); + goto fail; + } + rc = nbd_receive_options(client); + if (rc < 0) { + LOG("option negotiation failed"); + goto fail; + } + + assert ((client->exp->nbdflags & ~65535) == 0); + cpu_to_be64w((uint64_t*)(buf + 18), client->exp->size); + cpu_to_be16w((uint16_t*)(buf + 26), client->exp->nbdflags | myflags); + if (write_sync(csock, buf + 18, sizeof(buf) - 18) != sizeof(buf) - 18) { + LOG("write failed"); + goto fail; + } } TRACE("Negotiation succeeded."); @@ -295,7 +469,7 @@ int nbd_receive_negotiate(int csock, const char *name, uint32_t *flags, uint32_t namesize; TRACE("Checking magic (opts_magic)"); - if (magic != 0x49484156454F5054LL) { + if (magic != NBD_OPTS_MAGIC) { LOG("Bad magic received"); goto fail; } @@ -334,7 +508,7 @@ int nbd_receive_negotiate(int csock, const char *name, uint32_t *flags, } else { TRACE("Checking magic (cli_magic)"); - if (magic != 0x00420281861253LL) { + if (magic != NBD_CLIENT_MAGIC) { LOG("Bad magic received"); goto fail; } @@ -477,7 +651,7 @@ int nbd_client(int fd) ssize_t nbd_send_request(int csock, struct nbd_request *request) { - uint8_t buf[4 + 4 + 8 + 8 + 4]; + uint8_t buf[NBD_REQUEST_SIZE]; ssize_t ret; cpu_to_be32w((uint32_t*)buf, NBD_REQUEST_MAGIC); @@ -504,7 +678,7 @@ ssize_t nbd_send_request(int csock, struct nbd_request *request) static ssize_t nbd_receive_request(int csock, struct nbd_request *request) { - uint8_t buf[4 + 4 + 8 + 8 + 4]; + uint8_t buf[NBD_REQUEST_SIZE]; uint32_t magic; ssize_t ret; @@ -582,7 +756,7 @@ ssize_t nbd_receive_reply(int csock, struct nbd_reply *reply) static ssize_t nbd_send_reply(int csock, struct nbd_reply *reply) { - uint8_t buf[4 + 4 + 8]; + uint8_t buf[NBD_REPLY_SIZE]; ssize_t ret; /* Reply @@ -610,58 +784,47 @@ static ssize_t nbd_send_reply(int csock, struct nbd_reply *reply) #define MAX_NBD_REQUESTS 16 -typedef struct NBDRequest NBDRequest; - -struct NBDRequest { - QSIMPLEQ_ENTRY(NBDRequest) entry; - NBDClient *client; - uint8_t *data; -}; - -struct NBDExport { - BlockDriverState *bs; - off_t dev_offset; - off_t size; - uint32_t nbdflags; - QSIMPLEQ_HEAD(, NBDRequest) requests; -}; - -struct NBDClient { - int refcount; - void (*close)(NBDClient *client); - - NBDExport *exp; - int sock; - - Coroutine *recv_coroutine; - - CoMutex send_lock; - Coroutine *send_coroutine; - - int nb_requests; -}; - -static void nbd_client_get(NBDClient *client) +void nbd_client_get(NBDClient *client) { client->refcount++; } -static void nbd_client_put(NBDClient *client) +void nbd_client_put(NBDClient *client) { if (--client->refcount == 0) { + /* The last reference should be dropped by client->close, + * which is called by nbd_client_close. + */ + assert(client->closing); + + qemu_set_fd_handler2(client->sock, NULL, NULL, NULL, NULL); + close(client->sock); + client->sock = -1; + if (client->exp) { + QTAILQ_REMOVE(&client->exp->clients, client, next); + nbd_export_put(client->exp); + } g_free(client); } } -static void nbd_client_close(NBDClient *client) +void nbd_client_close(NBDClient *client) { - qemu_set_fd_handler2(client->sock, NULL, NULL, NULL, NULL); - close(client->sock); - client->sock = -1; + if (client->closing) { + return; + } + + client->closing = true; + + /* Force requests to finish. They will drop their own references, + * then we'll close the socket and free the NBDClient. + */ + shutdown(client->sock, 2); + + /* Also tell the client, so that they release their reference. */ if (client->close) { client->close(client); } - nbd_client_put(client); } static NBDRequest *nbd_request_get(NBDClient *client) @@ -695,28 +858,109 @@ static void nbd_request_put(NBDRequest *req) } NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, - off_t size, uint32_t nbdflags) + off_t size, uint32_t nbdflags, + void (*close)(NBDExport *)) { NBDExport *exp = g_malloc0(sizeof(NBDExport)); QSIMPLEQ_INIT(&exp->requests); + exp->refcount = 1; + QTAILQ_INIT(&exp->clients); exp->bs = bs; exp->dev_offset = dev_offset; exp->nbdflags = nbdflags; exp->size = size == -1 ? bdrv_getlength(bs) : size; + exp->close = close; return exp; } +NBDExport *nbd_export_find(const char *name) +{ + NBDExport *exp; + QTAILQ_FOREACH(exp, &exports, next) { + if (strcmp(name, exp->name) == 0) { + return exp; + } + } + + return NULL; +} + +void nbd_export_set_name(NBDExport *exp, const char *name) +{ + if (exp->name == name) { + return; + } + + nbd_export_get(exp); + if (exp->name != NULL) { + g_free(exp->name); + exp->name = NULL; + QTAILQ_REMOVE(&exports, exp, next); + nbd_export_put(exp); + } + if (name != NULL) { + nbd_export_get(exp); + exp->name = g_strdup(name); + QTAILQ_INSERT_TAIL(&exports, exp, next); + } + nbd_export_put(exp); +} + void nbd_export_close(NBDExport *exp) { - while (!QSIMPLEQ_EMPTY(&exp->requests)) { - NBDRequest *first = QSIMPLEQ_FIRST(&exp->requests); - QSIMPLEQ_REMOVE_HEAD(&exp->requests, entry); - qemu_vfree(first->data); - g_free(first); + NBDClient *client, *next; + + nbd_export_get(exp); + QTAILQ_FOREACH_SAFE(client, &exp->clients, next, next) { + nbd_client_close(client); } + nbd_export_set_name(exp, NULL); + nbd_export_put(exp); +} - bdrv_close(exp->bs); - g_free(exp); +void nbd_export_get(NBDExport *exp) +{ + assert(exp->refcount > 0); + exp->refcount++; +} + +void nbd_export_put(NBDExport *exp) +{ + assert(exp->refcount > 0); + if (exp->refcount == 1) { + nbd_export_close(exp); + } + + if (--exp->refcount == 0) { + assert(exp->name == NULL); + + if (exp->close) { + exp->close(exp); + } + + while (!QSIMPLEQ_EMPTY(&exp->requests)) { + NBDRequest *first = QSIMPLEQ_FIRST(&exp->requests); + QSIMPLEQ_REMOVE_HEAD(&exp->requests, entry); + qemu_vfree(first->data); + g_free(first); + } + + g_free(exp); + } +} + +BlockDriverState *nbd_export_get_blockdev(NBDExport *exp) +{ + return exp->bs; +} + +void nbd_export_close_all(void) +{ + NBDExport *exp, *next; + + QTAILQ_FOREACH_SAFE(exp, &exports, next, next) { + nbd_export_close(exp); + } } static int nbd_can_read(void *opaque); @@ -805,14 +1049,18 @@ out: static void nbd_trip(void *opaque) { NBDClient *client = opaque; - NBDRequest *req = nbd_request_get(client); NBDExport *exp = client->exp; + NBDRequest *req; struct nbd_request request; struct nbd_reply reply; ssize_t ret; TRACE("Reading request."); + if (client->closing) { + return; + } + req = nbd_request_get(client); ret = nbd_co_receive_request(req, &request); if (ret == -EAGAIN) { goto done; @@ -974,15 +1222,21 @@ NBDClient *nbd_client_new(NBDExport *exp, int csock, void (*close)(NBDClient *)) { NBDClient *client; - if (nbd_send_negotiate(csock, exp->size, exp->nbdflags) < 0) { - return NULL; - } client = g_malloc0(sizeof(NBDClient)); client->refcount = 1; client->exp = exp; client->sock = csock; + if (nbd_send_negotiate(client) < 0) { + g_free(client); + return NULL; + } client->close = close; qemu_co_mutex_init(&client->send_lock); qemu_set_fd_handler2(csock, nbd_can_read, nbd_read, NULL, client); + + if (exp) { + QTAILQ_INSERT_TAIL(&exp->clients, client, next); + nbd_export_get(exp); + } return client; } @@ -79,9 +79,22 @@ typedef struct NBDExport NBDExport; typedef struct NBDClient NBDClient; NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, - off_t size, uint32_t nbdflags); + off_t size, uint32_t nbdflags, + void (*close)(NBDExport *)); void nbd_export_close(NBDExport *exp); +void nbd_export_get(NBDExport *exp); +void nbd_export_put(NBDExport *exp); + +BlockDriverState *nbd_export_get_blockdev(NBDExport *exp); + +NBDExport *nbd_export_find(const char *name); +void nbd_export_set_name(NBDExport *exp, const char *name); +void nbd_export_close_all(void); + NBDClient *nbd_client_new(NBDExport *exp, int csock, void (*close)(NBDClient *)); +void nbd_client_close(NBDClient *client); +void nbd_client_get(NBDClient *client); +void nbd_client_put(NBDClient *client); #endif diff --git a/oslib-win32.c b/oslib-win32.c index ffbc6d0c9f..51b33e8b20 100644 --- a/oslib-win32.c +++ b/oslib-win32.c @@ -74,6 +74,30 @@ void qemu_vfree(void *ptr) VirtualFree(ptr, 0, MEM_RELEASE); } +/* FIXME: add proper locking */ +struct tm *gmtime_r(const time_t *timep, struct tm *result) +{ + struct tm *p = gmtime(timep); + memset(result, 0, sizeof(*result)); + if (p) { + *result = *p; + p = result; + } + return p; +} + +/* FIXME: add proper locking */ +struct tm *localtime_r(const time_t *timep, struct tm *result) +{ + struct tm *p = localtime(timep); + memset(result, 0, sizeof(*result)); + if (p) { + *result = *p; + p = result; + } + return p; +} + void socket_set_block(int fd) { unsigned long opt = 0; diff --git a/qemu-nbd.c b/qemu-nbd.c index 1c1cf6a463..15bcd08123 100644 --- a/qemu-nbd.c +++ b/qemu-nbd.c @@ -41,8 +41,8 @@ static NBDExport *exp; static int verbose; static char *srcpath; static char *sockpath; -static bool sigterm_reported; -static bool nbd_started; +static int persistent = 0; +static enum { RUNNING, TERMINATE, TERMINATING, TERMINATED } state; static int shared = 1; static int nb_fds; @@ -186,7 +186,7 @@ static int find_partition(BlockDriverState *bs, int partition, static void termsig_handler(int signum) { - sigterm_reported = true; + state = TERMINATE; qemu_notify_event(); } @@ -269,10 +269,20 @@ static int nbd_can_accept(void *opaque) return nb_fds < shared; } +static void nbd_export_closed(NBDExport *exp) +{ + assert(state == TERMINATING); + state = TERMINATED; +} + static void nbd_client_closed(NBDClient *client) { nb_fds--; + if (nb_fds == 0 && !persistent && state == RUNNING) { + state = TERMINATE; + } qemu_notify_event(); + nbd_client_put(client); } static void nbd_accept(void *opaque) @@ -282,7 +292,11 @@ static void nbd_accept(void *opaque) socklen_t addr_len = sizeof(addr); int fd = accept(server_fd, (struct sockaddr *)&addr, &addr_len); - nbd_started = true; + if (state >= TERMINATE) { + close(fd); + return; + } + if (fd >= 0 && nbd_client_new(exp, fd, nbd_client_closed)) { nb_fds++; } @@ -329,7 +343,6 @@ int main(int argc, char **argv) int partition = -1; int ret; int fd; - int persistent = 0; bool seen_cache = false; #ifdef CONFIG_LINUX_AIO bool seen_aio = false; @@ -546,7 +559,7 @@ int main(int argc, char **argv) } } - exp = nbd_export_new(bs, dev_offset, fd_size, nbdflags); + exp = nbd_export_new(bs, dev_offset, fd_size, nbdflags, nbd_export_closed); if (sockpath) { fd = unix_socket_incoming(sockpath); @@ -581,11 +594,18 @@ int main(int argc, char **argv) err(EXIT_FAILURE, "Could not chdir to root directory"); } + state = RUNNING; do { main_loop_wait(false); - } while (!sigterm_reported && (persistent || !nbd_started || nb_fds > 0)); + if (state == TERMINATE) { + state = TERMINATING; + nbd_export_close(exp); + nbd_export_put(exp); + exp = NULL; + } + } while (state != TERMINATED); - nbd_export_close(exp); + bdrv_close(bs); if (sockpath) { unlink(sockpath); } diff --git a/qemu-os-win32.h b/qemu-os-win32.h index 753679b194..3b5a35b6c2 100644 --- a/qemu-os-win32.h +++ b/qemu-os-win32.h @@ -68,6 +68,12 @@ /* Declaration of ffs() is missing in MinGW's strings.h. */ int ffs(int i); +/* Missing POSIX functions. Don't use MinGW-w64 macros. */ +#undef gmtime_r +struct tm *gmtime_r(const time_t *timep, struct tm *result); +#undef localtime_r +struct tm *localtime_r(const time_t *timep, struct tm *result); + static inline void os_setup_signal_handling(void) {} static inline void os_daemonize(void) {} static inline void os_setup_post(void) {} diff --git a/qemu-timer.h b/qemu-timer.h index f8af595f13..da7e97cd5a 100644 --- a/qemu-timer.h +++ b/qemu-timer.h @@ -218,7 +218,7 @@ static inline int64_t cpu_get_real_ticks(void) return val; } -#elif defined(__sparc_v8plus__) || defined(__sparc_v8plusa__) || defined(__sparc_v9__) +#elif defined(__sparc__) static inline int64_t cpu_get_real_ticks (void) { @@ -227,6 +227,8 @@ static inline int64_t cpu_get_real_ticks (void) asm volatile("rd %%tick,%0" : "=r"(rval)); return rval; #else + /* We need an %o or %g register for this. For recent enough gcc + there is an "h" constraint for that. Don't bother with that. */ union { uint64_t i64; struct { @@ -234,8 +236,8 @@ static inline int64_t cpu_get_real_ticks (void) uint32_t low; } i32; } rval; - asm volatile("rd %%tick,%1; srlx %1,32,%0" - : "=r"(rval.i32.high), "=r"(rval.i32.low)); + asm volatile("rd %%tick,%%g1; srlx %%g1,32,%0; mov %%g1,%1" + : "=r"(rval.i32.high), "=r"(rval.i32.low) : : "g1"); return rval.i64; #endif } diff --git a/target-alpha/translate.c b/target-alpha/translate.c index 93063fbe73..3f9aee12d4 100644 --- a/target-alpha/translate.c +++ b/target-alpha/translate.c @@ -426,27 +426,15 @@ static ExitStatus gen_bcond_internal(DisasContext *ctx, TCGCond cond, return EXIT_GOTO_TB; } else { - int lab_over = gen_new_label(); - - /* ??? Consider using either - movi pc, next - addi tmp, pc, disp - movcond pc, cond, 0, tmp, pc - or - setcond tmp, cond, 0 - movi pc, next - neg tmp, tmp - andi tmp, tmp, disp - add pc, pc, tmp - The current diamond subgraph surely isn't efficient. */ + TCGv_i64 z = tcg_const_i64(0); + TCGv_i64 d = tcg_const_i64(dest); + TCGv_i64 p = tcg_const_i64(ctx->pc); - tcg_gen_brcondi_i64(cond, cmp, 0, lab_true); - tcg_gen_movi_i64(cpu_pc, ctx->pc); - tcg_gen_br(lab_over); - gen_set_label(lab_true); - tcg_gen_movi_i64(cpu_pc, dest); - gen_set_label(lab_over); + tcg_gen_movcond_i64(cond, cpu_pc, cmp, z, d, p); + tcg_temp_free_i64(z); + tcg_temp_free_i64(d); + tcg_temp_free_i64(p); return EXIT_PC_UPDATED; } } @@ -521,61 +509,67 @@ static ExitStatus gen_fbcond(DisasContext *ctx, TCGCond cond, int ra, static void gen_cmov(TCGCond cond, int ra, int rb, int rc, int islit, uint8_t lit, int mask) { - TCGCond inv_cond = tcg_invert_cond(cond); - int l1; + TCGv_i64 c1, z, v1; - if (unlikely(rc == 31)) + if (unlikely(rc == 31)) { return; + } - l1 = gen_new_label(); - - if (ra != 31) { - if (mask) { - TCGv tmp = tcg_temp_new(); - tcg_gen_andi_i64(tmp, cpu_ir[ra], 1); - tcg_gen_brcondi_i64(inv_cond, tmp, 0, l1); - tcg_temp_free(tmp); - } else - tcg_gen_brcondi_i64(inv_cond, cpu_ir[ra], 0, l1); - } else { + if (ra == 31) { /* Very uncommon case - Do not bother to optimize. */ - TCGv tmp = tcg_const_i64(0); - tcg_gen_brcondi_i64(inv_cond, tmp, 0, l1); - tcg_temp_free(tmp); + c1 = tcg_const_i64(0); + } else if (mask) { + c1 = tcg_const_i64(1); + tcg_gen_and_i64(c1, c1, cpu_ir[ra]); + } else { + c1 = cpu_ir[ra]; } + if (islit) { + v1 = tcg_const_i64(lit); + } else { + v1 = cpu_ir[rb]; + } + z = tcg_const_i64(0); - if (islit) - tcg_gen_movi_i64(cpu_ir[rc], lit); - else - tcg_gen_mov_i64(cpu_ir[rc], cpu_ir[rb]); - gen_set_label(l1); + tcg_gen_movcond_i64(cond, cpu_ir[rc], c1, z, v1, cpu_ir[rc]); + + tcg_temp_free_i64(z); + if (ra == 31 || mask) { + tcg_temp_free_i64(c1); + } + if (islit) { + tcg_temp_free_i64(v1); + } } static void gen_fcmov(TCGCond cond, int ra, int rb, int rc) { - TCGv cmp_tmp; - int l1; + TCGv_i64 c1, z, v1; if (unlikely(rc == 31)) { return; } - cmp_tmp = tcg_temp_new(); + c1 = tcg_temp_new_i64(); if (unlikely(ra == 31)) { - tcg_gen_movi_i64(cmp_tmp, 0); + tcg_gen_movi_i64(c1, 0); + } else { + gen_fold_mzero(cond, c1, cpu_fir[ra]); + } + if (rb == 31) { + v1 = tcg_const_i64(0); } else { - gen_fold_mzero(cond, cmp_tmp, cpu_fir[ra]); + v1 = cpu_fir[rb]; } + z = tcg_const_i64(0); - l1 = gen_new_label(); - tcg_gen_brcondi_i64(tcg_invert_cond(cond), cmp_tmp, 0, l1); - tcg_temp_free(cmp_tmp); + tcg_gen_movcond_i64(cond, cpu_fir[rc], c1, z, v1, cpu_fir[rc]); - if (rb != 31) - tcg_gen_mov_i64(cpu_fir[rc], cpu_fir[rb]); - else - tcg_gen_movi_i64(cpu_fir[rc], 0); - gen_set_label(l1); + tcg_temp_free_i64(z); + tcg_temp_free_i64(c1); + if (rb == 31) { + tcg_temp_free_i64(v1); + } } #define QUAL_RM_N 0x080 /* Round mode nearest even */ diff --git a/target-sh4/helper.h b/target-sh4/helper.h index 6e4f108124..6c1a47da9f 100644 --- a/target-sh4/helper.h +++ b/target-sh4/helper.h @@ -1,30 +1,26 @@ #include "def-helper.h" DEF_HELPER_1(ldtlb, void, env) -DEF_HELPER_1(raise_illegal_instruction, void, env) -DEF_HELPER_1(raise_slot_illegal_instruction, void, env) -DEF_HELPER_1(raise_fpu_disable, void, env) -DEF_HELPER_1(raise_slot_fpu_disable, void, env) -DEF_HELPER_1(debug, void, env) -DEF_HELPER_2(sleep, void, env, i32) -DEF_HELPER_2(trapa, void, env, i32) +DEF_HELPER_1(raise_illegal_instruction, noreturn, env) +DEF_HELPER_1(raise_slot_illegal_instruction, noreturn, env) +DEF_HELPER_1(raise_fpu_disable, noreturn, env) +DEF_HELPER_1(raise_slot_fpu_disable, noreturn, env) +DEF_HELPER_1(debug, noreturn, env) +DEF_HELPER_1(sleep, noreturn, env) +DEF_HELPER_2(trapa, noreturn, env, i32) DEF_HELPER_3(movcal, void, env, i32, i32) DEF_HELPER_1(discard_movcal_backup, void, env) DEF_HELPER_2(ocbi, void, env, i32) -DEF_HELPER_3(addv, i32, env, i32, i32) -DEF_HELPER_3(addc, i32, env, i32, i32) -DEF_HELPER_3(subv, i32, env, i32, i32) -DEF_HELPER_3(subc, i32, env, i32, i32) DEF_HELPER_3(div1, i32, env, i32, i32) DEF_HELPER_3(macl, void, env, i32, i32) DEF_HELPER_3(macw, void, env, i32, i32) DEF_HELPER_2(ld_fpscr, void, env, i32) -DEF_HELPER_1(fabs_FT, f32, f32) -DEF_HELPER_1(fabs_DT, f64, f64) +DEF_HELPER_FLAGS_1(fabs_FT, TCG_CALL_CONST | TCG_CALL_PURE, f32, f32) +DEF_HELPER_FLAGS_1(fabs_DT, TCG_CALL_CONST | TCG_CALL_PURE, f64, f64) DEF_HELPER_3(fadd_FT, f32, env, f32, f32) DEF_HELPER_3(fadd_DT, f64, env, f64, f64) DEF_HELPER_2(fcnvsd_FT_DT, f64, env, f32) @@ -41,7 +37,7 @@ DEF_HELPER_2(float_DT, f64, env, i32) DEF_HELPER_4(fmac_FT, f32, env, f32, f32, f32) DEF_HELPER_3(fmul_FT, f32, env, f32, f32) DEF_HELPER_3(fmul_DT, f64, env, f64, f64) -DEF_HELPER_1(fneg_T, f32, f32) +DEF_HELPER_FLAGS_1(fneg_T, TCG_CALL_CONST | TCG_CALL_PURE, f32, f32) DEF_HELPER_3(fsub_FT, f32, env, f32, f32) DEF_HELPER_3(fsub_DT, f64, env, f64, f64) DEF_HELPER_2(fsqrt_FT, f32, env, f32) diff --git a/target-sh4/op_helper.c b/target-sh4/op_helper.c index 9b4328de37..60ec4cbc4d 100644 --- a/target-sh4/op_helper.c +++ b/target-sh4/op_helper.c @@ -21,7 +21,8 @@ #include "cpu.h" #include "helper.h" -static void cpu_restore_state_from_retaddr(CPUSH4State *env, uintptr_t retaddr) +static inline void cpu_restore_state_from_retaddr(CPUSH4State *env, + uintptr_t retaddr) { TranslationBlock *tb; @@ -77,8 +78,8 @@ void helper_ldtlb(CPUSH4State *env) #endif } -static inline void raise_exception(CPUSH4State *env, int index, - uintptr_t retaddr) +static inline void QEMU_NORETURN raise_exception(CPUSH4State *env, int index, + uintptr_t retaddr) { env->exception_index = index; cpu_restore_state_from_retaddr(env, retaddr); @@ -87,43 +88,40 @@ static inline void raise_exception(CPUSH4State *env, int index, void helper_raise_illegal_instruction(CPUSH4State *env) { - raise_exception(env, 0x180, GETPC()); + raise_exception(env, 0x180, 0); } void helper_raise_slot_illegal_instruction(CPUSH4State *env) { - raise_exception(env, 0x1a0, GETPC()); + raise_exception(env, 0x1a0, 0); } void helper_raise_fpu_disable(CPUSH4State *env) { - raise_exception(env, 0x800, GETPC()); + raise_exception(env, 0x800, 0); } void helper_raise_slot_fpu_disable(CPUSH4State *env) { - raise_exception(env, 0x820, GETPC()); + raise_exception(env, 0x820, 0); } void helper_debug(CPUSH4State *env) { - env->exception_index = EXCP_DEBUG; - cpu_loop_exit(env); + raise_exception(env, EXCP_DEBUG, 0); } -void helper_sleep(CPUSH4State *env, uint32_t next_pc) +void helper_sleep(CPUSH4State *env) { env->halted = 1; env->in_sleep = 1; - env->exception_index = EXCP_HLT; - env->pc = next_pc; - cpu_loop_exit(env); + raise_exception(env, EXCP_HLT, 0); } void helper_trapa(CPUSH4State *env, uint32_t tra) { env->tra = tra << 2; - raise_exception(env, 0x160, GETPC()); + raise_exception(env, 0x160, 0); } void helper_movcal(CPUSH4State *env, uint32_t address, uint32_t value) @@ -177,51 +175,6 @@ void helper_ocbi(CPUSH4State *env, uint32_t address) } } -uint32_t helper_addc(CPUSH4State *env, uint32_t arg0, uint32_t arg1) -{ - uint32_t tmp0, tmp1; - - tmp1 = arg0 + arg1; - tmp0 = arg1; - arg1 = tmp1 + (env->sr & 1); - if (tmp0 > tmp1) - env->sr |= SR_T; - else - env->sr &= ~SR_T; - if (tmp1 > arg1) - env->sr |= SR_T; - return arg1; -} - -uint32_t helper_addv(CPUSH4State *env, uint32_t arg0, uint32_t arg1) -{ - uint32_t dest, src, ans; - - if ((int32_t) arg1 >= 0) - dest = 0; - else - dest = 1; - if ((int32_t) arg0 >= 0) - src = 0; - else - src = 1; - src += dest; - arg1 += arg0; - if ((int32_t) arg1 >= 0) - ans = 0; - else - ans = 1; - ans += dest; - if (src == 0 || src == 2) { - if (ans == 1) - env->sr |= SR_T; - else - env->sr &= ~SR_T; - } else - env->sr &= ~SR_T; - return arg1; -} - #define T (env->sr & SR_T) #define Q (env->sr & SR_Q ? 1 : 0) #define M (env->sr & SR_M ? 1 : 0) @@ -375,51 +328,6 @@ void helper_macw(CPUSH4State *env, uint32_t arg0, uint32_t arg1) } } -uint32_t helper_subc(CPUSH4State *env, uint32_t arg0, uint32_t arg1) -{ - uint32_t tmp0, tmp1; - - tmp1 = arg1 - arg0; - tmp0 = arg1; - arg1 = tmp1 - (env->sr & SR_T); - if (tmp0 < tmp1) - env->sr |= SR_T; - else - env->sr &= ~SR_T; - if (tmp1 < arg1) - env->sr |= SR_T; - return arg1; -} - -uint32_t helper_subv(CPUSH4State *env, uint32_t arg0, uint32_t arg1) -{ - int32_t dest, src, ans; - - if ((int32_t) arg1 >= 0) - dest = 0; - else - dest = 1; - if ((int32_t) arg0 >= 0) - src = 0; - else - src = 1; - src += dest; - arg1 -= arg0; - if ((int32_t) arg1 >= 0) - ans = 0; - else - ans = 1; - ans += dest; - if (src == 1) { - if (ans == 1) - env->sr |= SR_T; - else - env->sr &= ~SR_T; - } else - env->sr &= ~SR_T; - return arg1; -} - static inline void set_t(CPUSH4State *env) { env->sr |= SR_T; @@ -475,9 +383,7 @@ static void update_fpscr(CPUSH4State *env, uintptr_t retaddr) cause = (env->fpscr & FPSCR_CAUSE_MASK) >> FPSCR_CAUSE_SHIFT; enable = (env->fpscr & FPSCR_ENABLE_MASK) >> FPSCR_ENABLE_SHIFT; if (cause & enable) { - cpu_restore_state_from_retaddr(env, retaddr); - env->exception_index = 0x120; - cpu_loop_exit(env); + raise_exception(env, 0x120, retaddr); } } } @@ -623,8 +529,7 @@ float64 helper_float_DT(CPUSH4State *env, uint32_t t0) float32 helper_fmac_FT(CPUSH4State *env, float32 t0, float32 t1, float32 t2) { set_float_exception_flags(0, &env->fp_status); - t0 = float32_mul(t0, t1, &env->fp_status); - t0 = float32_add(t0, t2, &env->fp_status); + t0 = float32_muladd(t0, t1, t2, 0, &env->fp_status); update_fpscr(env, GETPC()); return t0; } diff --git a/target-sh4/translate.c b/target-sh4/translate.c index d05c74c8c3..0fa83cab99 100644 --- a/target-sh4/translate.c +++ b/target-sh4/translate.c @@ -18,7 +18,6 @@ */ #define DEBUG_DISAS -#define SH4_DEBUG_DISAS //#define SH4_SINGLE_STEP #include "cpu.h" @@ -32,8 +31,6 @@ typedef struct DisasContext { struct TranslationBlock *tb; target_ulong pc; - uint32_t sr; - uint32_t fpscr; uint16_t opcode; uint32_t flags; int bstate; @@ -47,7 +44,7 @@ typedef struct DisasContext { #if defined(CONFIG_USER_ONLY) #define IS_USER(ctx) 1 #else -#define IS_USER(ctx) (!(ctx->sr & SR_MD)) +#define IS_USER(ctx) (!(ctx->flags & SR_MD)) #endif enum { @@ -339,16 +336,6 @@ static void gen_delayed_conditional_jump(DisasContext * ctx) gen_jump(ctx); } -static inline void gen_set_t(void) -{ - tcg_gen_ori_i32(cpu_sr, cpu_sr, SR_T); -} - -static inline void gen_clr_t(void) -{ - tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); -} - static inline void gen_cmp(int cond, TCGv t0, TCGv t1) { TCGv t; @@ -423,44 +410,47 @@ static inline void gen_store_fpr64 (TCGv_i64 t, int reg) #define B11_8 ((ctx->opcode >> 8) & 0xf) #define B15_12 ((ctx->opcode >> 12) & 0xf) -#define REG(x) ((x) < 8 && (ctx->sr & (SR_MD | SR_RB)) == (SR_MD | SR_RB) ? \ - (cpu_gregs[x + 16]) : (cpu_gregs[x])) +#define REG(x) ((x) < 8 && (ctx->flags & (SR_MD | SR_RB)) == (SR_MD | SR_RB) \ + ? (cpu_gregs[x + 16]) : (cpu_gregs[x])) -#define ALTREG(x) ((x) < 8 && (ctx->sr & (SR_MD | SR_RB)) != (SR_MD | SR_RB) \ +#define ALTREG(x) ((x) < 8 && (ctx->flags & (SR_MD | SR_RB)) != (SR_MD | SR_RB)\ ? (cpu_gregs[x + 16]) : (cpu_gregs[x])) -#define FREG(x) (ctx->fpscr & FPSCR_FR ? (x) ^ 0x10 : (x)) +#define FREG(x) (ctx->flags & FPSCR_FR ? (x) ^ 0x10 : (x)) #define XHACK(x) ((((x) & 1 ) << 4) | ((x) & 0xe)) -#define XREG(x) (ctx->fpscr & FPSCR_FR ? XHACK(x) ^ 0x10 : XHACK(x)) +#define XREG(x) (ctx->flags & FPSCR_FR ? XHACK(x) ^ 0x10 : XHACK(x)) #define DREG(x) FREG(x) /* Assumes lsb of (x) is always 0 */ #define CHECK_NOT_DELAY_SLOT \ if (ctx->flags & (DELAY_SLOT | DELAY_SLOT_CONDITIONAL)) \ { \ + tcg_gen_movi_i32(cpu_pc, ctx->pc); \ gen_helper_raise_slot_illegal_instruction(cpu_env); \ - ctx->bstate = BS_EXCP; \ + ctx->bstate = BS_BRANCH; \ return; \ } #define CHECK_PRIVILEGED \ if (IS_USER(ctx)) { \ + tcg_gen_movi_i32(cpu_pc, ctx->pc); \ if (ctx->flags & (DELAY_SLOT | DELAY_SLOT_CONDITIONAL)) { \ gen_helper_raise_slot_illegal_instruction(cpu_env); \ } else { \ gen_helper_raise_illegal_instruction(cpu_env); \ } \ - ctx->bstate = BS_EXCP; \ + ctx->bstate = BS_BRANCH; \ return; \ } #define CHECK_FPU_ENABLED \ if (ctx->flags & SR_FD) { \ + tcg_gen_movi_i32(cpu_pc, ctx->pc); \ if (ctx->flags & (DELAY_SLOT | DELAY_SLOT_CONDITIONAL)) { \ gen_helper_raise_slot_fpu_disable(cpu_env); \ } else { \ gen_helper_raise_fpu_disable(cpu_env); \ } \ - ctx->bstate = BS_EXCP; \ + ctx->bstate = BS_BRANCH; \ return; \ } @@ -519,7 +509,7 @@ static void _decode_opc(DisasContext * ctx) tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_S); return; case 0x0008: /* clrt */ - gen_clr_t(); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); return; case 0x0038: /* ldtlb */ CHECK_PRIVILEGED @@ -537,21 +527,22 @@ static void _decode_opc(DisasContext * ctx) tcg_gen_ori_i32(cpu_sr, cpu_sr, SR_S); return; case 0x0018: /* sett */ - gen_set_t(); + tcg_gen_ori_i32(cpu_sr, cpu_sr, SR_T); return; case 0xfbfd: /* frchg */ tcg_gen_xori_i32(cpu_fpscr, cpu_fpscr, FPSCR_FR); ctx->bstate = BS_STOP; return; case 0xf3fd: /* fschg */ - tcg_gen_xori_i32(cpu_fpscr, cpu_fpscr, FPSCR_SZ); + tcg_gen_xori_i32(cpu_fpscr, cpu_fpscr, FPSCR_SZ); ctx->bstate = BS_STOP; return; case 0x0009: /* nop */ return; case 0x001b: /* sleep */ CHECK_PRIVILEGED - gen_helper_sleep(cpu_env, tcg_const_i32(ctx->pc + 2)); + tcg_gen_movi_i32(cpu_pc, ctx->pc + 2); + gen_helper_sleep(cpu_env); return; } @@ -732,17 +723,7 @@ static void _decode_opc(DisasContext * ctx) } return; case 0x6009: /* swap.w Rm,Rn */ - { - TCGv high, low; - high = tcg_temp_new(); - tcg_gen_shli_i32(high, REG(B7_4), 16); - low = tcg_temp_new(); - tcg_gen_shri_i32(low, REG(B7_4), 16); - tcg_gen_ext16u_i32(low, low); - tcg_gen_or_i32(REG(B11_8), high, low); - tcg_temp_free(low); - tcg_temp_free(high); - } + tcg_gen_rotli_i32(REG(B11_8), REG(B7_4), 16); return; case 0x200d: /* xtrct Rm,Rn */ { @@ -751,7 +732,6 @@ static void _decode_opc(DisasContext * ctx) tcg_gen_shli_i32(high, REG(B7_4), 16); low = tcg_temp_new(); tcg_gen_shri_i32(low, REG(B11_8), 16); - tcg_gen_ext16u_i32(low, low); tcg_gen_or_i32(REG(B11_8), high, low); tcg_temp_free(low); tcg_temp_free(high); @@ -761,10 +741,43 @@ static void _decode_opc(DisasContext * ctx) tcg_gen_add_i32(REG(B11_8), REG(B11_8), REG(B7_4)); return; case 0x300e: /* addc Rm,Rn */ - gen_helper_addc(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8)); + { + TCGv t0, t1, t2; + t0 = tcg_temp_new(); + tcg_gen_andi_i32(t0, cpu_sr, SR_T); + t1 = tcg_temp_new(); + tcg_gen_add_i32(t1, REG(B7_4), REG(B11_8)); + tcg_gen_add_i32(t0, t0, t1); + t2 = tcg_temp_new(); + tcg_gen_setcond_i32(TCG_COND_GTU, t2, REG(B11_8), t1); + tcg_gen_setcond_i32(TCG_COND_GTU, t1, t1, t0); + tcg_gen_or_i32(t1, t1, t2); + tcg_temp_free(t2); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); + tcg_gen_or_i32(cpu_sr, cpu_sr, t1); + tcg_temp_free(t1); + tcg_gen_mov_i32(REG(B11_8), t0); + tcg_temp_free(t0); + } return; case 0x300f: /* addv Rm,Rn */ - gen_helper_addv(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8)); + { + TCGv t0, t1, t2; + t0 = tcg_temp_new(); + tcg_gen_add_i32(t0, REG(B7_4), REG(B11_8)); + t1 = tcg_temp_new(); + tcg_gen_xor_i32(t1, t0, REG(B11_8)); + t2 = tcg_temp_new(); + tcg_gen_xor_i32(t2, REG(B7_4), REG(B11_8)); + tcg_gen_andc_i32(t1, t1, t2); + tcg_temp_free(t2); + tcg_gen_shri_i32(t1, t1, 31); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); + tcg_gen_or_i32(cpu_sr, cpu_sr, t1); + tcg_temp_free(t1); + tcg_gen_mov_i32(REG(B7_4), t0); + tcg_temp_free(t0); + } return; case 0x2009: /* and Rm,Rn */ tcg_gen_and_i32(REG(B11_8), REG(B11_8), REG(B7_4)); @@ -1013,10 +1026,43 @@ static void _decode_opc(DisasContext * ctx) tcg_gen_sub_i32(REG(B11_8), REG(B11_8), REG(B7_4)); return; case 0x300a: /* subc Rm,Rn */ - gen_helper_subc(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8)); + { + TCGv t0, t1, t2; + t0 = tcg_temp_new(); + tcg_gen_andi_i32(t0, cpu_sr, SR_T); + t1 = tcg_temp_new(); + tcg_gen_sub_i32(t1, REG(B11_8), REG(B7_4)); + tcg_gen_sub_i32(t0, t1, t0); + t2 = tcg_temp_new(); + tcg_gen_setcond_i32(TCG_COND_LTU, t2, REG(B11_8), t1); + tcg_gen_setcond_i32(TCG_COND_LTU, t1, t1, t0); + tcg_gen_or_i32(t1, t1, t2); + tcg_temp_free(t2); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); + tcg_gen_or_i32(cpu_sr, cpu_sr, t1); + tcg_temp_free(t1); + tcg_gen_mov_i32(REG(B11_8), t0); + tcg_temp_free(t0); + } return; case 0x300b: /* subv Rm,Rn */ - gen_helper_subv(REG(B11_8), cpu_env, REG(B7_4), REG(B11_8)); + { + TCGv t0, t1, t2; + t0 = tcg_temp_new(); + tcg_gen_sub_i32(t0, REG(B11_8), REG(B7_4)); + t1 = tcg_temp_new(); + tcg_gen_xor_i32(t1, t0, REG(B7_4)); + t2 = tcg_temp_new(); + tcg_gen_xor_i32(t2, REG(B11_8), REG(B7_4)); + tcg_gen_and_i32(t1, t1, t2); + tcg_temp_free(t2); + tcg_gen_shri_i32(t1, t1, 31); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); + tcg_gen_or_i32(cpu_sr, cpu_sr, t1); + tcg_temp_free(t1); + tcg_gen_mov_i32(REG(B11_8), t0); + tcg_temp_free(t0); + } return; case 0x2008: /* tst Rm,Rn */ { @@ -1031,7 +1077,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf00c: /* fmov {F,D,X}Rm,{F,D,X}Rn - FPSCR: Nothing */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { TCGv_i64 fp = tcg_temp_new_i64(); gen_load_fpr64(fp, XREG(B7_4)); gen_store_fpr64(fp, XREG(B11_8)); @@ -1042,7 +1088,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf00a: /* fmov {F,D,X}Rm,@Rn - FPSCR: Nothing */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { TCGv addr_hi = tcg_temp_new(); int fr = XREG(B7_4); tcg_gen_addi_i32(addr_hi, REG(B11_8), 4); @@ -1055,7 +1101,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf008: /* fmov @Rm,{F,D,X}Rn - FPSCR: Nothing */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { TCGv addr_hi = tcg_temp_new(); int fr = XREG(B11_8); tcg_gen_addi_i32(addr_hi, REG(B7_4), 4); @@ -1068,7 +1114,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf009: /* fmov @Rm+,{F,D,X}Rn - FPSCR: Nothing */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { TCGv addr_hi = tcg_temp_new(); int fr = XREG(B11_8); tcg_gen_addi_i32(addr_hi, REG(B7_4), 4); @@ -1083,7 +1129,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf00b: /* fmov {F,D,X}Rm,@-Rn - FPSCR: Nothing */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { TCGv addr = tcg_temp_new_i32(); int fr = XREG(B7_4); tcg_gen_subi_i32(addr, REG(B11_8), 4); @@ -1106,7 +1152,7 @@ static void _decode_opc(DisasContext * ctx) { TCGv addr = tcg_temp_new_i32(); tcg_gen_add_i32(addr, REG(B7_4), REG(0)); - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { int fr = XREG(B11_8); tcg_gen_qemu_ld32u(cpu_fregs[fr ], addr, ctx->memidx); tcg_gen_addi_i32(addr, addr, 4); @@ -1122,7 +1168,7 @@ static void _decode_opc(DisasContext * ctx) { TCGv addr = tcg_temp_new(); tcg_gen_add_i32(addr, REG(B11_8), REG(0)); - if (ctx->fpscr & FPSCR_SZ) { + if (ctx->flags & FPSCR_SZ) { int fr = XREG(B7_4); tcg_gen_qemu_ld32u(cpu_fregs[fr ], addr, ctx->memidx); tcg_gen_addi_i32(addr, addr, 4); @@ -1141,7 +1187,7 @@ static void _decode_opc(DisasContext * ctx) case 0xf005: /* fcmp/gt Rm,Rn - FPSCR: R[PR,Enable.V]/W[Cause,Flag] */ { CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { TCGv_i64 fp0, fp1; if (ctx->opcode & 0x0110) @@ -1210,7 +1256,7 @@ static void _decode_opc(DisasContext * ctx) case 0xf00e: /* fmac FR0,RM,Rn */ { CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { break; /* illegal instruction */ } else { gen_helper_fmac_FT(cpu_fregs[FREG(B11_8)], cpu_env, @@ -1366,6 +1412,7 @@ static void _decode_opc(DisasContext * ctx) { TCGv imm; CHECK_NOT_DELAY_SLOT + tcg_gen_movi_i32(cpu_pc, ctx->pc); imm = tcg_const_i32(B7_0); gen_helper_trapa(cpu_env, imm); tcg_temp_free(imm); @@ -1605,7 +1652,7 @@ static void _decode_opc(DisasContext * ctx) */ if (ctx->features & SH_FEATURE_SH4A) { int label = gen_new_label(); - gen_clr_t(); + tcg_gen_andi_i32(cpu_sr, cpu_sr, ~SR_T); tcg_gen_or_i32(cpu_sr, cpu_sr, cpu_ldst); tcg_gen_brcondi_i32(TCG_COND_EQ, cpu_ldst, 0, label); tcg_gen_qemu_st32(REG(0), REG(B11_8), ctx->memidx); @@ -1739,7 +1786,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf02d: /* float FPUL,FRn/DRn - FPSCR: R[PR,Enable.I]/W[Cause,Flag] */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { TCGv_i64 fp; if (ctx->opcode & 0x0100) break; /* illegal instruction */ @@ -1754,7 +1801,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf03d: /* ftrc FRm/DRm,FPUL - FPSCR: R[PR,Enable.V]/W[Cause,Flag] */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { TCGv_i64 fp; if (ctx->opcode & 0x0100) break; /* illegal instruction */ @@ -1775,7 +1822,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf05d: /* fabs FRn/DRn */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { if (ctx->opcode & 0x0100) break; /* illegal instruction */ TCGv_i64 fp = tcg_temp_new_i64(); @@ -1789,7 +1836,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf06d: /* fsqrt FRn */ CHECK_FPU_ENABLED - if (ctx->fpscr & FPSCR_PR) { + if (ctx->flags & FPSCR_PR) { if (ctx->opcode & 0x0100) break; /* illegal instruction */ TCGv_i64 fp = tcg_temp_new_i64(); @@ -1807,13 +1854,13 @@ static void _decode_opc(DisasContext * ctx) break; case 0xf08d: /* fldi0 FRn - FPSCR: R[PR] */ CHECK_FPU_ENABLED - if (!(ctx->fpscr & FPSCR_PR)) { + if (!(ctx->flags & FPSCR_PR)) { tcg_gen_movi_i32(cpu_fregs[FREG(B11_8)], 0); } return; case 0xf09d: /* fldi1 FRn - FPSCR: R[PR] */ CHECK_FPU_ENABLED - if (!(ctx->fpscr & FPSCR_PR)) { + if (!(ctx->flags & FPSCR_PR)) { tcg_gen_movi_i32(cpu_fregs[FREG(B11_8)], 0x3f800000); } return; @@ -1837,7 +1884,7 @@ static void _decode_opc(DisasContext * ctx) return; case 0xf0ed: /* fipr FVm,FVn */ CHECK_FPU_ENABLED - if ((ctx->fpscr & FPSCR_PR) == 0) { + if ((ctx->flags & FPSCR_PR) == 0) { TCGv m, n; m = tcg_const_i32((ctx->opcode >> 8) & 3); n = tcg_const_i32((ctx->opcode >> 10) & 3); @@ -1850,7 +1897,7 @@ static void _decode_opc(DisasContext * ctx) case 0xf0fd: /* ftrv XMTRX,FVn */ CHECK_FPU_ENABLED if ((ctx->opcode & 0x0300) == 0x0100 && - (ctx->fpscr & FPSCR_PR) == 0) { + (ctx->flags & FPSCR_PR) == 0) { TCGv n; n = tcg_const_i32((ctx->opcode >> 10) & 3); gen_helper_ftrv(cpu_env, n); @@ -1864,12 +1911,13 @@ static void _decode_opc(DisasContext * ctx) ctx->opcode, ctx->pc); fflush(stderr); #endif + tcg_gen_movi_i32(cpu_pc, ctx->pc); if (ctx->flags & (DELAY_SLOT | DELAY_SLOT_CONDITIONAL)) { gen_helper_raise_slot_illegal_instruction(cpu_env); } else { gen_helper_raise_illegal_instruction(cpu_env); } - ctx->bstate = BS_EXCP; + ctx->bstate = BS_BRANCH; } static void decode_opc(DisasContext * ctx) @@ -1923,16 +1971,14 @@ gen_intermediate_code_internal(CPUSH4State * env, TranslationBlock * tb, ctx.pc = pc_start; ctx.flags = (uint32_t)tb->flags; ctx.bstate = BS_NONE; - ctx.sr = env->sr; - ctx.fpscr = env->fpscr; - ctx.memidx = (env->sr & SR_MD) == 0 ? 1 : 0; + ctx.memidx = (ctx.flags & SR_MD) == 0 ? 1 : 0; /* We don't know if the delayed pc came from a dynamic or static branch, so assume it is a dynamic branch. */ ctx.delayed_pc = -1; /* use delayed pc from env pointer */ ctx.tb = tb; ctx.singlestep_enabled = env->singlestep_enabled; ctx.features = env->features; - ctx.has_movcal = (tb->flags & TB_FLAG_PENDING_MOVCA); + ctx.has_movcal = (ctx.flags & TB_FLAG_PENDING_MOVCA); ii = -1; num_insns = 0; @@ -1947,7 +1993,7 @@ gen_intermediate_code_internal(CPUSH4State * env, TranslationBlock * tb, /* We have hit a breakpoint - make sure PC is up-to-date */ tcg_gen_movi_i32(cpu_pc, ctx.pc); gen_helper_debug(cpu_env); - ctx.bstate = BS_EXCP; + ctx.bstate = BS_BRANCH; break; } } @@ -2022,9 +2068,6 @@ gen_intermediate_code_internal(CPUSH4State * env, TranslationBlock * tb, } #ifdef DEBUG_DISAS -#ifdef SH4_DEBUG_DISAS - qemu_log_mask(CPU_LOG_TB_IN_ASM, "\n"); -#endif if (qemu_loglevel_mask(CPU_LOG_TB_IN_ASM)) { qemu_log("IN:\n"); /* , lookup_symbol(pc_start)); */ log_target_disas(pc_start, ctx.pc - pc_start, 0); diff --git a/target-xtensa/cpu.h b/target-xtensa/cpu.h index 177094ae9a..7348277edc 100644 --- a/target-xtensa/cpu.h +++ b/target-xtensa/cpu.h @@ -36,6 +36,7 @@ #include "config.h" #include "qemu-common.h" #include "cpu-defs.h" +#include "fpu/softfloat.h" #define TARGET_HAS_ICE 1 @@ -325,6 +326,8 @@ typedef struct CPUXtensaState { uint32_t sregs[256]; uint32_t uregs[256]; uint32_t phys_regs[MAX_NAREG]; + float32 fregs[16]; + float_status fp_status; xtensa_tlb_entry itlb[7][MAX_TLB_WAY_SIZE]; xtensa_tlb_entry dtlb[10][MAX_TLB_WAY_SIZE]; @@ -465,6 +468,8 @@ static inline int cpu_mmu_index(CPUXtensaState *env) #define XTENSA_TBFLAG_LITBASE 0x8 #define XTENSA_TBFLAG_DEBUG 0x10 #define XTENSA_TBFLAG_ICOUNT 0x20 +#define XTENSA_TBFLAG_CPENABLE_MASK 0x3fc0 +#define XTENSA_TBFLAG_CPENABLE_SHIFT 6 static inline void cpu_get_tb_cpu_state(CPUXtensaState *env, target_ulong *pc, target_ulong *cs_base, int *flags) @@ -488,6 +493,9 @@ static inline void cpu_get_tb_cpu_state(CPUXtensaState *env, target_ulong *pc, *flags |= XTENSA_TBFLAG_ICOUNT; } } + if (xtensa_option_enabled(env->config, XTENSA_OPTION_COPROCESSOR)) { + *flags |= env->sregs[CPENABLE] << XTENSA_TBFLAG_CPENABLE_SHIFT; + } } #include "cpu-all.h" diff --git a/target-xtensa/helper.h b/target-xtensa/helper.h index 152fec044d..4cc0088c36 100644 --- a/target-xtensa/helper.h +++ b/target-xtensa/helper.h @@ -36,4 +36,25 @@ DEF_HELPER_3(wsr_ibreaka, void, env, i32, i32) DEF_HELPER_3(wsr_dbreaka, void, env, i32, i32) DEF_HELPER_3(wsr_dbreakc, void, env, i32, i32) +DEF_HELPER_2(wur_fcr, void, env, i32) +DEF_HELPER_FLAGS_1(abs_s, TCG_CALL_CONST | TCG_CALL_PURE, f32, f32) +DEF_HELPER_FLAGS_1(neg_s, TCG_CALL_CONST | TCG_CALL_PURE, f32, f32) +DEF_HELPER_3(add_s, f32, env, f32, f32) +DEF_HELPER_3(sub_s, f32, env, f32, f32) +DEF_HELPER_3(mul_s, f32, env, f32, f32) +DEF_HELPER_4(madd_s, f32, env, f32, f32, f32) +DEF_HELPER_4(msub_s, f32, env, f32, f32, f32) +DEF_HELPER_FLAGS_3(ftoi, TCG_CALL_CONST | TCG_CALL_PURE, i32, f32, i32, i32) +DEF_HELPER_FLAGS_3(ftoui, TCG_CALL_CONST | TCG_CALL_PURE, i32, f32, i32, i32) +DEF_HELPER_3(itof, f32, env, i32, i32) +DEF_HELPER_3(uitof, f32, env, i32, i32) + +DEF_HELPER_4(un_s, void, env, i32, f32, f32) +DEF_HELPER_4(oeq_s, void, env, i32, f32, f32) +DEF_HELPER_4(ueq_s, void, env, i32, f32, f32) +DEF_HELPER_4(olt_s, void, env, i32, f32, f32) +DEF_HELPER_4(ult_s, void, env, i32, f32, f32) +DEF_HELPER_4(ole_s, void, env, i32, f32, f32) +DEF_HELPER_4(ule_s, void, env, i32, f32, f32) + #include "def-helper.h" diff --git a/target-xtensa/op_helper.c b/target-xtensa/op_helper.c index 2659c0e00f..ae0c09977b 100644 --- a/target-xtensa/op_helper.c +++ b/target-xtensa/op_helper.c @@ -771,3 +771,137 @@ void HELPER(wsr_dbreakc)(CPUXtensaState *env, uint32_t i, uint32_t v) } env->sregs[DBREAKC + i] = v; } + +void HELPER(wur_fcr)(CPUXtensaState *env, uint32_t v) +{ + static const int rounding_mode[] = { + float_round_nearest_even, + float_round_to_zero, + float_round_up, + float_round_down, + }; + + env->uregs[FCR] = v & 0xfffff07f; + set_float_rounding_mode(rounding_mode[v & 3], &env->fp_status); +} + +float32 HELPER(abs_s)(float32 v) +{ + return float32_abs(v); +} + +float32 HELPER(neg_s)(float32 v) +{ + return float32_chs(v); +} + +float32 HELPER(add_s)(CPUXtensaState *env, float32 a, float32 b) +{ + return float32_add(a, b, &env->fp_status); +} + +float32 HELPER(sub_s)(CPUXtensaState *env, float32 a, float32 b) +{ + return float32_sub(a, b, &env->fp_status); +} + +float32 HELPER(mul_s)(CPUXtensaState *env, float32 a, float32 b) +{ + return float32_mul(a, b, &env->fp_status); +} + +float32 HELPER(madd_s)(CPUXtensaState *env, float32 a, float32 b, float32 c) +{ + return float32_muladd(b, c, a, 0, + &env->fp_status); +} + +float32 HELPER(msub_s)(CPUXtensaState *env, float32 a, float32 b, float32 c) +{ + return float32_muladd(b, c, a, float_muladd_negate_product, + &env->fp_status); +} + +uint32_t HELPER(ftoi)(float32 v, uint32_t rounding_mode, uint32_t scale) +{ + float_status fp_status = {0}; + + set_float_rounding_mode(rounding_mode, &fp_status); + return float32_to_int32( + float32_scalbn(v, scale, &fp_status), &fp_status); +} + +uint32_t HELPER(ftoui)(float32 v, uint32_t rounding_mode, uint32_t scale) +{ + float_status fp_status = {0}; + float32 res; + + set_float_rounding_mode(rounding_mode, &fp_status); + + res = float32_scalbn(v, scale, &fp_status); + + if (float32_is_neg(v) && !float32_is_any_nan(v)) { + return float32_to_int32(res, &fp_status); + } else { + return float32_to_uint32(res, &fp_status); + } +} + +float32 HELPER(itof)(CPUXtensaState *env, uint32_t v, uint32_t scale) +{ + return float32_scalbn(int32_to_float32(v, &env->fp_status), + (int32_t)scale, &env->fp_status); +} + +float32 HELPER(uitof)(CPUXtensaState *env, uint32_t v, uint32_t scale) +{ + return float32_scalbn(uint32_to_float32(v, &env->fp_status), + (int32_t)scale, &env->fp_status); +} + +static inline void set_br(CPUXtensaState *env, bool v, uint32_t br) +{ + if (v) { + env->sregs[BR] |= br; + } else { + env->sregs[BR] &= ~br; + } +} + +void HELPER(un_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + set_br(env, float32_unordered_quiet(a, b, &env->fp_status), br); +} + +void HELPER(oeq_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + set_br(env, float32_eq_quiet(a, b, &env->fp_status), br); +} + +void HELPER(ueq_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + int v = float32_compare_quiet(a, b, &env->fp_status); + set_br(env, v == float_relation_equal || v == float_relation_unordered, br); +} + +void HELPER(olt_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + set_br(env, float32_lt_quiet(a, b, &env->fp_status), br); +} + +void HELPER(ult_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + int v = float32_compare_quiet(a, b, &env->fp_status); + set_br(env, v == float_relation_less || v == float_relation_unordered, br); +} + +void HELPER(ole_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + set_br(env, float32_le_quiet(a, b, &env->fp_status), br); +} + +void HELPER(ule_s)(CPUXtensaState *env, uint32_t br, float32 a, float32 b) +{ + int v = float32_compare_quiet(a, b, &env->fp_status); + set_br(env, v != float_relation_greater, br); +} diff --git a/target-xtensa/overlay_tool.h b/target-xtensa/overlay_tool.h index a3a5650fb0..e39505316b 100644 --- a/target-xtensa/overlay_tool.h +++ b/target-xtensa/overlay_tool.h @@ -58,6 +58,7 @@ XCHAL_OPTION(XCHAL_HAVE_SEXT, XTENSA_OPTION_MISC_OP_SEXT) | \ XCHAL_OPTION(XCHAL_HAVE_CLAMPS, XTENSA_OPTION_MISC_OP_CLAMPS) | \ XCHAL_OPTION(XCHAL_HAVE_CP, XTENSA_OPTION_COPROCESSOR) | \ + XCHAL_OPTION(XCHAL_HAVE_BOOLEANS, XTENSA_OPTION_BOOLEAN) | \ XCHAL_OPTION(XCHAL_HAVE_FP, XTENSA_OPTION_FP_COPROCESSOR) | \ XCHAL_OPTION(XCHAL_HAVE_RELEASE_SYNC, XTENSA_OPTION_MP_SYNCHRO) | \ XCHAL_OPTION(XCHAL_HAVE_S32C1I, XTENSA_OPTION_CONDITIONAL_STORE) | \ diff --git a/target-xtensa/translate.c b/target-xtensa/translate.c index b6643eb818..ba3ffcb7a4 100644 --- a/target-xtensa/translate.c +++ b/target-xtensa/translate.c @@ -65,11 +65,14 @@ typedef struct DisasContext { bool debug; bool icount; TCGv_i32 next_icount; + + unsigned cpenable; } DisasContext; static TCGv_ptr cpu_env; static TCGv_i32 cpu_pc; static TCGv_i32 cpu_R[16]; +static TCGv_i32 cpu_FR[16]; static TCGv_i32 cpu_SR[256]; static TCGv_i32 cpu_UR[256]; @@ -155,6 +158,12 @@ void xtensa_translate_init(void) "ar8", "ar9", "ar10", "ar11", "ar12", "ar13", "ar14", "ar15", }; + static const char * const fregnames[] = { + "f0", "f1", "f2", "f3", + "f4", "f5", "f6", "f7", + "f8", "f9", "f10", "f11", + "f12", "f13", "f14", "f15", + }; int i; cpu_env = tcg_global_reg_new_ptr(TCG_AREG0, "env"); @@ -167,6 +176,12 @@ void xtensa_translate_init(void) regnames[i]); } + for (i = 0; i < 16; i++) { + cpu_FR[i] = tcg_global_mem_new_i32(TCG_AREG0, + offsetof(CPUXtensaState, fregs[i]), + fregnames[i]); + } + for (i = 0; i < 256; ++i) { if (sregnames[i]) { cpu_SR[i] = tcg_global_mem_new_i32(TCG_AREG0, @@ -318,6 +333,15 @@ static void gen_check_privilege(DisasContext *dc) } } +static void gen_check_cpenable(DisasContext *dc, unsigned cp) +{ + if (option_enabled(dc, XTENSA_OPTION_COPROCESSOR) && + !(dc->cpenable & (1 << cp))) { + gen_exception_cause(dc, COPROCESSOR0_DISABLED + cp); + dc->is_jmp = DISAS_UPDATE; + } +} + static void gen_jump_slot(DisasContext *dc, TCGv dest, int slot) { tcg_gen_mov_i32(cpu_pc, dest); @@ -566,6 +590,13 @@ static void gen_wsr_dbreakc(DisasContext *dc, uint32_t sr, TCGv_i32 v) } } +static void gen_wsr_cpenable(DisasContext *dc, uint32_t sr, TCGv_i32 v) +{ + tcg_gen_andi_i32(cpu_SR[sr], v, 0xff); + /* This can change tb->flags, so exit tb */ + gen_jumpi_check_loop_end(dc, -1); +} + static void gen_wsr_intset(DisasContext *dc, uint32_t sr, TCGv_i32 v) { tcg_gen_andi_i32(cpu_SR[sr], v, @@ -668,6 +699,7 @@ static void gen_wsr(DisasContext *dc, uint32_t sr, TCGv_i32 s) [DBREAKA + 1] = gen_wsr_dbreaka, [DBREAKC] = gen_wsr_dbreakc, [DBREAKC + 1] = gen_wsr_dbreakc, + [CPENABLE] = gen_wsr_cpenable, [INTSET] = gen_wsr_intset, [INTCLEAR] = gen_wsr_intclear, [INTENABLE] = gen_wsr_intenable, @@ -692,6 +724,23 @@ static void gen_wsr(DisasContext *dc, uint32_t sr, TCGv_i32 s) } } +static void gen_wur(uint32_t ur, TCGv_i32 s) +{ + switch (ur) { + case FCR: + gen_helper_wur_fcr(cpu_env, s); + break; + + case FSR: + tcg_gen_andi_i32(cpu_UR[ur], s, 0xffffff80); + break; + + default: + tcg_gen_mov_i32(cpu_UR[ur], s); + break; + } +} + static void gen_load_store_alignment(DisasContext *dc, int shift, TCGv_i32 addr, bool no_hw_alignment) { @@ -1761,13 +1810,11 @@ static void disas_xtensa_insn(DisasContext *dc) case 15: /*WUR*/ gen_window_check1(dc, RRR_T); - { - if (uregnames[RSR_SR]) { - tcg_gen_mov_i32(cpu_UR[RSR_SR], cpu_R[RRR_T]); - } else { - qemu_log("WUR %d not implemented, ", RSR_SR); - TBD(); - } + if (uregnames[RSR_SR]) { + gen_wur(RSR_SR, cpu_R[RRR_T]); + } else { + qemu_log("WUR %d not implemented, ", RSR_SR); + TBD(); } break; @@ -1815,8 +1862,34 @@ static void disas_xtensa_insn(DisasContext *dc) break; case 8: /*LSCXp*/ - HAS_OPTION(XTENSA_OPTION_COPROCESSOR); - TBD(); + switch (OP2) { + case 0: /*LSXf*/ + case 1: /*LSXUf*/ + case 4: /*SSXf*/ + case 5: /*SSXUf*/ + HAS_OPTION(XTENSA_OPTION_FP_COPROCESSOR); + gen_window_check2(dc, RRR_S, RRR_T); + gen_check_cpenable(dc, 0); + { + TCGv_i32 addr = tcg_temp_new_i32(); + tcg_gen_add_i32(addr, cpu_R[RRR_S], cpu_R[RRR_T]); + gen_load_store_alignment(dc, 2, addr, false); + if (OP2 & 0x4) { + tcg_gen_qemu_st32(cpu_FR[RRR_R], addr, dc->cring); + } else { + tcg_gen_qemu_ld32u(cpu_FR[RRR_R], addr, dc->cring); + } + if (OP2 & 0x1) { + tcg_gen_mov_i32(cpu_R[RRR_S], addr); + } + tcg_temp_free(addr); + } + break; + + default: /*reserved*/ + RESERVED(); + break; + } break; case 9: /*LSC4*/ @@ -1854,12 +1927,213 @@ static void disas_xtensa_insn(DisasContext *dc) case 10: /*FP0*/ HAS_OPTION(XTENSA_OPTION_FP_COPROCESSOR); - TBD(); + switch (OP2) { + case 0: /*ADD.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_add_s(cpu_FR[RRR_R], cpu_env, + cpu_FR[RRR_S], cpu_FR[RRR_T]); + break; + + case 1: /*SUB.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_sub_s(cpu_FR[RRR_R], cpu_env, + cpu_FR[RRR_S], cpu_FR[RRR_T]); + break; + + case 2: /*MUL.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_mul_s(cpu_FR[RRR_R], cpu_env, + cpu_FR[RRR_S], cpu_FR[RRR_T]); + break; + + case 4: /*MADD.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_madd_s(cpu_FR[RRR_R], cpu_env, + cpu_FR[RRR_R], cpu_FR[RRR_S], cpu_FR[RRR_T]); + break; + + case 5: /*MSUB.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_msub_s(cpu_FR[RRR_R], cpu_env, + cpu_FR[RRR_R], cpu_FR[RRR_S], cpu_FR[RRR_T]); + break; + + case 8: /*ROUND.Sf*/ + case 9: /*TRUNC.Sf*/ + case 10: /*FLOOR.Sf*/ + case 11: /*CEIL.Sf*/ + case 14: /*UTRUNC.Sf*/ + gen_window_check1(dc, RRR_R); + gen_check_cpenable(dc, 0); + { + static const unsigned rounding_mode_const[] = { + float_round_nearest_even, + float_round_to_zero, + float_round_down, + float_round_up, + [6] = float_round_to_zero, + }; + TCGv_i32 rounding_mode = tcg_const_i32( + rounding_mode_const[OP2 & 7]); + TCGv_i32 scale = tcg_const_i32(RRR_T); + + if (OP2 == 14) { + gen_helper_ftoui(cpu_R[RRR_R], cpu_FR[RRR_S], + rounding_mode, scale); + } else { + gen_helper_ftoi(cpu_R[RRR_R], cpu_FR[RRR_S], + rounding_mode, scale); + } + + tcg_temp_free(rounding_mode); + tcg_temp_free(scale); + } + break; + + case 12: /*FLOAT.Sf*/ + case 13: /*UFLOAT.Sf*/ + gen_window_check1(dc, RRR_S); + gen_check_cpenable(dc, 0); + { + TCGv_i32 scale = tcg_const_i32(-RRR_T); + + if (OP2 == 13) { + gen_helper_uitof(cpu_FR[RRR_R], cpu_env, + cpu_R[RRR_S], scale); + } else { + gen_helper_itof(cpu_FR[RRR_R], cpu_env, + cpu_R[RRR_S], scale); + } + tcg_temp_free(scale); + } + break; + + case 15: /*FP1OP*/ + switch (RRR_T) { + case 0: /*MOV.Sf*/ + gen_check_cpenable(dc, 0); + tcg_gen_mov_i32(cpu_FR[RRR_R], cpu_FR[RRR_S]); + break; + + case 1: /*ABS.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_abs_s(cpu_FR[RRR_R], cpu_FR[RRR_S]); + break; + + case 4: /*RFRf*/ + gen_window_check1(dc, RRR_R); + gen_check_cpenable(dc, 0); + tcg_gen_mov_i32(cpu_R[RRR_R], cpu_FR[RRR_S]); + break; + + case 5: /*WFRf*/ + gen_window_check1(dc, RRR_S); + gen_check_cpenable(dc, 0); + tcg_gen_mov_i32(cpu_FR[RRR_R], cpu_R[RRR_S]); + break; + + case 6: /*NEG.Sf*/ + gen_check_cpenable(dc, 0); + gen_helper_neg_s(cpu_FR[RRR_R], cpu_FR[RRR_S]); + break; + + default: /*reserved*/ + RESERVED(); + break; + } + break; + + default: /*reserved*/ + RESERVED(); + break; + } break; case 11: /*FP1*/ HAS_OPTION(XTENSA_OPTION_FP_COPROCESSOR); - TBD(); + +#define gen_compare(rel, br, a, b) \ + do { \ + TCGv_i32 bit = tcg_const_i32(1 << br); \ + \ + gen_check_cpenable(dc, 0); \ + gen_helper_##rel(cpu_env, bit, cpu_FR[a], cpu_FR[b]); \ + tcg_temp_free(bit); \ + } while (0) + + switch (OP2) { + case 1: /*UN.Sf*/ + gen_compare(un_s, RRR_R, RRR_S, RRR_T); + break; + + case 2: /*OEQ.Sf*/ + gen_compare(oeq_s, RRR_R, RRR_S, RRR_T); + break; + + case 3: /*UEQ.Sf*/ + gen_compare(ueq_s, RRR_R, RRR_S, RRR_T); + break; + + case 4: /*OLT.Sf*/ + gen_compare(olt_s, RRR_R, RRR_S, RRR_T); + break; + + case 5: /*ULT.Sf*/ + gen_compare(ult_s, RRR_R, RRR_S, RRR_T); + break; + + case 6: /*OLE.Sf*/ + gen_compare(ole_s, RRR_R, RRR_S, RRR_T); + break; + + case 7: /*ULE.Sf*/ + gen_compare(ule_s, RRR_R, RRR_S, RRR_T); + break; + +#undef gen_compare + + case 8: /*MOVEQZ.Sf*/ + case 9: /*MOVNEZ.Sf*/ + case 10: /*MOVLTZ.Sf*/ + case 11: /*MOVGEZ.Sf*/ + gen_window_check1(dc, RRR_T); + gen_check_cpenable(dc, 0); + { + static const TCGCond cond[] = { + TCG_COND_NE, + TCG_COND_EQ, + TCG_COND_GE, + TCG_COND_LT + }; + int label = gen_new_label(); + tcg_gen_brcondi_i32(cond[OP2 - 8], cpu_R[RRR_T], 0, label); + tcg_gen_mov_i32(cpu_FR[RRR_R], cpu_FR[RRR_S]); + gen_set_label(label); + } + break; + + case 12: /*MOVF.Sf*/ + case 13: /*MOVT.Sf*/ + HAS_OPTION(XTENSA_OPTION_BOOLEAN); + gen_check_cpenable(dc, 0); + { + int label = gen_new_label(); + TCGv_i32 tmp = tcg_temp_new_i32(); + + tcg_gen_andi_i32(tmp, cpu_SR[BR], 1 << RRR_T); + tcg_gen_brcondi_i32( + OP2 & 1 ? TCG_COND_EQ : TCG_COND_NE, + tmp, 0, label); + tcg_gen_mov_i32(cpu_FR[RRR_R], cpu_FR[RRR_S]); + gen_set_label(label); + tcg_temp_free(tmp); + } + break; + + default: /*reserved*/ + RESERVED(); + break; + } break; default: /*reserved*/ @@ -2090,8 +2364,34 @@ static void disas_xtensa_insn(DisasContext *dc) break; case 3: /*LSCIp*/ - HAS_OPTION(XTENSA_OPTION_COPROCESSOR); - TBD(); + switch (RRI8_R) { + case 0: /*LSIf*/ + case 4: /*SSIf*/ + case 8: /*LSIUf*/ + case 12: /*SSIUf*/ + HAS_OPTION(XTENSA_OPTION_FP_COPROCESSOR); + gen_window_check1(dc, RRI8_S); + gen_check_cpenable(dc, 0); + { + TCGv_i32 addr = tcg_temp_new_i32(); + tcg_gen_addi_i32(addr, cpu_R[RRI8_S], RRI8_IMM8 << 2); + gen_load_store_alignment(dc, 2, addr, false); + if (RRI8_R & 0x4) { + tcg_gen_qemu_st32(cpu_FR[RRI8_T], addr, dc->cring); + } else { + tcg_gen_qemu_ld32u(cpu_FR[RRI8_T], addr, dc->cring); + } + if (RRI8_R & 0x8) { + tcg_gen_mov_i32(cpu_R[RRI8_S], addr); + } + tcg_temp_free(addr); + } + break; + + default: /*reserved*/ + RESERVED(); + break; + } break; case 4: /*MAC16d*/ @@ -2589,6 +2889,8 @@ static void gen_intermediate_code_internal( dc.ccount_delta = 0; dc.debug = tb->flags & XTENSA_TBFLAG_DEBUG; dc.icount = tb->flags & XTENSA_TBFLAG_ICOUNT; + dc.cpenable = (tb->flags & XTENSA_TBFLAG_CPENABLE_MASK) >> + XTENSA_TBFLAG_CPENABLE_SHIFT; init_litbase(&dc); init_sar_tracker(&dc); @@ -2730,6 +3032,16 @@ void cpu_dump_state(CPUXtensaState *env, FILE *f, fprintf_function cpu_fprintf, cpu_fprintf(f, "AR%02d=%08x%c", i, env->phys_regs[i], (i % 4) == 3 ? '\n' : ' '); } + + if (xtensa_option_enabled(env->config, XTENSA_OPTION_FP_COPROCESSOR)) { + cpu_fprintf(f, "\n"); + + for (i = 0; i < 16; ++i) { + cpu_fprintf(f, "F%02d=%08x (%+10.8e)%c", i, + float32_val(env->fregs[i]), + *(float *)&env->fregs[i], (i % 2) == 1 ? '\n' : ' '); + } + } } void restore_state_to_opc(CPUXtensaState *env, TranslationBlock *tb, int pc_pos) diff --git a/tcg/README b/tcg/README index cfdfd96d09..33783ee17b 100644 --- a/tcg/README +++ b/tcg/README @@ -307,6 +307,12 @@ dest = (t1 cond t2) Set DEST to 1 if (T1 cond T2) is true, otherwise set to 0. +* movcond_i32/i64 cond, dest, c1, c2, v1, v2 + +dest = (c1 cond c2 ? v1 : v2) + +Set DEST to V1 if (C1 cond C2) is true, otherwise set to V2. + ********* Type conversions * ext_i32_i64 t0, t1 @@ -386,7 +392,8 @@ Exit the current TB and return the value t0 (word type). Exit the current TB and jump to the TB index 'index' (constant) if the current TB was linked to this TB. Otherwise execute the next -instructions. +instructions. Only indices 0 and 1 are valid and tcg_gen_goto_tb may be issued +at most once with each slot index per TB. * qemu_ld8u t0, t1, flags qemu_ld8s t0, t1, flags diff --git a/tcg/arm/tcg-target.c b/tcg/arm/tcg-target.c index aed3b53247..2bad0a2b17 100644 --- a/tcg/arm/tcg-target.c +++ b/tcg/arm/tcg-target.c @@ -145,12 +145,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - return 4; -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { diff --git a/tcg/arm/tcg-target.h b/tcg/arm/tcg-target.h index c0b8f7274d..e2299cadd3 100644 --- a/tcg/arm/tcg-target.h +++ b/tcg/arm/tcg-target.h @@ -73,6 +73,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 0 +#define TCG_TARGET_HAS_movcond_i32 0 #define TCG_TARGET_HAS_GUEST_BASE diff --git a/tcg/hppa/tcg-target.c b/tcg/hppa/tcg-target.c index 8b81b70f17..2c79c1081e 100644 --- a/tcg/hppa/tcg-target.c +++ b/tcg/hppa/tcg-target.c @@ -175,12 +175,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, *insn_ptr = insn; } -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - return 4; -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { @@ -820,19 +814,34 @@ static void tcg_out_comclr(TCGContext *s, int cond, TCGArg ret, tcg_out32(s, op); } +static TCGCond const tcg_high_cond[] = { + [TCG_COND_EQ] = TCG_COND_EQ, + [TCG_COND_NE] = TCG_COND_NE, + [TCG_COND_LT] = TCG_COND_LT, + [TCG_COND_LE] = TCG_COND_LT, + [TCG_COND_GT] = TCG_COND_GT, + [TCG_COND_GE] = TCG_COND_GT, + [TCG_COND_LTU] = TCG_COND_LTU, + [TCG_COND_LEU] = TCG_COND_LTU, + [TCG_COND_GTU] = TCG_COND_GTU, + [TCG_COND_GEU] = TCG_COND_GTU +}; + static void tcg_out_brcond2(TCGContext *s, int cond, TCGArg al, TCGArg ah, TCGArg bl, int blconst, TCGArg bh, int bhconst, int label_index) { switch (cond) { case TCG_COND_EQ: + tcg_out_comclr(s, TCG_COND_NE, TCG_REG_R0, al, bl, blconst); + tcg_out_brcond(s, TCG_COND_EQ, ah, bh, bhconst, label_index); + break; case TCG_COND_NE: - tcg_out_comclr(s, tcg_invert_cond(cond), TCG_REG_R0, al, bl, blconst); - tcg_out_brcond(s, cond, ah, bh, bhconst, label_index); + tcg_out_brcond(s, TCG_COND_NE, al, bl, bhconst, label_index); + tcg_out_brcond(s, TCG_COND_NE, ah, bh, bhconst, label_index); break; - default: - tcg_out_brcond(s, cond, ah, bh, bhconst, label_index); + tcg_out_brcond(s, tcg_high_cond[cond], ah, bh, bhconst, label_index); tcg_out_comclr(s, TCG_COND_NE, TCG_REG_R0, ah, bh, bhconst); tcg_out_brcond(s, tcg_unsigned_cond(cond), al, bl, blconst, label_index); @@ -853,9 +862,8 @@ static void tcg_out_setcond2(TCGContext *s, int cond, TCGArg ret, { int scratch = TCG_REG_R20; - if (ret != al && ret != ah - && (blconst || ret != bl) - && (bhconst || ret != bh)) { + /* Note that the low parts are fully consumed before scratch is set. */ + if (ret != ah && (bhconst || ret != bh)) { scratch = ret; } @@ -867,18 +875,49 @@ static void tcg_out_setcond2(TCGContext *s, int cond, TCGArg ret, tcg_out_movi(s, TCG_TYPE_I32, scratch, cond == TCG_COND_NE); break; - default: + case TCG_COND_GE: + case TCG_COND_GEU: + case TCG_COND_LT: + case TCG_COND_LTU: + /* Optimize compares with low part zero. */ + if (bl == 0) { + tcg_out_setcond(s, cond, ret, ah, bh, bhconst); + return; + } + /* FALLTHRU */ + + case TCG_COND_LE: + case TCG_COND_LEU: + case TCG_COND_GT: + case TCG_COND_GTU: + /* <= : ah < bh | (ah == bh && al <= bl) */ tcg_out_setcond(s, tcg_unsigned_cond(cond), scratch, al, bl, blconst); tcg_out_comclr(s, TCG_COND_EQ, TCG_REG_R0, ah, bh, bhconst); tcg_out_movi(s, TCG_TYPE_I32, scratch, 0); - tcg_out_comclr(s, cond, TCG_REG_R0, ah, bh, bhconst); + tcg_out_comclr(s, tcg_invert_cond(tcg_high_cond[cond]), + TCG_REG_R0, ah, bh, bhconst); tcg_out_movi(s, TCG_TYPE_I32, scratch, 1); break; + + default: + tcg_abort(); } tcg_out_mov(s, TCG_TYPE_I32, ret, scratch); } +static void tcg_out_movcond(TCGContext *s, int cond, TCGArg ret, + TCGArg c1, TCGArg c2, int c2const, + TCGArg v1, int v1const) +{ + tcg_out_comclr(s, tcg_invert_cond(cond), TCG_REG_R0, c1, c2, c2const); + if (v1const) { + tcg_out_movi(s, TCG_TYPE_I32, ret, v1); + } else { + tcg_out_mov(s, TCG_TYPE_I32, ret, v1); + } +} + #if defined(CONFIG_SOFTMMU) #include "../../softmmu_defs.h" @@ -943,10 +982,11 @@ static int tcg_out_tlb_read(TCGContext *s, int r0, int r1, int addrlo, tcg_out_ld(s, TCG_TYPE_PTR, TCG_REG_R20, r1, offset); } - /* Compute the value that ought to appear in the TLB for a hit, namely, the page - of the address. We include the low N bits of the address to catch unaligned - accesses and force them onto the slow path. Do this computation after having - issued the load from the TLB slot to give the load time to complete. */ + /* Compute the value that ought to appear in the TLB for a hit, namely, + the page of the address. We include the low N bits of the address + to catch unaligned accesses and force them onto the slow path. Do + this computation after having issued the load from the TLB slot to + give the load time to complete. */ tcg_out_andi(s, r0, addrlo, TARGET_PAGE_MASK | ((1 << s_bits) - 1)); /* If not equal, jump to lab_miss. */ @@ -959,6 +999,36 @@ static int tcg_out_tlb_read(TCGContext *s, int r0, int r1, int addrlo, return ret; } + +static int tcg_out_arg_reg32(TCGContext *s, int argno, TCGArg v, bool vconst) +{ + if (argno < 4) { + if (vconst) { + tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[argno], v); + } else { + tcg_out_mov(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[argno], v); + } + } else { + if (vconst && v != 0) { + tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_R20, v); + v = TCG_REG_R20; + } + tcg_out_st(s, TCG_TYPE_I32, v, TCG_REG_CALL_STACK, + TCG_TARGET_CALL_STACK_OFFSET - ((argno - 3) * 4)); + } + return argno + 1; +} + +static int tcg_out_arg_reg64(TCGContext *s, int argno, TCGArg vl, TCGArg vh) +{ + /* 64-bit arguments must go in even reg pairs and stack slots. */ + if (argno & 1) { + argno++; + } + argno = tcg_out_arg_reg32(s, argno, vl, false); + argno = tcg_out_arg_reg32(s, argno, vh, false); + return argno; +} #endif static void tcg_out_qemu_ld_direct(TCGContext *s, int datalo_reg, int datahi_reg, @@ -1039,39 +1109,36 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, int opc) /* Note that addrhi_reg is only used for 64-bit guests. */ int addrhi_reg = (TARGET_LONG_BITS == 64 ? *args++ : TCG_REG_R0); int mem_index = *args; - int lab1, lab2, argreg, offset; + int lab1, lab2, argno, offset; lab1 = gen_new_label(); lab2 = gen_new_label(); offset = offsetof(CPUArchState, tlb_table[mem_index][0].addr_read); - offset = tcg_out_tlb_read(s, TCG_REG_R26, TCG_REG_R25, addrlo_reg, addrhi_reg, - opc & 3, lab1, offset); + offset = tcg_out_tlb_read(s, TCG_REG_R26, TCG_REG_R25, addrlo_reg, + addrhi_reg, opc & 3, lab1, offset); /* TLB Hit. */ - tcg_out_ld(s, TCG_TYPE_PTR, TCG_REG_R20, (offset ? TCG_REG_R1 : TCG_REG_R25), + tcg_out_ld(s, TCG_TYPE_PTR, TCG_REG_R20, + (offset ? TCG_REG_R1 : TCG_REG_R25), offsetof(CPUArchState, tlb_table[mem_index][0].addend) - offset); - tcg_out_qemu_ld_direct(s, datalo_reg, datahi_reg, addrlo_reg, TCG_REG_R20, opc); + tcg_out_qemu_ld_direct(s, datalo_reg, datahi_reg, addrlo_reg, + TCG_REG_R20, opc); tcg_out_branch(s, lab2, 1); /* TLB Miss. */ /* label1: */ tcg_out_label(s, lab1, s->code_ptr); - argreg = TCG_REG_R26; - tcg_out_mov(s, TCG_TYPE_I32, argreg--, addrlo_reg); + argno = 0; + argno = tcg_out_arg_reg32(s, argno, TCG_AREG0, false); if (TARGET_LONG_BITS == 64) { - tcg_out_mov(s, TCG_TYPE_I32, argreg--, addrhi_reg); + argno = tcg_out_arg_reg64(s, argno, addrlo_reg, addrhi_reg); + } else { + argno = tcg_out_arg_reg32(s, argno, addrlo_reg, false); } - tcg_out_movi(s, TCG_TYPE_I32, argreg, mem_index); - - /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_TL, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], - TCG_AREG0); + argno = tcg_out_arg_reg32(s, argno, mem_index, true); + tcg_out_call(s, qemu_ld_helpers[opc & 3]); switch (opc) { @@ -1107,8 +1174,8 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, int opc) #endif } -static void tcg_out_qemu_st_direct(TCGContext *s, int datalo_reg, int datahi_reg, - int addr_reg, int opc) +static void tcg_out_qemu_st_direct(TCGContext *s, int datalo_reg, + int datahi_reg, int addr_reg, int opc) { #ifdef TARGET_WORDS_BIGENDIAN const int bswap = 0; @@ -1161,17 +1228,18 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, int opc) /* Note that addrhi_reg is only used for 64-bit guests. */ int addrhi_reg = (TARGET_LONG_BITS == 64 ? *args++ : TCG_REG_R0); int mem_index = *args; - int lab1, lab2, argreg, offset; + int lab1, lab2, argno, next, offset; lab1 = gen_new_label(); lab2 = gen_new_label(); offset = offsetof(CPUArchState, tlb_table[mem_index][0].addr_write); - offset = tcg_out_tlb_read(s, TCG_REG_R26, TCG_REG_R25, addrlo_reg, addrhi_reg, - opc, lab1, offset); + offset = tcg_out_tlb_read(s, TCG_REG_R26, TCG_REG_R25, addrlo_reg, + addrhi_reg, opc, lab1, offset); /* TLB Hit. */ - tcg_out_ld(s, TCG_TYPE_PTR, TCG_REG_R20, (offset ? TCG_REG_R1 : TCG_REG_R25), + tcg_out_ld(s, TCG_TYPE_PTR, TCG_REG_R20, + (offset ? TCG_REG_R1 : TCG_REG_R25), offsetof(CPUArchState, tlb_table[mem_index][0].addend) - offset); /* There are no indexed stores, so we must do this addition explitly. @@ -1184,63 +1252,46 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, int opc) /* label1: */ tcg_out_label(s, lab1, s->code_ptr); - argreg = TCG_REG_R26; - tcg_out_mov(s, TCG_TYPE_I32, argreg--, addrlo_reg); + argno = 0; + argno = tcg_out_arg_reg32(s, argno, TCG_AREG0, false); if (TARGET_LONG_BITS == 64) { - tcg_out_mov(s, TCG_TYPE_I32, argreg--, addrhi_reg); + argno = tcg_out_arg_reg64(s, argno, addrlo_reg, addrhi_reg); + } else { + argno = tcg_out_arg_reg32(s, argno, addrlo_reg, false); } + next = (argno < 4 ? tcg_target_call_iarg_regs[argno] : TCG_REG_R20); switch(opc) { case 0: - tcg_out_andi(s, argreg--, datalo_reg, 0xff); - tcg_out_movi(s, TCG_TYPE_I32, argreg, mem_index); + tcg_out_andi(s, next, datalo_reg, 0xff); + argno = tcg_out_arg_reg32(s, argno, next, false); break; case 1: - tcg_out_andi(s, argreg--, datalo_reg, 0xffff); - tcg_out_movi(s, TCG_TYPE_I32, argreg, mem_index); + tcg_out_andi(s, next, datalo_reg, 0xffff); + argno = tcg_out_arg_reg32(s, argno, next, false); break; case 2: - tcg_out_mov(s, TCG_TYPE_I32, argreg--, datalo_reg); - tcg_out_movi(s, TCG_TYPE_I32, argreg, mem_index); + argno = tcg_out_arg_reg32(s, argno, datalo_reg, false); break; case 3: - /* Because of the alignment required by the 64-bit data argument, - we will always use R23/R24. Also, we will always run out of - argument registers for storing mem_index, so that will have - to go on the stack. */ - if (mem_index == 0) { - argreg = TCG_REG_R0; - } else { - argreg = TCG_REG_R20; - tcg_out_movi(s, TCG_TYPE_I32, argreg, mem_index); - } - tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R23, datahi_reg); - tcg_out_mov(s, TCG_TYPE_I32, TCG_REG_R24, datalo_reg); - tcg_out_st(s, TCG_TYPE_I32, argreg, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - 4); + argno = tcg_out_arg_reg64(s, argno, datalo_reg, datahi_reg); break; default: tcg_abort(); } + argno = tcg_out_arg_reg32(s, argno, mem_index, true); - /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[3], - tcg_target_call_iarg_regs[2]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_TL, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], - TCG_AREG0); tcg_out_call(s, qemu_st_helpers[opc]); /* label2: */ tcg_out_label(s, lab2, s->code_ptr); #else - /* There are no indexed stores, so if GUEST_BASE is set we must do the add - explicitly. Careful to avoid R20, which is used for the bswaps to follow. */ + /* There are no indexed stores, so if GUEST_BASE is set we must do + the add explicitly. Careful to avoid R20, which is used for the + bswaps to follow. */ if (GUEST_BASE != 0) { - tcg_out_arith(s, TCG_REG_R31, addrlo_reg, TCG_GUEST_BASE_REG, INSN_ADDL); + tcg_out_arith(s, TCG_REG_R31, addrlo_reg, + TCG_GUEST_BASE_REG, INSN_ADDL); addrlo_reg = TCG_REG_R31; } tcg_out_qemu_st_direct(s, datalo_reg, datahi_reg, addrlo_reg, opc); @@ -1475,6 +1526,11 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, args[3], const_args[3], args[4], const_args[4]); break; + case INDEX_op_movcond_i32: + tcg_out_movcond(s, args[5], args[0], args[1], args[2], const_args[2], + args[3], const_args[3]); + break; + case INDEX_op_add2_i32: tcg_out_add2(s, args[0], args[1], args[2], args[3], args[4], args[5], const_args[4]); @@ -1583,6 +1639,10 @@ static const TCGTargetOpDef hppa_op_defs[] = { { INDEX_op_setcond_i32, { "r", "rZ", "rI" } }, { INDEX_op_setcond2_i32, { "r", "rZ", "rZ", "rI", "rI" } }, + /* ??? We can actually support a signed 14-bit arg3, but we + only have existing constraints for a signed 11-bit. */ + { INDEX_op_movcond_i32, { "r", "rZ", "rI", "rI", "0" } }, + { INDEX_op_add2_i32, { "r", "r", "rZ", "rZ", "rI", "rZ" } }, { INDEX_op_sub2_i32, { "r", "r", "rI", "rZ", "rK", "rZ" } }, diff --git a/tcg/hppa/tcg-target.h b/tcg/hppa/tcg-target.h index 01ef9605f7..5351353719 100644 --- a/tcg/hppa/tcg-target.h +++ b/tcg/hppa/tcg-target.h @@ -96,6 +96,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 1 +#define TCG_TARGET_HAS_movcond_i32 1 /* optional instructions automatically implemented */ #define TCG_TARGET_HAS_neg_i32 0 /* sub rd, 0, rs */ diff --git a/tcg/i386/tcg-target.c b/tcg/i386/tcg-target.c index 3017858751..122d63630c 100644 --- a/tcg/i386/tcg-target.c +++ b/tcg/i386/tcg-target.c @@ -75,9 +75,7 @@ static const int tcg_target_call_iarg_regs[] = { TCG_REG_R8, TCG_REG_R9, #else - TCG_REG_EAX, - TCG_REG_EDX, - TCG_REG_ECX + /* 32 bit mode uses stack based calling convention (GCC default). */ #endif }; @@ -88,6 +86,18 @@ static const int tcg_target_call_oarg_regs[] = { #endif }; +/* Registers used with L constraint, which are the first argument + registers on x86_64, and two random call clobbered registers on + i386. */ +#if TCG_TARGET_REG_BITS == 64 +# define TCG_REG_L0 tcg_target_call_iarg_regs[0] +# define TCG_REG_L1 tcg_target_call_iarg_regs[1] +# define TCG_REG_L2 tcg_target_call_iarg_regs[2] +#else +# define TCG_REG_L0 TCG_REG_EAX +# define TCG_REG_L1 TCG_REG_EDX +#endif + static uint8_t *tb_ret_addr; static void patch_reloc(uint8_t *code_ptr, int type, @@ -114,16 +124,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - if (TCG_TARGET_REG_BITS == 64) { - return 6; - } - - return 0; -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { @@ -179,16 +179,16 @@ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) /* qemu_ld/st address constraint */ case 'L': ct->ct |= TCG_CT_REG; - if (TCG_TARGET_REG_BITS == 64) { +#if TCG_TARGET_REG_BITS == 64 tcg_regset_set32(ct->u.regs, 0, 0xffff); - tcg_regset_reset_reg(ct->u.regs, tcg_target_call_iarg_regs[0]); - tcg_regset_reset_reg(ct->u.regs, tcg_target_call_iarg_regs[1]); - tcg_regset_reset_reg(ct->u.regs, tcg_target_call_iarg_regs[2]); - } else { + tcg_regset_reset_reg(ct->u.regs, TCG_REG_L0); + tcg_regset_reset_reg(ct->u.regs, TCG_REG_L1); + tcg_regset_reset_reg(ct->u.regs, TCG_REG_L2); +#else tcg_regset_set32(ct->u.regs, 0, 0xff); - tcg_regset_reset_reg(ct->u.regs, TCG_REG_EAX); - tcg_regset_reset_reg(ct->u.regs, TCG_REG_EDX); - } + tcg_regset_reset_reg(ct->u.regs, TCG_REG_L0); + tcg_regset_reset_reg(ct->u.regs, TCG_REG_L1); +#endif break; case 'e': @@ -249,6 +249,7 @@ static inline int tcg_target_const_match(tcg_target_long val, #define OPC_ADD_GvEv (OPC_ARITH_GvEv | (ARITH_ADD << 3)) #define OPC_BSWAP (0xc8 | P_EXT) #define OPC_CALL_Jz (0xe8) +#define OPC_CMOVCC (0x40 | P_EXT) /* ... plus condition code */ #define OPC_CMP_GvEv (OPC_ARITH_GvEv | (ARITH_CMP << 3)) #define OPC_DEC_r32 (0x48) #define OPC_IMUL_GvEv (0xaf | P_EXT) @@ -936,6 +937,24 @@ static void tcg_out_setcond2(TCGContext *s, const TCGArg *args, } #endif +static void tcg_out_movcond32(TCGContext *s, TCGCond cond, TCGArg dest, + TCGArg c1, TCGArg c2, int const_c2, + TCGArg v1) +{ + tcg_out_cmp(s, c1, c2, const_c2, 0); + tcg_out_modrm(s, OPC_CMOVCC | tcg_cond_to_jcc[cond], dest, v1); +} + +#if TCG_TARGET_REG_BITS == 64 +static void tcg_out_movcond64(TCGContext *s, TCGCond cond, TCGArg dest, + TCGArg c1, TCGArg c2, int const_c2, + TCGArg v1) +{ + tcg_out_cmp(s, c1, c2, const_c2, P_REXW); + tcg_out_modrm(s, OPC_CMOVCC | tcg_cond_to_jcc[cond] | P_REXW, dest, v1); +} +#endif + static void tcg_out_branch(TCGContext *s, int call, tcg_target_long dest) { tcg_target_long disp = dest - (tcg_target_long)s->code_ptr - 5; @@ -1010,8 +1029,8 @@ static inline void tcg_out_tlb_load(TCGContext *s, int addrlo_idx, uint8_t **label_ptr, int which) { const int addrlo = args[addrlo_idx]; - const int r0 = tcg_target_call_iarg_regs[0]; - const int r1 = tcg_target_call_iarg_regs[1]; + const int r0 = TCG_REG_L0; + const int r1 = TCG_REG_L1; TCGType type = TCG_TYPE_I32; int rexw = 0; @@ -1173,8 +1192,7 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, label_ptr, offsetof(CPUTLBEntry, addr_read)); /* TLB Hit. */ - tcg_out_qemu_ld_direct(s, data_reg, data_reg2, - tcg_target_call_iarg_regs[0], 0, opc); + tcg_out_qemu_ld_direct(s, data_reg, data_reg2, TCG_REG_L0, 0, opc); /* jmp label2 */ tcg_out8(s, OPC_JMP_short); @@ -1207,14 +1225,10 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[arg_idx], mem_index); /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[3], - tcg_target_call_iarg_regs[2]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[0], - TCG_AREG0); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[3], TCG_REG_L2); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], TCG_REG_L1); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[1], TCG_REG_L0); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[0], TCG_AREG0); #endif tcg_out_calli(s, (tcg_target_long)qemu_ld_helpers[s_bits]); @@ -1280,11 +1294,9 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, use the ADDR32 prefix. For now, do nothing. */ if (offset != GUEST_BASE) { - tcg_out_movi(s, TCG_TYPE_I64, - tcg_target_call_iarg_regs[0], GUEST_BASE); - tgen_arithr(s, ARITH_ADD + P_REXW, - tcg_target_call_iarg_regs[0], base); - base = tcg_target_call_iarg_regs[0]; + tcg_out_movi(s, TCG_TYPE_I64, TCG_REG_L0, GUEST_BASE); + tgen_arithr(s, ARITH_ADD + P_REXW, TCG_REG_L0, base); + base = TCG_REG_L0; offset = 0; } } @@ -1305,8 +1317,8 @@ static void tcg_out_qemu_st_direct(TCGContext *s, int datalo, int datahi, /* ??? Ideally we wouldn't need a scratch register. For user-only, we could perform the bswap twice to restore the original value instead of moving to the scratch. But as it is, the L constraint - means that the second argument reg is definitely free here. */ - int scratch = tcg_target_call_iarg_regs[1]; + means that TCG_REG_L1 is definitely free here. */ + const int scratch = TCG_REG_L1; switch (sizeop) { case 0: @@ -1379,8 +1391,7 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, label_ptr, offsetof(CPUTLBEntry, addr_write)); /* TLB Hit. */ - tcg_out_qemu_st_direct(s, data_reg, data_reg2, - tcg_target_call_iarg_regs[0], 0, opc); + tcg_out_qemu_st_direct(s, data_reg, data_reg2, TCG_REG_L0, 0, opc); /* jmp label2 */ tcg_out8(s, OPC_JMP_short); @@ -1415,18 +1426,14 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, stack_adjust += 4; #else tcg_out_mov(s, (opc == 3 ? TCG_TYPE_I64 : TCG_TYPE_I32), - tcg_target_call_iarg_regs[1], data_reg); - tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[2], mem_index); + TCG_REG_L1, data_reg); + tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_L2, mem_index); stack_adjust = 0; /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[3], - tcg_target_call_iarg_regs[2]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[0], - TCG_AREG0); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[3], TCG_REG_L2); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], TCG_REG_L1); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[1], TCG_REG_L0); + tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[0], TCG_AREG0); #endif tcg_out_calli(s, (tcg_target_long)qemu_st_helpers[s_bits]); @@ -1453,11 +1460,9 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, use the ADDR32 prefix. For now, do nothing. */ if (offset != GUEST_BASE) { - tcg_out_movi(s, TCG_TYPE_I64, - tcg_target_call_iarg_regs[0], GUEST_BASE); - tgen_arithr(s, ARITH_ADD + P_REXW, - tcg_target_call_iarg_regs[0], base); - base = tcg_target_call_iarg_regs[0]; + tcg_out_movi(s, TCG_TYPE_I64, TCG_REG_L0, GUEST_BASE); + tgen_arithr(s, ARITH_ADD + P_REXW, TCG_REG_L0, base); + base = TCG_REG_L0; offset = 0; } } @@ -1668,6 +1673,10 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_setcond32(s, args[3], args[0], args[1], args[2], const_args[2]); break; + case INDEX_op_movcond_i32: + tcg_out_movcond32(s, args[5], args[0], args[1], + args[2], const_args[2], args[3]); + break; OP_32_64(bswap16): tcg_out_rolw_8(s, args[0]); @@ -1796,6 +1805,10 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_setcond64(s, args[3], args[0], args[1], args[2], const_args[2]); break; + case INDEX_op_movcond_i64: + tcg_out_movcond64(s, args[5], args[0], args[1], + args[2], const_args[2], args[3]); + break; case INDEX_op_bswap64_i64: tcg_out_bswap64(s, args[0]); @@ -1880,6 +1893,7 @@ static const TCGTargetOpDef x86_op_defs[] = { { INDEX_op_setcond_i32, { "q", "r", "ri" } }, { INDEX_op_deposit_i32, { "Q", "0", "Q" } }, + { INDEX_op_movcond_i32, { "r", "r", "ri", "r", "0" } }, #if TCG_TARGET_REG_BITS == 32 { INDEX_op_mulu2_i32, { "a", "d", "a", "r" } }, @@ -1934,6 +1948,7 @@ static const TCGTargetOpDef x86_op_defs[] = { { INDEX_op_ext32u_i64, { "r", "r" } }, { INDEX_op_deposit_i64, { "Q", "0", "Q" } }, + { INDEX_op_movcond_i64, { "r", "r", "re", "r", "0" } }, #endif #if TCG_TARGET_REG_BITS == 64 @@ -2032,15 +2047,17 @@ static void tcg_target_qemu_prologue(TCGContext *s) #if TCG_TARGET_REG_BITS == 32 tcg_out_ld(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_ESP, (ARRAY_SIZE(tcg_target_callee_save_regs) + 1) * 4); - tcg_out_ld(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[1], TCG_REG_ESP, - (ARRAY_SIZE(tcg_target_callee_save_regs) + 2) * 4); + tcg_out_addi(s, TCG_REG_ESP, -stack_addend); + /* jmp *tb. */ + tcg_out_modrm_offset(s, OPC_GRP5, EXT5_JMPN_Ev, TCG_REG_ESP, + (ARRAY_SIZE(tcg_target_callee_save_regs) + 2) * 4 + + stack_addend); #else tcg_out_mov(s, TCG_TYPE_PTR, TCG_AREG0, tcg_target_call_iarg_regs[0]); -#endif tcg_out_addi(s, TCG_REG_ESP, -stack_addend); - /* jmp *tb. */ tcg_out_modrm(s, OPC_GRP5, EXT5_JMPN_Ev, tcg_target_call_iarg_regs[1]); +#endif /* TB epilogue */ tb_ret_addr = s->code_ptr; diff --git a/tcg/i386/tcg-target.h b/tcg/i386/tcg-target.h index 8be42f3ccb..ace63ba37b 100644 --- a/tcg/i386/tcg-target.h +++ b/tcg/i386/tcg-target.h @@ -67,7 +67,11 @@ typedef enum { /* used for function call generation */ #define TCG_REG_CALL_STACK TCG_REG_ESP #define TCG_TARGET_STACK_ALIGN 16 +#if defined(_WIN64) +#define TCG_TARGET_CALL_STACK_OFFSET 32 +#else #define TCG_TARGET_CALL_STACK_OFFSET 0 +#endif /* optional instructions */ #define TCG_TARGET_HAS_div2_i32 1 @@ -86,6 +90,12 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 1 +#if defined(__x86_64__) || defined(__i686__) +/* Use cmov only if the compiler is already doing so. */ +#define TCG_TARGET_HAS_movcond_i32 1 +#else +#define TCG_TARGET_HAS_movcond_i32 0 +#endif #if TCG_TARGET_REG_BITS == 64 #define TCG_TARGET_HAS_div2_i64 1 @@ -107,6 +117,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 #define TCG_TARGET_HAS_deposit_i64 1 +#define TCG_TARGET_HAS_movcond_i64 1 #endif #define TCG_TARGET_deposit_i32_valid(ofs, len) \ diff --git a/tcg/ia64/tcg-target.c b/tcg/ia64/tcg-target.c index 1745038f14..dc9c12cf18 100644 --- a/tcg/ia64/tcg-target.c +++ b/tcg/ia64/tcg-target.c @@ -176,12 +176,6 @@ static const int tcg_target_call_oarg_regs[] = { TCG_REG_R8 }; -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - return 8; -} - /* * opcode formation */ diff --git a/tcg/ia64/tcg-target.h b/tcg/ia64/tcg-target.h index c22962ac12..368aee4196 100644 --- a/tcg/ia64/tcg-target.h +++ b/tcg/ia64/tcg-target.h @@ -133,6 +133,8 @@ typedef enum { #define TCG_TARGET_HAS_rot_i64 1 #define TCG_TARGET_HAS_deposit_i32 0 #define TCG_TARGET_HAS_deposit_i64 0 +#define TCG_TARGET_HAS_movcond_i32 0 +#define TCG_TARGET_HAS_movcond_i64 0 /* optional instructions automatically implemented */ #define TCG_TARGET_HAS_neg_i32 0 /* sub r1, r0, r3 */ diff --git a/tcg/mips/tcg-target.c b/tcg/mips/tcg-target.c index 74db83d440..f70910a90f 100644 --- a/tcg/mips/tcg-target.c +++ b/tcg/mips/tcg-target.c @@ -68,7 +68,7 @@ static const char * const tcg_target_reg_names[TCG_TARGET_NB_REGS] = { #endif /* check if we really need so many registers :P */ -static const int tcg_target_reg_alloc_order[] = { +static const TCGReg tcg_target_reg_alloc_order[] = { TCG_REG_S0, TCG_REG_S1, TCG_REG_S2, @@ -94,14 +94,14 @@ static const int tcg_target_reg_alloc_order[] = { TCG_REG_V1 }; -static const int tcg_target_call_iarg_regs[4] = { +static const TCGReg tcg_target_call_iarg_regs[4] = { TCG_REG_A0, TCG_REG_A1, TCG_REG_A2, TCG_REG_A3 }; -static const int tcg_target_call_oarg_regs[2] = { +static const TCGReg tcg_target_call_oarg_regs[2] = { TCG_REG_V0, TCG_REG_V1 }; @@ -185,12 +185,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - return 4; -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { @@ -278,6 +272,8 @@ static inline int tcg_target_const_match(tcg_target_long val, enum { OPC_BEQ = 0x04 << 26, OPC_BNE = 0x05 << 26, + OPC_BLEZ = 0x06 << 26, + OPC_BGTZ = 0x07 << 26, OPC_ADDIU = 0x09 << 26, OPC_SLTI = 0x0A << 26, OPC_SLTIU = 0x0B << 26, @@ -298,12 +294,16 @@ enum { OPC_SPECIAL = 0x00 << 26, OPC_SLL = OPC_SPECIAL | 0x00, OPC_SRL = OPC_SPECIAL | 0x02, + OPC_ROTR = OPC_SPECIAL | (0x01 << 21) | 0x02, OPC_SRA = OPC_SPECIAL | 0x03, OPC_SLLV = OPC_SPECIAL | 0x04, OPC_SRLV = OPC_SPECIAL | 0x06, + OPC_ROTRV = OPC_SPECIAL | (0x01 << 6) | 0x06, OPC_SRAV = OPC_SPECIAL | 0x07, OPC_JR = OPC_SPECIAL | 0x08, OPC_JALR = OPC_SPECIAL | 0x09, + OPC_MOVZ = OPC_SPECIAL | 0x0A, + OPC_MOVN = OPC_SPECIAL | 0x0B, OPC_MFHI = OPC_SPECIAL | 0x10, OPC_MFLO = OPC_SPECIAL | 0x12, OPC_MULT = OPC_SPECIAL | 0x18, @@ -319,7 +319,13 @@ enum { OPC_SLT = OPC_SPECIAL | 0x2A, OPC_SLTU = OPC_SPECIAL | 0x2B, + OPC_REGIMM = 0x01 << 26, + OPC_BLTZ = OPC_REGIMM | (0x00 << 16), + OPC_BGEZ = OPC_REGIMM | (0x01 << 16), + OPC_SPECIAL3 = 0x1f << 26, + OPC_INS = OPC_SPECIAL3 | 0x004, + OPC_WSBH = OPC_SPECIAL3 | 0x0a0, OPC_SEB = OPC_SPECIAL3 | 0x420, OPC_SEH = OPC_SPECIAL3 | 0x620, }; @@ -327,7 +333,8 @@ enum { /* * Type reg */ -static inline void tcg_out_opc_reg(TCGContext *s, int opc, int rd, int rs, int rt) +static inline void tcg_out_opc_reg(TCGContext *s, int opc, + TCGReg rd, TCGReg rs, TCGReg rt) { int32_t inst; @@ -341,7 +348,8 @@ static inline void tcg_out_opc_reg(TCGContext *s, int opc, int rd, int rs, int r /* * Type immediate */ -static inline void tcg_out_opc_imm(TCGContext *s, int opc, int rt, int rs, int imm) +static inline void tcg_out_opc_imm(TCGContext *s, int opc, + TCGReg rt, TCGReg rs, TCGArg imm) { int32_t inst; @@ -355,7 +363,8 @@ static inline void tcg_out_opc_imm(TCGContext *s, int opc, int rt, int rs, int i /* * Type branch */ -static inline void tcg_out_opc_br(TCGContext *s, int opc, int rt, int rs) +static inline void tcg_out_opc_br(TCGContext *s, int opc, + TCGReg rt, TCGReg rs) { /* We pay attention here to not modify the branch target by reading the existing value and using it again. This ensure that caches and @@ -368,7 +377,8 @@ static inline void tcg_out_opc_br(TCGContext *s, int opc, int rt, int rs) /* * Type sa */ -static inline void tcg_out_opc_sa(TCGContext *s, int opc, int rd, int rt, int sa) +static inline void tcg_out_opc_sa(TCGContext *s, int opc, + TCGReg rd, TCGReg rt, TCGArg sa) { int32_t inst; @@ -407,38 +417,47 @@ static inline void tcg_out_movi(TCGContext *s, TCGType type, } } -static inline void tcg_out_bswap16(TCGContext *s, int ret, int arg) +static inline void tcg_out_bswap16(TCGContext *s, TCGReg ret, TCGReg arg) { +#ifdef _MIPS_ARCH_MIPS32R2 + tcg_out_opc_reg(s, OPC_WSBH, ret, 0, arg); +#else /* ret and arg can't be register at */ if (ret == TCG_REG_AT || arg == TCG_REG_AT) { tcg_abort(); } tcg_out_opc_sa(s, OPC_SRL, TCG_REG_AT, arg, 8); - tcg_out_opc_imm(s, OPC_ANDI, TCG_REG_AT, TCG_REG_AT, 0x00ff); - tcg_out_opc_sa(s, OPC_SLL, ret, arg, 8); tcg_out_opc_imm(s, OPC_ANDI, ret, ret, 0xff00); tcg_out_opc_reg(s, OPC_OR, ret, ret, TCG_REG_AT); +#endif } -static inline void tcg_out_bswap16s(TCGContext *s, int ret, int arg) +static inline void tcg_out_bswap16s(TCGContext *s, TCGReg ret, TCGReg arg) { +#ifdef _MIPS_ARCH_MIPS32R2 + tcg_out_opc_reg(s, OPC_WSBH, ret, 0, arg); + tcg_out_opc_reg(s, OPC_SEH, ret, 0, ret); +#else /* ret and arg can't be register at */ if (ret == TCG_REG_AT || arg == TCG_REG_AT) { tcg_abort(); } tcg_out_opc_sa(s, OPC_SRL, TCG_REG_AT, arg, 8); - tcg_out_opc_imm(s, OPC_ANDI, TCG_REG_AT, TCG_REG_AT, 0xff); - tcg_out_opc_sa(s, OPC_SLL, ret, arg, 24); tcg_out_opc_sa(s, OPC_SRA, ret, ret, 16); tcg_out_opc_reg(s, OPC_OR, ret, ret, TCG_REG_AT); +#endif } -static inline void tcg_out_bswap32(TCGContext *s, int ret, int arg) +static inline void tcg_out_bswap32(TCGContext *s, TCGReg ret, TCGReg arg) { +#ifdef _MIPS_ARCH_MIPS32R2 + tcg_out_opc_reg(s, OPC_WSBH, ret, 0, arg); + tcg_out_opc_sa(s, OPC_ROTR, ret, ret, 16); +#else /* ret and arg must be different and can't be register at */ if (ret == arg || ret == TCG_REG_AT || arg == TCG_REG_AT) { tcg_abort(); @@ -456,9 +475,10 @@ static inline void tcg_out_bswap32(TCGContext *s, int ret, int arg) tcg_out_opc_sa(s, OPC_SRL, TCG_REG_AT, arg, 8); tcg_out_opc_imm(s, OPC_ANDI, TCG_REG_AT, TCG_REG_AT, 0xff00); tcg_out_opc_reg(s, OPC_OR, ret, ret, TCG_REG_AT); +#endif } -static inline void tcg_out_ext8s(TCGContext *s, int ret, int arg) +static inline void tcg_out_ext8s(TCGContext *s, TCGReg ret, TCGReg arg) { #ifdef _MIPS_ARCH_MIPS32R2 tcg_out_opc_reg(s, OPC_SEB, ret, 0, arg); @@ -468,7 +488,7 @@ static inline void tcg_out_ext8s(TCGContext *s, int ret, int arg) #endif } -static inline void tcg_out_ext16s(TCGContext *s, int ret, int arg) +static inline void tcg_out_ext16s(TCGContext *s, TCGReg ret, TCGReg arg) { #ifdef _MIPS_ARCH_MIPS32R2 tcg_out_opc_reg(s, OPC_SEH, ret, 0, arg); @@ -478,8 +498,8 @@ static inline void tcg_out_ext16s(TCGContext *s, int ret, int arg) #endif } -static inline void tcg_out_ldst(TCGContext *s, int opc, int arg, - int arg1, tcg_target_long arg2) +static inline void tcg_out_ldst(TCGContext *s, int opc, TCGArg arg, + TCGReg arg1, TCGArg arg2) { if (arg2 == (int16_t) arg2) { tcg_out_opc_imm(s, opc, arg, arg1, arg2); @@ -502,7 +522,7 @@ static inline void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg, tcg_out_ldst(s, OPC_SW, arg, arg1, arg2); } -static inline void tcg_out_addi(TCGContext *s, int reg, tcg_target_long val) +static inline void tcg_out_addi(TCGContext *s, TCGReg reg, TCGArg val) { if (val == (int16_t)val) { tcg_out_opc_imm(s, OPC_ADDIU, reg, reg, val); @@ -543,7 +563,7 @@ DEFINE_TCG_OUT_CALL_IARG(tcg_out_call_iarg_reg16, TCGReg arg) #undef DEFINE_TCG_OUT_CALL_IARG_GET_ARG #define DEFINE_TCG_OUT_CALL_IARG_GET_ARG(A) \ tcg_out_movi(s, TCG_TYPE_I32, A, arg); -DEFINE_TCG_OUT_CALL_IARG(tcg_out_call_iarg_imm32, uint32_t arg) +DEFINE_TCG_OUT_CALL_IARG(tcg_out_call_iarg_imm32, TCGArg arg) #undef DEFINE_TCG_OUT_CALL_IARG_GET_ARG /* We don't use the macro for this one to avoid an unnecessary reg-reg @@ -573,8 +593,8 @@ static inline void tcg_out_call_iarg_reg64(TCGContext *s, int *arg_num, #endif } -static void tcg_out_brcond(TCGContext *s, TCGCond cond, int arg1, - int arg2, int label_index) +static void tcg_out_brcond(TCGContext *s, TCGCond cond, TCGArg arg1, + TCGArg arg2, int label_index) { TCGLabel *l = &s->labels[label_index]; @@ -586,32 +606,48 @@ static void tcg_out_brcond(TCGContext *s, TCGCond cond, int arg1, tcg_out_opc_br(s, OPC_BNE, arg1, arg2); break; case TCG_COND_LT: - tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg1, arg2); - tcg_out_opc_br(s, OPC_BNE, TCG_REG_AT, TCG_REG_ZERO); + if (arg2 == 0) { + tcg_out_opc_br(s, OPC_BLTZ, 0, arg1); + } else { + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg1, arg2); + tcg_out_opc_br(s, OPC_BNE, TCG_REG_AT, TCG_REG_ZERO); + } break; case TCG_COND_LTU: tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, arg1, arg2); tcg_out_opc_br(s, OPC_BNE, TCG_REG_AT, TCG_REG_ZERO); break; case TCG_COND_GE: - tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg1, arg2); - tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); + if (arg2 == 0) { + tcg_out_opc_br(s, OPC_BGEZ, 0, arg1); + } else { + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg1, arg2); + tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); + } break; case TCG_COND_GEU: tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, arg1, arg2); tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); break; case TCG_COND_LE: - tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg2, arg1); - tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); + if (arg2 == 0) { + tcg_out_opc_br(s, OPC_BLEZ, 0, arg1); + } else { + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg2, arg1); + tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); + } break; case TCG_COND_LEU: tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, arg2, arg1); tcg_out_opc_br(s, OPC_BEQ, TCG_REG_AT, TCG_REG_ZERO); break; case TCG_COND_GT: - tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg2, arg1); - tcg_out_opc_br(s, OPC_BNE, TCG_REG_AT, TCG_REG_ZERO); + if (arg2 == 0) { + tcg_out_opc_br(s, OPC_BGTZ, 0, arg1); + } else { + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, arg2, arg1); + tcg_out_opc_br(s, OPC_BNE, TCG_REG_AT, TCG_REG_ZERO); + } break; case TCG_COND_GTU: tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, arg2, arg1); @@ -631,8 +667,9 @@ static void tcg_out_brcond(TCGContext *s, TCGCond cond, int arg1, /* XXX: we implement it at the target level to avoid having to handle cross basic blocks temporaries */ -static void tcg_out_brcond2(TCGContext *s, TCGCond cond, int arg1, - int arg2, int arg3, int arg4, int label_index) +static void tcg_out_brcond2(TCGContext *s, TCGCond cond, TCGArg arg1, + TCGArg arg2, TCGArg arg3, TCGArg arg4, + int label_index) { void *label_ptr; @@ -694,8 +731,70 @@ static void tcg_out_brcond2(TCGContext *s, TCGCond cond, int arg1, reloc_pc16(label_ptr, (tcg_target_long) s->code_ptr); } -static void tcg_out_setcond(TCGContext *s, TCGCond cond, int ret, - int arg1, int arg2) +static void tcg_out_movcond(TCGContext *s, TCGCond cond, TCGReg ret, + TCGArg c1, TCGArg c2, TCGArg v) +{ + switch (cond) { + case TCG_COND_EQ: + if (c1 == 0) { + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, c2); + } else if (c2 == 0) { + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, c1); + } else { + tcg_out_opc_reg(s, OPC_XOR, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, TCG_REG_AT); + } + break; + case TCG_COND_NE: + if (c1 == 0) { + tcg_out_opc_reg(s, OPC_MOVN, ret, v, c2); + } else if (c2 == 0) { + tcg_out_opc_reg(s, OPC_MOVN, ret, v, c1); + } else { + tcg_out_opc_reg(s, OPC_XOR, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVN, ret, v, TCG_REG_AT); + } + break; + case TCG_COND_LT: + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVN, ret, v, TCG_REG_AT); + break; + case TCG_COND_LTU: + tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVN, ret, v, TCG_REG_AT); + break; + case TCG_COND_GE: + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, TCG_REG_AT); + break; + case TCG_COND_GEU: + tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, c1, c2); + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, TCG_REG_AT); + break; + case TCG_COND_LE: + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, c2, c1); + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, TCG_REG_AT); + break; + case TCG_COND_LEU: + tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, c2, c1); + tcg_out_opc_reg(s, OPC_MOVZ, ret, v, TCG_REG_AT); + break; + case TCG_COND_GT: + tcg_out_opc_reg(s, OPC_SLT, TCG_REG_AT, c2, c1); + tcg_out_opc_reg(s, OPC_MOVN, ret, v, TCG_REG_AT); + break; + case TCG_COND_GTU: + tcg_out_opc_reg(s, OPC_SLTU, TCG_REG_AT, c2, c1); + tcg_out_opc_reg(s, OPC_MOVN, ret, v, TCG_REG_AT); + break; + default: + tcg_abort(); + break; + } +} + +static void tcg_out_setcond(TCGContext *s, TCGCond cond, TCGReg ret, + TCGArg arg1, TCGArg arg2) { switch (cond) { case TCG_COND_EQ: @@ -754,8 +853,8 @@ static void tcg_out_setcond(TCGContext *s, TCGCond cond, int ret, /* XXX: we implement it at the target level to avoid having to handle cross basic blocks temporaries */ -static void tcg_out_setcond2(TCGContext *s, TCGCond cond, int ret, - int arg1, int arg2, int arg3, int arg4) +static void tcg_out_setcond2(TCGContext *s, TCGCond cond, TCGReg ret, + TCGArg arg1, TCGArg arg2, TCGArg arg3, TCGArg arg4) { switch (cond) { case TCG_COND_EQ: @@ -842,18 +941,17 @@ static const void * const qemu_st_helpers[4] = { static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, int opc) { - int addr_regl, addr_meml; - int data_regl, data_regh, data_reg1, data_reg2; - int mem_index, s_bits; + TCGReg addr_regl, data_regl, data_regh, data_reg1, data_reg2; #if defined(CONFIG_SOFTMMU) void *label1_ptr, *label2_ptr; int arg_num; -#endif -#if TARGET_LONG_BITS == 64 -# if defined(CONFIG_SOFTMMU) + int mem_index, s_bits; + int addr_meml; +# if TARGET_LONG_BITS == 64 uint8_t *label3_ptr; + TCGReg addr_regh; + int addr_memh; # endif - int addr_regh, addr_memh; #endif data_regl = *args++; if (opc == 3) @@ -861,11 +959,22 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, else data_regh = 0; addr_regl = *args++; -#if TARGET_LONG_BITS == 64 +#if defined(CONFIG_SOFTMMU) +# if TARGET_LONG_BITS == 64 addr_regh = *args++; -#endif +# if defined(TCG_TARGET_WORDS_BIGENDIAN) + addr_memh = 0; + addr_meml = 4; +# else + addr_memh = 4; + addr_meml = 0; +# endif +# else + addr_meml = 0; +# endif mem_index = *args; s_bits = opc & 3; +#endif if (opc == 3) { #if defined(TCG_TARGET_WORDS_BIGENDIAN) @@ -879,18 +988,6 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, data_reg1 = data_regl; data_reg2 = 0; } -#if TARGET_LONG_BITS == 64 -# if defined(TCG_TARGET_WORDS_BIGENDIAN) - addr_memh = 0; - addr_meml = 4; -# else - addr_memh = 4; - addr_meml = 0; -# endif -#else - addr_meml = 0; -#endif - #if defined(CONFIG_SOFTMMU) tcg_out_opc_sa(s, OPC_SRL, TCG_REG_A0, addr_regl, TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS); tcg_out_opc_imm(s, OPC_ANDI, TCG_REG_A0, TCG_REG_A0, (CPU_TLB_SIZE - 1) << CPU_TLB_ENTRY_BITS); @@ -1029,50 +1126,56 @@ static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, int opc) { - int addr_regl, addr_meml; - int data_regl, data_regh, data_reg1, data_reg2; - int mem_index, s_bits; + TCGReg addr_regl, data_regl, data_regh, data_reg1, data_reg2; #if defined(CONFIG_SOFTMMU) uint8_t *label1_ptr, *label2_ptr; int arg_num; + int mem_index, s_bits; + int addr_meml; #endif #if TARGET_LONG_BITS == 64 # if defined(CONFIG_SOFTMMU) uint8_t *label3_ptr; + TCGReg addr_regh; + int addr_memh; # endif - int addr_regh, addr_memh; #endif - data_regl = *args++; if (opc == 3) { data_regh = *args++; -#if defined(TCG_TARGET_WORDS_BIGENDIAN) - data_reg1 = data_regh; - data_reg2 = data_regl; -#else - data_reg1 = data_regl; - data_reg2 = data_regh; -#endif } else { - data_reg1 = data_regl; - data_reg2 = 0; data_regh = 0; } addr_regl = *args++; -#if TARGET_LONG_BITS == 64 +#if defined(CONFIG_SOFTMMU) +# if TARGET_LONG_BITS == 64 addr_regh = *args++; -# if defined(TCG_TARGET_WORDS_BIGENDIAN) +# if defined(TCG_TARGET_WORDS_BIGENDIAN) addr_memh = 0; addr_meml = 4; -# else +# else addr_memh = 4; addr_meml = 0; -# endif -#else +# endif +# else addr_meml = 0; -#endif +# endif mem_index = *args; s_bits = opc; +#endif + + if (opc == 3) { +#if defined(TCG_TARGET_WORDS_BIGENDIAN) + data_reg1 = data_regh; + data_reg2 = data_regl; +#else + data_reg1 = data_regl; + data_reg2 = data_regh; +#endif + } else { + data_reg1 = data_regl; + data_reg2 = 0; + } #if defined(CONFIG_SOFTMMU) tcg_out_opc_sa(s, OPC_SRL, TCG_REG_A0, addr_regl, TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS); @@ -1157,7 +1260,8 @@ static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, break; case 1: if (TCG_NEED_BSWAP) { - tcg_out_bswap16(s, TCG_REG_T0, data_reg1); + tcg_out_opc_imm(s, OPC_ANDI, TCG_REG_T0, data_reg1, 0xffff); + tcg_out_bswap16(s, TCG_REG_T0, TCG_REG_T0); tcg_out_opc_imm(s, OPC_SH, TCG_REG_T0, TCG_REG_A0, 0); } else { tcg_out_opc_imm(s, OPC_SH, data_reg1, TCG_REG_A0, 0); @@ -1377,6 +1481,31 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_opc_reg(s, OPC_SRLV, args[0], args[2], args[1]); } break; + case INDEX_op_rotl_i32: + if (const_args[2]) { + tcg_out_opc_sa(s, OPC_ROTR, args[0], args[1], 0x20 - args[2]); + } else { + tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_AT, 32); + tcg_out_opc_reg(s, OPC_SUBU, TCG_REG_AT, TCG_REG_AT, args[2]); + tcg_out_opc_reg(s, OPC_ROTRV, args[0], TCG_REG_AT, args[1]); + } + break; + case INDEX_op_rotr_i32: + if (const_args[2]) { + tcg_out_opc_sa(s, OPC_ROTR, args[0], args[1], args[2]); + } else { + tcg_out_opc_reg(s, OPC_ROTRV, args[0], args[2], args[1]); + } + break; + + /* The bswap routines do not work on non-R2 CPU. In that case + we let TCG generating the corresponding code. */ + case INDEX_op_bswap16_i32: + tcg_out_bswap16(s, args[0], args[1]); + break; + case INDEX_op_bswap32_i32: + tcg_out_bswap32(s, args[0], args[1]); + break; case INDEX_op_ext8s_i32: tcg_out_ext8s(s, args[0], args[1]); @@ -1385,6 +1514,11 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_ext16s(s, args[0], args[1]); break; + case INDEX_op_deposit_i32: + tcg_out_opc_imm(s, OPC_INS, args[0], args[2], + ((args[3] + args[4] - 1) << 11) | (args[3] << 6)); + break; + case INDEX_op_brcond_i32: tcg_out_brcond(s, args[2], args[0], args[1], args[3]); break; @@ -1392,6 +1526,10 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, tcg_out_brcond2(s, args[4], args[0], args[1], args[2], args[3], args[5]); break; + case INDEX_op_movcond_i32: + tcg_out_movcond(s, args[5], args[0], args[1], args[2], args[3]); + break; + case INDEX_op_setcond_i32: tcg_out_setcond(s, args[3], args[0], args[1], args[2]); break; @@ -1453,34 +1591,42 @@ static const TCGTargetOpDef mips_op_defs[] = { { INDEX_op_st16_i32, { "rZ", "r" } }, { INDEX_op_st_i32, { "rZ", "r" } }, - { INDEX_op_add_i32, { "r", "rZ", "rJZ" } }, + { INDEX_op_add_i32, { "r", "rZ", "rJ" } }, { INDEX_op_mul_i32, { "r", "rZ", "rZ" } }, { INDEX_op_mulu2_i32, { "r", "r", "rZ", "rZ" } }, { INDEX_op_div_i32, { "r", "rZ", "rZ" } }, { INDEX_op_divu_i32, { "r", "rZ", "rZ" } }, { INDEX_op_rem_i32, { "r", "rZ", "rZ" } }, { INDEX_op_remu_i32, { "r", "rZ", "rZ" } }, - { INDEX_op_sub_i32, { "r", "rZ", "rJZ" } }, + { INDEX_op_sub_i32, { "r", "rZ", "rJ" } }, - { INDEX_op_and_i32, { "r", "rZ", "rIZ" } }, + { INDEX_op_and_i32, { "r", "rZ", "rI" } }, { INDEX_op_nor_i32, { "r", "rZ", "rZ" } }, { INDEX_op_not_i32, { "r", "rZ" } }, { INDEX_op_or_i32, { "r", "rZ", "rIZ" } }, { INDEX_op_xor_i32, { "r", "rZ", "rIZ" } }, - { INDEX_op_shl_i32, { "r", "rZ", "riZ" } }, - { INDEX_op_shr_i32, { "r", "rZ", "riZ" } }, - { INDEX_op_sar_i32, { "r", "rZ", "riZ" } }, + { INDEX_op_shl_i32, { "r", "rZ", "ri" } }, + { INDEX_op_shr_i32, { "r", "rZ", "ri" } }, + { INDEX_op_sar_i32, { "r", "rZ", "ri" } }, + { INDEX_op_rotr_i32, { "r", "rZ", "ri" } }, + { INDEX_op_rotl_i32, { "r", "rZ", "ri" } }, + + { INDEX_op_bswap16_i32, { "r", "r" } }, + { INDEX_op_bswap32_i32, { "r", "r" } }, { INDEX_op_ext8s_i32, { "r", "rZ" } }, { INDEX_op_ext16s_i32, { "r", "rZ" } }, + { INDEX_op_deposit_i32, { "r", "0", "rZ" } }, + { INDEX_op_brcond_i32, { "rZ", "rZ" } }, + { INDEX_op_movcond_i32, { "r", "rZ", "rZ", "rZ", "0" } }, { INDEX_op_setcond_i32, { "r", "rZ", "rZ" } }, { INDEX_op_setcond2_i32, { "r", "rZ", "rZ", "rZ", "rZ" } }, - { INDEX_op_add2_i32, { "r", "r", "rZ", "rZ", "rJZ", "rJZ" } }, - { INDEX_op_sub2_i32, { "r", "r", "rZ", "rZ", "rJZ", "rJZ" } }, + { INDEX_op_add2_i32, { "r", "r", "rZ", "rZ", "rJ", "rJ" } }, + { INDEX_op_sub2_i32, { "r", "r", "rZ", "rZ", "rJ", "rJ" } }, { INDEX_op_brcond2_i32, { "rZ", "rZ", "rZ", "rZ" } }, #if TARGET_LONG_BITS == 32 @@ -1520,7 +1666,6 @@ static int tcg_target_callee_save_regs[] = { TCG_REG_S5, TCG_REG_S6, TCG_REG_S7, - TCG_REG_GP, TCG_REG_FP, TCG_REG_RA, /* should be last for ABI compliance */ }; @@ -1530,11 +1675,15 @@ static void tcg_target_qemu_prologue(TCGContext *s) { int i, frame_size; - /* reserve some stack space */ + /* reserve some stack space, also for TCG temps. */ frame_size = ARRAY_SIZE(tcg_target_callee_save_regs) * 4 - + TCG_STATIC_CALL_ARGS_SIZE; + + TCG_STATIC_CALL_ARGS_SIZE + + CPU_TEMP_BUF_NLONGS * sizeof(long); frame_size = (frame_size + TCG_TARGET_STACK_ALIGN - 1) & ~(TCG_TARGET_STACK_ALIGN - 1); + tcg_set_frame(s, TCG_REG_SP, ARRAY_SIZE(tcg_target_callee_save_regs) * 4 + + TCG_STATIC_CALL_ARGS_SIZE, + CPU_TEMP_BUF_NLONGS * sizeof(long)); /* TB prologue */ tcg_out_addi(s, TCG_REG_SP, -frame_size); @@ -1586,8 +1735,7 @@ static void tcg_target_init(TCGContext *s) tcg_regset_set_reg(s->reserved_regs, TCG_REG_T0); /* internal use */ tcg_regset_set_reg(s->reserved_regs, TCG_REG_RA); /* return address */ tcg_regset_set_reg(s->reserved_regs, TCG_REG_SP); /* stack pointer */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_GP); /* global pointer */ tcg_add_target_add_op_defs(mips_op_defs); - tcg_set_frame(s, TCG_AREG0, offsetof(CPUArchState, temp_buf), - CPU_TEMP_BUF_NLONGS * sizeof(long)); } diff --git a/tcg/mips/tcg-target.h b/tcg/mips/tcg-target.h index 1c6193180c..d147e70eb1 100644 --- a/tcg/mips/tcg-target.h +++ b/tcg/mips/tcg-target.h @@ -80,16 +80,34 @@ typedef enum { #define TCG_TARGET_HAS_div_i32 1 #define TCG_TARGET_HAS_not_i32 1 #define TCG_TARGET_HAS_nor_i32 1 -#define TCG_TARGET_HAS_rot_i32 0 #define TCG_TARGET_HAS_ext8s_i32 1 #define TCG_TARGET_HAS_ext16s_i32 1 -#define TCG_TARGET_HAS_bswap32_i32 0 -#define TCG_TARGET_HAS_bswap16_i32 0 #define TCG_TARGET_HAS_andc_i32 0 #define TCG_TARGET_HAS_orc_i32 0 #define TCG_TARGET_HAS_eqv_i32 0 #define TCG_TARGET_HAS_nand_i32 0 + +/* optional instructions only implemented on MIPS4, MIPS32 and Loongson 2 */ +#if defined(_MIPS_ARCH_MIPS4) || defined(_MIPS_ARCH_MIPS32) || \ + defined(_MIPS_ARCH_MIPS32R2) || defined(_MIPS_ARCH_LOONGSON2E) || \ + defined(_MIPS_ARCH_LOONGSON2F) +#define TCG_TARGET_HAS_movcond_i32 1 +#else +#define TCG_TARGET_HAS_movcond_i32 0 +#endif + +/* optional instructions only implemented on MIPS32R2 */ +#ifdef _MIPS_ARCH_MIPS32R2 +#define TCG_TARGET_HAS_bswap16_i32 1 +#define TCG_TARGET_HAS_bswap32_i32 1 +#define TCG_TARGET_HAS_rot_i32 1 +#define TCG_TARGET_HAS_deposit_i32 1 +#else +#define TCG_TARGET_HAS_bswap16_i32 0 +#define TCG_TARGET_HAS_bswap32_i32 0 +#define TCG_TARGET_HAS_rot_i32 0 #define TCG_TARGET_HAS_deposit_i32 0 +#endif /* optional instructions automatically implemented */ #define TCG_TARGET_HAS_neg_i32 0 /* sub rd, zero, rt */ diff --git a/tcg/optimize.c b/tcg/optimize.c index 9da333c303..35532a1e03 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -39,8 +39,6 @@ typedef enum { TCG_TEMP_UNDEF = 0, TCG_TEMP_CONST, TCG_TEMP_COPY, - TCG_TEMP_HAS_COPY, - TCG_TEMP_ANY } tcg_temp_state; struct tcg_temp_info { @@ -52,39 +50,19 @@ struct tcg_temp_info { static struct tcg_temp_info temps[TCG_MAX_TEMPS]; -/* Reset TEMP's state to TCG_TEMP_ANY. If TEMP was a representative of some - class of equivalent temp's, a new representative should be chosen in this - class. */ -static void reset_temp(TCGArg temp, int nb_temps, int nb_globals) +/* Reset TEMP's state to TCG_TEMP_UNDEF. If TEMP only had one copy, remove + the copy flag from the left temp. */ +static void reset_temp(TCGArg temp) { - int i; - TCGArg new_base = (TCGArg)-1; - if (temps[temp].state == TCG_TEMP_HAS_COPY) { - for (i = temps[temp].next_copy; i != temp; i = temps[i].next_copy) { - if (i >= nb_globals) { - temps[i].state = TCG_TEMP_HAS_COPY; - new_base = i; - break; - } + if (temps[temp].state == TCG_TEMP_COPY) { + if (temps[temp].prev_copy == temps[temp].next_copy) { + temps[temps[temp].next_copy].state = TCG_TEMP_UNDEF; + } else { + temps[temps[temp].next_copy].prev_copy = temps[temp].prev_copy; + temps[temps[temp].prev_copy].next_copy = temps[temp].next_copy; } - for (i = temps[temp].next_copy; i != temp; i = temps[i].next_copy) { - if (new_base == (TCGArg)-1) { - temps[i].state = TCG_TEMP_ANY; - } else { - temps[i].val = new_base; - } - } - temps[temps[temp].next_copy].prev_copy = temps[temp].prev_copy; - temps[temps[temp].prev_copy].next_copy = temps[temp].next_copy; - } else if (temps[temp].state == TCG_TEMP_COPY) { - temps[temps[temp].next_copy].prev_copy = temps[temp].prev_copy; - temps[temps[temp].prev_copy].next_copy = temps[temp].next_copy; - new_base = temps[temp].val; - } - temps[temp].state = TCG_TEMP_ANY; - if (new_base != (TCGArg)-1 && temps[new_base].next_copy == new_base) { - temps[new_base].state = TCG_TEMP_ANY; } + temps[temp].state = TCG_TEMP_UNDEF; } static int op_bits(TCGOpcode op) @@ -107,33 +85,83 @@ static TCGOpcode op_to_movi(TCGOpcode op) } } -static void tcg_opt_gen_mov(TCGArg *gen_args, TCGArg dst, TCGArg src, - int nb_temps, int nb_globals) +static TCGArg find_better_copy(TCGContext *s, TCGArg temp) +{ + TCGArg i; + + /* If this is already a global, we can't do better. */ + if (temp < s->nb_globals) { + return temp; + } + + /* Search for a global first. */ + for (i = temps[temp].next_copy ; i != temp ; i = temps[i].next_copy) { + if (i < s->nb_globals) { + return i; + } + } + + /* If it is a temp, search for a temp local. */ + if (!s->temps[temp].temp_local) { + for (i = temps[temp].next_copy ; i != temp ; i = temps[i].next_copy) { + if (s->temps[i].temp_local) { + return i; + } + } + } + + /* Failure to find a better representation, return the same temp. */ + return temp; +} + +static bool temps_are_copies(TCGArg arg1, TCGArg arg2) +{ + TCGArg i; + + if (arg1 == arg2) { + return true; + } + + if (temps[arg1].state != TCG_TEMP_COPY + || temps[arg2].state != TCG_TEMP_COPY) { + return false; + } + + for (i = temps[arg1].next_copy ; i != arg1 ; i = temps[i].next_copy) { + if (i == arg2) { + return true; + } + } + + return false; +} + +static void tcg_opt_gen_mov(TCGContext *s, TCGArg *gen_args, + TCGArg dst, TCGArg src) { - reset_temp(dst, nb_temps, nb_globals); - assert(temps[src].state != TCG_TEMP_COPY); - if (src >= nb_globals) { - assert(temps[src].state != TCG_TEMP_CONST); - if (temps[src].state != TCG_TEMP_HAS_COPY) { - temps[src].state = TCG_TEMP_HAS_COPY; + reset_temp(dst); + assert(temps[src].state != TCG_TEMP_CONST); + + if (s->temps[src].type == s->temps[dst].type) { + if (temps[src].state != TCG_TEMP_COPY) { + temps[src].state = TCG_TEMP_COPY; temps[src].next_copy = src; temps[src].prev_copy = src; } temps[dst].state = TCG_TEMP_COPY; - temps[dst].val = src; temps[dst].next_copy = temps[src].next_copy; temps[dst].prev_copy = src; temps[temps[dst].next_copy].prev_copy = dst; temps[src].next_copy = dst; } + gen_args[0] = dst; gen_args[1] = src; } -static void tcg_opt_gen_movi(TCGArg *gen_args, TCGArg dst, TCGArg val, - int nb_temps, int nb_globals) +static void tcg_opt_gen_movi(TCGArg *gen_args, TCGArg dst, TCGArg val) { - reset_temp(dst, nb_temps, nb_globals); + reset_temp(dst); temps[dst].state = TCG_TEMP_CONST; temps[dst].val = val; gen_args[0] = dst; @@ -264,58 +292,88 @@ static TCGArg do_constant_folding(TCGOpcode op, TCGArg x, TCGArg y) return res; } +/* Return 2 if the condition can't be simplified, and the result + of the condition (0 or 1) if it can */ static TCGArg do_constant_folding_cond(TCGOpcode op, TCGArg x, TCGArg y, TCGCond c) { - switch (op_bits(op)) { - case 32: + if (temps[x].state == TCG_TEMP_CONST && temps[y].state == TCG_TEMP_CONST) { + switch (op_bits(op)) { + case 32: + switch (c) { + case TCG_COND_EQ: + return (uint32_t)temps[x].val == (uint32_t)temps[y].val; + case TCG_COND_NE: + return (uint32_t)temps[x].val != (uint32_t)temps[y].val; + case TCG_COND_LT: + return (int32_t)temps[x].val < (int32_t)temps[y].val; + case TCG_COND_GE: + return (int32_t)temps[x].val >= (int32_t)temps[y].val; + case TCG_COND_LE: + return (int32_t)temps[x].val <= (int32_t)temps[y].val; + case TCG_COND_GT: + return (int32_t)temps[x].val > (int32_t)temps[y].val; + case TCG_COND_LTU: + return (uint32_t)temps[x].val < (uint32_t)temps[y].val; + case TCG_COND_GEU: + return (uint32_t)temps[x].val >= (uint32_t)temps[y].val; + case TCG_COND_LEU: + return (uint32_t)temps[x].val <= (uint32_t)temps[y].val; + case TCG_COND_GTU: + return (uint32_t)temps[x].val > (uint32_t)temps[y].val; + } + break; + case 64: + switch (c) { + case TCG_COND_EQ: + return (uint64_t)temps[x].val == (uint64_t)temps[y].val; + case TCG_COND_NE: + return (uint64_t)temps[x].val != (uint64_t)temps[y].val; + case TCG_COND_LT: + return (int64_t)temps[x].val < (int64_t)temps[y].val; + case TCG_COND_GE: + return (int64_t)temps[x].val >= (int64_t)temps[y].val; + case TCG_COND_LE: + return (int64_t)temps[x].val <= (int64_t)temps[y].val; + case TCG_COND_GT: + return (int64_t)temps[x].val > (int64_t)temps[y].val; + case TCG_COND_LTU: + return (uint64_t)temps[x].val < (uint64_t)temps[y].val; + case TCG_COND_GEU: + return (uint64_t)temps[x].val >= (uint64_t)temps[y].val; + case TCG_COND_LEU: + return (uint64_t)temps[x].val <= (uint64_t)temps[y].val; + case TCG_COND_GTU: + return (uint64_t)temps[x].val > (uint64_t)temps[y].val; + } + break; + } + } else if (temps_are_copies(x, y)) { switch (c) { - case TCG_COND_EQ: - return (uint32_t)x == (uint32_t)y; - case TCG_COND_NE: - return (uint32_t)x != (uint32_t)y; - case TCG_COND_LT: - return (int32_t)x < (int32_t)y; - case TCG_COND_GE: - return (int32_t)x >= (int32_t)y; - case TCG_COND_LE: - return (int32_t)x <= (int32_t)y; case TCG_COND_GT: - return (int32_t)x > (int32_t)y; case TCG_COND_LTU: - return (uint32_t)x < (uint32_t)y; - case TCG_COND_GEU: - return (uint32_t)x >= (uint32_t)y; - case TCG_COND_LEU: - return (uint32_t)x <= (uint32_t)y; + case TCG_COND_LT: case TCG_COND_GTU: - return (uint32_t)x > (uint32_t)y; - } - break; - case 64: - switch (c) { - case TCG_COND_EQ: - return (uint64_t)x == (uint64_t)y; case TCG_COND_NE: - return (uint64_t)x != (uint64_t)y; - case TCG_COND_LT: - return (int64_t)x < (int64_t)y; + return 0; case TCG_COND_GE: - return (int64_t)x >= (int64_t)y; + case TCG_COND_GEU: case TCG_COND_LE: - return (int64_t)x <= (int64_t)y; - case TCG_COND_GT: - return (int64_t)x > (int64_t)y; + case TCG_COND_LEU: + case TCG_COND_EQ: + return 1; + } + } else if (temps[y].state == TCG_TEMP_CONST && temps[y].val == 0) { + switch (c) { case TCG_COND_LTU: - return (uint64_t)x < (uint64_t)y; + return 0; case TCG_COND_GEU: - return (uint64_t)x >= (uint64_t)y; - case TCG_COND_LEU: - return (uint64_t)x <= (uint64_t)y; - case TCG_COND_GTU: - return (uint64_t)x > (uint64_t)y; + return 1; + default: + return 2; } - break; + } else { + return 2; } fprintf(stderr, @@ -324,7 +382,6 @@ static TCGArg do_constant_folding_cond(TCGOpcode op, TCGArg x, tcg_abort(); } - /* Propagate constants and copies, fold constant expressions. */ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, TCGArg *args, TCGOpDef *tcg_op_defs) @@ -334,12 +391,12 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, const TCGOpDef *def; TCGArg *gen_args; TCGArg tmp; + TCGCond cond; + /* Array VALS has an element for each temp. If this temp holds a constant then its value is kept in VALS' element. - If this temp is a copy of other ones then this equivalence class' - representative is kept in VALS' element. - If this temp is neither copy nor constant then corresponding VALS' - element is unused. */ + If this temp is a copy of other ones then the other copies are + available through the doubly linked circular list. */ nb_temps = s->nb_temps; nb_globals = s->nb_globals; @@ -351,11 +408,18 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, op = gen_opc_buf[op_index]; def = &tcg_op_defs[op]; /* Do copy propagation */ - if (!(def->flags & (TCG_OPF_CALL_CLOBBER | TCG_OPF_SIDE_EFFECTS))) { - assert(op != INDEX_op_call); + if (op == INDEX_op_call) { + int nb_oargs = args[0] >> 16; + int nb_iargs = args[0] & 0xffff; + for (i = nb_oargs + 1; i < nb_oargs + nb_iargs + 1; i++) { + if (temps[args[i]].state == TCG_TEMP_COPY) { + args[i] = find_better_copy(s, args[i]); + } + } + } else { for (i = def->nb_oargs; i < def->nb_oargs + def->nb_iargs; i++) { if (temps[args[i]].state == TCG_TEMP_COPY) { - args[i] = temps[args[i]].val; + args[i] = find_better_copy(s, args[i]); } } } @@ -370,7 +434,10 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, CASE_OP_32_64(eqv): CASE_OP_32_64(nand): CASE_OP_32_64(nor): - if (temps[args[1]].state == TCG_TEMP_CONST) { + /* Prefer the constant in second argument, and then the form + op a, a, b, which is better handled on non-RISC hosts. */ + if (temps[args[1]].state == TCG_TEMP_CONST || (args[0] == args[2] + && temps[args[2]].state != TCG_TEMP_CONST)) { tmp = args[1]; args[1] = args[2]; args[2] = tmp; @@ -394,6 +461,25 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, args[3] = tcg_swap_cond(args[3]); } break; + CASE_OP_32_64(movcond): + cond = args[5]; + if (temps[args[1]].state == TCG_TEMP_CONST + && temps[args[2]].state != TCG_TEMP_CONST) { + tmp = args[1]; + args[1] = args[2]; + args[2] = tmp; + cond = tcg_swap_cond(cond); + } + /* For movcond, we canonicalize the "false" input reg to match + the destination reg so that the tcg backend can implement + a "move if true" operation. */ + if (args[0] == args[3]) { + tmp = args[3]; + args[3] = args[4]; + args[4] = tmp; + cond = tcg_invert_cond(cond); + } + args[5] = cond; default: break; } @@ -408,7 +494,7 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, if (temps[args[1]].state == TCG_TEMP_CONST && temps[args[1]].val == 0) { gen_opc_buf[op_index] = op_to_movi(op); - tcg_opt_gen_movi(gen_args, args[0], 0, nb_temps, nb_globals); + tcg_opt_gen_movi(gen_args, args[0], 0); args += 3; gen_args += 2; continue; @@ -435,14 +521,11 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, } if (temps[args[2]].state == TCG_TEMP_CONST && temps[args[2]].val == 0) { - if ((temps[args[0]].state == TCG_TEMP_COPY - && temps[args[0]].val == args[1]) - || args[0] == args[1]) { + if (temps_are_copies(args[0], args[1])) { gen_opc_buf[op_index] = INDEX_op_nop; } else { gen_opc_buf[op_index] = op_to_mov(op); - tcg_opt_gen_mov(gen_args, args[0], args[1], - nb_temps, nb_globals); + tcg_opt_gen_mov(s, gen_args, args[0], args[1]); gen_args += 2; } args += 3; @@ -460,7 +543,7 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, if ((temps[args[2]].state == TCG_TEMP_CONST && temps[args[2]].val == 0)) { gen_opc_buf[op_index] = op_to_movi(op); - tcg_opt_gen_movi(gen_args, args[0], 0, nb_temps, nb_globals); + tcg_opt_gen_movi(gen_args, args[0], 0); args += 3; gen_args += 2; continue; @@ -474,13 +557,12 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, switch (op) { CASE_OP_32_64(or): CASE_OP_32_64(and): - if (args[1] == args[2]) { - if (args[1] == args[0]) { + if (temps_are_copies(args[1], args[2])) { + if (temps_are_copies(args[0], args[1])) { gen_opc_buf[op_index] = INDEX_op_nop; } else { gen_opc_buf[op_index] = op_to_mov(op); - tcg_opt_gen_mov(gen_args, args[0], args[1], nb_temps, - nb_globals); + tcg_opt_gen_mov(s, gen_args, args[0], args[1]); gen_args += 2; } args += 3; @@ -491,21 +573,34 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, break; } + /* Simplify expression for "op r, a, a => movi r, 0" cases */ + switch (op) { + CASE_OP_32_64(sub): + CASE_OP_32_64(xor): + if (temps_are_copies(args[1], args[2])) { + gen_opc_buf[op_index] = op_to_movi(op); + tcg_opt_gen_movi(gen_args, args[0], 0); + gen_args += 2; + args += 3; + continue; + } + break; + default: + break; + } + /* Propagate constants through copy operations and do constant folding. Constants will be substituted to arguments by register allocator where needed and possible. Also detect copies. */ switch (op) { CASE_OP_32_64(mov): - if ((temps[args[1]].state == TCG_TEMP_COPY - && temps[args[1]].val == args[0]) - || args[0] == args[1]) { + if (temps_are_copies(args[0], args[1])) { args += 2; gen_opc_buf[op_index] = INDEX_op_nop; break; } if (temps[args[1]].state != TCG_TEMP_CONST) { - tcg_opt_gen_mov(gen_args, args[0], args[1], - nb_temps, nb_globals); + tcg_opt_gen_mov(s, gen_args, args[0], args[1]); gen_args += 2; args += 2; break; @@ -517,7 +612,7 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, args[1] = temps[args[1]].val; /* fallthrough */ CASE_OP_32_64(movi): - tcg_opt_gen_movi(gen_args, args[0], args[1], nb_temps, nb_globals); + tcg_opt_gen_movi(gen_args, args[0], args[1]); gen_args += 2; args += 2; break; @@ -532,9 +627,9 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, if (temps[args[1]].state == TCG_TEMP_CONST) { gen_opc_buf[op_index] = op_to_movi(op); tmp = do_constant_folding(op, temps[args[1]].val, 0); - tcg_opt_gen_movi(gen_args, args[0], tmp, nb_temps, nb_globals); + tcg_opt_gen_movi(gen_args, args[0], tmp); } else { - reset_temp(args[0], nb_temps, nb_globals); + reset_temp(args[0]); gen_args[0] = args[0]; gen_args[1] = args[1]; } @@ -562,10 +657,10 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, gen_opc_buf[op_index] = op_to_movi(op); tmp = do_constant_folding(op, temps[args[1]].val, temps[args[2]].val); - tcg_opt_gen_movi(gen_args, args[0], tmp, nb_temps, nb_globals); + tcg_opt_gen_movi(gen_args, args[0], tmp); gen_args += 2; } else { - reset_temp(args[0], nb_temps, nb_globals); + reset_temp(args[0]); gen_args[0] = args[0]; gen_args[1] = args[1]; gen_args[2] = args[2]; @@ -573,16 +668,34 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, } args += 3; break; - CASE_OP_32_64(setcond): + CASE_OP_32_64(deposit): if (temps[args[1]].state == TCG_TEMP_CONST && temps[args[2]].state == TCG_TEMP_CONST) { gen_opc_buf[op_index] = op_to_movi(op); - tmp = do_constant_folding_cond(op, temps[args[1]].val, - temps[args[2]].val, args[3]); - tcg_opt_gen_movi(gen_args, args[0], tmp, nb_temps, nb_globals); + tmp = ((1ull << args[4]) - 1); + tmp = (temps[args[1]].val & ~(tmp << args[3])) + | ((temps[args[2]].val & tmp) << args[3]); + tcg_opt_gen_movi(gen_args, args[0], tmp); + gen_args += 2; + } else { + reset_temp(args[0]); + gen_args[0] = args[0]; + gen_args[1] = args[1]; + gen_args[2] = args[2]; + gen_args[3] = args[3]; + gen_args[4] = args[4]; + gen_args += 5; + } + args += 5; + break; + CASE_OP_32_64(setcond): + tmp = do_constant_folding_cond(op, args[1], args[2], args[3]); + if (tmp != 2) { + gen_opc_buf[op_index] = op_to_movi(op); + tcg_opt_gen_movi(gen_args, args[0], tmp); gen_args += 2; } else { - reset_temp(args[0], nb_temps, nb_globals); + reset_temp(args[0]); gen_args[0] = args[0]; gen_args[1] = args[1]; gen_args[2] = args[2]; @@ -592,10 +705,9 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, args += 4; break; CASE_OP_32_64(brcond): - if (temps[args[0]].state == TCG_TEMP_CONST - && temps[args[1]].state == TCG_TEMP_CONST) { - if (do_constant_folding_cond(op, temps[args[0]].val, - temps[args[1]].val, args[2])) { + tmp = do_constant_folding_cond(op, args[0], args[1], args[2]); + if (tmp != 2) { + if (tmp) { memset(temps, 0, nb_temps * sizeof(struct tcg_temp_info)); gen_opc_buf[op_index] = INDEX_op_br; gen_args[0] = args[3]; @@ -605,7 +717,7 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, } } else { memset(temps, 0, nb_temps * sizeof(struct tcg_temp_info)); - reset_temp(args[0], nb_temps, nb_globals); + reset_temp(args[0]); gen_args[0] = args[0]; gen_args[1] = args[1]; gen_args[2] = args[2]; @@ -614,15 +726,41 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, } args += 4; break; + CASE_OP_32_64(movcond): + tmp = do_constant_folding_cond(op, args[1], args[2], args[5]); + if (tmp != 2) { + if (temps_are_copies(args[0], args[4-tmp])) { + gen_opc_buf[op_index] = INDEX_op_nop; + } else if (temps[args[4-tmp]].state == TCG_TEMP_CONST) { + gen_opc_buf[op_index] = op_to_movi(op); + tcg_opt_gen_movi(gen_args, args[0], temps[args[4-tmp]].val); + gen_args += 2; + } else { + gen_opc_buf[op_index] = op_to_mov(op); + tcg_opt_gen_mov(s, gen_args, args[0], args[4-tmp]); + gen_args += 2; + } + } else { + reset_temp(args[0]); + gen_args[0] = args[0]; + gen_args[1] = args[1]; + gen_args[2] = args[2]; + gen_args[3] = args[3]; + gen_args[4] = args[4]; + gen_args[5] = args[5]; + gen_args += 6; + } + args += 6; + break; case INDEX_op_call: nb_call_args = (args[0] >> 16) + (args[0] & 0xffff); if (!(args[nb_call_args + 1] & (TCG_CALL_CONST | TCG_CALL_PURE))) { for (i = 0; i < nb_globals; i++) { - reset_temp(i, nb_temps, nb_globals); + reset_temp(i); } } for (i = 0; i < (args[0] >> 16); i++) { - reset_temp(args[i + 1], nb_temps, nb_globals); + reset_temp(args[i + 1]); } i = nb_call_args + 3; while (i) { @@ -641,7 +779,7 @@ static TCGArg *tcg_constant_folding(TCGContext *s, uint16_t *tcg_opc_ptr, memset(temps, 0, nb_temps * sizeof(struct tcg_temp_info)); } else { for (i = 0; i < def->nb_oargs; i++) { - reset_temp(args[i], nb_temps, nb_globals); + reset_temp(args[i]); } } for (i = 0; i < def->nb_args; i++) { diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c index 26c4b33e60..90c275d698 100644 --- a/tcg/ppc/tcg-target.c +++ b/tcg/ppc/tcg-target.c @@ -221,12 +221,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -/* 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); -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { @@ -390,6 +384,7 @@ static int tcg_target_const_match(tcg_target_long val, #define ORC XO31(412) #define EQV XO31(284) #define NAND XO31(476) +#define ISEL XO31( 15) #define LBZX XO31( 87) #define LHZX XO31(279) @@ -1269,6 +1264,72 @@ static void tcg_out_setcond2 (TCGContext *s, const TCGArg *args, ); } +static void tcg_out_movcond (TCGContext *s, TCGCond cond, + TCGArg dest, + TCGArg c1, TCGArg c2, + TCGArg v1, TCGArg v2, + int const_c2) +{ + tcg_out_cmp (s, cond, c1, c2, const_c2, 7); + + if (1) { + /* At least here on 7747A bit twiddling hacks are outperformed + by jumpy code (the testing was not scientific) */ + if (dest == v2) { + cond = tcg_invert_cond (cond); + v2 = v1; + } + else { + if (dest != v1) { + tcg_out_mov (s, TCG_TYPE_I32, dest, v1); + } + } + /* Branch forward over one insn */ + tcg_out32 (s, tcg_to_bc[cond] | 8); + tcg_out_mov (s, TCG_TYPE_I32, dest, v2); + } + else { + /* isel version, "if (1)" above should be replaced once a way + to figure out availability of isel on the underlying + hardware is found */ + int tab, bc; + + switch (cond) { + case TCG_COND_EQ: + tab = TAB (dest, v1, v2); + bc = CR_EQ; + break; + case TCG_COND_NE: + tab = TAB (dest, v2, v1); + bc = CR_EQ; + break; + case TCG_COND_LTU: + case TCG_COND_LT: + tab = TAB (dest, v1, v2); + bc = CR_LT; + break; + case TCG_COND_GEU: + case TCG_COND_GE: + tab = TAB (dest, v2, v1); + bc = CR_LT; + break; + case TCG_COND_LEU: + case TCG_COND_LE: + tab = TAB (dest, v2, v1); + bc = CR_GT; + break; + case TCG_COND_GTU: + case TCG_COND_GT: + tab = TAB (dest, v1, v2); + bc = CR_GT; + break; + default: + tcg_abort (); + } + tcg_out32 (s, ISEL | tab | ((bc + 28) << 6)); + } +} + static void tcg_out_brcond (TCGContext *s, TCGCond cond, TCGArg arg1, TCGArg arg2, int const_arg2, int label_index) @@ -1826,6 +1887,13 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, ); break; + case INDEX_op_movcond_i32: + tcg_out_movcond (s, args[5], args[0], + args[1], args[2], + args[3], args[4], + const_args[2]); + break; + default: tcg_dump_ops (s); tcg_abort (); @@ -1922,6 +1990,7 @@ static const TCGTargetOpDef ppc_op_defs[] = { { INDEX_op_ext16u_i32, { "r", "r" } }, { INDEX_op_deposit_i32, { "r", "0", "r" } }, + { INDEX_op_movcond_i32, { "r", "r", "ri", "r", "r" } }, { -1 }, }; diff --git a/tcg/ppc/tcg-target.h b/tcg/ppc/tcg-target.h index 2f37fd289b..3259d898ab 100644 --- a/tcg/ppc/tcg-target.h +++ b/tcg/ppc/tcg-target.h @@ -92,6 +92,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 1 #define TCG_TARGET_HAS_nor_i32 1 #define TCG_TARGET_HAS_deposit_i32 1 +#define TCG_TARGET_HAS_movcond_i32 1 #define TCG_AREG0 TCG_REG_R27 diff --git a/tcg/ppc64/tcg-target.c b/tcg/ppc64/tcg-target.c index 337cd419f8..19944bc427 100644 --- a/tcg/ppc64/tcg-target.c +++ b/tcg/ppc64/tcg-target.c @@ -208,12 +208,6 @@ static void patch_reloc (uint8_t *code_ptr, int type, } } -/* 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); -} - /* parse target specific constraints */ static int target_parse_constraint (TCGArgConstraint *ct, const char **pct_str) { diff --git a/tcg/ppc64/tcg-target.h b/tcg/ppc64/tcg-target.h index 97eec0843d..57569e8938 100644 --- a/tcg/ppc64/tcg-target.h +++ b/tcg/ppc64/tcg-target.h @@ -83,6 +83,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 0 +#define TCG_TARGET_HAS_movcond_i32 0 #define TCG_TARGET_HAS_div_i64 1 #define TCG_TARGET_HAS_rot_i64 0 @@ -103,6 +104,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 #define TCG_TARGET_HAS_deposit_i64 0 +#define TCG_TARGET_HAS_movcond_i64 0 #define TCG_AREG0 TCG_REG_R27 diff --git a/tcg/s390/tcg-target.c b/tcg/s390/tcg-target.c index aac11d9b61..3b90605fb1 100644 --- a/tcg/s390/tcg-target.c +++ b/tcg/s390/tcg-target.c @@ -356,11 +356,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -static int tcg_target_get_call_iarg_regs_count(int flags) -{ - return sizeof(tcg_target_call_iarg_regs) / sizeof(int); -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { diff --git a/tcg/s390/tcg-target.h b/tcg/s390/tcg-target.h index 4f7dfaba50..ed55c331c6 100644 --- a/tcg/s390/tcg-target.h +++ b/tcg/s390/tcg-target.h @@ -63,6 +63,7 @@ typedef enum TCGReg { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 0 +#define TCG_TARGET_HAS_movcond_i32 0 #if TCG_TARGET_REG_BITS == 64 #define TCG_TARGET_HAS_div2_i64 1 @@ -84,6 +85,7 @@ typedef enum TCGReg { #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 #define TCG_TARGET_HAS_deposit_i64 0 +#define TCG_TARGET_HAS_movcond_i64 0 #endif #define TCG_TARGET_HAS_GUEST_BASE diff --git a/tcg/sparc/tcg-target.c b/tcg/sparc/tcg-target.c index baed3b49ff..8fd7f86de0 100644 --- a/tcg/sparc/tcg-target.c +++ b/tcg/sparc/tcg-target.c @@ -59,7 +59,15 @@ static const char * const tcg_target_reg_names[TCG_TARGET_NB_REGS] = { }; #endif -#define ARG_OFFSET 1 +/* Define some temporary registers. T2 is used for constant generation. */ +#define TCG_REG_T1 TCG_REG_G1 +#define TCG_REG_T2 TCG_REG_O7 + +#ifdef CONFIG_USE_GUEST_BASE +# define TCG_GUEST_BASE_REG TCG_REG_I5 +#else +# define TCG_GUEST_BASE_REG TCG_REG_G0 +#endif static const int tcg_target_reg_alloc_order[] = { TCG_REG_L0, @@ -70,11 +78,25 @@ static const int tcg_target_reg_alloc_order[] = { TCG_REG_L5, TCG_REG_L6, TCG_REG_L7, + TCG_REG_I0, TCG_REG_I1, TCG_REG_I2, TCG_REG_I3, TCG_REG_I4, + TCG_REG_I5, + + TCG_REG_G2, + TCG_REG_G3, + TCG_REG_G4, + TCG_REG_G5, + + TCG_REG_O0, + TCG_REG_O1, + TCG_REG_O2, + TCG_REG_O3, + TCG_REG_O4, + TCG_REG_O5, }; static const int tcg_target_call_iarg_regs[6] = { @@ -133,12 +155,6 @@ static void patch_reloc(uint8_t *code_ptr, int type, } } -/* maximum number of register used for input function arguments */ -static inline int tcg_target_get_call_iarg_regs_count(int flags) -{ - return 6; -} - /* parse target specific constraints */ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) { @@ -157,7 +173,6 @@ static int target_parse_constraint(TCGArgConstraint *ct, const char **pct_str) tcg_regset_reset_reg(ct->u.regs, TCG_REG_O0); tcg_regset_reset_reg(ct->u.regs, TCG_REG_O1); tcg_regset_reset_reg(ct->u.regs, TCG_REG_O2); - tcg_regset_reset_reg(ct->u.regs, TCG_REG_O3); break; case 'I': ct->ct |= TCG_CT_CONST_S11; @@ -236,7 +251,7 @@ static inline int tcg_target_const_match(tcg_target_long val, #define ARITH_XOR (INSN_OP(2) | INSN_OP3(0x03)) #define ARITH_SUB (INSN_OP(2) | INSN_OP3(0x04)) #define ARITH_SUBCC (INSN_OP(2) | INSN_OP3(0x14)) -#define ARITH_ADDX (INSN_OP(2) | INSN_OP3(0x10)) +#define ARITH_ADDX (INSN_OP(2) | INSN_OP3(0x08)) #define ARITH_SUBX (INSN_OP(2) | INSN_OP3(0x0c)) #define ARITH_UMUL (INSN_OP(2) | INSN_OP3(0x0a)) #define ARITH_UDIV (INSN_OP(2) | INSN_OP3(0x0e)) @@ -288,6 +303,16 @@ static inline int tcg_target_const_match(tcg_target_long val, #define ASI_PRIMARY_LITTLE 0x88 #endif +#define LDUH_LE (LDUHA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define LDSH_LE (LDSHA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define LDUW_LE (LDUWA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define LDSW_LE (LDSWA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define LDX_LE (LDXA | INSN_ASI(ASI_PRIMARY_LITTLE)) + +#define STH_LE (STHA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define STW_LE (STWA | INSN_ASI(ASI_PRIMARY_LITTLE)) +#define STX_LE (STXA | INSN_ASI(ASI_PRIMARY_LITTLE)) + static inline void tcg_out_arith(TCGContext *s, int rd, int rs1, int rs2, int op) { @@ -353,71 +378,50 @@ static inline void tcg_out_movi(TCGContext *s, TCGType type, tcg_out_sethi(s, ret, ~arg); tcg_out_arithi(s, ret, ret, (arg & 0x3ff) | -0x400, ARITH_XOR); } else { - tcg_out_movi_imm32(s, TCG_REG_I4, arg >> (TCG_TARGET_REG_BITS / 2)); - tcg_out_arithi(s, TCG_REG_I4, TCG_REG_I4, 32, SHIFT_SLLX); - tcg_out_movi_imm32(s, ret, arg); - tcg_out_arith(s, ret, ret, TCG_REG_I4, ARITH_OR); + tcg_out_movi_imm32(s, ret, arg >> (TCG_TARGET_REG_BITS / 2)); + tcg_out_arithi(s, ret, ret, 32, SHIFT_SLLX); + tcg_out_movi_imm32(s, TCG_REG_T2, arg); + tcg_out_arith(s, ret, ret, TCG_REG_T2, ARITH_OR); } } -static inline void tcg_out_ld_raw(TCGContext *s, int ret, - tcg_target_long arg) -{ - tcg_out_sethi(s, ret, arg); - tcg_out32(s, LDUW | INSN_RD(ret) | INSN_RS1(ret) | - INSN_IMM13(arg & 0x3ff)); -} - -static inline void tcg_out_ld_ptr(TCGContext *s, int ret, - tcg_target_long arg) +static inline void tcg_out_ldst_rr(TCGContext *s, int data, int a1, + int a2, int op) { - if (!check_fit_tl(arg, 10)) - tcg_out_movi(s, TCG_TYPE_PTR, ret, arg & ~0x3ffULL); - if (TCG_TARGET_REG_BITS == 64) { - tcg_out32(s, LDX | INSN_RD(ret) | INSN_RS1(ret) | - INSN_IMM13(arg & 0x3ff)); - } else { - tcg_out32(s, LDUW | INSN_RD(ret) | INSN_RS1(ret) | - INSN_IMM13(arg & 0x3ff)); - } + tcg_out32(s, op | INSN_RD(data) | INSN_RS1(a1) | INSN_RS2(a2)); } -static inline void tcg_out_ldst(TCGContext *s, int ret, int addr, int offset, int op) +static inline void tcg_out_ldst(TCGContext *s, int ret, int addr, + int offset, int op) { - if (check_fit_tl(offset, 13)) + if (check_fit_tl(offset, 13)) { tcg_out32(s, op | INSN_RD(ret) | INSN_RS1(addr) | INSN_IMM13(offset)); - else { - tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_I5, offset); - tcg_out32(s, op | INSN_RD(ret) | INSN_RS1(TCG_REG_I5) | - INSN_RS2(addr)); + } else { + tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_T1, offset); + tcg_out_ldst_rr(s, ret, addr, TCG_REG_T1, op); } } -static inline void tcg_out_ldst_asi(TCGContext *s, int ret, int addr, - int offset, int op, int asi) -{ - tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_I5, offset); - tcg_out32(s, op | INSN_RD(ret) | INSN_RS1(TCG_REG_I5) | - INSN_ASI(asi) | INSN_RS2(addr)); -} - static inline void tcg_out_ld(TCGContext *s, TCGType type, TCGReg ret, TCGReg arg1, tcg_target_long arg2) { - if (type == TCG_TYPE_I32) - tcg_out_ldst(s, ret, arg1, arg2, LDUW); - else - tcg_out_ldst(s, ret, arg1, arg2, LDX); + tcg_out_ldst(s, ret, arg1, arg2, (type == TCG_TYPE_I32 ? LDUW : LDX)); } static inline void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg, TCGReg arg1, tcg_target_long arg2) { - if (type == TCG_TYPE_I32) - tcg_out_ldst(s, arg, arg1, arg2, STW); - else - tcg_out_ldst(s, arg, arg1, arg2, STX); + tcg_out_ldst(s, arg, arg1, arg2, (type == TCG_TYPE_I32 ? STW : STX)); +} + +static inline void tcg_out_ld_ptr(TCGContext *s, int ret, + tcg_target_long arg) +{ + if (!check_fit_tl(arg, 10)) { + tcg_out_movi(s, TCG_TYPE_PTR, ret, arg & ~0x3ff); + } + tcg_out_ld(s, TCG_TYPE_PTR, ret, ret, arg & 0x3ff); } static inline void tcg_out_sety(TCGContext *s, int rs) @@ -436,20 +440,21 @@ static inline void tcg_out_addi(TCGContext *s, int reg, tcg_target_long val) if (check_fit_tl(val, 13)) tcg_out_arithi(s, reg, reg, val, ARITH_ADD); else { - tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_I5, val); - tcg_out_arith(s, reg, reg, TCG_REG_I5, ARITH_ADD); + tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_T1, val); + tcg_out_arith(s, reg, reg, TCG_REG_T1, ARITH_ADD); } } } -static inline void tcg_out_andi(TCGContext *s, int reg, tcg_target_long val) +static inline void tcg_out_andi(TCGContext *s, int rd, int rs, + tcg_target_long val) { if (val != 0) { if (check_fit_tl(val, 13)) - tcg_out_arithi(s, reg, reg, val, ARITH_AND); + tcg_out_arithi(s, rd, rs, val, ARITH_AND); else { - tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_I5, val); - tcg_out_arith(s, reg, reg, TCG_REG_I5, ARITH_AND); + tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_T1, val); + tcg_out_arith(s, rd, rs, TCG_REG_T1, ARITH_AND); } } } @@ -461,8 +466,8 @@ static void tcg_out_div32(TCGContext *s, int rd, int rs1, if (uns) { tcg_out_sety(s, TCG_REG_G0); } else { - tcg_out_arithi(s, TCG_REG_I5, rs1, 31, SHIFT_SRA); - tcg_out_sety(s, TCG_REG_I5); + tcg_out_arithi(s, TCG_REG_T1, rs1, 31, SHIFT_SRA); + tcg_out_sety(s, TCG_REG_T1); } tcg_out_arithc(s, rd, rs1, val2, val2const, @@ -477,30 +482,33 @@ static inline void tcg_out_nop(TCGContext *s) static void tcg_out_branch_i32(TCGContext *s, int opc, int label_index) { TCGLabel *l = &s->labels[label_index]; + uint32_t off22; if (l->has_value) { - tcg_out32(s, (INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x2) - | INSN_OFF22(l->u.value - (unsigned long)s->code_ptr))); + off22 = INSN_OFF22(l->u.value - (unsigned long)s->code_ptr); } else { + /* Make sure to preserve destinations during retranslation. */ + off22 = *(uint32_t *)s->code_ptr & INSN_OFF22(-1); tcg_out_reloc(s, s->code_ptr, R_SPARC_WDISP22, label_index, 0); - tcg_out32(s, (INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x2) | 0)); } + tcg_out32(s, INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x2) | off22); } #if TCG_TARGET_REG_BITS == 64 static void tcg_out_branch_i64(TCGContext *s, int opc, int label_index) { TCGLabel *l = &s->labels[label_index]; + uint32_t off19; if (l->has_value) { - tcg_out32(s, (INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x1) | - (0x5 << 19) | - INSN_OFF19(l->u.value - (unsigned long)s->code_ptr))); + off19 = INSN_OFF19(l->u.value - (unsigned long)s->code_ptr); } else { + /* Make sure to preserve destinations during retranslation. */ + off19 = *(uint32_t *)s->code_ptr & INSN_OFF19(-1); tcg_out_reloc(s, s->code_ptr, R_SPARC_WDISP19, label_index, 0); - tcg_out32(s, (INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x1) | - (0x5 << 19) | 0)); } + tcg_out32(s, (INSN_OP(0) | INSN_COND(opc, 0) | INSN_OP2(0x1) | + (0x5 << 19) | off19)); } #endif @@ -608,8 +616,8 @@ static void tcg_out_setcond_i32(TCGContext *s, TCGCond cond, TCGArg ret, case TCG_COND_GTU: case TCG_COND_GEU: if (c2const && c2 != 0) { - tcg_out_movi_imm13(s, TCG_REG_I5, c2); - c2 = TCG_REG_I5; + tcg_out_movi_imm13(s, TCG_REG_T1, c2); + c2 = TCG_REG_T1; } t = c1, c1 = c2, c2 = t, c2const = 0; cond = tcg_swap_cond(cond); @@ -621,18 +629,10 @@ static void tcg_out_setcond_i32(TCGContext *s, TCGCond cond, TCGArg ret, default: tcg_out_cmp(s, c1, c2, c2const); -#if defined(__sparc_v9__) || defined(__sparc_v8plus__) - tcg_out_movi_imm13(s, ret, 0); - tcg_out32 (s, ARITH_MOVCC | INSN_RD(ret) - | INSN_RS1(tcg_cond_to_bcond[cond]) - | MOVCC_ICC | INSN_IMM11(1)); -#else - t = gen_new_label(); - tcg_out_branch_i32(s, INSN_COND(tcg_cond_to_bcond[cond], 1), t); - tcg_out_movi_imm13(s, ret, 1); tcg_out_movi_imm13(s, ret, 0); - tcg_out_label(s, t, s->code_ptr); -#endif + tcg_out32(s, ARITH_MOVCC | INSN_RD(ret) + | INSN_RS1(tcg_cond_to_bcond[cond]) + | MOVCC_ICC | INSN_IMM11(1)); return; } @@ -664,15 +664,15 @@ static void tcg_out_setcond2_i32(TCGContext *s, TCGCond cond, TCGArg ret, switch (cond) { case TCG_COND_EQ: - tcg_out_setcond_i32(s, TCG_COND_EQ, TCG_REG_I5, al, bl, blconst); + tcg_out_setcond_i32(s, TCG_COND_EQ, TCG_REG_T1, al, bl, blconst); tcg_out_setcond_i32(s, TCG_COND_EQ, ret, ah, bh, bhconst); - tcg_out_arith(s, ret, ret, TCG_REG_I5, ARITH_AND); + tcg_out_arith(s, ret, ret, TCG_REG_T1, ARITH_AND); break; case TCG_COND_NE: - tcg_out_setcond_i32(s, TCG_COND_NE, TCG_REG_I5, al, al, blconst); + tcg_out_setcond_i32(s, TCG_COND_NE, TCG_REG_T1, al, al, blconst); tcg_out_setcond_i32(s, TCG_COND_NE, ret, ah, bh, bhconst); - tcg_out_arith(s, ret, ret, TCG_REG_I5, ARITH_OR); + tcg_out_arith(s, ret, ret, TCG_REG_T1, ARITH_OR); break; default: @@ -695,14 +695,36 @@ static void tcg_out_setcond2_i32(TCGContext *s, TCGCond cond, TCGArg ret, /* Generate global QEMU prologue and epilogue code */ static void tcg_target_qemu_prologue(TCGContext *s) { - tcg_set_frame(s, TCG_REG_I6, TCG_TARGET_CALL_STACK_OFFSET, - CPU_TEMP_BUF_NLONGS * (int)sizeof(long)); + int tmp_buf_size, frame_size; + + /* The TCG temp buffer is at the top of the frame, immediately + below the frame pointer. */ + tmp_buf_size = CPU_TEMP_BUF_NLONGS * (int)sizeof(long); + tcg_set_frame(s, TCG_REG_I6, TCG_TARGET_STACK_BIAS - tmp_buf_size, + tmp_buf_size); + + /* TCG_TARGET_CALL_STACK_OFFSET includes the stack bias, but is + otherwise the minimal frame usable by callees. */ + frame_size = TCG_TARGET_CALL_STACK_OFFSET - TCG_TARGET_STACK_BIAS; + frame_size += TCG_STATIC_CALL_ARGS_SIZE + tmp_buf_size; + frame_size += TCG_TARGET_STACK_ALIGN - 1; + frame_size &= -TCG_TARGET_STACK_ALIGN; tcg_out32(s, SAVE | INSN_RD(TCG_REG_O6) | INSN_RS1(TCG_REG_O6) | - INSN_IMM13(-(TCG_TARGET_STACK_MINFRAME + - CPU_TEMP_BUF_NLONGS * (int)sizeof(long)))); + INSN_IMM13(-frame_size)); + +#ifdef CONFIG_USE_GUEST_BASE + if (GUEST_BASE != 0) { + tcg_out_movi(s, TCG_TYPE_PTR, TCG_GUEST_BASE_REG, GUEST_BASE); + tcg_regset_set_reg(s->reserved_regs, TCG_GUEST_BASE_REG); + } +#endif + tcg_out32(s, JMPL | INSN_RD(TCG_REG_G0) | INSN_RS1(TCG_REG_I1) | INSN_RS2(TCG_REG_G0)); - tcg_out_mov(s, TCG_TYPE_PTR, TCG_AREG0, TCG_REG_I0); + /* delay slot */ + tcg_out_nop(s); + + /* No epilogue required. We issue ret + restore directly in the TB. */ } #if defined(CONFIG_SOFTMMU) @@ -726,418 +748,309 @@ static const void * const qemu_st_helpers[4] = { helper_stl_mmu, helper_stq_mmu, }; -#endif -#if TARGET_LONG_BITS == 32 -#define TARGET_LD_OP LDUW -#else -#define TARGET_LD_OP LDX -#endif +/* Perform the TLB load and compare. -#if defined(CONFIG_SOFTMMU) -#if HOST_LONG_BITS == 32 -#define TARGET_ADDEND_LD_OP LDUW + Inputs: + ADDRLO_IDX contains the index into ARGS of the low part of the + address; the high part of the address is at ADDR_LOW_IDX+1. + + MEM_INDEX and S_BITS are the memory context and log2 size of the load. + + WHICH is the offset into the CPUTLBEntry structure of the slot to read. + This should be offsetof addr_read or addr_write. + + The result of the TLB comparison is in %[ix]cc. The sanitized address + is in the returned register, maybe %o0. The TLB addend is in %o1. */ + +static int tcg_out_tlb_load(TCGContext *s, int addrlo_idx, int mem_index, + int s_bits, const TCGArg *args, int which) +{ + const int addrlo = args[addrlo_idx]; + const int r0 = TCG_REG_O0; + const int r1 = TCG_REG_O1; + const int r2 = TCG_REG_O2; + int addr = addrlo; + int tlb_ofs; + + if (TCG_TARGET_REG_BITS == 32 && TARGET_LONG_BITS == 64) { + /* Assemble the 64-bit address in R0. */ + tcg_out_arithi(s, r0, addrlo, 0, SHIFT_SRL); + tcg_out_arithi(s, r1, args[addrlo_idx + 1], 32, SHIFT_SLLX); + tcg_out_arith(s, r0, r0, r1, ARITH_OR); + } + + /* Shift the page number down to tlb-entry. */ + tcg_out_arithi(s, r1, addrlo, + TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS, SHIFT_SRL); + + /* Mask out the page offset, except for the required alignment. */ + tcg_out_andi(s, r0, addr, TARGET_PAGE_MASK | ((1 << s_bits) - 1)); + + /* Compute tlb index, modulo tlb size. */ + tcg_out_andi(s, r1, r1, (CPU_TLB_SIZE - 1) << CPU_TLB_ENTRY_BITS); + + /* Relative to the current ENV. */ + tcg_out_arith(s, r1, TCG_AREG0, r1, ARITH_ADD); + + /* Find a base address that can load both tlb comparator and addend. */ + tlb_ofs = offsetof(CPUArchState, tlb_table[mem_index][0]); + if (!check_fit_tl(tlb_ofs + sizeof(CPUTLBEntry), 13)) { + tcg_out_addi(s, r1, tlb_ofs); + tlb_ofs = 0; + } + + /* Load the tlb comparator and the addend. */ + tcg_out_ld(s, TCG_TYPE_TL, r2, r1, tlb_ofs + which); + tcg_out_ld(s, TCG_TYPE_PTR, r1, r1, tlb_ofs+offsetof(CPUTLBEntry, addend)); + + /* subcc arg0, arg2, %g0 */ + tcg_out_cmp(s, r0, r2, 0); + + /* If the guest address must be zero-extended, do so now. */ + if (TCG_TARGET_REG_BITS == 64 && TARGET_LONG_BITS == 32) { + tcg_out_arithi(s, r0, addrlo, 0, SHIFT_SRL); + return r0; + } + return addrlo; +} +#endif /* CONFIG_SOFTMMU */ + +static const int qemu_ld_opc[8] = { +#ifdef TARGET_WORDS_BIGENDIAN + LDUB, LDUH, LDUW, LDX, LDSB, LDSH, LDSW, LDX #else -#define TARGET_ADDEND_LD_OP LDX -#endif + LDUB, LDUH_LE, LDUW_LE, LDX_LE, LDSB, LDSH_LE, LDSW_LE, LDX_LE #endif +}; -#ifdef __arch64__ -#define HOST_LD_OP LDX -#define HOST_ST_OP STX -#define HOST_SLL_OP SHIFT_SLLX -#define HOST_SRA_OP SHIFT_SRAX +static const int qemu_st_opc[4] = { +#ifdef TARGET_WORDS_BIGENDIAN + STB, STH, STW, STX #else -#define HOST_LD_OP LDUW -#define HOST_ST_OP STW -#define HOST_SLL_OP SHIFT_SLL -#define HOST_SRA_OP SHIFT_SRA + STB, STH_LE, STW_LE, STX_LE #endif +}; -static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, - int opc) +static void tcg_out_qemu_ld(TCGContext *s, const TCGArg *args, int sizeop) { - int addr_reg, data_reg, arg0, arg1, arg2, mem_index, s_bits; + int addrlo_idx = 1, datalo, datahi, addr_reg; #if defined(CONFIG_SOFTMMU) - uint32_t *label1_ptr, *label2_ptr; + int memi_idx, memi, s_bits, n; + uint32_t *label_ptr[2]; #endif - data_reg = *args++; - addr_reg = *args++; - mem_index = *args; - s_bits = opc & 3; - - arg0 = TCG_REG_O0; - arg1 = TCG_REG_O1; - arg2 = TCG_REG_O2; + datahi = datalo = args[0]; + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + datahi = args[1]; + addrlo_idx = 2; + } #if defined(CONFIG_SOFTMMU) - /* srl addr_reg, x, arg1 */ - tcg_out_arithi(s, arg1, addr_reg, TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS, - SHIFT_SRL); - /* and addr_reg, x, arg0 */ - tcg_out_arithi(s, arg0, addr_reg, TARGET_PAGE_MASK | ((1 << s_bits) - 1), - ARITH_AND); - - /* and arg1, x, arg1 */ - tcg_out_andi(s, arg1, (CPU_TLB_SIZE - 1) << CPU_TLB_ENTRY_BITS); + memi_idx = addrlo_idx + 1 + (TARGET_LONG_BITS > TCG_TARGET_REG_BITS); + memi = args[memi_idx]; + s_bits = sizeop & 3; + + addr_reg = tcg_out_tlb_load(s, addrlo_idx, memi, s_bits, args, + offsetof(CPUTLBEntry, addr_read)); + + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + int reg64; + + /* bne,pn %[xi]cc, label0 */ + label_ptr[0] = (uint32_t *)s->code_ptr; + tcg_out32(s, (INSN_OP(0) | INSN_COND(COND_NE, 0) | INSN_OP2(0x1) + | ((TARGET_LONG_BITS == 64) << 21))); + + /* TLB Hit. */ + /* Load all 64-bits into an O/G register. */ + reg64 = (datalo < 16 ? datalo : TCG_REG_O0); + tcg_out_ldst_rr(s, reg64, addr_reg, TCG_REG_O1, qemu_ld_opc[sizeop]); + + /* Move the two 32-bit pieces into the destination registers. */ + tcg_out_arithi(s, datahi, reg64, 32, SHIFT_SRLX); + if (reg64 != datalo) { + tcg_out_mov(s, TCG_TYPE_I32, datalo, reg64); + } - /* add arg1, x, arg1 */ - tcg_out_addi(s, arg1, offsetof(CPUArchState, - tlb_table[mem_index][0].addr_read)); + /* b,a,pt label1 */ + label_ptr[1] = (uint32_t *)s->code_ptr; + tcg_out32(s, (INSN_OP(0) | INSN_COND(COND_A, 0) | INSN_OP2(0x1) + | (1 << 29) | (1 << 19))); + } else { + /* The fast path is exactly one insn. Thus we can perform the + entire TLB Hit in the (annulled) delay slot of the branch + over the TLB Miss case. */ + + /* beq,a,pt %[xi]cc, label0 */ + label_ptr[0] = NULL; + label_ptr[1] = (uint32_t *)s->code_ptr; + tcg_out32(s, (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x1) + | ((TARGET_LONG_BITS == 64) << 21) + | (1 << 29) | (1 << 19))); + /* delay slot */ + tcg_out_ldst_rr(s, datalo, addr_reg, TCG_REG_O1, qemu_ld_opc[sizeop]); + } - /* add env, arg1, arg1 */ - tcg_out_arith(s, arg1, TCG_AREG0, arg1, ARITH_ADD); + /* TLB Miss. */ - /* ld [arg1], arg2 */ - tcg_out32(s, TARGET_LD_OP | INSN_RD(arg2) | INSN_RS1(arg1) | - INSN_RS2(TCG_REG_G0)); + if (label_ptr[0]) { + *label_ptr[0] |= INSN_OFF19((unsigned long)s->code_ptr - + (unsigned long)label_ptr[0]); + } + n = 0; + tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[n++], TCG_AREG0); + if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) { + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], + args[addrlo_idx + 1]); + } + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], + args[addrlo_idx]); - /* subcc arg0, arg2, %g0 */ - tcg_out_arith(s, TCG_REG_G0, arg0, arg2, ARITH_SUBCC); - - /* will become: - be label1 - or - be,pt %xcc label1 */ - label1_ptr = (uint32_t *)s->code_ptr; - tcg_out32(s, 0); - - /* mov (delay slot) */ - tcg_out_mov(s, TCG_TYPE_PTR, arg0, addr_reg); - - /* mov */ - tcg_out_movi(s, TCG_TYPE_I32, arg1, mem_index); - /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[3], - tcg_target_call_iarg_regs[2]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_TL, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], - TCG_AREG0); - - /* XXX: move that code at the end of the TB */ /* qemu_ld_helper[s_bits](arg0, arg1) */ tcg_out32(s, CALL | ((((tcg_target_ulong)qemu_ld_helpers[s_bits] - (tcg_target_ulong)s->code_ptr) >> 2) & 0x3fffffff)); - /* Store AREG0 in stack to avoid ugly glibc bugs that mangle - global registers */ - // delay slot - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_ST_OP); - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_LD_OP); - - /* data_reg = sign_extend(arg0) */ - switch(opc) { + /* delay slot */ + tcg_out_movi(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[n], memi); + + n = tcg_target_call_oarg_regs[0]; + /* datalo = sign_extend(arg0) */ + switch (sizeop) { case 0 | 4: - /* sll arg0, 24/56, data_reg */ - tcg_out_arithi(s, data_reg, arg0, (int)sizeof(tcg_target_long) * 8 - 8, - HOST_SLL_OP); - /* sra data_reg, 24/56, data_reg */ - tcg_out_arithi(s, data_reg, data_reg, - (int)sizeof(tcg_target_long) * 8 - 8, HOST_SRA_OP); + /* Recall that SRA sign extends from bit 31 through bit 63. */ + tcg_out_arithi(s, datalo, n, 24, SHIFT_SLL); + tcg_out_arithi(s, datalo, datalo, 24, SHIFT_SRA); break; case 1 | 4: - /* sll arg0, 16/48, data_reg */ - tcg_out_arithi(s, data_reg, arg0, - (int)sizeof(tcg_target_long) * 8 - 16, HOST_SLL_OP); - /* sra data_reg, 16/48, data_reg */ - tcg_out_arithi(s, data_reg, data_reg, - (int)sizeof(tcg_target_long) * 8 - 16, HOST_SRA_OP); + tcg_out_arithi(s, datalo, n, 16, SHIFT_SLL); + tcg_out_arithi(s, datalo, datalo, 16, SHIFT_SRA); break; case 2 | 4: - /* sll arg0, 32, data_reg */ - tcg_out_arithi(s, data_reg, arg0, 32, HOST_SLL_OP); - /* sra data_reg, 32, data_reg */ - tcg_out_arithi(s, data_reg, data_reg, 32, HOST_SRA_OP); + tcg_out_arithi(s, datalo, n, 0, SHIFT_SRA); break; + case 3: + if (TCG_TARGET_REG_BITS == 32) { + tcg_out_mov(s, TCG_TYPE_REG, datahi, n); + tcg_out_mov(s, TCG_TYPE_REG, datalo, n + 1); + break; + } + /* FALLTHRU */ case 0: case 1: case 2: - case 3: default: /* mov */ - tcg_out_mov(s, TCG_TYPE_REG, data_reg, arg0); + tcg_out_mov(s, TCG_TYPE_REG, datalo, n); break; } - /* will become: - ba label2 */ - label2_ptr = (uint32_t *)s->code_ptr; - tcg_out32(s, 0); - - /* nop (delay slot */ - tcg_out_nop(s); - - /* label1: */ -#if TARGET_LONG_BITS == 32 - /* be label1 */ - *label1_ptr = (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x2) | - INSN_OFF22((unsigned long)s->code_ptr - - (unsigned long)label1_ptr)); -#else - /* be,pt %xcc label1 */ - *label1_ptr = (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x1) | - (0x5 << 19) | INSN_OFF19((unsigned long)s->code_ptr - - (unsigned long)label1_ptr)); -#endif - - /* ld [arg1 + x], arg1 */ - tcg_out_ldst(s, arg1, arg1, offsetof(CPUTLBEntry, addend) - - offsetof(CPUTLBEntry, addr_read), TARGET_ADDEND_LD_OP); - -#if TARGET_LONG_BITS == 32 - /* and addr_reg, x, arg0 */ - tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_I5, 0xffffffff); - tcg_out_arith(s, arg0, addr_reg, TCG_REG_I5, ARITH_AND); - /* add arg0, arg1, arg0 */ - tcg_out_arith(s, arg0, arg0, arg1, ARITH_ADD); + *label_ptr[1] |= INSN_OFF19((unsigned long)s->code_ptr - + (unsigned long)label_ptr[1]); #else - /* add addr_reg, arg1, arg0 */ - tcg_out_arith(s, arg0, addr_reg, arg1, ARITH_ADD); -#endif + addr_reg = args[addrlo_idx]; + if (TCG_TARGET_REG_BITS == 64 && TARGET_LONG_BITS == 32) { + tcg_out_arithi(s, TCG_REG_T1, addr_reg, 0, SHIFT_SRL); + addr_reg = TCG_REG_T1; + } + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + int reg64 = (datalo < 16 ? datalo : TCG_REG_O0); -#else - arg0 = addr_reg; -#endif + tcg_out_ldst_rr(s, reg64, addr_reg, + (GUEST_BASE ? TCG_GUEST_BASE_REG : TCG_REG_G0), + qemu_ld_opc[sizeop]); - switch(opc) { - case 0: - /* ldub [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDUB); - break; - case 0 | 4: - /* ldsb [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDSB); - break; - case 1: -#ifdef TARGET_WORDS_BIGENDIAN - /* lduh [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDUH); -#else - /* lduha [arg0] ASI_PRIMARY_LITTLE, data_reg */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, LDUHA, ASI_PRIMARY_LITTLE); -#endif - break; - case 1 | 4: -#ifdef TARGET_WORDS_BIGENDIAN - /* ldsh [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDSH); -#else - /* ldsha [arg0] ASI_PRIMARY_LITTLE, data_reg */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, LDSHA, ASI_PRIMARY_LITTLE); -#endif - break; - case 2: -#ifdef TARGET_WORDS_BIGENDIAN - /* lduw [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDUW); -#else - /* lduwa [arg0] ASI_PRIMARY_LITTLE, data_reg */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, LDUWA, ASI_PRIMARY_LITTLE); -#endif - break; - case 2 | 4: -#ifdef TARGET_WORDS_BIGENDIAN - /* ldsw [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDSW); -#else - /* ldswa [arg0] ASI_PRIMARY_LITTLE, data_reg */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, LDSWA, ASI_PRIMARY_LITTLE); -#endif - break; - case 3: -#ifdef TARGET_WORDS_BIGENDIAN - /* ldx [arg0], data_reg */ - tcg_out_ldst(s, data_reg, arg0, 0, LDX); -#else - /* ldxa [arg0] ASI_PRIMARY_LITTLE, data_reg */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, LDXA, ASI_PRIMARY_LITTLE); -#endif - break; - default: - tcg_abort(); + tcg_out_arithi(s, datahi, reg64, 32, SHIFT_SRLX); + if (reg64 != datalo) { + tcg_out_mov(s, TCG_TYPE_I32, datalo, reg64); + } + } else { + tcg_out_ldst_rr(s, datalo, addr_reg, + (GUEST_BASE ? TCG_GUEST_BASE_REG : TCG_REG_G0), + qemu_ld_opc[sizeop]); } - -#if defined(CONFIG_SOFTMMU) - /* label2: */ - *label2_ptr = (INSN_OP(0) | INSN_COND(COND_A, 0) | INSN_OP2(0x2) | - INSN_OFF22((unsigned long)s->code_ptr - - (unsigned long)label2_ptr)); -#endif +#endif /* CONFIG_SOFTMMU */ } -static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, - int opc) +static void tcg_out_qemu_st(TCGContext *s, const TCGArg *args, int sizeop) { - int addr_reg, data_reg, arg0, arg1, arg2, mem_index, s_bits; + int addrlo_idx = 1, datalo, datahi, addr_reg; #if defined(CONFIG_SOFTMMU) - uint32_t *label1_ptr, *label2_ptr; + int memi_idx, memi, n; + uint32_t *label_ptr; #endif - data_reg = *args++; - addr_reg = *args++; - mem_index = *args; - - s_bits = opc; - - arg0 = TCG_REG_O0; - arg1 = TCG_REG_O1; - arg2 = TCG_REG_O2; + datahi = datalo = args[0]; + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + datahi = args[1]; + addrlo_idx = 2; + } #if defined(CONFIG_SOFTMMU) - /* srl addr_reg, x, arg1 */ - tcg_out_arithi(s, arg1, addr_reg, TARGET_PAGE_BITS - CPU_TLB_ENTRY_BITS, - SHIFT_SRL); - - /* and addr_reg, x, arg0 */ - tcg_out_arithi(s, arg0, addr_reg, TARGET_PAGE_MASK | ((1 << s_bits) - 1), - ARITH_AND); - - /* and arg1, x, arg1 */ - tcg_out_andi(s, arg1, (CPU_TLB_SIZE - 1) << CPU_TLB_ENTRY_BITS); - - /* add arg1, x, arg1 */ - tcg_out_addi(s, arg1, offsetof(CPUArchState, - tlb_table[mem_index][0].addr_write)); - - /* add env, arg1, arg1 */ - tcg_out_arith(s, arg1, TCG_AREG0, arg1, ARITH_ADD); + memi_idx = addrlo_idx + 1 + (TARGET_LONG_BITS > TCG_TARGET_REG_BITS); + memi = args[memi_idx]; + + addr_reg = tcg_out_tlb_load(s, addrlo_idx, memi, sizeop, args, + offsetof(CPUTLBEntry, addr_write)); + + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + /* Reconstruct the full 64-bit value. */ + tcg_out_arithi(s, TCG_REG_T1, datalo, 0, SHIFT_SRL); + tcg_out_arithi(s, TCG_REG_O2, datahi, 32, SHIFT_SLLX); + tcg_out_arith(s, TCG_REG_O2, TCG_REG_T1, TCG_REG_O2, ARITH_OR); + datalo = TCG_REG_O2; + } - /* ld [arg1], arg2 */ - tcg_out32(s, TARGET_LD_OP | INSN_RD(arg2) | INSN_RS1(arg1) | - INSN_RS2(TCG_REG_G0)); + /* The fast path is exactly one insn. Thus we can perform the entire + TLB Hit in the (annulled) delay slot of the branch over TLB Miss. */ + /* beq,a,pt %[xi]cc, label0 */ + label_ptr = (uint32_t *)s->code_ptr; + tcg_out32(s, (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x1) + | ((TARGET_LONG_BITS == 64) << 21) + | (1 << 29) | (1 << 19))); + /* delay slot */ + tcg_out_ldst_rr(s, datalo, addr_reg, TCG_REG_O1, qemu_st_opc[sizeop]); + + /* TLB Miss. */ + + n = 0; + tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[n++], TCG_AREG0); + if (TARGET_LONG_BITS > TCG_TARGET_REG_BITS) { + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], + args[addrlo_idx + 1]); + } + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], + args[addrlo_idx]); + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], datahi); + } + tcg_out_mov(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n++], datalo); - /* subcc arg0, arg2, %g0 */ - tcg_out_arith(s, TCG_REG_G0, arg0, arg2, ARITH_SUBCC); - - /* will become: - be label1 - or - be,pt %xcc label1 */ - label1_ptr = (uint32_t *)s->code_ptr; - tcg_out32(s, 0); - - /* mov (delay slot) */ - tcg_out_mov(s, TCG_TYPE_PTR, arg0, addr_reg); - - /* mov */ - tcg_out_mov(s, TCG_TYPE_REG, arg1, data_reg); - - /* mov */ - tcg_out_movi(s, TCG_TYPE_I32, arg2, mem_index); - - /* XXX/FIXME: suboptimal */ - tcg_out_mov(s, TCG_TYPE_I32, tcg_target_call_iarg_regs[3], - tcg_target_call_iarg_regs[2]); - tcg_out_mov(s, TCG_TYPE_I64, tcg_target_call_iarg_regs[2], - tcg_target_call_iarg_regs[1]); - tcg_out_mov(s, TCG_TYPE_TL, tcg_target_call_iarg_regs[1], - tcg_target_call_iarg_regs[0]); - tcg_out_mov(s, TCG_TYPE_PTR, tcg_target_call_iarg_regs[0], - TCG_AREG0); - /* XXX: move that code at the end of the TB */ /* qemu_st_helper[s_bits](arg0, arg1, arg2) */ - tcg_out32(s, CALL | ((((tcg_target_ulong)qemu_st_helpers[s_bits] + tcg_out32(s, CALL | ((((tcg_target_ulong)qemu_st_helpers[sizeop] - (tcg_target_ulong)s->code_ptr) >> 2) & 0x3fffffff)); - /* Store AREG0 in stack to avoid ugly glibc bugs that mangle - global registers */ - // delay slot - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_ST_OP); - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_LD_OP); - - /* will become: - ba label2 */ - label2_ptr = (uint32_t *)s->code_ptr; - tcg_out32(s, 0); - - /* nop (delay slot) */ - tcg_out_nop(s); - -#if TARGET_LONG_BITS == 32 - /* be label1 */ - *label1_ptr = (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x2) | - INSN_OFF22((unsigned long)s->code_ptr - - (unsigned long)label1_ptr)); -#else - /* be,pt %xcc label1 */ - *label1_ptr = (INSN_OP(0) | INSN_COND(COND_E, 0) | INSN_OP2(0x1) | - (0x5 << 19) | INSN_OFF19((unsigned long)s->code_ptr - - (unsigned long)label1_ptr)); -#endif - - /* ld [arg1 + x], arg1 */ - tcg_out_ldst(s, arg1, arg1, offsetof(CPUTLBEntry, addend) - - offsetof(CPUTLBEntry, addr_write), TARGET_ADDEND_LD_OP); - -#if TARGET_LONG_BITS == 32 - /* and addr_reg, x, arg0 */ - tcg_out_movi(s, TCG_TYPE_I32, TCG_REG_I5, 0xffffffff); - tcg_out_arith(s, arg0, addr_reg, TCG_REG_I5, ARITH_AND); - /* add arg0, arg1, arg0 */ - tcg_out_arith(s, arg0, arg0, arg1, ARITH_ADD); -#else - /* add addr_reg, arg1, arg0 */ - tcg_out_arith(s, arg0, addr_reg, arg1, ARITH_ADD); -#endif + /* delay slot */ + tcg_out_movi(s, TCG_TYPE_REG, tcg_target_call_iarg_regs[n], memi); + *label_ptr |= INSN_OFF19((unsigned long)s->code_ptr - + (unsigned long)label_ptr); #else - arg0 = addr_reg; -#endif - - switch(opc) { - case 0: - /* stb data_reg, [arg0] */ - tcg_out_ldst(s, data_reg, arg0, 0, STB); - break; - case 1: -#ifdef TARGET_WORDS_BIGENDIAN - /* sth data_reg, [arg0] */ - tcg_out_ldst(s, data_reg, arg0, 0, STH); -#else - /* stha data_reg, [arg0] ASI_PRIMARY_LITTLE */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, STHA, ASI_PRIMARY_LITTLE); -#endif - break; - case 2: -#ifdef TARGET_WORDS_BIGENDIAN - /* stw data_reg, [arg0] */ - tcg_out_ldst(s, data_reg, arg0, 0, STW); -#else - /* stwa data_reg, [arg0] ASI_PRIMARY_LITTLE */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, STWA, ASI_PRIMARY_LITTLE); -#endif - break; - case 3: -#ifdef TARGET_WORDS_BIGENDIAN - /* stx data_reg, [arg0] */ - tcg_out_ldst(s, data_reg, arg0, 0, STX); -#else - /* stxa data_reg, [arg0] ASI_PRIMARY_LITTLE */ - tcg_out_ldst_asi(s, data_reg, arg0, 0, STXA, ASI_PRIMARY_LITTLE); -#endif - break; - default: - tcg_abort(); + addr_reg = args[addrlo_idx]; + if (TCG_TARGET_REG_BITS == 64 && TARGET_LONG_BITS == 32) { + tcg_out_arithi(s, TCG_REG_T1, addr_reg, 0, SHIFT_SRL); + addr_reg = TCG_REG_T1; } - -#if defined(CONFIG_SOFTMMU) - /* label2: */ - *label2_ptr = (INSN_OP(0) | INSN_COND(COND_A, 0) | INSN_OP2(0x2) | - INSN_OFF22((unsigned long)s->code_ptr - - (unsigned long)label2_ptr)); -#endif + if (TCG_TARGET_REG_BITS == 32 && sizeop == 3) { + tcg_out_arithi(s, TCG_REG_T1, datalo, 0, SHIFT_SRL); + tcg_out_arithi(s, TCG_REG_O2, datahi, 32, SHIFT_SLLX); + tcg_out_arith(s, TCG_REG_O2, TCG_REG_T1, TCG_REG_O2, ARITH_OR); + datalo = TCG_REG_O2; + } + tcg_out_ldst_rr(s, datalo, addr_reg, + (GUEST_BASE ? TCG_GUEST_BASE_REG : TCG_REG_G0), + qemu_st_opc[sizeop]); +#endif /* CONFIG_SOFTMMU */ } static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, @@ -1156,39 +1069,33 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, case INDEX_op_goto_tb: if (s->tb_jmp_offset) { /* direct jump method */ - tcg_out_sethi(s, TCG_REG_I5, args[0] & 0xffffe000); - tcg_out32(s, JMPL | INSN_RD(TCG_REG_G0) | INSN_RS1(TCG_REG_I5) | - INSN_IMM13((args[0] & 0x1fff))); + uint32_t old_insn = *(uint32_t *)s->code_ptr; s->tb_jmp_offset[args[0]] = s->code_ptr - s->code_buf; + /* Make sure to preserve links during retranslation. */ + tcg_out32(s, CALL | (old_insn & ~INSN_OP(-1))); } else { /* indirect jump method */ - tcg_out_ld_ptr(s, TCG_REG_I5, (tcg_target_long)(s->tb_next + args[0])); - tcg_out32(s, JMPL | INSN_RD(TCG_REG_G0) | INSN_RS1(TCG_REG_I5) | + tcg_out_ld_ptr(s, TCG_REG_T1, + (tcg_target_long)(s->tb_next + args[0])); + tcg_out32(s, JMPL | INSN_RD(TCG_REG_G0) | INSN_RS1(TCG_REG_T1) | INSN_RS2(TCG_REG_G0)); } tcg_out_nop(s); s->tb_next_offset[args[0]] = s->code_ptr - s->code_buf; break; case INDEX_op_call: - if (const_args[0]) + if (const_args[0]) { tcg_out32(s, CALL | ((((tcg_target_ulong)args[0] - (tcg_target_ulong)s->code_ptr) >> 2) & 0x3fffffff)); - else { - tcg_out_ld_ptr(s, TCG_REG_I5, + } else { + tcg_out_ld_ptr(s, TCG_REG_T1, (tcg_target_long)(s->tb_next + args[0])); - tcg_out32(s, JMPL | INSN_RD(TCG_REG_O7) | INSN_RS1(TCG_REG_I5) | + tcg_out32(s, JMPL | INSN_RD(TCG_REG_O7) | INSN_RS1(TCG_REG_T1) | INSN_RS2(TCG_REG_G0)); } - /* Store AREG0 in stack to avoid ugly glibc bugs that mangle - global registers */ - // delay slot - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_ST_OP); - tcg_out_ldst(s, TCG_AREG0, TCG_REG_CALL_STACK, - TCG_TARGET_CALL_STACK_OFFSET - TCG_STATIC_CALL_ARGS_SIZE - - sizeof(long), HOST_LD_OP); + /* delay slot */ + tcg_out_nop(s); break; case INDEX_op_jmp: case INDEX_op_br: @@ -1260,13 +1167,16 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, goto gen_arith; case INDEX_op_shl_i32: c = SHIFT_SLL; - goto gen_arith; + do_shift32: + /* Limit immediate shift count lest we create an illegal insn. */ + tcg_out_arithc(s, args[0], args[1], args[2] & 31, const_args[2], c); + break; case INDEX_op_shr_i32: c = SHIFT_SRL; - goto gen_arith; + goto do_shift32; case INDEX_op_sar_i32: c = SHIFT_SRA; - goto gen_arith; + goto do_shift32; case INDEX_op_mul_i32: c = ARITH_UMUL; goto gen_arith; @@ -1287,11 +1197,11 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, case INDEX_op_rem_i32: case INDEX_op_remu_i32: - tcg_out_div32(s, TCG_REG_I5, args[1], args[2], const_args[2], + tcg_out_div32(s, TCG_REG_T1, args[1], args[2], const_args[2], opc == INDEX_op_remu_i32); - tcg_out_arithc(s, TCG_REG_I5, TCG_REG_I5, args[2], const_args[2], + tcg_out_arithc(s, TCG_REG_T1, TCG_REG_T1, args[2], const_args[2], ARITH_UMUL); - tcg_out_arith(s, args[0], args[1], TCG_REG_I5, ARITH_SUB); + tcg_out_arith(s, args[0], args[1], TCG_REG_T1, ARITH_SUB); break; case INDEX_op_brcond_i32: @@ -1356,6 +1266,9 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, tcg_out_qemu_ld(s, args, 2 | 4); break; #endif + case INDEX_op_qemu_ld64: + tcg_out_qemu_ld(s, args, 3); + break; case INDEX_op_qemu_st8: tcg_out_qemu_st(s, args, 0); break; @@ -1365,6 +1278,9 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, case INDEX_op_qemu_st32: tcg_out_qemu_st(s, args, 2); break; + case INDEX_op_qemu_st64: + tcg_out_qemu_st(s, args, 3); + break; #if TCG_TARGET_REG_BITS == 64 case INDEX_op_movi_i64: @@ -1381,13 +1297,16 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, break; case INDEX_op_shl_i64: c = SHIFT_SLLX; - goto gen_arith; + do_shift64: + /* Limit immediate shift count lest we create an illegal insn. */ + tcg_out_arithc(s, args[0], args[1], args[2] & 63, const_args[2], c); + break; case INDEX_op_shr_i64: c = SHIFT_SRLX; - goto gen_arith; + goto do_shift64; case INDEX_op_sar_i64: c = SHIFT_SRAX; - goto gen_arith; + goto do_shift64; case INDEX_op_mul_i64: c = ARITH_MULX; goto gen_arith; @@ -1399,11 +1318,11 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, goto gen_arith; case INDEX_op_rem_i64: case INDEX_op_remu_i64: - tcg_out_arithc(s, TCG_REG_I5, args[1], args[2], const_args[2], + tcg_out_arithc(s, TCG_REG_T1, args[1], args[2], const_args[2], opc == INDEX_op_rem_i64 ? ARITH_SDIVX : ARITH_UDIVX); - tcg_out_arithc(s, TCG_REG_I5, TCG_REG_I5, args[2], const_args[2], + tcg_out_arithc(s, TCG_REG_T1, TCG_REG_T1, args[2], const_args[2], ARITH_MULX); - tcg_out_arith(s, args[0], args[1], TCG_REG_I5, ARITH_SUB); + tcg_out_arith(s, args[0], args[1], TCG_REG_T1, ARITH_SUB); break; case INDEX_op_ext32s_i64: if (const_args[1]) { @@ -1429,13 +1348,6 @@ static inline void tcg_out_op(TCGContext *s, TCGOpcode opc, const TCGArg *args, args[2], const_args[2]); break; - case INDEX_op_qemu_ld64: - tcg_out_qemu_ld(s, args, 3); - break; - case INDEX_op_qemu_st64: - tcg_out_qemu_st(s, args, 3); - break; - #endif gen_arith: tcg_out_arithc(s, args[0], args[1], args[2], const_args[2], c); @@ -1500,20 +1412,6 @@ static const TCGTargetOpDef sparc_op_defs[] = { { INDEX_op_mulu2_i32, { "r", "r", "r", "rJ" } }, #endif - { 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_st8, { "L", "L" } }, - { INDEX_op_qemu_st16, { "L", "L" } }, - { INDEX_op_qemu_st32, { "L", "L" } }, - #if TCG_TARGET_REG_BITS == 64 { INDEX_op_mov_i64, { "r", "r" } }, { INDEX_op_movi_i64, { "r" } }, @@ -1528,8 +1426,6 @@ static const TCGTargetOpDef sparc_op_defs[] = { { INDEX_op_st16_i64, { "r", "r" } }, { INDEX_op_st32_i64, { "r", "r" } }, { INDEX_op_st_i64, { "r", "r" } }, - { INDEX_op_qemu_ld64, { "L", "L" } }, - { INDEX_op_qemu_st64, { "L", "L" } }, { INDEX_op_add_i64, { "r", "r", "rJ" } }, { INDEX_op_mul_i64, { "r", "r", "rJ" } }, @@ -1557,6 +1453,47 @@ static const TCGTargetOpDef sparc_op_defs[] = { { INDEX_op_brcond_i64, { "r", "rJ" } }, { INDEX_op_setcond_i64, { "r", "r", "rJ" } }, #endif + +#if 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" } }, + { INDEX_op_qemu_ld32u, { "r", "L" } }, + { INDEX_op_qemu_ld32s, { "r", "L" } }, + { INDEX_op_qemu_ld64, { "r", "L" } }, + + { INDEX_op_qemu_st8, { "L", "L" } }, + { INDEX_op_qemu_st16, { "L", "L" } }, + { INDEX_op_qemu_st32, { "L", "L" } }, + { INDEX_op_qemu_st64, { "L", "L" } }, +#elif TARGET_LONG_BITS <= TCG_TARGET_REG_BITS + { 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" } }, + { INDEX_op_qemu_ld64, { "r", "r", "L" } }, + + { INDEX_op_qemu_st8, { "L", "L" } }, + { INDEX_op_qemu_st16, { "L", "L" } }, + { INDEX_op_qemu_st32, { "L", "L" } }, + { INDEX_op_qemu_st64, { "L", "L", "L" } }, +#else + { INDEX_op_qemu_ld8u, { "r", "L", "L" } }, + { INDEX_op_qemu_ld8s, { "r", "L", "L" } }, + { INDEX_op_qemu_ld16u, { "r", "L", "L" } }, + { INDEX_op_qemu_ld16s, { "r", "L", "L" } }, + { INDEX_op_qemu_ld32, { "r", "L", "L" } }, + { INDEX_op_qemu_ld64, { "L", "L", "L", "L" } }, + + { INDEX_op_qemu_st8, { "L", "L", "L" } }, + { INDEX_op_qemu_st16, { "L", "L", "L" } }, + { INDEX_op_qemu_st32, { "L", "L", "L" } }, + { INDEX_op_qemu_st64, { "L", "L", "L", "L" } }, +#endif + { -1 }, }; @@ -1583,25 +1520,23 @@ static void tcg_target_init(TCGContext *s) (1 << TCG_REG_O7)); tcg_regset_clear(s->reserved_regs); - tcg_regset_set_reg(s->reserved_regs, TCG_REG_G0); -#if TCG_TARGET_REG_BITS == 64 - tcg_regset_set_reg(s->reserved_regs, TCG_REG_I4); // for internal use -#endif - tcg_regset_set_reg(s->reserved_regs, TCG_REG_I5); // for internal use - tcg_regset_set_reg(s->reserved_regs, TCG_REG_I6); - tcg_regset_set_reg(s->reserved_regs, TCG_REG_I7); - tcg_regset_set_reg(s->reserved_regs, TCG_REG_O6); - tcg_regset_set_reg(s->reserved_regs, TCG_REG_O7); + tcg_regset_set_reg(s->reserved_regs, TCG_REG_G0); /* zero */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_G6); /* reserved for os */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_G7); /* thread pointer */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_I6); /* frame pointer */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_I7); /* return address */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_O6); /* stack pointer */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_T1); /* for internal use */ + tcg_regset_set_reg(s->reserved_regs, TCG_REG_T2); /* for internal use */ + tcg_add_target_add_op_defs(sparc_op_defs); } #if TCG_TARGET_REG_BITS == 64 # define ELF_HOST_MACHINE EM_SPARCV9 -#elif defined(__sparc_v8plus__) +#else # define ELF_HOST_MACHINE EM_SPARC32PLUS # define ELF_HOST_FLAGS EF_SPARC_32PLUS -#else -# define ELF_HOST_MACHINE EM_SPARC #endif typedef struct { @@ -1657,3 +1592,18 @@ void tcg_register_jit(void *buf, size_t buf_size) tcg_register_jit_int(buf, buf_size, &debug_frame, sizeof(debug_frame)); } + +void tb_set_jmp_target1(uintptr_t jmp_addr, uintptr_t addr) +{ + uint32_t *ptr = (uint32_t *)jmp_addr; + tcg_target_long disp = (tcg_target_long)(addr - jmp_addr) >> 2; + + /* We can reach the entire address space for 32-bit. For 64-bit + the code_gen_buffer can't be larger than 2GB. */ + if (TCG_TARGET_REG_BITS == 64 && !check_fit_tl(disp, 30)) { + tcg_abort(); + } + + *ptr = CALL | (disp & 0x3fffffff); + flush_icache_range(jmp_addr, jmp_addr + 4); +} diff --git a/tcg/sparc/tcg-target.h b/tcg/sparc/tcg-target.h index 0ea87bef72..6314ffb303 100644 --- a/tcg/sparc/tcg-target.h +++ b/tcg/sparc/tcg-target.h @@ -66,22 +66,19 @@ typedef enum { #define TCG_CT_CONST_S13 0x200 /* used for function call generation */ -#define TCG_REG_CALL_STACK TCG_REG_I6 -#ifdef __arch64__ -// Reserve space for AREG0 -#define TCG_TARGET_STACK_MINFRAME (176 + 4 * (int)sizeof(long) + \ - TCG_STATIC_CALL_ARGS_SIZE) -#define TCG_TARGET_CALL_STACK_OFFSET (2047 - 16) -#define TCG_TARGET_STACK_ALIGN 16 +#define TCG_REG_CALL_STACK TCG_REG_O6 + +#if TCG_TARGET_REG_BITS == 64 +#define TCG_TARGET_STACK_BIAS 2047 +#define TCG_TARGET_STACK_ALIGN 16 +#define TCG_TARGET_CALL_STACK_OFFSET (128 + 6*8 + TCG_TARGET_STACK_BIAS) #else -// AREG0 + one word for alignment -#define TCG_TARGET_STACK_MINFRAME (92 + (2 + 1) * (int)sizeof(long) + \ - TCG_STATIC_CALL_ARGS_SIZE) -#define TCG_TARGET_CALL_STACK_OFFSET TCG_TARGET_STACK_MINFRAME -#define TCG_TARGET_STACK_ALIGN 8 +#define TCG_TARGET_STACK_BIAS 0 +#define TCG_TARGET_STACK_ALIGN 8 +#define TCG_TARGET_CALL_STACK_OFFSET (64 + 4 + 6*4) #endif -#ifdef __arch64__ +#if TCG_TARGET_REG_BITS == 64 #define TCG_TARGET_EXTEND_ARGS 1 #endif @@ -102,6 +99,7 @@ typedef enum { #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 #define TCG_TARGET_HAS_deposit_i32 0 +#define TCG_TARGET_HAS_movcond_i32 0 #if TCG_TARGET_REG_BITS == 64 #define TCG_TARGET_HAS_div_i64 1 @@ -123,15 +121,12 @@ typedef enum { #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 #define TCG_TARGET_HAS_deposit_i64 0 +#define TCG_TARGET_HAS_movcond_i64 0 #endif -#ifdef CONFIG_SOLARIS -#define TCG_AREG0 TCG_REG_G2 -#elif defined(__sparc_v9__) -#define TCG_AREG0 TCG_REG_G5 -#else -#define TCG_AREG0 TCG_REG_G6 -#endif +#define TCG_TARGET_HAS_GUEST_BASE + +#define TCG_AREG0 TCG_REG_I0 static inline void flush_icache_range(tcg_target_ulong start, tcg_target_ulong stop) diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h index 169d3b2b0d..6d28f82ad8 100644 --- a/tcg/tcg-op.h +++ b/tcg/tcg-op.h @@ -2118,6 +2118,44 @@ static inline void tcg_gen_deposit_i64(TCGv_i64 ret, TCGv_i64 arg1, tcg_temp_free_i64(t1); } +static inline void tcg_gen_movcond_i32(TCGCond cond, TCGv_i32 ret, + TCGv_i32 c1, TCGv_i32 c2, + TCGv_i32 v1, TCGv_i32 v2) +{ + if (TCG_TARGET_HAS_movcond_i32) { + tcg_gen_op6i_i32(INDEX_op_movcond_i32, ret, c1, c2, v1, v2, cond); + } else { + TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_new_i32(); + tcg_gen_setcond_i32(cond, t0, c1, c2); + tcg_gen_neg_i32(t0, t0); + tcg_gen_and_i32(t1, v1, t0); + tcg_gen_andc_i32(ret, v2, t0); + tcg_gen_or_i32(ret, ret, t1); + tcg_temp_free_i32(t0); + tcg_temp_free_i32(t1); + } +} + +static inline void tcg_gen_movcond_i64(TCGCond cond, TCGv_i64 ret, + TCGv_i64 c1, TCGv_i64 c2, + TCGv_i64 v1, TCGv_i64 v2) +{ + if (TCG_TARGET_HAS_movcond_i64) { + tcg_gen_op6i_i64(INDEX_op_movcond_i64, ret, c1, c2, v1, v2, cond); + } else { + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + tcg_gen_setcond_i64(cond, t0, c1, c2); + tcg_gen_neg_i64(t0, t0); + tcg_gen_and_i64(t1, v1, t0); + tcg_gen_andc_i64(ret, v2, t0); + tcg_gen_or_i64(ret, ret, t1); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + } +} + /***************************************/ /* QEMU specific operations. Their type depend on the QEMU CPU type. */ @@ -2434,6 +2472,7 @@ static inline void tcg_gen_qemu_st64(TCGv_i64 arg, TCGv addr, int mem_index) #define tcg_gen_deposit_tl tcg_gen_deposit_i64 #define tcg_const_tl tcg_const_i64 #define tcg_const_local_tl tcg_const_local_i64 +#define tcg_gen_movcond_tl tcg_gen_movcond_i64 #else #define tcg_gen_movi_tl tcg_gen_movi_i32 #define tcg_gen_mov_tl tcg_gen_mov_i32 @@ -2505,6 +2544,7 @@ static inline void tcg_gen_qemu_st64(TCGv_i64 arg, TCGv addr, int mem_index) #define tcg_gen_deposit_tl tcg_gen_deposit_i32 #define tcg_const_tl tcg_const_i32 #define tcg_const_local_tl tcg_const_local_i32 +#define tcg_gen_movcond_tl tcg_gen_movcond_i32 #endif #if TCG_TARGET_REG_BITS == 32 diff --git a/tcg/tcg-opc.h b/tcg/tcg-opc.h index d12e8d01e9..dbb0e3916a 100644 --- a/tcg/tcg-opc.h +++ b/tcg/tcg-opc.h @@ -51,6 +51,7 @@ DEF(br, 0, 0, 1, TCG_OPF_BB_END | TCG_OPF_SIDE_EFFECTS) DEF(mov_i32, 1, 1, 0, 0) DEF(movi_i32, 1, 0, 1, 0) DEF(setcond_i32, 1, 2, 1, 0) +DEF(movcond_i32, 1, 4, 1, IMPL(TCG_TARGET_HAS_movcond_i32)) /* load/store */ DEF(ld8u_i32, 1, 1, 1, 0) DEF(ld8s_i32, 1, 1, 1, 0) @@ -107,6 +108,7 @@ DEF(nor_i32, 1, 2, 0, IMPL(TCG_TARGET_HAS_nor_i32)) DEF(mov_i64, 1, 1, 0, IMPL64) DEF(movi_i64, 1, 0, 1, IMPL64) DEF(setcond_i64, 1, 2, 1, IMPL64) +DEF(movcond_i64, 1, 4, 1, IMPL64 | IMPL(TCG_TARGET_HAS_movcond_i64)) /* load/store */ DEF(ld8u_i64, 1, 1, 1, IMPL64) DEF(ld8s_i64, 1, 1, 1, IMPL64) @@ -89,7 +89,6 @@ static void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg, TCGReg arg1, tcg_target_long arg2); static int tcg_target_const_match(tcg_target_long val, const TCGArgConstraint *arg_ct); -static int tcg_target_get_call_iarg_regs_count(int flags); TCGOpDef tcg_op_defs[] = { #define DEF(s, oargs, iargs, cargs, flags) { #s, oargs, iargs, cargs, iargs + oargs + cargs, flags }, @@ -937,11 +936,7 @@ void tcg_dump_ops(TCGContext *s) args[nb_oargs + i])); } } - } else if (c == INDEX_op_movi_i32 -#if TCG_TARGET_REG_BITS == 64 - || c == INDEX_op_movi_i64 -#endif - ) { + } else if (c == INDEX_op_movi_i32 || c == INDEX_op_movi_i64) { tcg_target_ulong val; TCGHelperInfo *th; @@ -991,17 +986,13 @@ void tcg_dump_ops(TCGContext *s) } switch (c) { case INDEX_op_brcond_i32: -#if TCG_TARGET_REG_BITS == 32 - case INDEX_op_brcond2_i32: -#elif TCG_TARGET_REG_BITS == 64 - case INDEX_op_brcond_i64: -#endif case INDEX_op_setcond_i32: -#if TCG_TARGET_REG_BITS == 32 + case INDEX_op_movcond_i32: + case INDEX_op_brcond2_i32: case INDEX_op_setcond2_i32: -#elif TCG_TARGET_REG_BITS == 64 + case INDEX_op_brcond_i64: case INDEX_op_setcond_i64: -#endif + case INDEX_op_movcond_i64: if (args[k] < ARRAY_SIZE(cond_name) && cond_name[args[k]]) { qemu_log(",%s", cond_name[args[k++]]); } else { @@ -1458,7 +1449,8 @@ static void temp_allocate_frame(TCGContext *s, int temp) { TCGTemp *ts; ts = &s->temps[temp]; -#ifndef __sparc_v9__ /* Sparc64 stack is accessed with offset of 2047 */ +#if !(defined(__sparc__) && TCG_TARGET_REG_BITS == 64) + /* Sparc64 stack is accessed with offset of 2047 */ s->current_frame_offset = (s->current_frame_offset + (tcg_target_long)sizeof(tcg_target_long) - 1) & ~(sizeof(tcg_target_long) - 1); @@ -1861,7 +1853,7 @@ static int tcg_reg_alloc_call(TCGContext *s, const TCGOpDef *def, flags = args[nb_oargs + nb_iargs]; - nb_regs = tcg_target_get_call_iarg_regs_count(flags); + nb_regs = ARRAY_SIZE(tcg_target_call_iarg_regs); if (nb_regs > nb_params) nb_regs = nb_params; @@ -2103,16 +2095,12 @@ static inline int tcg_gen_code_common(TCGContext *s, uint8_t *gen_code_buf, #endif switch(opc) { case INDEX_op_mov_i32: -#if TCG_TARGET_REG_BITS == 64 case INDEX_op_mov_i64: -#endif dead_args = s->op_dead_args[op_index]; tcg_reg_alloc_mov(s, def, args, dead_args); break; case INDEX_op_movi_i32: -#if TCG_TARGET_REG_BITS == 64 case INDEX_op_movi_i64: -#endif tcg_reg_alloc_movi(s, args); break; case INDEX_op_debug_insn_start: @@ -79,6 +79,7 @@ typedef uint64_t TCGRegSet; #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 #define TCG_TARGET_HAS_deposit_i64 0 +#define TCG_TARGET_HAS_movcond_i64 0 #endif #ifndef TCG_TARGET_deposit_i32_valid @@ -343,7 +344,7 @@ struct TCGContext { /* goto_tb support */ uint8_t *code_buf; - unsigned long *tb_next; + uintptr_t *tb_next; uint16_t *tb_next_offset; uint16_t *tb_jmp_offset; /* != NULL if USE_DIRECT_JUMP */ diff --git a/tcg/tci/tcg-target.c b/tcg/tci/tcg-target.c index 003244cb0b..3f4a24bb8b 100644 --- a/tcg/tci/tcg-target.c +++ b/tcg/tci/tcg-target.c @@ -863,12 +863,6 @@ static int tcg_target_const_match(tcg_target_long val, 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) diff --git a/tcg/tci/tcg-target.h b/tcg/tci/tcg-target.h index 30a0f21596..6d894953aa 100644 --- a/tcg/tci/tcg-target.h +++ b/tcg/tci/tcg-target.h @@ -75,6 +75,7 @@ #define TCG_TARGET_HAS_not_i32 1 #define TCG_TARGET_HAS_orc_i32 0 #define TCG_TARGET_HAS_rot_i32 1 +#define TCG_TARGET_HAS_movcond_i32 0 #if TCG_TARGET_REG_BITS == 64 #define TCG_TARGET_HAS_bswap16_i64 1 @@ -98,6 +99,7 @@ #define TCG_TARGET_HAS_not_i64 1 #define TCG_TARGET_HAS_orc_i64 0 #define TCG_TARGET_HAS_rot_i64 1 +#define TCG_TARGET_HAS_movcond_i64 0 #endif /* TCG_TARGET_REG_BITS == 64 */ /* Offset to user memory in user mode. */ diff --git a/trace-events b/trace-events index b48fe2d15a..f5b5097552 100644 --- a/trace-events +++ b/trace-events @@ -243,9 +243,12 @@ usb_port_release(int bus, const char *port) "bus %d, port %s" # hw/usb/hcd-ehci.c usb_ehci_reset(void) "=== RESET ===" -usb_ehci_mmio_readl(uint32_t addr, const char *str, uint32_t val) "rd mmio %04x [%s] = %x" -usb_ehci_mmio_writel(uint32_t addr, const char *str, uint32_t val) "wr mmio %04x [%s] = %x" -usb_ehci_mmio_change(uint32_t addr, const char *str, uint32_t new, uint32_t old) "ch mmio %04x [%s] = %x (old: %x)" +usb_ehci_opreg_read(uint32_t addr, const char *str, uint32_t val) "rd mmio %04x [%s] = %x" +usb_ehci_opreg_write(uint32_t addr, const char *str, uint32_t val) "wr mmio %04x [%s] = %x" +usb_ehci_opreg_change(uint32_t addr, const char *str, uint32_t new, uint32_t old) "ch mmio %04x [%s] = %x (old: %x)" +usb_ehci_portsc_read(uint32_t addr, uint32_t port, uint32_t val) "rd mmio %04x [port %d] = %x" +usb_ehci_portsc_write(uint32_t addr, uint32_t port, uint32_t val) "wr mmio %04x [port %d] = %x" +usb_ehci_portsc_change(uint32_t addr, uint32_t port, uint32_t new, uint32_t old) "ch mmio %04x [port %d] = %x (old: %x)" usb_ehci_usbsts(const char *sts, int state) "usbsts %s %d" usb_ehci_state(const char *schedule, const char *state) "%s schedule %s" usb_ehci_qh_ptrs(void *q, uint32_t addr, uint32_t nxt, uint32_t c_qtd, uint32_t n_qtd, uint32_t a_qtd) "q %p - QH @ %08x: next %08x qtds %08x,%08x,%08x" |