/* Definitions of target machine for GNU compiler, for Sun SPARC.
- Copyright (C) 1987, 1988, 1989, 1992 Free Software Foundation, Inc.
+ Copyright (C) 1987, 88, 89, 92, 94, 95, 96, 1997 Free Software Foundation,
+ Inc.
Contributed by Michael Tiemann (tiemann@cygnus.com).
+ 64 bit SPARC V9 support by Michael Tiemann, Jim Wilson, and Doug Evans,
+ at Cygnus Support.
This file is part of GNU CC.
You should have received a copy of the GNU General Public License
along with GNU CC; see the file COPYING. If not, write to
-the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */
+the Free Software Foundation, 59 Temple Place - Suite 330,
+Boston, MA 02111-1307, USA. */
/* Note that some other tm.h files include this one and then override
many of the definitions that relate to assembler syntax. */
-#define LIB_SPEC "%{!p:%{!pg:-lc}}%{p:-lc_p}%{pg:-lc_p} %{g:-lg} \
- %{a:/usr/lib/bb_link.o}"
+/* Sparc64 support has been added by trying to allow for a day when one
+ compiler can handle both v8 and v9. There are a few cases where this
+ isn't doable, but keep them to a minimum!
+
+ TARGET_V9 is used to select at runtime the sparc64 chip.
+ TARGET_ARCH64 is used to select at runtime a 64 bit environment.
+ SPARC_V9 is defined as 0 or 1 (so it may be used inside and outside
+ #define's), and says whether the cpu is a sparc64 chip (which may be
+ running in a 32 or 64 bit environment).
+ SPARC_ARCH64 is defined as 0 for a 32 bit environment and 1 for a 64 bit
+ environment.
+
+ In places where it is possible to choose at runtime, use TARGET_V9 and
+ TARGET_ARCH64. In places where it is currently not possible to select
+ between the two at runtime use SPARC_{V9,ARCH64}. Again, keep uses of
+ SPARC_{V9,ARCH64} to a minimum. No attempt is made to support both v8
+ and v9 in the v9 compiler.
+
+ ??? All uses of SPARC_V9 have been removed. Try not to add new ones.
+*/
+
+#ifndef SPARC_V9
+#define SPARC_V9 0
+#endif
+#ifndef SPARC_ARCH64
+#define SPARC_ARCH64 0
+#endif
-/* Provide required defaults for linker -e and -d switches. */
+/* Values of TARGET_CPU_DEFAULT, set via -D in the Makefile. */
+#define TARGET_CPU_sparc 0
+#define TARGET_CPU_v7 0 /* alias for previous */
+#define TARGET_CPU_sparclet 1
+#define TARGET_CPU_sparclite 2
+#define TARGET_CPU_v8 3
+#define TARGET_CPU_supersparc 4
+#define TARGET_CPU_ultrasparc 5
+#define TARGET_CPU_sparc64 5 /* alias for ultrasparc */
+
+#if TARGET_CPU_DEFAULT == TARGET_CPU_sparc || TARGET_CPU_DEFAULT == TARGET_CPU_v8 || TARGET_CPU_DEFAULT == TARGET_CPU_supersparc
+#define CPP_DEFAULT_SPEC ""
+#define ASM_DEFAULT_SPEC ""
+#else
+#if TARGET_CPU_DEFAULT == TARGET_CPU_sparclet
+#define CPP_DEFAULT_SPEC "-D__sparclet__"
+#define ASM_DEFAULT_SPEC "-Asparclet"
+#else
+#if TARGET_CPU_DEFAULT == TARGET_CPU_sparclite
+#define CPP_DEFAULT_SPEC "-D__sparclite__"
+#define ASM_DEFAULT_SPEC "-Asparclite"
+#else
+#if TARGET_CPU_DEFAULT == TARGET_CPU_sparc64
+/* ??? What does Sun's CC pass? */
+#define CPP_DEFAULT_SPEC "-D__sparc_v9__"
+/* ??? It's not clear how other assemblers will handle this, so by default
+ use GAS. Sun's Solaris assembler recognizes -xarch=v8plus, but this case
+ is handled in sol2.h. */
+#define ASM_DEFAULT_SPEC "-Av9"
+#else
+Unrecognized value in TARGET_CPU_DEFAULT.
+#endif
+#endif
+#endif
+#endif
-#define LINK_SPEC \
- "%{!nostdlib:%{!r*:%{!e*:-e start}}} -dc -dp %{static:-Bstatic} %{assert*}"
+/* Names to predefine in the preprocessor for this target machine. */
-/* Special flags to the Sun-4 assembler when using pipe for input. */
+/* ??? The GCC_NEW_VARARGS macro is now obsolete, because gcc always uses
+ the right varags.h file when bootstrapping. */
+/* ??? It's not clear what value we want to use for -Acpu/machine for
+ sparc64 in 32 bit environments, so for now we only use `sparc64' in
+ 64 bit environments. */
+/* ??? __arch64__ is subject to change. */
-#define ASM_SPEC " %| %{fpic:-k} %{fPIC:-k}"
+#if SPARC_ARCH64
+#define CPP_PREDEFINES \
+ "-Dsparc -Dsun -Dunix -D__arch64__ \
+ -Asystem(unix) -Asystem(bsd) -Acpu(sparc64) -Amachine(sparc64)"
+#else
+#define CPP_PREDEFINES \
+ "-Dsparc -Dsun -Dunix -D__GCC_NEW_VARARGS__ \
+ -Asystem(unix) -Asystem(bsd) -Acpu(sparc) -Amachine(sparc)"
+#endif
/* Define macros to distinguish architectures. */
-#define CPP_SPEC "%{msparclite:-D__sparclite__} %{mf930:-D__sparclite__} \
-%{mf934:-D__sparclite__} %{mv8:-D__sparc_v8__}"
+
+#if SPARC_ARCH64
+#define CPP_SPEC "\
+%{mint64:-D__INT_MAX__=9223372036854775807LL -D__LONG_MAX__=9223372036854775807LL} \
+%{mlong64:-D__LONG_MAX__=9223372036854775807LL} \
+"
+#else
+#define CPP_SPEC "%(cpp_cpu)"
+#endif
+
+/* Common CPP definitions used by CPP_SPEC amongst the various targets
+ for handling -mcpu=xxx switches. */
+/* ??? v8plus/v9/ultrasparc handling is tentative */
+#define CPP_CPU_SPEC "\
+%{mcypress:} \
+%{msparclite:-D__sparclite__} \
+%{mf930:-D__sparclite__} %{mf934:-D__sparclite__} \
+%{mv8:-D__sparc_v8__} \
+%{msupersparc:-D__supersparc__ -D__sparc_v8__} \
+%{mcpu=sparclet:-D__sparclet__} %{mcpu=tsc701:-D__sparclet__} \
+%{mcpu=sparclite:-D__sparclite__} \
+%{mcpu=f930:-D__sparclite__} %{mcpu=f934:-D__sparclite__} \
+%{mcpu=v8:-D__sparc_v8__} \
+%{mcpu=supersparc:-D__supersparc__ -D__sparc_v8__} \
+%{mcpu=v8plus:-D__sparc_v9__} \
+%{mcpu=v9:-D__sparc_v9__} \
+%{mcpu=ultrasparc:-D__sparc_v9__} \
+%{!mcpu*:%{!mcypress:%{!msparclite:%{!mf930:%{!mf934:%{!mv8:%{!msupersparc:%(cpp_default)}}}}}}} \
+"
/* Prevent error on `-sun4' and `-target sun4' options. */
/* This used to translate -dalign to -malign, but that is no good
because it can't turn off the usual meaning of making debugging dumps. */
+/* Translate old style -m<cpu> into new style -mcpu=<cpu>.
+ ??? Delete support for -m<cpu> for 2.9. */
+
+#define CC1_SPEC "\
+%{sun4:} %{target:} \
+%{mcypress:-mcpu=cypress} \
+%{msparclite:-mcpu=sparclite} %{mf930:-mcpu=f930} %{mf934:-mcpu=f934} \
+%{mv8:-mcpu=v8} %{msupersparc:-mcpu=supersparc} \
+"
+
+#define LIB_SPEC "%{!shared:%{!p:%{!pg:-lc}}%{p:-lc_p}%{pg:-lc_p} %{g:-lg}}"
+
+/* Provide required defaults for linker -e and -d switches. */
+
+#define LINK_SPEC \
+ "%{!shared:%{!nostdlib:%{!r*:%{!e*:-e start}}} -dc -dp} %{static:-Bstatic} \
+ %{assert*} %{shared:%{!mimpure-text:-assert pure-text}}"
-#define CC1_SPEC "%{sun4:} %{target:}"
+/* Special flags to the Sun-4 assembler when using pipe for input. */
+#define ASM_SPEC "\
+%| %{R} %{!pg:%{!p:%{fpic:-k} %{fPIC:-k}}} %{keep-local-as-symbols:-L} \
+%(asm_cpu) \
+"
+
+/* Override in target specific files. */
+#define ASM_CPU_SPEC "\
+%{mcpu=sparclet:-Asparclet} %{mcpu=tsc701:-Asparclet} \
+%{msparclite:-Asparclite} \
+%{mf930:-Asparclite} %{mf934:-Asparclite} \
+%{mcpu=sparclite:-Asparclite} \
+%{mcpu=f930:-Asparclite} %{mcpu=f934:-Asparclite} \
+%{mcpu=v8plus:-Av9} \
+%{mcpu=v9:-Av9} \
+%{mcpu=ultrasparc:-Av9} \
+%{!mcpu*:%{!mcypress:%{!msparclite:%{!mf930:%{!mf934:%{!mv8:%{!msupersparc:%(asm_default)}}}}}}} \
+"
+
+/* This macro defines names of additional specifications to put in the specs
+ that can be used in various specifications like CC1_SPEC. Its definition
+ is an initializer with a subgrouping for each command option.
+
+ Each subgrouping contains a string constant, that defines the
+ specification name, and a string constant that used by the GNU CC driver
+ program.
+
+ Do not define this macro if it does not need to do anything. */
+
+#define EXTRA_SPECS \
+ { "cpp_cpu", CPP_CPU_SPEC }, \
+ { "cpp_default", CPP_DEFAULT_SPEC }, \
+ { "asm_cpu", ASM_CPU_SPEC }, \
+ { "asm_default", ASM_DEFAULT_SPEC }, \
+ SUBTARGET_EXTRA_SPECS
+
+#define SUBTARGET_EXTRA_SPECS
+\f
+#if SPARC_ARCH64
+#define PTRDIFF_TYPE "long long int"
+#define SIZE_TYPE "long long unsigned int"
+#else
#define PTRDIFF_TYPE "int"
-/* In 2.4 it should work to delete this.
- #define SIZE_TYPE "int" */
+/* The default value for SIZE_TYPE is "unsigned int" which is what we want. */
+#endif
+
+/* ??? This should be 32 bits for v9 but what can we do? */
#define WCHAR_TYPE "short unsigned int"
#define WCHAR_TYPE_SIZE 16
+#define MAX_WCHAR_TYPE_SIZE 16
-/* Omit frame pointer at high optimization levels. */
-
-#define OPTIMIZATION_OPTIONS(OPTIMIZE) \
-{ \
- if (OPTIMIZE >= 2) \
- { \
- flag_omit_frame_pointer = 1; \
- } \
-}
+/* Show we can debug even without a frame pointer. */
+#define CAN_DEBUG_WITHOUT_FP
/* To make profiling work with -f{pic,PIC}, we need to emit the profiling
code into the rtl. Also, if we are profiling, we cannot eliminate
the frame pointer (because the return address will get smashed). */
+void sparc_override_options ();
+
#define OVERRIDE_OPTIONS \
-{ \
- if (profile_flag || profile_block_flag) \
- flag_omit_frame_pointer = 0, flag_pic = 0; \
- SUBTARGET_OVERRIDE_OPTIONS \
- }
+ do { \
+ if (profile_flag || profile_block_flag || profile_arc_flag) \
+ { \
+ if (flag_pic) \
+ { \
+ char *pic_string = (flag_pic == 1) ? "-fpic" : "-fPIC"; \
+ warning ("%s and profiling conflict: disabling %s", \
+ pic_string, pic_string); \
+ flag_pic = 0; \
+ } \
+ flag_omit_frame_pointer = 0; \
+ } \
+ sparc_override_options (); \
+ SUBTARGET_OVERRIDE_OPTIONS; \
+ } while (0)
-/* This is meant to be redefined in the host dependent files */
+/* This is meant to be redefined in the host dependent files. */
#define SUBTARGET_OVERRIDE_OPTIONS
/* These compiler options take an argument. We ignore -target for now. */
(DEFAULT_WORD_SWITCH_TAKES_ARG (STR) \
|| !strcmp (STR, "target") || !strcmp (STR, "assert"))
-/* Names to predefine in the preprocessor for this target machine. */
-
-/* The GCC_NEW_VARARGS macro is so that old versions of gcc can compile
- new versions, which have an incompatible va-sparc.h file. This matters
- because gcc does "gvarargs.h" instead of <varargs.h>, and thus gets the
- wrong varargs file when it is compiled with a different version of gcc. */
-
-#define CPP_PREDEFINES \
- "-Dsparc -Dsun -Dunix -D__GCC_NEW_VARARGS__ \
- -Asystem(unix) -Asystem(bsd) -Acpu(sparc) -Amachine(sparc)"
-
/* Print subsidiary information on the compiler version in use. */
#define TARGET_VERSION fprintf (stderr, " (sparc)");
/* Generate DBX debugging information. */
#define DBX_DEBUGGING_INFO
-
+\f
/* Run-time compilation parameters selecting different hardware subsets. */
extern int target_flags;
/* Nonzero if we should generate code to use the fpu. */
-#define TARGET_FPU (target_flags & 1)
+#define MASK_FPU 1
+#define TARGET_FPU (target_flags & MASK_FPU)
/* Nonzero if we should use FUNCTION_EPILOGUE. Otherwise, we
use fast return insns, but lose some generality. */
-#define TARGET_EPILOGUE (target_flags & 2)
+#define MASK_EPILOGUE 2
+#define TARGET_EPILOGUE (target_flags & MASK_EPILOGUE)
/* Nonzero if we should assume that double pointers might be unaligned.
This can happen when linking gcc compiled code with other compilers,
because the ABI only guarantees 4 byte alignment. */
-#define TARGET_UNALIGNED_DOUBLES (target_flags & 4)
+#define MASK_UNALIGNED_DOUBLES 4
+#define TARGET_UNALIGNED_DOUBLES (target_flags & MASK_UNALIGNED_DOUBLES)
/* Nonzero means that we should generate code for a v8 sparc. */
-#define TARGET_V8 (target_flags & 64)
+#define MASK_V8 0x8
+#define TARGET_V8 (target_flags & MASK_V8)
/* Nonzero means that we should generate code for a sparclite.
This enables the sparclite specific instructions, but does not affect
whether FPU instructions are emitted. */
-#define TARGET_SPARCLITE (target_flags & 128)
+#define MASK_SPARCLITE 0x10
+#define TARGET_SPARCLITE (target_flags & MASK_SPARCLITE)
+
+/* Nonzero if we're compiling for the sparclet. */
+#define MASK_SPARCLET 0x20
+#define TARGET_SPARCLET (target_flags & MASK_SPARCLET)
+
+/* Nonzero if we're compiling for v9 sparc.
+ Note that v9's can run in 32 bit mode so this doesn't necessarily mean
+ the word size is 64. */
+#define MASK_V9 0x40
+#define TARGET_V9 (target_flags & MASK_V9)
+
+/* Non-zero to generate code that uses the instructions deprecated in
+ the v9 architecture. This option only applies to v9 systems. */
+/* ??? This isn't user selectable yet. It's used to enable such insns
+ on 32 bit v9 systems and for the moment they're permanently disabled
+ on 64 bit v9 systems. */
+#define MASK_DEPRECATED_V8_INSNS 0x80
+#define TARGET_DEPRECATED_V8_INSNS (target_flags & MASK_DEPRECATED_V8_INSNS)
+
+/* Mask of all CPU selection flags. */
+#define MASK_ISA \
+(MASK_V8 + MASK_SPARCLITE + MASK_SPARCLET + MASK_V9 + MASK_DEPRECATED_V8_INSNS)
+
+/* Non-zero means don't pass `-assert pure-text' to the linker. */
+#define MASK_IMPURE_TEXT 0x100
+#define TARGET_IMPURE_TEXT (target_flags & MASK_IMPURE_TEXT)
/* Nonzero means that we should generate code using a flat register window
- model, i.e. no save/restore instructions are generated, in the most
- efficient manner. This code is not compatible with normal sparc code. */
-/* This is not a user selectable option yet, because it requires changes
- that are not yet switchable via command line arguments. */
-#define TARGET_FRW (target_flags & 256)
-
-/* Nonzero means that we should generate code using a flat register window
- model, i.e. no save/restore instructions are generated, but which is
- compatible with normal sparc code. This is the same as above, except
- that the frame pointer is %l6 instead of %fp. This code is not as efficient
- as TARGET_FRW, because it has one less allocatable register. */
-/* This is not a user selectable option yet, because it requires changes
- that are not yet switchable via command line arguments. */
-#define TARGET_FRW_COMPAT (target_flags & 512)
+ model, i.e. no save/restore instructions are generated, which is
+ compatible with normal sparc code.
+ The frame pointer is %i7 instead of %fp. */
+#define MASK_FLAT 0x200
+#define TARGET_FLAT (target_flags & MASK_FLAT)
+
+/* Nonzero means use the registers that the Sparc ABI reserves for
+ application software. This must be the default to coincide with the
+ setting in FIXED_REGISTERS. */
+#define MASK_APP_REGS 0x400
+#define TARGET_APP_REGS (target_flags & MASK_APP_REGS)
+
+/* Option to select how quad word floating point is implemented.
+ When TARGET_HARD_QUAD is true, we use the hardware quad instructions.
+ Otherwise, we use the SPARC ABI quad library functions. */
+#define MASK_HARD_QUAD 0x800
+#define TARGET_HARD_QUAD (target_flags & MASK_HARD_QUAD)
+
+/* Non-zero on little-endian machines. */
+/* ??? Little endian support currently only exists for sparclet-aout and
+ sparc64-elf configurations. May eventually want to expand the support
+ to all targets, but for now it's kept local to only those two. */
+#define MASK_LITTLE_ENDIAN 0x1000
+#define TARGET_LITTLE_ENDIAN (target_flags & MASK_LITTLE_ENDIAN)
+
+/* Nonzero if ints are 64 bits.
+ This automatically implies longs are 64 bits too.
+ This option is for v9 only. */
+#define MASK_INT64 0x2000
+#define TARGET_INT64 (target_flags & MASK_INT64)
+
+/* Nonzero if longs are 64 bits.
+ This option is for v9 only. */
+#define MASK_LONG64 0x4000
+#define TARGET_LONG64 (target_flags & MASK_LONG64)
+
+/* Nonzero if pointers are 64 bits.
+ This is not a user selectable option, though it may be one day -
+ so it is used to determine pointer size instead of an architecture flag. */
+#define MASK_PTR64 0x8000
+#define TARGET_PTR64 (target_flags & MASK_PTR64)
+
+/* Nonzero if generating code to run in a 64 bit environment. */
+#define MASK_ARCH64 0x10000
+#define TARGET_ARCH64 (target_flags & MASK_ARCH64)
+#define TARGET_ARCH32 (! TARGET_ARCH64)
+
+/* SPARC64 memory models.
+ TARGET_MEDLOW: 32 bit address space, top 32 bits = 0,
+ avoid generating %uhi and %ulo terms.
+ (pointers can be 32 or 64 bits)
+ TARGET_MEDANY: 64 bit address space, data segment restricted to 4G, but
+ can be loaded anywhere (use %g4 as offset).
+ TARGET_FULLANY: 64 bit address space, no restrictions.
+ This option is not fully supported yet.
+ These options are for v9 only. All mask values are nonzero so the v8
+ compiler can assume this stuff won't interfere. */
+#define MASK_MEDLOW 0x20000
+#define MASK_MEDANY 0x40000
+#define MASK_FULLANY 0x60000
+#define MASK_CODE_MODEL (MASK_MEDLOW + MASK_MEDANY)
+#define TARGET_MEDLOW ((target_flags & MASK_CODE_MODEL) == MASK_MEDLOW)
+#define TARGET_MEDANY ((target_flags & MASK_CODE_MODEL) == MASK_MEDANY)
+#define TARGET_FULLANY ((target_flags & MASK_CODE_MODEL) == MASK_FULLANY)
+
+/* ??? There are hardcoded references to this reg in the .md file. */
+#define MEDANY_BASE_REG "%g4"
+
+/* Non-zero means use a stack bias of 2047. Stack offsets are obtained by
+ adding 2047 to %sp. This option is for v9 only and is the default. */
+#define MASK_STACK_BIAS 0x80000
+#define TARGET_STACK_BIAS (target_flags & MASK_STACK_BIAS)
+
+/* Non-zero means %g0 is a normal register.
+ We still clobber it as necessary, but we can't rely on it always having
+ a zero value.
+ We don't bother to support this in true 64 bit mode. */
+#define MASK_LIVE_G0 0x100000
+#define TARGET_LIVE_G0 (target_flags & MASK_LIVE_G0)
+
+/* Non-zero means the cpu has broken `save' and `restore' insns, only
+ the trivial versions work (save %g0,%g0,%g0; restore %g0,%g0,%g0).
+ We assume the environment will properly handle or otherwise avoid
+ trouble associated with an interrupt occuring after the `save' or trap
+ occuring during it. */
+#define MASK_BROKEN_SAVERESTORE 0x200000
+#define TARGET_BROKEN_SAVERESTORE (target_flags & MASK_BROKEN_SAVERESTORE)
+
+/* Non-zero means -m{,no-}fpu was passed on the command line. */
+#define MASK_FPU_SET 0x400000
+#define TARGET_FPU_SET (target_flags & MASK_FPU_SET)
/* Macro to define tables used to set the flags.
This is a list in braces of pairs in braces,
where VALUE is the bits to set or minus the bits to clear.
An empty string NAME is used to identify the default VALUE. */
-/* The Fujitsu MB86930 is the original sparclite chip, with no fpu.
- The Fujitsu MB86934 is the recent sparclite chip, with an fup.
- We use -mf930 and -mf934 options to choose which.
- ??? These should perhaps be -mcpu= options. */
-
#define TARGET_SWITCHES \
- { {"fpu", 1}, \
- {"no-fpu", -1}, \
- {"hard-float", 1}, \
- {"soft-float", -1}, \
- {"epilogue", 2}, \
- {"no-epilogue", -2}, \
- {"unaligned-doubles", 4}, \
- {"no-unaligned-doubles", -4},\
- {"v8", 64}, \
- {"no-v8", -64}, \
- {"sparclite", 128}, \
- {"no-sparclite", -128}, \
-/* {"frw", 256}, */ \
-/* {"no-frw", -256}, */ \
-/* {"frw-compat", 256+512}, */ \
-/* {"no-frw-compat", -(256+512)}, */ \
- {"f930", 128}, \
- {"f930", -1}, \
- {"f934", 128}, \
- SUBTARGET_SWITCHES \
+ { {"fpu", MASK_FPU | MASK_FPU_SET}, \
+ {"no-fpu", -MASK_FPU}, \
+ {"no-fpu", MASK_FPU_SET}, \
+ {"hard-float", MASK_FPU | MASK_FPU_SET}, \
+ {"soft-float", -MASK_FPU}, \
+ {"soft-float", MASK_FPU_SET}, \
+ {"epilogue", MASK_EPILOGUE}, \
+ {"no-epilogue", -MASK_EPILOGUE}, \
+ {"unaligned-doubles", MASK_UNALIGNED_DOUBLES}, \
+ {"no-unaligned-doubles", -MASK_UNALIGNED_DOUBLES}, \
+ {"impure-text", MASK_IMPURE_TEXT}, \
+ {"no-impure-text", -MASK_IMPURE_TEXT}, \
+ {"flat", MASK_FLAT}, \
+ {"no-flat", -MASK_FLAT}, \
+ {"app-regs", MASK_APP_REGS}, \
+ {"no-app-regs", -MASK_APP_REGS}, \
+ {"hard-quad-float", MASK_HARD_QUAD}, \
+ {"soft-quad-float", -MASK_HARD_QUAD}, \
+ /* ??? These are coerced to -mcpu=. Delete in 2.9. */ \
+ {"cypress", 0}, \
+ {"sparclite", 0}, \
+ {"f930", 0}, \
+ {"f934", 0}, \
+ {"v8", 0}, \
+ {"supersparc", 0}, \
+ SUBTARGET_SWITCHES \
+ ARCH64_SWITCHES \
{ "", TARGET_DEFAULT}}
-#define TARGET_DEFAULT 3
+/* MASK_APP_REGS must always be the default because that's what
+ FIXED_REGISTERS is set to and -ffixed- is processed before
+ CONDITIONAL_REGISTER_USAGE is called (where we process -mno-app-regs). */
+#define TARGET_DEFAULT (MASK_APP_REGS + MASK_EPILOGUE + MASK_FPU)
-/* This is meant to be redefined in the host dependent files */
+/* This is meant to be redefined in target specific files. */
#define SUBTARGET_SWITCHES
-\f
-/* target machine storage layout */
-#if 0
-/* ??? This does not work in SunOS 4.x, so it is not enabled here.
- Instead, it is enabled in sol2.h, because it does work under Solaris. */
-/* Define for support of TFmode long double and REAL_ARITHMETIC.
- Sparc ABI says that long double is 4 words. */
-#define LONG_DOUBLE_TYPE_SIZE 128
+/* ??? Until we support a combination 32/64 bit compiler, these options
+ are only defined for the v9 compiler in a true 64 bit environment. */
+#if SPARC_ARCH64
+#define ARCH64_SWITCHES \
+/* {"arch32", -MASK_ARCH64}, */ \
+/* {"arch64", MASK_ARCH64}, */ \
+ {"int64", MASK_INT64+MASK_LONG64}, \
+ {"int32", -MASK_INT64}, \
+ {"int32", MASK_LONG64}, \
+ {"long64", -MASK_INT64}, \
+ {"long64", MASK_LONG64}, \
+ {"long32", -(MASK_INT64+MASK_LONG64)}, \
+/* {"ptr64", MASK_PTR64}, */ \
+/* {"ptr32", -MASK_PTR64}, */ \
+ {"stack-bias", MASK_STACK_BIAS}, \
+ {"no-stack-bias", -MASK_STACK_BIAS}, \
+ {"medlow", -MASK_CODE_MODEL}, \
+ {"medlow", MASK_MEDLOW}, \
+ {"medany", -MASK_CODE_MODEL}, \
+ {"medany", MASK_MEDANY}, \
+ {"fullany", -MASK_CODE_MODEL}, \
+ {"fullany", MASK_FULLANY},
+#else
+#define ARCH64_SWITCHES
#endif
+/* Processor type.
+ These must match the values for the cpu attribute in sparc.md. */
+enum processor_type {
+ PROCESSOR_V7,
+ PROCESSOR_CYPRESS,
+ PROCESSOR_V8,
+ PROCESSOR_SUPERSPARC,
+ PROCESSOR_SPARCLITE,
+ PROCESSOR_F930,
+ PROCESSOR_F934,
+ PROCESSOR_SPARCLET,
+ PROCESSOR_TSC701,
+ PROCESSOR_V8PLUS,
+ PROCESSOR_V9,
+ PROCESSOR_ULTRASPARC
+};
+
+/* This is set from -m{cpu,tune}=xxx. */
+extern enum processor_type sparc_cpu;
+
+/* Recast the cpu class to be the cpu attribute.
+ Every file includes us, but not every file includes insn-attr.h. */
+#define sparc_cpu_attr ((enum attr_cpu) sparc_cpu)
+
+/* This macro is similar to `TARGET_SWITCHES' but defines names of
+ command options that have values. Its definition is an
+ initializer with a subgrouping for each command option.
+
+ Each subgrouping contains a string constant, that defines the
+ fixed part of the option name, and the address of a variable.
+ The variable, type `char *', is set to the variable part of the
+ given option if the fixed part matches. The actual option name
+ is made by appending `-m' to the specified name.
+
+ Here is an example which defines `-mshort-data-NUMBER'. If the
+ given option is `-mshort-data-512', the variable `m88k_short_data'
+ will be set to the string `"512"'.
+
+ extern char *m88k_short_data;
+ #define TARGET_OPTIONS { { "short-data-", &m88k_short_data } } */
+
+#define TARGET_OPTIONS \
+{ \
+ {"cpu=", &sparc_select[1].string}, \
+ {"tune=", &sparc_select[2].string}, \
+ SUBTARGET_OPTIONS \
+}
+
+/* This is meant to be redefined in target specific files. */
+#define SUBTARGET_OPTIONS
+
+/* sparc_select[0] is reserved for the default cpu. */
+struct sparc_cpu_select
+{
+ char *string;
+ char *name;
+ int set_tune_p;
+ int set_arch_p;
+};
+
+extern struct sparc_cpu_select sparc_select[];
+\f
+/* target machine storage layout */
+
/* Define for cross-compilation to a sparc target with no TFmode from a host
with a different float format (e.g. VAX). */
#define REAL_ARITHMETIC
#define BITS_BIG_ENDIAN 1
/* Define this if most significant byte of a word is the lowest numbered. */
-/* This is true on the SPARC. */
#define BYTES_BIG_ENDIAN 1
/* Define this if most significant word of a multiword number is the lowest
numbered. */
-/* Doubles are stored in memory with the high order word first. This
- matters when cross-compiling. */
#define WORDS_BIG_ENDIAN 1
+/* Define this to set the endianness to use in libgcc2.c, which can
+ not depend on target_flags. */
+#if defined (__LITTLE_ENDIAN__)
+#define LIBGCC2_WORDS_BIG_ENDIAN 0
+#else
+#define LIBGCC2_WORDS_BIG_ENDIAN 1
+#endif
+
/* number of bits in an addressable storage unit */
#define BITS_PER_UNIT 8
Note that this is not necessarily the width of data type `int';
if using 16-bit ints on a 68000, this would still be 32.
But on a machine with 16-bit registers, this would be 16. */
-#define BITS_PER_WORD 32
-#define MAX_BITS_PER_WORD 32
+#define BITS_PER_WORD (TARGET_ARCH64 ? 64 : 32)
+#define MAX_BITS_PER_WORD 64
/* Width of a word, in units (bytes). */
-#define UNITS_PER_WORD 4
+#define UNITS_PER_WORD (TARGET_ARCH64 ? 8 : 4)
+#define MIN_UNITS_PER_WORD 4
+
+/* Now define the sizes of the C data types. */
+
+#define SHORT_TYPE_SIZE 16
+#define INT_TYPE_SIZE (TARGET_INT64 ? 64 : 32)
+#define LONG_TYPE_SIZE (TARGET_LONG64 ? 64 : 32)
+#define LONG_LONG_TYPE_SIZE 64
+#define FLOAT_TYPE_SIZE 32
+#define DOUBLE_TYPE_SIZE 64
+
+#define MAX_INT_TYPE_SIZE 64
+#define MAX_LONG_TYPE_SIZE 64
+
+#if SPARC_ARCH64
+/* ??? This does not work in SunOS 4.x, so it is not enabled here.
+ Instead, it is enabled in sol2.h, because it does work under Solaris. */
+/* Define for support of TFmode long double and REAL_ARITHMETIC.
+ Sparc ABI says that long double is 4 words. */
+#define LONG_DOUBLE_TYPE_SIZE 128
+#endif
/* Width in bits of a pointer.
See also the macro `Pmode' defined below. */
-#define POINTER_SIZE 32
+#define POINTER_SIZE (TARGET_PTR64 ? 64 : 32)
/* Allocation boundary (in *bits*) for storing arguments in argument list. */
-#define PARM_BOUNDARY 32
+#define PARM_BOUNDARY (TARGET_ARCH64 ? 64 : 32)
/* Boundary (in *bits*) on which stack pointer should be aligned. */
-#define STACK_BOUNDARY 64
+#define STACK_BOUNDARY (TARGET_ARCH64 ? 128 : 64)
/* ALIGN FRAMES on double word boundaries */
-#define SPARC_STACK_ALIGN(LOC) (((LOC)+7) & 0xfffffff8)
+#define SPARC_STACK_ALIGN(LOC) \
+ (TARGET_ARCH64 ? (((LOC)+15) & ~15) : (((LOC)+7) & ~7))
/* Allocation boundary (in *bits*) for the code of a function. */
#define FUNCTION_BOUNDARY 32
/* Alignment of field after `int : 0' in a structure. */
-#define EMPTY_FIELD_BOUNDARY 32
+/* ??? Should this be based on TARGET_INT64? */
+#define EMPTY_FIELD_BOUNDARY (TARGET_ARCH64 ? 64 : 32)
/* Every structure's size must be a multiple of this. */
#define STRUCTURE_SIZE_BOUNDARY 8
#define PCC_BITFIELD_TYPE_MATTERS 1
/* No data type wants to be aligned rounder than this. */
-#define BIGGEST_ALIGNMENT 64
+#define BIGGEST_ALIGNMENT (TARGET_ARCH64 ? 128 : 64)
/* The best alignment to use in cases where we have a choice. */
#define FASTEST_ALIGNMENT 64
/* Things that must be doubleword aligned cannot go in the text section,
because the linker fails to align the text section enough!
- Put them in the data section. */
+ Put them in the data section. This macro is only used in this file. */
#define MAX_TEXT_ALIGN 32
+/* This forces all variables and constants to the data section when PIC.
+ This is because the SunOS 4 shared library scheme thinks everything in
+ text is a function, and patches the address to point to a loader stub. */
+/* This is defined to zero for every system which doesn't use the a.out object
+ file format. */
+#ifndef SUNOS4_SHARED_LIBRARIES
+#define SUNOS4_SHARED_LIBRARIES 0
+#endif
+
+/* This is defined differently for v9 in a cover file. */
#define SELECT_SECTION(T,RELOC) \
{ \
if (TREE_CODE (T) == VAR_DECL) \
{ \
if (TREE_READONLY (T) && ! TREE_SIDE_EFFECTS (T) \
+ && DECL_INITIAL (T) \
+ && (DECL_INITIAL (T) == error_mark_node \
+ || TREE_CONSTANT (DECL_INITIAL (T))) \
&& DECL_ALIGN (T) <= MAX_TEXT_ALIGN \
- && ! (flag_pic && (RELOC))) \
+ && ! (flag_pic && ((RELOC) || SUNOS4_SHARED_LIBRARIES))) \
text_section (); \
else \
data_section (); \
} \
else if (TREE_CODE (T) == CONSTRUCTOR) \
{ \
- if (flag_pic != 0 && (RELOC) != 0) \
+ if (flag_pic && ((RELOC) || SUNOS4_SHARED_LIBRARIES)) \
data_section (); \
} \
- else if (*tree_code_type[(int) TREE_CODE (T)] == 'c') \
+ else if (TREE_CODE_CLASS (TREE_CODE (T)) == 'c') \
{ \
if ((TREE_CODE (T) == STRING_CST && flag_writable_strings) \
- || TYPE_ALIGN (TREE_TYPE (T)) > MAX_TEXT_ALIGN) \
+ || TYPE_ALIGN (TREE_TYPE (T)) > MAX_TEXT_ALIGN \
+ || (flag_pic && ((RELOC) || SUNOS4_SHARED_LIBRARIES))) \
data_section (); \
else \
text_section (); \
/* Use text section for a constant
unless we need more alignment than that offers. */
+/* This is defined differently for v9 in a cover file. */
#define SELECT_RTX_SECTION(MODE, X) \
{ \
if (GET_MODE_BITSIZE (MODE) <= MAX_TEXT_ALIGN \
- && ! (flag_pic && symbolic_operand (X))) \
+ && ! (flag_pic && (symbolic_operand (X) || SUNOS4_SHARED_LIBRARIES))) \
text_section (); \
else \
data_section (); \
All registers that the compiler knows about must be given numbers,
even those that are not normally considered general registers.
- SPARC has 32 integer registers and 32 floating point registers. */
-
-#define FIRST_PSEUDO_REGISTER 64
+ SPARC has 32 integer registers and 32 floating point registers.
+ 64 bit SPARC has 32 additional fp regs, but the odd numbered ones are not
+ accessible. We still account for them to simplify register computations
+ (eg: in CLASS_MAX_NREGS). There are also 4 fp condition code registers, so
+ 32+32+32+4 == 100.
+ Register 100 is used as the integer condition code register. */
+
+#define FIRST_PSEUDO_REGISTER 101
+
+/* Additional V9 fp regs. */
+#define SPARC_FIRST_V9_FP_REG 64
+#define SPARC_LAST_V9_FP_REG 95
+/* V9 %fcc[0123]. V8 uses (figuratively) %fcc0. */
+#define SPARC_FIRST_V9_FCC_REG 96
+#define SPARC_LAST_V9_FCC_REG 99
+/* V8 fcc reg. */
+#define SPARC_FCC_REG 96
+/* Integer CC reg. We don't distinguish %icc from %xcc. */
+#define SPARC_ICC_REG 100
/* 1 for registers that have pervasive standard uses
and are not available for the register allocator.
- g0 is used for the condition code and not to represent %g0, which is
- hardwired to 0, so reg 0 is *not* fixed.
- g1 through g4 are free to use as temporaries.
- g5 through g7 are reserved for the operating system. */
+ On non-v9 systems:
+ g1 is free to use as temporary.
+ g2-g4 are reserved for applications. Gcc normally uses them as
+ temporaries, but this can be disabled via the -mno-app-regs option.
+ g5 through g7 are reserved for the operating system.
+ On v9 systems:
+ g1 and g5 are free to use as temporaries.
+ g2-g4 are reserved for applications. Gcc normally uses them as
+ temporaries, but this can be disabled via the -mno-app-regs option.
+ g6-g7 are reserved for the operating system.
+ ??? Register 1 is used as a temporary by the 64 bit sethi pattern, so must
+ currently be a fixed register until this pattern is rewritten.
+ Register 1 is also used when restoring call-preserved registers in large
+ stack frames.
+
+ Registers fixed in arch32 and not arch64 (or vice-versa) are marked in
+ CONDITIONAL_REGISTER_USAGE in order to properly handle -ffixed-.
+*/
+
#define FIXED_REGISTERS \
- {0, 0, 0, 0, 0, 1, 1, 1, \
+ {1, 0, 0, 0, 0, 0, 1, 1, \
0, 0, 0, 0, 0, 0, 1, 0, \
0, 0, 0, 0, 0, 0, 0, 0, \
0, 0, 0, 0, 0, 0, 1, 1, \
0, 0, 0, 0, 0, 0, 0, 0, \
0, 0, 0, 0, 0, 0, 0, 0, \
0, 0, 0, 0, 0, 0, 0, 0, \
- 0, 0, 0, 0, 0, 0, 0, 0}
+ 0, 0, 0, 0, 0, 0, 0, 0, \
+ \
+ 0, 0, 0, 0, 0, 0, 0, 0, \
+ 0, 0, 0, 0, 0, 0, 0, 0, \
+ 0, 0, 0, 0, 0, 0, 0, 0, \
+ 0, 0, 0, 0, 0, 0, 0, 0, \
+ \
+ 0, 0, 0, 0, 0}
/* 1 for registers not available across function calls.
These must include the FIXED_REGISTERS and also any
The latter must include the registers where values are returned
and the register where structure-value addresses are passed.
Aside from that, you can include as many other registers as you like. */
+
#define CALL_USED_REGISTERS \
{1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
- 1, 1, 1, 1, 1, 1, 1, 1}
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ \
+ 1, 1, 1, 1, 1}
-/* If !TARGET_FPU, then make the fp registers fixed so that they won't
- be allocated. */
+/* If !TARGET_FPU, then make the fp registers and fp cc regs fixed so that
+ they won't be allocated. */
#define CONDITIONAL_REGISTER_USAGE \
do \
{ \
+ if (! SPARC_ARCH64) \
+ { \
+ fixed_regs[5] = 1; \
+ } \
+ if (SPARC_ARCH64) \
+ { \
+ int regno; \
+ fixed_regs[1] = 1; \
+ /* ??? We need to scan argv for -fcall-used-. */ \
+ for (regno = 48; regno < 80; regno++) \
+ call_used_regs[regno] = 0; \
+ } \
+ if (! TARGET_V9) \
+ { \
+ int regno; \
+ for (regno = SPARC_FIRST_V9_FP_REG; \
+ regno <= SPARC_LAST_V9_FP_REG; \
+ regno++) \
+ fixed_regs[regno] = 1; \
+ /* %fcc0 is used by v8 and v9. */ \
+ for (regno = SPARC_FIRST_V9_FCC_REG + 1; \
+ regno <= SPARC_LAST_V9_FCC_REG; \
+ regno++) \
+ fixed_regs[regno] = 1; \
+ } \
if (! TARGET_FPU) \
{ \
int regno; \
- for (regno = 32; regno < 64; regno++) \
+ for (regno = 32; regno < SPARC_LAST_V9_FCC_REG; regno++) \
fixed_regs[regno] = 1; \
} \
+ /* Don't unfix g2-g4 if they were fixed with -ffixed-. */ \
+ fixed_regs[2] |= ! TARGET_APP_REGS; \
+ fixed_regs[3] |= ! TARGET_APP_REGS; \
+ fixed_regs[4] |= ! TARGET_APP_REGS || TARGET_MEDANY; \
+ if (TARGET_FLAT) \
+ { \
+ /* Let the compiler believe the frame pointer is still \
+ %fp, but output it as %i7. */ \
+ fixed_regs[31] = 1; \
+ reg_names[FRAME_POINTER_REGNUM] = "%i7"; \
+ /* ??? This is a hack to disable leaf functions. */ \
+ global_regs[7] = 1; \
+ } \
+ if (profile_block_flag) \
+ { \
+ /* %g1 and %g2 must be fixed, because BLOCK_PROFILER \
+ uses them. */ \
+ fixed_regs[1] = 1; \
+ fixed_regs[2] = 1; \
+ } \
} \
while (0)
On SPARC, ordinary registers hold 32 bits worth;
this means both integer and floating point registers.
+ On v9, integer regs hold 64 bits worth; floating point regs hold
+ 32 bits worth (this includes the new fp regs as even the odd ones are
+ included in the hard register count). */
- We use vectors to keep this information about registers. */
-
-/* How many hard registers it takes to make a register of this mode. */
-extern int hard_regno_nregs[];
-
-#define HARD_REGNO_NREGS(REGNO, MODE) \
- ((GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
-
-/* Value is 1 if register/mode pair is acceptable on sparc. */
-extern int hard_regno_mode_ok[FIRST_PSEUDO_REGISTER];
+#define HARD_REGNO_NREGS(REGNO, MODE) \
+ (TARGET_ARCH64 \
+ ? ((REGNO) < 32 \
+ ? (GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD \
+ : (GET_MODE_SIZE (MODE) + 3) / 4) \
+ : ((GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD))
/* Value is 1 if hard register REGNO can hold a value of machine-mode MODE.
- On SPARC, the cpu registers can hold any mode but the float registers
- can only hold SFmode or DFmode. See sparc.c for how we
- initialize this. */
+ See sparc.c for how we initialize this. */
+extern int *hard_regno_mode_classes;
+extern int sparc_mode_class[];
#define HARD_REGNO_MODE_OK(REGNO, MODE) \
- ((hard_regno_mode_ok[REGNO] & (1<<(int)(MODE))) != 0)
+ ((hard_regno_mode_classes[REGNO] & sparc_mode_class[MODE]) != 0)
/* Value is 1 if it is a good idea to tie two pseudo registers
when one has mode MODE1 and one has mode MODE2.
If HARD_REGNO_MODE_OK could produce different values for MODE1 and MODE2,
- for any hard reg, then this must be 0 for correct output. */
+ for any hard reg, then this must be 0 for correct output.
+
+ For V9: SFmode can't be combined with other float modes, because they can't
+ be allocated to the %d registers. Also, DFmode won't fit in odd %f
+ registers, but SFmode will. */
#define MODES_TIEABLE_P(MODE1, MODE2) \
- ((MODE1) == (MODE2) || GET_MODE_CLASS (MODE1) == GET_MODE_CLASS (MODE2))
+ ((MODE1) == (MODE2) \
+ || (GET_MODE_CLASS (MODE1) == GET_MODE_CLASS (MODE2) \
+ && (! TARGET_V9 \
+ || (GET_MODE_CLASS (MODE1) != MODE_FLOAT \
+ || (MODE1 != SFmode && MODE2 != SFmode)))))
/* Specify the registers used for certain standard purposes.
The values of these macros are register numbers. */
/* Register to use for pushing function arguments. */
#define STACK_POINTER_REGNUM 14
-/* Actual top-of-stack address is 92 greater than the contents
- of the stack pointer register. 92 = 68 + 24. 64 bytes reserving space
- for the ins and local registers, 4 byte for structure return address, and
- 24 bytes for the 6 register parameters. */
+/* Actual top-of-stack address is 92/136 greater than the contents of the
+ stack pointer register for !v9/v9. That is:
+ - !v9: 64 bytes for the in and local registers, 4 bytes for structure return
+ address, and 24 bytes for the 6 register parameters.
+ - v9: 128 bytes for the in and local registers + 8 bytes reserved. */
#define STACK_POINTER_OFFSET FIRST_PARM_OFFSET(0)
+/* The stack bias (amount by which the hardware register is offset by). */
+#define SPARC_STACK_BIAS (TARGET_STACK_BIAS ? 2047 : 0)
+
/* Base register for access to local variables of the function. */
#define FRAME_POINTER_REGNUM 30
#if 0
-/* Register that is used for the return address. */
+/* Register that is used for the return address for the flat model. */
#define RETURN_ADDR_REGNUM 15
#endif
Zero means the frame pointer need not be set up (and parms
may be accessed via the stack pointer) in functions that seem suitable.
This is computed in `reload', in reload1.c.
+ Used in flow.c, global.c, and reload1.c.
- Used in flow.c, global.c, and reload1.c. */
-extern int leaf_function;
-
+ Being a non-leaf function does not mean a frame pointer is needed in the
+ flat window model. However, the debugger won't be able to backtrace through
+ us with out it. */
#define FRAME_POINTER_REQUIRED \
- (! (leaf_function_p () && only_leaf_regs_used ()))
+ (TARGET_FLAT ? (current_function_calls_alloca || current_function_varargs \
+ || !leaf_function_p ()) \
+ : ! (leaf_function_p () && only_leaf_regs_used ()))
/* C statement to store the difference between the frame pointer
and the stack pointer values immediately after the function prologue.
it's not, there's no point in trying to eliminate the
frame pointer. If it is a leaf function, we guessed right! */
#define INITIAL_FRAME_POINTER_OFFSET(VAR) \
- ((VAR) = (TARGET_FRW ? sparc_frw_compute_frame_size (get_frame_size ()) \
+ ((VAR) = (TARGET_FLAT ? sparc_flat_compute_frame_size (get_frame_size ()) \
: compute_frame_size (get_frame_size (), 1)))
/* Base register for access to arguments of the function. */
-#define ARG_POINTER_REGNUM 30
+#define ARG_POINTER_REGNUM FRAME_POINTER_REGNUM
-/* Register in which static-chain is passed to a function. */
-/* ??? */
-#define STATIC_CHAIN_REGNUM 1
+/* Register in which static-chain is passed to a function. This must
+ not be a register used by the prologue. */
+#define STATIC_CHAIN_REGNUM (TARGET_ARCH64 ? 5 : 2)
/* Register which holds offset table for position-independent
data references. */
#define FINALIZE_PIC finalize_pic ()
/* Sparc ABI says that quad-precision floats and all structures are returned
- in memory. */
-#define RETURN_IN_MEMORY(TYPE) \
- (TYPE_MODE (TYPE) == BLKmode || TYPE_MODE (TYPE) == TFmode)
+ in memory.
+ For v9, all aggregates are returned in memory. */
+#define RETURN_IN_MEMORY(TYPE) \
+ (TYPE_MODE (TYPE) == BLKmode \
+ || (! TARGET_ARCH64 && (TYPE_MODE (TYPE) == TFmode \
+ || TYPE_MODE (TYPE) == TCmode)))
/* Functions which return large structures get the address
to place the wanted value at offset 64 from the frame.
- Must reserve 64 bytes for the in and local registers. */
+ Must reserve 64 bytes for the in and local registers.
+ v9: Functions which return large structures get the address to place the
+ wanted value from an invisible first argument. */
/* Used only in other #defines in this file. */
#define STRUCT_VALUE_OFFSET 64
#define STRUCT_VALUE \
- gen_rtx (MEM, Pmode, \
- gen_rtx (PLUS, Pmode, stack_pointer_rtx, \
- gen_rtx (CONST_INT, VOIDmode, STRUCT_VALUE_OFFSET)))
+ (TARGET_ARCH64 \
+ ? 0 \
+ : gen_rtx (MEM, Pmode, \
+ gen_rtx (PLUS, Pmode, stack_pointer_rtx, \
+ gen_rtx (CONST_INT, VOIDmode, STRUCT_VALUE_OFFSET))))
#define STRUCT_VALUE_INCOMING \
- gen_rtx (MEM, Pmode, \
- gen_rtx (PLUS, Pmode, frame_pointer_rtx, \
- gen_rtx (CONST_INT, VOIDmode, STRUCT_VALUE_OFFSET)))
+ (TARGET_ARCH64 \
+ ? 0 \
+ : gen_rtx (MEM, Pmode, \
+ gen_rtx (PLUS, Pmode, frame_pointer_rtx, \
+ gen_rtx (CONST_INT, VOIDmode, STRUCT_VALUE_OFFSET))))
\f
/* Define the classes of registers for register constraints in the
machine description. Also define ranges of constants.
For any two classes, it is very desirable that there be another
class that represents their union. */
-/* The SPARC has two kinds of registers, general and floating point. */
-
-enum reg_class { NO_REGS, GENERAL_REGS, FP_REGS, ALL_REGS, LIM_REG_CLASSES };
+/* The SPARC has two kinds of registers, general and floating point.
+
+ For v9 we must distinguish between the upper and lower floating point
+ registers because the upper ones can't hold SFmode values.
+ HARD_REGNO_MODE_OK won't help here because reload assumes that register(s)
+ satisfying a group need for a class will also satisfy a single need for
+ that class. EXTRA_FP_REGS is a bit of a misnomer as it covers all 64 fp
+ regs.
+
+ It is important that one class contains all the general and all the standard
+ fp regs. Otherwise find_reg() won't properly allocate int regs for moves,
+ because reg_class_record() will bias the selection in favor of fp regs,
+ because reg_class_subunion[GENERAL_REGS][FP_REGS] will yield FP_REGS,
+ because FP_REGS > GENERAL_REGS.
+
+ It is also important that one class contain all the general and all the
+ fp regs. Otherwise when spilling a DFmode reg, it may be from EXTRA_FP_REGS
+ but find_reloads() may use class GENERAL_OR_FP_REGS. This will cause
+ allocate_reload_reg() to bypass it causing an abort because the compiler
+ thinks it doesn't have a spill reg when in fact it does.
+
+ v9 also has 4 floating point condition code registers. Since we don't
+ have a class that is the union of FPCC_REGS with either of the others,
+ it is important that it appear first. Otherwise the compiler will die
+ trying to compile _fixunsdfsi because fix_truncdfsi2 won't match its
+ constraints.
+
+ It is important that SPARC_ICC_REG have class NO_REGS. Otherwise combine
+ may try to use it to hold an SImode value. See register_operand.
+ ??? Should %fcc[0123] be handled similarily?
+*/
+
+enum reg_class { NO_REGS, FPCC_REGS, GENERAL_REGS, FP_REGS, EXTRA_FP_REGS,
+ GENERAL_OR_FP_REGS, GENERAL_OR_EXTRA_FP_REGS,
+ ALL_REGS, LIM_REG_CLASSES };
#define N_REG_CLASSES (int) LIM_REG_CLASSES
/* Give names of register classes as strings for dump file. */
#define REG_CLASS_NAMES \
- {"NO_REGS", "GENERAL_REGS", "FP_REGS", "ALL_REGS" }
+ { "NO_REGS", "FPCC_REGS", "GENERAL_REGS", "FP_REGS", "EXTRA_FP_REGS", \
+ "GENERAL_OR_FP_REGS", "GENERAL_OR_EXTRA_FP_REGS", "ALL_REGS" }
/* Define which registers fit in which classes.
This is an initializer for a vector of HARD_REG_SET
of length N_REG_CLASSES. */
-#if 0 && defined (__GNUC__)
-#define REG_CLASS_CONTENTS {0LL, 0xfffffffeLL, 0xffffffff00000000LL, 0xfffffffffffffffeLL}
-#else
-#define REG_CLASS_CONTENTS {{0, 0}, {-2, 0}, {0, -1}, {-2, -1}}
-#endif
+#define REG_CLASS_CONTENTS \
+ {{0, 0, 0, 0}, {0, 0, 0, 0xf}, \
+ {-1, 0, 0, 0}, {0, -1, 0, 0}, {0, -1, -1, 0}, \
+ {-1, -1, 0, 0}, {-1, -1, -1, 0}, {-1, -1, -1, 0x1f}}
/* The same information, inverted:
Return the class number of the smallest class containing
reg number REGNO. This could be a conditional expression
or could index an array. */
-#define REGNO_REG_CLASS(REGNO) \
- ((REGNO) >= 32 ? FP_REGS : (REGNO) == 0 ? NO_REGS : GENERAL_REGS)
+extern enum reg_class sparc_regno_reg_class[];
-/* This is the order in which to allocate registers
- normally.
+#define REGNO_REG_CLASS(REGNO) sparc_regno_reg_class[(REGNO)]
+
+/* This is the order in which to allocate registers normally.
We put %f0/%f1 last among the float registers, so as to make it more
likely that a pseudo-register which dies in the float return register
will get allocated to the float return register, thus saving a move
- instruction at the end of the function. */
+ instruction at the end of the function.
+
+ The float registers are ordered a little "funny" because in the 64 bit
+ architecture, some of them (%f16-%f47) are call-preserved. */
+
#define REG_ALLOC_ORDER \
-{ 8, 9, 10, 11, 12, 13, 2, 3, \
- 15, 16, 17, 18, 19, 20, 21, 22, \
+{ 8, 9, 10, 11, 12, 13, 2, 3, \
+ 15, 16, 17, 18, 19, 20, 21, 22, \
23, 24, 25, 26, 27, 28, 29, 31, \
- 34, 35, 36, 37, 38, 39, \
- 40, 41, 42, 43, 44, 45, 46, 47, \
- 48, 49, 50, 51, 52, 53, 54, 55, \
- 56, 57, 58, 59, 60, 61, 62, 63, \
- 32, 33, \
+ 34, 35, 36, 37, 38, 39, /* %f2-%f7 */ \
+ 40, 41, 42, 43, 44, 45, 46, 47, /* %f8-%f15 */ \
+ 80, 81, 82, 83, 84, 85, 86, 87, /* %f48-%f55 */ \
+ 88, 89, 90, 91, 92, 93, 94, 95, /* %f56-%f63 */ \
+ 48, 49, 50, 51, 52, 53, 54, 55, /* %f16-%f23 */ \
+ 56, 57, 58, 59, 60, 61, 62, 63, /* %f24-%f31 */ \
+ 64, 65, 66, 67, 68, 69, 70, 71, /* %f32-%f39 */ \
+ 72, 73, 74, 75, 76, 77, 78, 79, /* %f40-%f47 */ \
+ 32, 33, /* %f0,%f1 */ \
+ 96, 97, 98, 99, 100, /* %fcc0-3, %icc */ \
1, 4, 5, 6, 7, 0, 14, 30}
/* This is the order in which to allocate registers for
leaf functions. If all registers can fit in the "i" registers,
- then we have the possibility of having a leaf function. */
+ then we have the possibility of having a leaf function.
+ The floating point registers are ordered a little "funny" because in the
+ 64 bit architecture some of them (%f16-%f47) are call-preserved. */
+
#define REG_LEAF_ALLOC_ORDER \
{ 2, 3, 24, 25, 26, 27, 28, 29, \
15, 8, 9, 10, 11, 12, 13, \
16, 17, 18, 19, 20, 21, 22, 23, \
34, 35, 36, 37, 38, 39, \
40, 41, 42, 43, 44, 45, 46, 47, \
+ 80, 81, 82, 83, 84, 85, 86, 87, \
+ 88, 89, 90, 91, 92, 93, 94, 95, \
48, 49, 50, 51, 52, 53, 54, 55, \
56, 57, 58, 59, 60, 61, 62, 63, \
+ 64, 65, 66, 67, 68, 69, 70, 71, \
+ 72, 73, 74, 75, 76, 77, 78, 79, \
32, 33, \
+ 96, 97, 98, 99, 100, \
1, 4, 5, 6, 7, 0, 14, 30, 31}
#define ORDER_REGS_FOR_LOCAL_ALLOC order_regs_for_local_alloc ()
+/* ??? %g7 is not a leaf register to effectively #undef LEAF_REGISTERS when
+ -mflat is used. Function only_leaf_regs_used will return 0 if a global
+ register is used and is not permitted in a leaf function. We make %g7
+ a global reg if -mflat and voila. Since %g7 is a system register and is
+ fixed it won't be used by gcc anyway. */
+
#define LEAF_REGISTERS \
-{ 1, 1, 1, 1, 1, 1, 1, 1, \
+{ 1, 1, 1, 1, 1, 1, 1, 0, \
0, 0, 0, 0, 0, 0, 1, 0, \
0, 0, 0, 0, 0, 0, 0, 0, \
1, 1, 1, 1, 1, 1, 0, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
1, 1, 1, 1, 1, 1, 1, 1, \
- 1, 1, 1, 1, 1, 1, 1, 1}
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1}
extern char leaf_reg_remap[];
#define LEAF_REG_REMAP(REGNO) (leaf_reg_remap[REGNO])
-extern char leaf_reg_backmap[];
-#define LEAF_REG_BACKMAP(REGNO) (leaf_reg_backmap[REGNO])
/* The class value for index registers, and the one for base regs. */
#define INDEX_REG_CLASS GENERAL_REGS
#define BASE_REG_CLASS GENERAL_REGS
-/* Get reg_class from a letter such as appears in the machine description. */
+/* Local macro to handle the two v9 classes of FP regs. */
+#define FP_REG_CLASS_P(CLASS) ((CLASS) == FP_REGS || (CLASS) == EXTRA_FP_REGS)
+
+/* Get reg_class from a letter such as appears in the machine description.
+ In the not-v9 case, coerce v9's 'e' class to 'f', so we can use 'e' in the
+ .md file for v8 and v9. */
#define REG_CLASS_FROM_LETTER(C) \
- ((C) == 'f' ? FP_REGS : (C) == 'r' ? GENERAL_REGS : NO_REGS)
+(TARGET_V9 \
+ ? ((C) == 'f' ? FP_REGS \
+ : (C) == 'e' ? EXTRA_FP_REGS \
+ : (C) == 'c' ? FPCC_REGS \
+ : NO_REGS) \
+ : ((C) == 'f' ? FP_REGS \
+ : (C) == 'e' ? FP_REGS \
+ : (C) == 'c' ? FPCC_REGS \
+ : NO_REGS))
/* The letters I, J, K, L and M in a register constraint string
can be used to stand for particular ranges of immediate operands.
C is the letter, and VALUE is a constant value.
Return 1 if VALUE is in the range specified by C.
- For SPARC, `I' is used for the range of constants an insn
- can actually contain.
+ `I' is used for the range of constants an insn can actually contain.
`J' is used for the range which is just zero (since that is R0).
- `K' is used for constants which can be loaded with a single sethi insn. */
+ `K' is used for constants which can be loaded with a single sethi insn.
+ `L' is used for the range of constants supported by the movcc insns.
+ `M' is used for the range of constants supported by the movrcc insns. */
-#define SMALL_INT(X) ((unsigned) (INTVAL (X) + 0x1000) < 0x2000)
+#define SPARC_SIMM10_P(X) ((unsigned HOST_WIDE_INT) ((X) + 0x200) < 0x400)
+#define SPARC_SIMM11_P(X) ((unsigned HOST_WIDE_INT) ((X) + 0x400) < 0x800)
+#define SPARC_SIMM13_P(X) ((unsigned HOST_WIDE_INT) ((X) + 0x1000) < 0x2000)
+/* 10 and 11 bit immediates are only used for a few specific insns.
+ SMALL_INT is used throughout the port so we continue to use it. */
+#define SMALL_INT(X) (SPARC_SIMM13_P (INTVAL (X)))
#define CONST_OK_FOR_LETTER_P(VALUE, C) \
- ((C) == 'I' ? (unsigned) ((VALUE) + 0x1000) < 0x2000 \
+ ((C) == 'I' ? SPARC_SIMM13_P (VALUE) \
: (C) == 'J' ? (VALUE) == 0 \
: (C) == 'K' ? ((VALUE) & 0x3ff) == 0 \
+ : (C) == 'L' ? SPARC_SIMM11_P (VALUE) \
+ : (C) == 'M' ? SPARC_SIMM10_P (VALUE) \
: 0)
/* Similar, but for floating constants, and defining letters G and H.
if an 'E' constraint fails to match it. */
#define PREFERRED_RELOAD_CLASS(X,CLASS) \
(CONSTANT_P (X) \
- && ((CLASS) == FP_REGS \
+ && (FP_REG_CLASS_P (CLASS) \
|| (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT \
&& (HOST_FLOAT_FORMAT != IEEE_FLOAT_FORMAT \
|| HOST_BITS_PER_INT != BITS_PER_WORD))) \
a paradoxical subreg in a float/fix conversion insn. */
#define SECONDARY_INPUT_RELOAD_CLASS(CLASS, MODE, IN) \
- (((CLASS) == FP_REGS && ((MODE) == HImode || (MODE) == QImode)\
- && (GET_CODE (IN) == MEM \
- || ((GET_CODE (IN) == REG || GET_CODE (IN) == SUBREG) \
- && true_regnum (IN) == -1))) ? GENERAL_REGS : NO_REGS)
+ ((FP_REG_CLASS_P (CLASS) && ((MODE) == HImode || (MODE) == QImode) \
+ && (GET_CODE (IN) == MEM \
+ || ((GET_CODE (IN) == REG || GET_CODE (IN) == SUBREG) \
+ && true_regnum (IN) == -1))) ? GENERAL_REGS : NO_REGS)
#define SECONDARY_OUTPUT_RELOAD_CLASS(CLASS, MODE, IN) \
- ((CLASS) == FP_REGS && ((MODE) == HImode || (MODE) == QImode) \
- && (GET_CODE (IN) == MEM \
- || ((GET_CODE (IN) == REG || GET_CODE (IN) == SUBREG) \
- && true_regnum (IN) == -1)) ? GENERAL_REGS : NO_REGS)
+ ((FP_REG_CLASS_P (CLASS) && ((MODE) == HImode || (MODE) == QImode) \
+ && (GET_CODE (IN) == MEM \
+ || ((GET_CODE (IN) == REG || GET_CODE (IN) == SUBREG) \
+ && true_regnum (IN) == -1))) ? GENERAL_REGS : NO_REGS)
/* On SPARC it is not possible to directly move data between
GENERAL_REGS and FP_REGS. */
-#define SECONDARY_MEMORY_NEEDED(CLASS1, CLASS2, MODE) \
- (((CLASS1) == FP_REGS && (CLASS2) == GENERAL_REGS) \
- || ((CLASS1) == GENERAL_REGS && (CLASS2) == FP_REGS))
+#define SECONDARY_MEMORY_NEEDED(CLASS1, CLASS2, MODE) \
+ (FP_REG_CLASS_P (CLASS1) != FP_REG_CLASS_P (CLASS2))
-/* Return the stack location to use for secondary memory needed reloads. */
+/* Return the stack location to use for secondary memory needed reloads.
+ We want to use the reserved location just below the frame pointer.
+ However, we must ensure that there is a frame, so use assign_stack_local
+ if the frame size is zero. */
#define SECONDARY_MEMORY_NEEDED_RTX(MODE) \
- gen_rtx (MEM, MODE, gen_rtx (PLUS, Pmode, frame_pointer_rtx, \
- GEN_INT (STARTING_FRAME_OFFSET)))
+ (get_frame_size () == 0 \
+ ? assign_stack_local (MODE, GET_MODE_SIZE (MODE), 0) \
+ : gen_rtx (MEM, MODE, gen_rtx (PLUS, Pmode, frame_pointer_rtx, \
+ GEN_INT (STARTING_FRAME_OFFSET))))
+
+/* Get_secondary_mem widens it's argument to BITS_PER_WORD which loses on v9
+ because the movsi and movsf patterns don't handle r/f moves.
+ For v8 we copy the default definition. */
+#define SECONDARY_MEMORY_NEEDED_MODE(MODE) \
+ (TARGET_ARCH64 \
+ ? (GET_MODE_BITSIZE (MODE) < 32 \
+ ? mode_for_size (32, GET_MODE_CLASS (MODE), 0) \
+ : MODE) \
+ : (GET_MODE_BITSIZE (MODE) < BITS_PER_WORD \
+ ? mode_for_size (BITS_PER_WORD, GET_MODE_CLASS (MODE), 0) \
+ : MODE))
/* Return the maximum number of consecutive registers
needed to represent mode MODE in a register of class CLASS. */
/* On SPARC, this is the size of MODE in words. */
#define CLASS_MAX_NREGS(CLASS, MODE) \
- ((GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
+ (FP_REG_CLASS_P (CLASS) ? (GET_MODE_SIZE (MODE) + 3) / 4 \
+ : (GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
\f
/* Stack layout; function entry, exit and calling. */
/* Define the number of register that can hold parameters.
- These two macros are used only in other macro definitions below. */
-#define NPARM_REGS 6
+ These two macros are used only in other macro definitions below.
+ MODE is the mode of the argument.
+ !v9: All args are passed in %o0-%o5.
+ v9: Non-float args are passed in %o0-5 and float args are passed in
+ %f0-%f15. */
+#define NPARM_REGS(MODE) \
+ (TARGET_ARCH64 ? (GET_MODE_CLASS (MODE) == MODE_FLOAT ? 16 : 6) : 6)
/* Define this if pushing a word on the stack
makes the stack pointer a smaller address. */
If FRAME_GROWS_DOWNWARD, this is the offset to the END of the
first local allocated. Otherwise, it is the offset to the BEGINNING
of the first local allocated. */
-/* This is 16 to allow space for one TFmode floating point value. */
-#define STARTING_FRAME_OFFSET (-16)
+/* This allows space for one TFmode floating point value. */
+#define STARTING_FRAME_OFFSET \
+ (TARGET_ARCH64 ? (SPARC_STACK_BIAS - 16) \
+ : (-SPARC_STACK_ALIGN (LONG_DOUBLE_TYPE_SIZE / BITS_PER_UNIT)))
/* If we generate an insn to push BYTES bytes,
this says how many the stack pointer really advances by.
/* #define PUSH_ROUNDING(BYTES) */
/* Offset of first parameter from the argument pointer register value.
- This is 64 for the ins and locals, plus 4 for the struct-return reg
- even if this function isn't going to use it. */
-#define FIRST_PARM_OFFSET(FNDECL) (STRUCT_VALUE_OFFSET + UNITS_PER_WORD)
+ !v9: This is 64 for the ins and locals, plus 4 for the struct-return reg
+ even if this function isn't going to use it.
+ v9: This is 128 for the ins and locals, plus a reserved space of 8. */
+#define FIRST_PARM_OFFSET(FNDECL) \
+ (TARGET_ARCH64 ? (SPARC_STACK_BIAS + 136) \
+ : (STRUCT_VALUE_OFFSET + UNITS_PER_WORD))
/* When a parameter is passed in a register, stack space is still
allocated for it. */
-#define REG_PARM_STACK_SPACE(DECL) (NPARM_REGS * UNITS_PER_WORD)
+#if ! SPARC_ARCH64
+#define REG_PARM_STACK_SPACE(DECL) (NPARM_REGS (SImode) * UNITS_PER_WORD)
+#endif
/* Keep the stack pointer constant throughout the function.
This is both an optimization and a necessity: longjmp
/* Value is the number of bytes of arguments automatically
popped when returning from a subroutine call.
+ FUNDECL is the declaration node of the function (as a tree),
FUNTYPE is the data type of the function (as a tree),
or for a library call it is an identifier node for the subroutine name.
SIZE is the number of bytes of arguments passed on the stack. */
-#define RETURN_POPS_ARGS(FUNTYPE,SIZE) 0
+#define RETURN_POPS_ARGS(FUNDECL,FUNTYPE,SIZE) 0
/* Some subroutine macros specific to this machine.
When !TARGET_FPU, put float return values in the general registers,
since we don't have any fp registers. */
#define BASE_RETURN_VALUE_REG(MODE) \
- (((MODE) == SFmode || (MODE) == DFmode) && TARGET_FPU ? 32 : 8)
+ (TARGET_ARCH64 \
+ ? (TARGET_FPU && GET_MODE_CLASS (MODE) == MODE_FLOAT ? 32 : 8) \
+ : (((MODE) == SFmode || (MODE) == DFmode) && TARGET_FPU ? 32 : 8))
#define BASE_OUTGOING_VALUE_REG(MODE) \
- (((MODE) == SFmode || (MODE) == DFmode) && TARGET_FPU ? 32 \
- : (TARGET_FRW ? 8 : 24))
-#define BASE_PASSING_ARG_REG(MODE) (8)
-#define BASE_INCOMING_ARG_REG(MODE) (TARGET_FRW ? 8 : 24)
+ (TARGET_ARCH64 \
+ ? (TARGET_FPU && GET_MODE_CLASS (MODE) == MODE_FLOAT ? 32 \
+ : TARGET_FLAT ? 8 : 24) \
+ : (((MODE) == SFmode || (MODE) == DFmode) && TARGET_FPU ? 32 \
+ : (TARGET_FLAT ? 8 : 24)))
+#define BASE_PASSING_ARG_REG(MODE) \
+ (TARGET_ARCH64 \
+ ? (TARGET_FPU && GET_MODE_CLASS (MODE) == MODE_FLOAT ? 32 : 8) \
+ : 8)
+#define BASE_INCOMING_ARG_REG(MODE) \
+ (TARGET_ARCH64 \
+ ? (TARGET_FPU && GET_MODE_CLASS (MODE) == MODE_FLOAT ? 32 \
+ : TARGET_FLAT ? 8 : 24) \
+ : (TARGET_FLAT ? 8 : 24))
/* Define this macro if the target machine has "register windows". This
C expression returns the register number as seen by the called function
Return OUT if register number OUT is not an outbound register. */
#define INCOMING_REGNO(OUT) \
- ((TARGET_FRW || (OUT) < 8 || (OUT) > 15) ? (OUT) : (OUT) + 16)
+ ((TARGET_FLAT || (OUT) < 8 || (OUT) > 15) ? (OUT) : (OUT) + 16)
/* Define this macro if the target machine has "register windows". This
C expression returns the register number as seen by the calling function
Return IN if register number IN is not an inbound register. */
#define OUTGOING_REGNO(IN) \
- ((TARGET_FRW || (IN) < 24 || (IN) > 31) ? (IN) : (IN) - 16)
+ ((TARGET_FLAT || (IN) < 24 || (IN) > 31) ? (IN) : (IN) - 16)
/* Define how to find the value returned by a function.
VALTYPE is the data type of the value (as a tree).
#define FUNCTION_VALUE_REGNO_P(N) ((N) == 8 || (N) == 32)
+/* Define the size of space to allocate for the return value of an
+ untyped_call. */
+
+#define APPLY_RESULT_SIZE 16
+
/* 1 if N is a possible register number for function argument passing.
- On SPARC, these are the "output" registers. */
+ On SPARC, these are the "output" registers. v9 also uses %f0-%f15. */
-#define FUNCTION_ARG_REGNO_P(N) ((N) < 14 && (N) > 7)
+#define FUNCTION_ARG_REGNO_P(N) \
+ (TARGET_ARCH64 ? (((N) < 14 && (N) > 7) || (N) > 31 && (N) < 48) \
+ : ((N) < 14 && (N) > 7))
\f
/* Define a data type for recording info about an argument list
during the scan of that argument list. This data type should
and about the args processed so far, enough to enable macros
such as FUNCTION_ARG to determine where the next arg should go.
- On SPARC, this is a single integer, which is a number of words
+ On SPARC (!v9), this is a single integer, which is a number of words
of arguments scanned so far (including the invisible argument,
if any, which holds the structure-value-address).
- Thus 7 or more means all following args should go on the stack. */
+ Thus 7 or more means all following args should go on the stack.
+
+ For v9, we record how many of each type has been passed. Different
+ types get passed differently.
+
+ - Float args are passed in %f0-15, after which they go to the stack
+ where floats and doubles are passed 8 byte aligned and long doubles
+ are passed 16 byte aligned.
+ - All aggregates are passed by reference. The callee copies
+ the structure if necessary, except if stdarg/varargs and the struct
+ matches the ellipse in which case the caller makes a copy.
+ - Any non-float argument might be split between memory and reg %o5.
+ ??? I don't think this can ever happen now that structs are no
+ longer passed in regs.
+
+ For v9 return values:
+
+ - For all aggregates, the caller allocates space for the return value,
+ and passes the pointer as an implicit first argument, which is
+ allocated like all other arguments.
+ - The unimp instruction stuff for structure returns is gone. */
+
+#if SPARC_ARCH64
+enum sparc_arg_class { SPARC_ARG_INT = 0, SPARC_ARG_FLOAT = 1 };
+struct sparc_args {
+ int arg_count[2]; /* must be int! (for __builtin_args_info) */
+};
+#define CUMULATIVE_ARGS struct sparc_args
+
+/* Return index into CUMULATIVE_ARGS. */
+
+#define GET_SPARC_ARG_CLASS(MODE) \
+ (GET_MODE_CLASS (MODE) == MODE_FLOAT ? SPARC_ARG_FLOAT : SPARC_ARG_INT)
+
+/* Round a register number up to a proper boundary for an arg of mode MODE.
+ This macro is only used in this file.
+
+ The "& (0x10000 - ...)" is used to round up to the next appropriate reg. */
+
+#define ROUND_REG(CUM, MODE) \
+ (GET_MODE_CLASS (MODE) != MODE_FLOAT \
+ ? (CUM).arg_count[(int) GET_SPARC_ARG_CLASS (MODE)] \
+ : ((CUM).arg_count[(int) GET_SPARC_ARG_CLASS (MODE)] \
+ + GET_MODE_UNIT_SIZE (MODE) / 4 - 1) \
+ & (0x10000 - GET_MODE_UNIT_SIZE (MODE) / 4))
+
+#define ROUND_ADVANCE(SIZE) \
+ (((SIZE) + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
+
+#else /* ! SPARC_ARCH64 */
#define CUMULATIVE_ARGS int
+#define ROUND_REG(CUM, MODE) (CUM)
+
#define ROUND_ADVANCE(SIZE) \
((SIZE + UNITS_PER_WORD - 1) / UNITS_PER_WORD)
+#endif /* ! SPARC_ARCH64 */
/* Initialize a variable CUM of type CUMULATIVE_ARGS
for a call to a function whose data type is FNTYPE.
On SPARC, the offset always starts at 0: the first parm reg is always
the same reg. */
-#define INIT_CUMULATIVE_ARGS(CUM,FNTYPE,LIBNAME) ((CUM) = 0)
+#if SPARC_ARCH64
+extern int sparc_arg_count,sparc_n_named_args;
+#define INIT_CUMULATIVE_ARGS(CUM,FNTYPE,LIBNAME,INDIRECT) \
+ do { \
+ (CUM).arg_count[(int) SPARC_ARG_INT] = 0; \
+ (CUM).arg_count[(int) SPARC_ARG_FLOAT] = 0; \
+ sparc_arg_count = 0; \
+ sparc_n_named_args = \
+ ((FNTYPE) && TYPE_ARG_TYPES (FNTYPE) \
+ ? (list_length (TYPE_ARG_TYPES (FNTYPE)) \
+ + (TREE_CODE (TREE_TYPE (FNTYPE)) == RECORD_TYPE \
+ || TREE_CODE (TREE_TYPE (FNTYPE)) == QUAL_UNION_TYPE\
+ || TREE_CODE (TREE_TYPE (FNTYPE)) == SET_TYPE \
+ || TREE_CODE (TREE_TYPE (FNTYPE)) == UNION_TYPE)) \
+ /* Can't tell, treat 'em all as named. */ \
+ : 10000); \
+ } while (0)
+#else
+#define INIT_CUMULATIVE_ARGS(CUM,FNTYPE,LIBNAME,INDIRECT) ((CUM) = 0)
+#endif
/* Update the data in CUM to advance over an argument
of mode MODE and data type TYPE.
(TYPE is null for libcalls where that information may not be available.) */
+#if SPARC_ARCH64
+#define FUNCTION_ARG_ADVANCE(CUM, MODE, TYPE, NAMED) \
+ do { \
+ (CUM).arg_count[(int) GET_SPARC_ARG_CLASS (MODE)] = \
+ ROUND_REG ((CUM), (MODE)) \
+ + (GET_MODE_CLASS (MODE) == MODE_FLOAT \
+ ? GET_MODE_SIZE (MODE) / 4 \
+ : ROUND_ADVANCE ((MODE) == BLKmode \
+ ? GET_MODE_SIZE (Pmode) \
+ : GET_MODE_SIZE (MODE))); \
+ sparc_arg_count++; \
+ } while (0)
+#else
#define FUNCTION_ARG_ADVANCE(CUM, MODE, TYPE, NAMED) \
((CUM) += ((MODE) != BLKmode \
? ROUND_ADVANCE (GET_MODE_SIZE (MODE)) \
: ROUND_ADVANCE (int_size_in_bytes (TYPE))))
+#endif
+
+/* Return boolean indicating arg of mode MODE will be passed in a reg.
+ This macro is only used in this file. */
+
+#if SPARC_ARCH64
+#define PASS_IN_REG_P(CUM, MODE, TYPE) \
+ (ROUND_REG ((CUM), (MODE)) < NPARM_REGS (MODE) \
+ && ((TYPE)==0 || ! TREE_ADDRESSABLE ((tree)(TYPE))) \
+ && ((TYPE)==0 || (MODE) != BLKmode))
+#else
+#define PASS_IN_REG_P(CUM, MODE, TYPE) \
+ ((CUM) < NPARM_REGS (SImode) \
+ && ((TYPE)==0 || ! TREE_ADDRESSABLE ((tree)(TYPE))) \
+ && ((TYPE)==0 || (MODE) != BLKmode \
+ || (TYPE_ALIGN (TYPE) % PARM_BOUNDARY == 0)))
+#endif
/* Determine where to put an argument to a function.
Value is zero to push the argument on the stack,
/* On SPARC the first six args are normally in registers
and the rest are pushed. Any arg that starts within the first 6 words
+ is at least partially passed in a register unless its data type forbids.
+ For v9, the first 6 int args are passed in regs and the first N
+ float args are passed in regs (where N is such that %f0-15 are filled).
+ The rest are pushed. Any arg that starts within the first 6 words
is at least partially passed in a register unless its data type forbids. */
#define FUNCTION_ARG(CUM, MODE, TYPE, NAMED) \
-((CUM) < NPARM_REGS \
- && ((TYPE)==0 || ! TREE_ADDRESSABLE ((tree)(TYPE))) \
- && ((TYPE)==0 || (MODE) != BLKmode \
- || (TYPE_ALIGN ((TYPE)) % PARM_BOUNDARY == 0)) \
- ? gen_rtx (REG, (MODE), (BASE_PASSING_ARG_REG (MODE) + (CUM))) \
- : 0)
+ (PASS_IN_REG_P ((CUM), (MODE), (TYPE)) \
+ ? gen_rtx (REG, (MODE), \
+ (BASE_PASSING_ARG_REG (MODE) + ROUND_REG ((CUM), (MODE))))\
+ : 0)
/* Define where a function finds its arguments.
This is different from FUNCTION_ARG because of register windows. */
#define FUNCTION_INCOMING_ARG(CUM, MODE, TYPE, NAMED) \
-((CUM) < NPARM_REGS \
- && ((TYPE)==0 || ! TREE_ADDRESSABLE ((tree)(TYPE))) \
- && ((TYPE)==0 || (MODE) != BLKmode \
- || (TYPE_ALIGN ((TYPE)) % PARM_BOUNDARY == 0)) \
- ? gen_rtx (REG, (MODE), (BASE_INCOMING_ARG_REG (MODE) + (CUM))) \
- : 0)
+ (PASS_IN_REG_P ((CUM), (MODE), (TYPE)) \
+ ? gen_rtx (REG, (MODE), \
+ (BASE_INCOMING_ARG_REG (MODE) + ROUND_REG ((CUM), (MODE))))\
+ : 0)
/* For an arg passed partly in registers and partly in memory,
this is the number of registers used.
For args passed entirely in registers or entirely in memory, zero.
Any arg that starts in the first 6 regs but won't entirely fit in them
- needs partial registers on the Sparc. */
+ needs partial registers on the Sparc (!v9). On v9, there are no arguments
+ that are passed partially in registers (??? complex values?). */
+#if ! SPARC_ARCH64
#define FUNCTION_ARG_PARTIAL_NREGS(CUM, MODE, TYPE, NAMED) \
- ((CUM) < NPARM_REGS \
- && ((TYPE)==0 || ! TREE_ADDRESSABLE ((tree)(TYPE))) \
- && ((TYPE)==0 || (MODE) != BLKmode \
- || (TYPE_ALIGN ((TYPE)) % PARM_BOUNDARY == 0)) \
- && ((CUM) + ((MODE) == BLKmode \
- ? ROUND_ADVANCE (int_size_in_bytes (TYPE)) \
- : ROUND_ADVANCE (GET_MODE_SIZE (MODE))) - NPARM_REGS > 0)\
- ? (NPARM_REGS - (CUM)) \
+ (PASS_IN_REG_P ((CUM), (MODE), (TYPE)) \
+ && ((CUM) + ((MODE) == BLKmode \
+ ? ROUND_ADVANCE (int_size_in_bytes (TYPE)) \
+ : ROUND_ADVANCE (GET_MODE_SIZE (MODE))) - NPARM_REGS (SImode) > 0)\
+ ? (NPARM_REGS (SImode) - (CUM)) \
: 0)
+#endif
/* The SPARC ABI stipulates passing struct arguments (of any size) and
- quad-precision floats by invisible reference. */
-#define FUNCTION_ARG_PASS_BY_REFERENCE(CUM, MODE, TYPE, NAMED) \
- ((TYPE && (TREE_CODE (TYPE) == RECORD_TYPE \
- || TREE_CODE (TYPE) == UNION_TYPE)) \
- || (MODE == TFmode))
+ (!v9) quad-precision floats by invisible reference.
+ For Pascal, also pass arrays by reference. */
+#define FUNCTION_ARG_PASS_BY_REFERENCE(CUM, MODE, TYPE, NAMED) \
+ ((TYPE && AGGREGATE_TYPE_P (TYPE)) \
+ || (!TARGET_ARCH64 && MODE == TFmode))
+
+/* A C expression that indicates when it is the called function's
+ responsibility to make copies of arguments passed by reference.
+ If the callee can determine that the argument won't be modified, it can
+ avoid the copy. */
+/* ??? We'd love to be able to use NAMED here. Unfortunately, it doesn't
+ include the last named argument so we keep track of the args ourselves. */
+
+#if SPARC_ARCH64
+#define FUNCTION_ARG_CALLEE_COPIES(CUM, MODE, TYPE, NAMED) \
+ (sparc_arg_count < sparc_n_named_args)
+#endif
+\f
+/* Initialize data used by insn expanders. This is called from
+ init_emit, once for each function, before code is generated.
+ For v9, clear the temp slot used by float/int DImode conversions.
+ ??? There is the 16 bytes at [%fp-16], however we'd like to delete this
+ space at some point.
+ ??? Use assign_stack_temp? */
+
+extern void sparc64_init_expanders ();
+extern struct rtx_def *sparc64_fpconv_stack_temp ();
+#if SPARC_ARCH64
+#define INIT_EXPANDERS sparc64_init_expanders ()
+#endif
/* Define the information needed to generate branch and scc insns. This is
stored from the compare operation. Note that we can't use "rtx" here
/* Define the function that build the compare insn for scc and bcc. */
extern struct rtx_def *gen_compare_reg ();
+
+/* This function handles all v9 scc insns */
+
+extern int gen_v9_scc ();
\f
/* Generate the special assembly code needed to tell the assembler whatever
it might need to know about the return value of a function.
to do a "save" insn. The decision about whether or not
to do this is made in regclass.c. */
-#define FUNCTION_PROLOGUE(FILE, SIZE) \
- (TARGET_FRW ? sparc_frw_output_function_prologue (FILE, SIZE, leaf_function)\
+extern int leaf_function;
+#define FUNCTION_PROLOGUE(FILE, SIZE) \
+ (TARGET_FLAT ? sparc_flat_output_function_prologue (FILE, SIZE) \
: output_function_prologue (FILE, SIZE, leaf_function))
-
+\f
/* Output assembler code to FILE to increment profiler label # LABELNO
- for profiling a function entry. */
+ for profiling a function entry.
+
+ 32 bit sparc uses %g2 as the STATIC_CHAIN_REGNUM which gets clobbered
+ during profiling so we need to save/restore it around the call to mcount.
+ We're guaranteed that a save has just been done, and we use the space
+ allocated for intreg/fpreg value passing. */
#define FUNCTION_PROFILER(FILE, LABELNO) \
do { \
- fputs ("\tsethi %hi(", (FILE)); \
- ASM_OUTPUT_INTERNAL_LABELREF (FILE, "LP", LABELNO); \
- fputs ("),%o0\n\tcall mcount\n\tor %lo(", (FILE)); \
- ASM_OUTPUT_INTERNAL_LABELREF (FILE, "LP", LABELNO); \
- fputs ("),%o0,%o0\n", (FILE)); \
+ char buf[20]; \
+ ASM_GENERATE_INTERNAL_LABEL (buf, "LP", (LABELNO)); \
+ if (! TARGET_ARCH64) \
+ fputs ("\tst %g2,[%fp-4]\n", FILE); \
+ fputs ("\tsethi %hi(", FILE); \
+ assemble_name (FILE, buf); \
+ fputs ("),%o0\n", FILE); \
+ if (TARGET_MEDANY) \
+ fprintf (FILE, "\tadd %o0,%s,%o0\n", MEDANY_BASE_REG); \
+ fputs ("\tcall mcount\n\tadd %o0,%lo(", FILE); \
+ assemble_name (FILE, buf); \
+ fputs ("),%o0\n", FILE); \
+ if (! TARGET_ARCH64) \
+ fputs ("\tld [%fp-4],%g2\n", FILE); \
} while (0)
-/* Output assembler code to FILE to initialize this source file's
- basic block profiling info, if that has not already been done. */
-/* FIXME -- this does not parameterize how it generates labels (like the
- above FUNCTION_PROFILER). Broken on Solaris-2. --gnu@cygnus.com */
-
-#define FUNCTION_BLOCK_PROFILER(FILE, LABELNO) \
- fprintf (FILE, "\tsethi %%hi(LPBX0),%%o0\n\tld [%%lo(LPBX0)+%%o0],%%o1\n\ttst %%o1\n\tbne LPY%d\n\tadd %%o0,%%lo(LPBX0),%%o0\n\tcall ___bb_init_func\n\tnop\nLPY%d:\n", \
- (LABELNO), (LABELNO))
-
-/* Output assembler code to FILE to increment the entry-count for
- the BLOCKNO'th basic block in this source file. */
-
-#define BLOCK_PROFILER(FILE, BLOCKNO) \
-{ \
- int blockn = (BLOCKNO); \
- fprintf (FILE, "\tsethi %%hi(LPBX2+%d),%%g1\n\tld [%%lo(LPBX2+%d)+%%g1],%%g2\n\
-\tadd %%g2,1,%%g2\n\tst %%g2,[%%lo(LPBX2+%d)+%%g1]\n", \
- 4 * blockn, 4 * blockn, 4 * blockn); \
-}
+/* There are three profiling modes for basic blocks available.
+ The modes are selected at compile time by using the options
+ -a or -ax of the gnu compiler.
+ The variable `profile_block_flag' will be set according to the
+ selected option.
+
+ profile_block_flag == 0, no option used:
+
+ No profiling done.
+
+ profile_block_flag == 1, -a option used.
+
+ Count frequency of execution of every basic block.
+
+ profile_block_flag == 2, -ax option used.
+
+ Generate code to allow several different profiling modes at run time.
+ Available modes are:
+ Produce a trace of all basic blocks.
+ Count frequency of jump instructions executed.
+ In every mode it is possible to start profiling upon entering
+ certain functions and to disable profiling of some other functions.
+
+ The result of basic-block profiling will be written to a file `bb.out'.
+ If the -ax option is used parameters for the profiling will be read
+ from file `bb.in'.
+
+*/
+
+/* The following macro shall output assembler code to FILE
+ to initialize basic-block profiling.
+
+ If profile_block_flag == 2
+
+ Output code to call the subroutine `__bb_init_trace_func'
+ and pass two parameters to it. The first parameter is
+ the address of a block allocated in the object module.
+ The second parameter is the number of the first basic block
+ of the function.
+
+ The name of the block is a local symbol made with this statement:
+
+ ASM_GENERATE_INTERNAL_LABEL (BUFFER, "LPBX", 0);
+
+ Of course, since you are writing the definition of
+ `ASM_GENERATE_INTERNAL_LABEL' as well as that of this macro, you
+ can take a short cut in the definition of this macro and use the
+ name that you know will result.
+
+ The number of the first basic block of the function is
+ passed to the macro in BLOCK_OR_LABEL.
+
+ If described in a virtual assembler language the code to be
+ output looks like:
+
+ parameter1 <- LPBX0
+ parameter2 <- BLOCK_OR_LABEL
+ call __bb_init_trace_func
+
+ else if profile_block_flag != 0
+
+ Output code to call the subroutine `__bb_init_func'
+ and pass one single parameter to it, which is the same
+ as the first parameter to `__bb_init_trace_func'.
+
+ The first word of this parameter is a flag which will be nonzero if
+ the object module has already been initialized. So test this word
+ first, and do not call `__bb_init_func' if the flag is nonzero.
+ Note: When profile_block_flag == 2 the test need not be done
+ but `__bb_init_trace_func' *must* be called.
+
+ BLOCK_OR_LABEL may be used to generate a label number as a
+ branch destination in case `__bb_init_func' will not be called.
+
+ If described in a virtual assembler language the code to be
+ output looks like:
+
+ cmp (LPBX0),0
+ jne local_label
+ parameter1 <- LPBX0
+ call __bb_init_func
+local_label:
+
+*/
+
+#define FUNCTION_BLOCK_PROFILER(FILE, BLOCK_OR_LABEL) \
+do \
+ { \
+ int bol = (BLOCK_OR_LABEL); \
+ switch (profile_block_flag) \
+ { \
+ case 2: \
+ if (TARGET_MEDANY) \
+ fprintf (FILE, "\tsethi %%hi(LPBX0),%%o0\n\tor %%0,%%lo(LPBX0),%%o0\n\tadd %%o0,%s,%%o0\n\tsethi %%hi(%d),%%o1\n\tcall ___bb_init_trace_func\n\tadd %g0,%%lo(%d),%%o1\n",\
+ MEDANY_BASE_REG, bol, bol); \
+ else \
+ fprintf (FILE, "\tsethi %%hi(LPBX0),%%o0\n\tor %%o0,%%lo(LPBX0),%%o0\n\tsethi %%hi(%d),%%o1\n\tcall ___bb_init_trace_func\n\tor %%o1,%%lo(%d),%%o1\n",\
+ bol, bol); \
+ break; \
+ default: \
+ if (TARGET_MEDANY) \
+ fprintf (FILE, "\tsethi %%hi(LPBX0),%%o0\n\tor %%0,%%lo(LPBX0),%%o0\n\tld [%s+%%o0],%%o1\n\ttst %%o1\n\tbne LPY%d\n\tadd %%o0,%s,%%o0\n\tcall ___bb_init_func\n\tnop\nLPY%d:\n",\
+ MEDANY_BASE_REG, bol, MEDANY_BASE_REG, bol);\
+ else \
+ fprintf (FILE, "\tsethi %%hi(LPBX0),%%o0\n\tld [%%lo(LPBX0)+%%o0],%%o1\n\ttst %%o1\n\tbne LPY%d\n\tadd %%o0,%%lo(LPBX0),%%o0\n\tcall ___bb_init_func\n\tnop\nLPY%d:\n",\
+ bol, bol); \
+ break; \
+ } \
+ } \
+while (0)
+/* The following macro shall output assembler code to FILE
+ to increment a counter associated with basic block number BLOCKNO.
+
+ If profile_block_flag == 2
+
+ Output code to initialize the global structure `__bb' and
+ call the function `__bb_trace_func' which will increment the
+ counter.
+
+ `__bb' consists of two words. In the first word the number
+ of the basic block has to be stored. In the second word
+ the address of a block allocated in the object module
+ has to be stored.
+
+ The basic block number is given by BLOCKNO.
+
+ The address of the block is given by the label created with
+
+ ASM_GENERATE_INTERNAL_LABEL (BUFFER, "LPBX", 0);
+
+ by FUNCTION_BLOCK_PROFILER.
+
+ Of course, since you are writing the definition of
+ `ASM_GENERATE_INTERNAL_LABEL' as well as that of this macro, you
+ can take a short cut in the definition of this macro and use the
+ name that you know will result.
+
+ If described in a virtual assembler language the code to be
+ output looks like:
+
+ move BLOCKNO -> (__bb)
+ move LPBX0 -> (__bb+4)
+ call __bb_trace_func
+
+ Note that function `__bb_trace_func' must not change the
+ machine state, especially the flag register. To grant
+ this, you must output code to save and restore registers
+ either in this macro or in the macros MACHINE_STATE_SAVE
+ and MACHINE_STATE_RESTORE. The last two macros will be
+ used in the function `__bb_trace_func', so you must make
+ sure that the function prologue does not change any
+ register prior to saving it with MACHINE_STATE_SAVE.
+
+ else if profile_block_flag != 0
+
+ Output code to increment the counter directly.
+ Basic blocks are numbered separately from zero within each
+ compiled object module. The count associated with block number
+ BLOCKNO is at index BLOCKNO in an array of words; the name of
+ this array is a local symbol made with this statement:
+
+ ASM_GENERATE_INTERNAL_LABEL (BUFFER, "LPBX", 2);
+
+ Of course, since you are writing the definition of
+ `ASM_GENERATE_INTERNAL_LABEL' as well as that of this macro, you
+ can take a short cut in the definition of this macro and use the
+ name that you know will result.
+
+ If described in a virtual assembler language, the code to be
+ output looks like:
+
+ inc (LPBX2+4*BLOCKNO)
+
+*/
+
+#define BLOCK_PROFILER(FILE, BLOCKNO) \
+do \
+ { \
+ int blockn = (BLOCKNO); \
+ switch (profile_block_flag) \
+ { \
+ case 2: \
+ if (TARGET_MEDANY) \
+ fprintf (FILE, "\tsethi %%hi(___bb),%%g1\n\tor %%0,%%lo(___bb),%%g1\n\tsethi %%hi(%d),%%g2\n\tor %%g2,%%lo(%d),%%g2\n\tst %%g2,[%s+%%g1]\n\tsethi %%hi(LPBX0),%%g2\n\tor %%0,%%lo(LPBX0),%%g2\n\tadd %%g2,%s,%%g2\n\tadd 4,%%g1,%%g1\n\tst %%g2,[%%g1+%%lo(___bb)]\n\tmov %%o7,%%g2\n\tcall ___bb_trace_func\n\tnop\n\tmov %%g2,%%o7\n",\
+ blockn, blockn, MEDANY_BASE_REG, MEDANY_BASE_REG); \
+ else \
+ fprintf (FILE, "\tsethi %%hi(___bb),%%g1\n\tsethi %%hi(%d),%%g2\n\tor %%g2,%%lo(%d),%%g2\n\tst %%g2,[%%lo(___bb)+%%g1]\n\tsethi %%hi(LPBX0),%%g2\n\tor %%g2,%%lo(LPBX0),%%g2\n\tadd 4,%%g1,%%g1\n\tst %%g2,[%%lo(___bb)+%%g1]\n\tmov %%o7,%%g2\n\tcall ___bb_trace_func\n\tnop\n\tmov %%g2,%%o7\n",\
+ blockn, blockn); \
+ break; \
+ default: \
+ if (TARGET_MEDANY) \
+ fprintf (FILE, "\tsethi %%hi(LPBX2+%d),%%g1\n\tor %%g1,%%lo(LPBX2+%d),%%g1\n\tld [%%g1+%s],%%g2\n\tadd %%g2,1,%%g2\n\tst %%g2,[%%g1+%s]\n", \
+ 4 * blockn, 4 * blockn, MEDANY_BASE_REG, MEDANY_BASE_REG); \
+ else \
+ fprintf (FILE, "\tsethi %%hi(LPBX2+%d),%%g1\n\tld [%%lo(LPBX2+%d)+%%g1],%%g2\n\
+\tadd %%g2,1,%%g2\n\tst %%g2,[%%lo(LPBX2+%d)+%%g1]\n", \
+ 4 * blockn, 4 * blockn, 4 * blockn); \
+ break; \
+ } \
+ } \
+while(0)
+
+/* The following macro shall output assembler code to FILE
+ to indicate a return from function during basic-block profiling.
+
+ If profiling_block_flag == 2:
+
+ Output assembler code to call function `__bb_trace_ret'.
+
+ Note that function `__bb_trace_ret' must not change the
+ machine state, especially the flag register. To grant
+ this, you must output code to save and restore registers
+ either in this macro or in the macros MACHINE_STATE_SAVE_RET
+ and MACHINE_STATE_RESTORE_RET. The last two macros will be
+ used in the function `__bb_trace_ret', so you must make
+ sure that the function prologue does not change any
+ register prior to saving it with MACHINE_STATE_SAVE_RET.
+
+ else if profiling_block_flag != 0:
+
+ The macro will not be used, so it need not distinguish
+ these cases.
+*/
+
+#define FUNCTION_BLOCK_PROFILER_EXIT(FILE) \
+ fprintf (FILE, "\tcall ___bb_trace_ret\n\tnop\n" );
+
+/* The function `__bb_trace_func' is called in every basic block
+ and is not allowed to change the machine state. Saving (restoring)
+ the state can either be done in the BLOCK_PROFILER macro,
+ before calling function (rsp. after returning from function)
+ `__bb_trace_func', or it can be done inside the function by
+ defining the macros:
+
+ MACHINE_STATE_SAVE(ID)
+ MACHINE_STATE_RESTORE(ID)
+
+ In the latter case care must be taken, that the prologue code
+ of function `__bb_trace_func' does not already change the
+ state prior to saving it with MACHINE_STATE_SAVE.
+
+ The parameter `ID' is a string identifying a unique macro use.
+
+ On sparc it is sufficient to save the psw register to memory.
+ Unfortunately the psw register can be read in supervisor mode only,
+ so we read only the condition codes by using branch instructions
+ and hope that this is enough. */
+
+#define MACHINE_STATE_SAVE(ID) \
+ asm (" mov %g0,%l0");\
+ asm (" be,a LFLGNZ" ID);\
+ asm (" or %l0,4,%l0");\
+ asm ("LFLGNZ" ID ": bcs,a LFLGNC" ID);\
+ asm (" or %l0,1,%l0");\
+ asm ("LFLGNC" ID ": bvs,a LFLGNV" ID);\
+ asm (" or %l0,2,%l0");\
+ asm ("LFLGNV" ID ": bneg,a LFLGNN" ID);\
+ asm (" or %l0,8,%l0");\
+ asm ("LFLGNN" ID ": sethi %hi(LFLAGS" ID "),%l1");\
+ asm (" st %l0,[%l1+%lo(LFLAGS" ID ")]"); \
+ asm (" st %g2,[%l1+%lo(LSAVRET" ID ")]");
+
+/* On sparc MACHINE_STATE_RESTORE restores the psw register from memory.
+ The psw register can be written in supervisor mode only,
+ which is true even for simple condition codes.
+ We use some combination of instructions to produce the
+ proper condition codes, but some flag combinations can not
+ be generated in this way. If this happens an unimplemented
+ instruction will be executed to abort the program. */
+
+#define MACHINE_STATE_RESTORE(ID) \
+ asm (" sethi %hi(LFLGTAB" ID "),%l1");\
+ asm (" ld [%l1+%lo(LFLGTAB" ID "-(LFLGTAB" ID "-LFLAGS" ID "))],%l0");\
+ asm (" ld [%l1+%lo(LFLGTAB" ID "-(LFLGTAB" ID "-LSAVRET" ID "))],%g2");\
+ asm (" sll %l0,2,%l0");\
+ asm (" add %l0,%l1,%l0");\
+ asm (" ld [%l0+%lo(LFLGTAB" ID ")],%l1");\
+ asm (" jmp %l1");\
+ asm (" nop");\
+ asm (".data");\
+ asm (" .align 4");\
+ asm ("LFLAGS" ID ":");\
+ asm (" .word 0");\
+ asm ("LSAVRET" ID ":");\
+ asm (" .word 0");\
+ asm ("LFLGTAB" ID ": ");\
+ asm (" .word LSFLG0" ID);\
+ asm (" .word LSFLGC" ID);\
+ asm (" .word LSFLGV" ID);\
+ asm (" .word LSFLGVC" ID);\
+ asm (" .word LSFLGZ" ID);\
+ asm (" .word LSFLGZC" ID);\
+ asm (" .word LSFLGZV" ID);\
+ asm (" .word LSFLGZVC" ID);\
+ asm (" .word LSFLGN" ID);\
+ asm (" .word LSFLGNC" ID);\
+ asm (" .word LSFLGNV" ID);\
+ asm (" .word LSFLGNVC" ID);\
+ asm (" .word LSFLGNZ" ID);\
+ asm (" .word LSFLGNZC" ID);\
+ asm (" .word LSFLGNZV" ID);\
+ asm (" .word LSFLGNZVC" ID);\
+ asm (".text");\
+ asm ("LSFLGVC" ID ": mov -1,%l0");\
+ asm (" addcc 2,%l0,%g0");\
+ asm (" sethi %hi(0x80000000),%l0");\
+ asm (" mov %l0,%l1");\
+ asm (" ba LFLGRET" ID);\
+ asm (" addxcc %l0,%l1,%l0");\
+ asm ("LSFLGC" ID ": mov -1,%l0");\
+ asm (" ba LFLGRET" ID);\
+ asm (" addcc 2,%l0,%g0");\
+ asm ("LSFLGZC" ID ": mov -1,%l0");\
+ asm (" ba LFLGRET" ID);\
+ asm (" addcc 1,%l0,%l0");\
+ asm ("LSFLGZVC" ID ": sethi %hi(0x80000000),%l0");\
+ asm (" mov %l0,%l1");\
+ asm (" ba LFLGRET" ID);\
+ asm (" addcc %l0,%l1,%l0");\
+ asm ("LSFLGZ" ID ": ba LFLGRET" ID);\
+ asm (" subcc %g0,%g0,%g0");\
+ asm ("LSFLGNC" ID ": add %g0,1,%l0");\
+ asm (" ba LFLGRET" ID);\
+ asm (" subcc %g0,%l0,%g0");\
+ asm ("LSFLG0" ID ": ba LFLGRET" ID);\
+ asm (" orcc 1,%g0,%g0");\
+ asm ("LSFLGN" ID ": ba LFLGRET" ID);\
+ asm (" orcc -1,%g0,%g0");\
+ asm ("LSFLGV" ID ":");\
+ asm ("LSFLGZV" ID ":");\
+ asm ("LSFLGNV" ID ":");\
+ asm ("LSFLGNVC" ID ":");\
+ asm ("LSFLGNZ" ID ":");\
+ asm ("LSFLGNZC" ID ":");\
+ asm ("LSFLGNZV" ID ":");\
+ asm ("LSFLGNZVC" ID ":");\
+ asm (" unimp");\
+ asm ("LFLGRET" ID ":");
+\f
/* EXIT_IGNORE_STACK should be nonzero if, when returning from a function,
the stack pointer does not matter. The value is tested only in
functions that have frame pointers.
because they occur inside of macros. Sigh. */
extern union tree_node *current_function_decl;
-#define FUNCTION_EPILOGUE(FILE, SIZE) \
- (TARGET_FRW ? sparc_frw_output_function_epilogue (FILE, SIZE, leaf_function)\
+#define FUNCTION_EPILOGUE(FILE, SIZE) \
+ (TARGET_FLAT ? sparc_flat_output_function_epilogue (FILE, SIZE) \
: output_function_epilogue (FILE, SIZE, leaf_function))
-#define DELAY_SLOTS_FOR_EPILOGUE \
- (TARGET_FRW ? sparc_frw_epilogue_delay_slots () : 1)
-#define ELIGIBLE_FOR_EPILOGUE_DELAY(trial, slots_filled) \
- (TARGET_FRW ? sparc_frw_eligible_for_epilogue_delay (trial, slots_filled) \
+#define DELAY_SLOTS_FOR_EPILOGUE \
+ (TARGET_FLAT ? sparc_flat_epilogue_delay_slots () : 1)
+#define ELIGIBLE_FOR_EPILOGUE_DELAY(trial, slots_filled) \
+ (TARGET_FLAT ? sparc_flat_eligible_for_epilogue_delay (trial, slots_filled) \
: eligible_for_epilogue_delay (trial, slots_filled))
+
+/* Define registers used by the epilogue and return instruction. */
+#define EPILOGUE_USES(REGNO) \
+ (!TARGET_FLAT && REGNO == 31)
\f
/* Output assembler code for a block containing the constant parts
of a trampoline, leaving space for the variable parts. */
-/* On the sparc, the trampoline contains five instructions:
- sethi #TOP_OF_FUNCTION,%g2
- or #BOTTOM_OF_FUNCTION,%g2,%g2
- sethi #TOP_OF_STATIC,%g1
- jmp g2
- or #BOTTOM_OF_STATIC,%g1,%g1 */
-#define TRAMPOLINE_TEMPLATE(FILE) \
-{ \
- ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
- ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
- ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
- ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x81C08000)); \
- ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
-}
+/* On 32 bit sparcs, the trampoline contains five instructions:
+ sethi #TOP_OF_FUNCTION,%g1
+ or #BOTTOM_OF_FUNCTION,%g1,%g1
+ sethi #TOP_OF_STATIC,%g2
+ jmp g1
+ or #BOTTOM_OF_STATIC,%g2,%g2
+
+ On 64 bit sparcs, the trampoline contains 4 insns and two pseudo-immediate
+ constants (plus some padding):
+ rd %pc,%g1
+ ldx[%g1+20],%g5
+ ldx[%g1+28],%g1
+ jmp %g1
+ nop
+ nop
+ .xword context
+ .xword function */
+
+#define TRAMPOLINE_TEMPLATE(FILE) \
+do { \
+ if (TARGET_ARCH64) \
+ { \
+ fprintf (FILE, "\trd %%pc,%%g1\n"); \
+ fprintf (FILE, "\tldx [%%g1+24],%%g5\n"); \
+ fprintf (FILE, "\tldx [%%g1+32],%%g1\n"); \
+ fprintf (FILE, "\tjmp %%g1\n"); \
+ fprintf (FILE, "\tnop\n"); \
+ fprintf (FILE, "\tnop\n"); \
+ /* -mmedlow shouldn't generate .xwords, so don't use them at all */ \
+ fprintf (FILE, "\t.word 0,0,0,0\n"); \
+ } \
+ else \
+ { \
+ ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
+ ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
+ ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
+ ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x81C04000)); \
+ ASM_OUTPUT_INT (FILE, gen_rtx (CONST_INT, VOIDmode, 0x00000000)); \
+ } \
+} while (0)
/* Length in units of the trampoline for entering a nested function. */
-#define TRAMPOLINE_SIZE 20
+#define TRAMPOLINE_SIZE (TARGET_ARCH64 ? 40 : 20)
/* Emit RTL insns to initialize the variable parts of a trampoline.
FNADDR is an RTX for the address of the function's pure code.
- CXT is an RTX for the static chain value for the function.
-
- This takes 16 insns: 2 shifts & 2 ands (to split up addresses), 4 sethi
- (to load in opcodes), 4 iors (to merge address and opcodes), and 4 writes
- (to store insns). This is a bit excessive. Perhaps a different
- mechanism would be better here.
+ CXT is an RTX for the static chain value for the function. */
- Emit 3 FLUSH instructions (UNSPEC_VOLATILE 2) to synchonize the data
- and instruction caches. */
-
-#define INITIALIZE_TRAMPOLINE(TRAMP, FNADDR, CXT) \
-{ \
- rtx high_cxt = expand_shift (RSHIFT_EXPR, SImode, CXT, \
- size_int (10), 0, 1); \
- rtx high_fn = expand_shift (RSHIFT_EXPR, SImode, FNADDR, \
- size_int (10), 0, 1); \
- rtx low_cxt = expand_and (CXT, gen_rtx (CONST_INT, VOIDmode, 0x3ff), 0); \
- rtx low_fn = expand_and (FNADDR, gen_rtx (CONST_INT, VOIDmode, 0x3ff), 0); \
- rtx g1_sethi = gen_rtx (HIGH, SImode, \
- gen_rtx (CONST_INT, VOIDmode, 0x03000000)); \
- rtx g2_sethi = gen_rtx (HIGH, SImode, \
- gen_rtx (CONST_INT, VOIDmode, 0x05000000)); \
- rtx g1_ori = gen_rtx (HIGH, SImode, \
- gen_rtx (CONST_INT, VOIDmode, 0x82106000)); \
- rtx g2_ori = gen_rtx (HIGH, SImode, \
- gen_rtx (CONST_INT, VOIDmode, 0x8410A000)); \
- rtx tem = gen_reg_rtx (SImode); \
- emit_move_insn (tem, g2_sethi); \
- emit_insn (gen_iorsi3 (high_fn, high_fn, tem)); \
- emit_move_insn (gen_rtx (MEM, SImode, plus_constant (TRAMP, 0)), high_fn);\
- emit_move_insn (tem, g2_ori); \
- emit_insn (gen_iorsi3 (low_fn, low_fn, tem)); \
- emit_move_insn (gen_rtx (MEM, SImode, plus_constant (TRAMP, 4)), low_fn);\
- emit_move_insn (tem, g1_sethi); \
- emit_insn (gen_iorsi3 (high_cxt, high_cxt, tem)); \
- emit_move_insn (gen_rtx (MEM, SImode, plus_constant (TRAMP, 8)), high_cxt);\
- emit_move_insn (tem, g1_ori); \
- emit_insn (gen_iorsi3 (low_cxt, low_cxt, tem)); \
- emit_move_insn (gen_rtx (MEM, SImode, plus_constant (TRAMP, 16)), low_cxt);\
- emit_insn (gen_rtx (UNSPEC_VOLATILE, VOIDmode, \
- gen_rtvec (1, plus_constant (TRAMP, 0)), \
- 2)); \
- emit_insn (gen_rtx (UNSPEC_VOLATILE, VOIDmode, \
- gen_rtvec (1, plus_constant (TRAMP, 8)), \
- 2)); \
- emit_insn (gen_rtx (UNSPEC_VOLATILE, VOIDmode, \
- gen_rtvec (1, plus_constant (TRAMP, 16)), \
- 2)); \
-}
+void sparc_initialize_trampoline ();
+void sparc64_initialize_trampoline ();
+#define INITIALIZE_TRAMPOLINE(TRAMP, FNADDR, CXT) \
+ do { \
+ if (TARGET_ARCH64) \
+ sparc64_initialize_trampoline (TRAMP, FNADDR, CXT); \
+ else \
+ sparc_initialize_trampoline (TRAMP, FNADDR, CXT); \
+ } while (0)
\f
/* Generate necessary RTL for __builtin_saveregs().
ARGLIST is the argument list; see expr.c. */
/* Given an rtx for the address of a frame,
return an rtx for the address of the word in the frame
- that holds the dynamic chain--the previous frame's address. */
+ that holds the dynamic chain--the previous frame's address.
+ ??? -mflat support? */
#define DYNAMIC_CHAIN_ADDRESS(frame) \
- gen_rtx (PLUS, Pmode, frame, gen_rtx (CONST_INT, VOIDmode, 56))
+ gen_rtx (PLUS, Pmode, frame, gen_rtx (CONST_INT, VOIDmode, 14 * UNITS_PER_WORD))
/* The return address isn't on the stack, it is in a register, so we can't
access it from the current frame pointer. We can access it from the
save area. */
#define RETURN_ADDR_IN_PREVIOUS_FRAME
+/* This is the offset of the return address to the true next instruction to be
+ executed for the current function. */
+#define RETURN_ADDR_OFFSET \
+ (8 + 4 * (! TARGET_ARCH64 && current_function_returns_struct))
+
/* The current return address is in %i7. The return address of anything
farther back is in the register window save area at [%fp+60]. */
/* ??? This ignores the fact that the actual return address is +8 for normal
#define RETURN_ADDR_RTX(count, frame) \
((count == -1) \
? gen_rtx (REG, Pmode, 31) \
- : copy_to_reg (gen_rtx (MEM, Pmode, \
- memory_address (Pmode, plus_constant (frame, 60)))))
+ : gen_rtx (MEM, Pmode, \
+ memory_address (Pmode, plus_constant (frame, 15 * UNITS_PER_WORD))))
+
+#define DOESNT_NEED_UNWINDER (! TARGET_FLAT)
\f
/* Addressing modes, and classification of registers for them. */
has been allocated, which happens in local-alloc.c. */
#define REGNO_OK_FOR_INDEX_P(REGNO) \
-(((REGNO) < 32 || (unsigned) reg_renumber[REGNO] < 32) && (REGNO) != 0)
+((REGNO) < 32 || (unsigned) reg_renumber[REGNO] < 32)
#define REGNO_OK_FOR_BASE_P(REGNO) \
-(((REGNO) < 32 || (unsigned) reg_renumber[REGNO] < 32) && (REGNO) != 0)
+((REGNO) < 32 || (unsigned) reg_renumber[REGNO] < 32)
#define REGNO_OK_FOR_FP_P(REGNO) \
-(((REGNO) ^ 0x20) < 32 \
- || (((REGNO) != 0) && (unsigned) (reg_renumber[REGNO] ^ 0x20) < 32))
+ (((unsigned) (REGNO) - 32 < (TARGET_V9 ? 64 : 32)) \
+ || ((unsigned) reg_renumber[REGNO] - 32 < (TARGET_V9 ? 64 : 32)))
+#define REGNO_OK_FOR_CCFP_P(REGNO) \
+ (TARGET_V9 \
+ && (((unsigned) (REGNO) - 96 < 4) \
+ || ((unsigned) reg_renumber[REGNO] - 96 < 4)))
/* Now macros that check whether X is a register and also,
strictly, whether it is in a specified class.
these things in insns and then not re-recognize the insns, causing
constrain_operands to fail.
- `S' handles constraints for calls. */
+ `S' handles constraints for calls. ??? So where is it? */
#ifndef REG_OK_STRICT
/* Nonzero if X is a hard reg that can be used as an index
or if it is a pseudo reg. */
-#define REG_OK_FOR_INDEX_P(X) (((unsigned) REGNO (X)) - 32 >= 32 && REGNO (X) != 0)
+#define REG_OK_FOR_INDEX_P(X) \
+ (((unsigned) REGNO (X)) - 32 >= (FIRST_PSEUDO_REGISTER - 32))
/* Nonzero if X is a hard reg that can be used as a base reg
or if it is a pseudo reg. */
-#define REG_OK_FOR_BASE_P(X) (((unsigned) REGNO (X)) - 32 >= 32 && REGNO (X) != 0)
+#define REG_OK_FOR_BASE_P(X) \
+ (((unsigned) REGNO (X)) - 32 >= (FIRST_PSEUDO_REGISTER - 32))
+
+/* 'T', 'U' are for aligned memory loads which aren't needed for v9. */
#define EXTRA_CONSTRAINT(OP, C) \
((C) == 'Q' \
? ((GET_CODE (OP) == MEM \
- && memory_address_p (GET_MODE (OP), XEXP (OP, 0)) \
+ && memory_address_p (GET_MODE (OP), XEXP (OP, 0)) \
&& ! symbolic_memory_operand (OP, VOIDmode)) \
|| (reload_in_progress && GET_CODE (OP) == REG \
&& REGNO (OP) >= FIRST_PSEUDO_REGISTER)) \
- : (C) == 'T' \
+ : (! TARGET_ARCH64 && (C) == 'T') \
? (mem_aligned_8 (OP)) \
- : (C) == 'U' \
+ : (! TARGET_ARCH64 && (C) == 'U') \
? (register_ok_for_ldd (OP)) \
: 0)
? (REGNO (OP) >= FIRST_PSEUDO_REGISTER \
&& reg_renumber[REGNO (OP)] < 0) \
: GET_CODE (OP) == MEM) \
- : (C) == 'T' \
+ : (! TARGET_ARCH64 && (C) == 'T') \
? mem_aligned_8 (OP) && strict_memory_address_p (Pmode, XEXP (OP, 0)) \
- : (C) == 'U' \
+ : (! TARGET_ARCH64 && (C) == 'U') \
? (GET_CODE (OP) == REG \
&& (REGNO (OP) < FIRST_PSEUDO_REGISTER \
- || reg_renumber[REGNO (OP)] > 0) \
- && register_ok_for_ldd (OP)) : 0)
+ || reg_renumber[REGNO (OP)] >= 0) \
+ && register_ok_for_ldd (OP)) \
+ : 0)
#endif
\f
/* GO_IF_LEGITIMATE_ADDRESS recognizes an RTL expression
register rtx op0 = XEXP (X, 0); \
register rtx op1 = XEXP (X, 1); \
if (RTX_OK_FOR_BASE_P (op0) \
- && CONSTANT_P (op1)) \
+ && CONSTANT_P (op1) \
+ /* We can't allow TFmode, because an offset \
+ greater than or equal to the alignment (8) \
+ may cause the LO_SUM to overflow. */ \
+ && MODE != TFmode) \
goto ADDR; \
} \
else if (GET_CODE (X) == CONST_INT && SMALL_INT (X)) \
On the SPARC this is never true. */
#define GO_IF_MODE_DEPENDENT_ADDRESS(ADDR,LABEL)
+
+/* If we are referencing a function make the SYMBOL_REF special.
+ In the Medium/Anywhere code model, %g4 points to the data segment so we
+ must not add it to function addresses. */
+
+#define ENCODE_SECTION_INFO(DECL) \
+ do { \
+ if (TARGET_MEDANY && TREE_CODE (DECL) == FUNCTION_DECL) \
+ SYMBOL_REF_FLAG (XEXP (DECL_RTL (DECL), 0)) = 1; \
+ } while (0)
\f
/* Specify the machine mode that this machine uses
for the index in the tablejump instruction. */
-#define CASE_VECTOR_MODE SImode
+#define CASE_VECTOR_MODE Pmode
/* Define this if the tablejump instruction expects the table
to contain offsets from the address of the table.
/* Specify the machine mode that pointers have.
After generation of rtl, the compiler makes no further distinction
between pointers and any other objects of this machine mode. */
-#define Pmode SImode
+#define Pmode (TARGET_PTR64 ? DImode : SImode)
/* Generate calls to memcpy, memcmp and memset. */
#define TARGET_MEM_FUNCTIONS
We also have two modes to indicate that the relevant condition code is
in the floating-point condition code register. One for comparisons which
will generate an exception if the result is unordered (CCFPEmode) and
- one for comparisons which will never trap (CCFPmode). This really should
- be a separate register, but we don't want to go to 65 registers. */
-#define EXTRA_CC_MODES CC_NOOVmode, CCFPmode, CCFPEmode
+ one for comparisons which will never trap (CCFPmode).
+
+ CCXmode and CCX_NOOVmode are only used by v9. */
+
+#define EXTRA_CC_MODES CCXmode, CC_NOOVmode, CCX_NOOVmode, CCFPmode, CCFPEmode
/* Define the names for the modes specified above. */
-#define EXTRA_CC_NAMES "CC_NOOV", "CCFP", "CCFPE"
+
+#define EXTRA_CC_NAMES "CCX", "CC_NOOV", "CCX_NOOV", "CCFP", "CCFPE"
/* Given a comparison code (EQ, NE, etc.) and the first operand of a COMPARE,
return the mode to be used for the comparison. For floating-point,
? ((OP == EQ || OP == NE) ? CCFPmode : CCFPEmode) \
: ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS \
|| GET_CODE (X) == NEG || GET_CODE (X) == ASHIFT) \
- ? CC_NOOVmode : CCmode))
+ ? (TARGET_ARCH64 && GET_MODE (X) == DImode ? CCX_NOOVmode : CC_NOOVmode) \
+ : (TARGET_ARCH64 && GET_MODE (X) == DImode ? CCXmode : CCmode)))
+
+/* Return non-zero if SELECT_CC_MODE will never return MODE for a
+ floating point inequality comparison. */
+
+#define REVERSIBLE_CC_MODE(MODE) ((MODE) != CCFPEmode)
/* A function address in a call instruction
is a byte address (for indexing purposes)
/* .umul is a little faster than .mul. */
#define MULSI3_LIBCALL "*.umul"
+/* Define library calls for quad FP operations. These are all part of the
+ SPARC ABI. */
+#define ADDTF3_LIBCALL "_Q_add"
+#define SUBTF3_LIBCALL "_Q_sub"
+#define NEGTF2_LIBCALL "_Q_neg"
+#define MULTF3_LIBCALL "_Q_mul"
+#define DIVTF3_LIBCALL "_Q_div"
+#define FLOATSITF2_LIBCALL "_Q_itoq"
+#define FIX_TRUNCTFSI2_LIBCALL "_Q_qtoi"
+#define FIXUNS_TRUNCTFSI2_LIBCALL "_Q_qtou"
+#define EXTENDSFTF2_LIBCALL "_Q_stoq"
+#define TRUNCTFSF2_LIBCALL "_Q_qtos"
+#define EXTENDDFTF2_LIBCALL "_Q_dtoq"
+#define TRUNCTFDF2_LIBCALL "_Q_qtod"
+#define EQTF2_LIBCALL "_Q_feq"
+#define NETF2_LIBCALL "_Q_fne"
+#define GTTF2_LIBCALL "_Q_fgt"
+#define GETF2_LIBCALL "_Q_fge"
+#define LTTF2_LIBCALL "_Q_flt"
+#define LETF2_LIBCALL "_Q_fle"
+
+/* We can define the TFmode sqrt optab only if TARGET_FPU. This is because
+ with soft-float, the SFmode and DFmode sqrt instructions will be absent,
+ and the compiler will notice and try to use the TFmode sqrt instruction
+ for calls to the builtin function sqrt, but this fails. */
+#define INIT_TARGET_OPTABS \
+ do { \
+ add_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, ADDTF3_LIBCALL); \
+ sub_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, SUBTF3_LIBCALL); \
+ neg_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, NEGTF2_LIBCALL); \
+ smul_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, MULTF3_LIBCALL); \
+ flodiv_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, DIVTF3_LIBCALL); \
+ eqtf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, EQTF2_LIBCALL); \
+ netf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, NETF2_LIBCALL); \
+ gttf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, GTTF2_LIBCALL); \
+ getf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, GETF2_LIBCALL); \
+ lttf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, LTTF2_LIBCALL); \
+ letf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, LETF2_LIBCALL); \
+ trunctfsf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, TRUNCTFSF2_LIBCALL); \
+ trunctfdf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, TRUNCTFDF2_LIBCALL); \
+ extendsftf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, EXTENDSFTF2_LIBCALL); \
+ extenddftf2_libfunc = gen_rtx (SYMBOL_REF, Pmode, EXTENDDFTF2_LIBCALL); \
+ floatsitf_libfunc = gen_rtx (SYMBOL_REF, Pmode, FLOATSITF2_LIBCALL); \
+ fixtfsi_libfunc = gen_rtx (SYMBOL_REF, Pmode, FIX_TRUNCTFSI2_LIBCALL); \
+ fixunstfsi_libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, FIXUNS_TRUNCTFSI2_LIBCALL); \
+ if (TARGET_FPU) \
+ sqrt_optab->handlers[(int) TFmode].libfunc \
+ = gen_rtx (SYMBOL_REF, Pmode, "_Q_sqrt"); \
+ INIT_SUBTARGET_OPTABS; \
+ } while (0)
+
+/* This is meant to be redefined in the host dependent files */
+#define INIT_SUBTARGET_OPTABS
+
/* Compute the cost of computing a constant rtl expression RTX
whose rtx-code is CODE. The body of this macro is a portion
of a switch statement. If the code is computed here,
return 0; \
return 8;
-/* SPARC offers addressing modes which are "as cheap as a register".
- See sparc.c (or gcc.texinfo) for details. */
+/* Compute the cost of an address. For the sparc, all valid addresses are
+ the same cost.
+ ??? Is this true for v9? */
-#define ADDRESS_COST(RTX) \
- (GET_CODE (RTX) == REG ? 1 : sparc_address_cost (RTX))
+#define ADDRESS_COST(RTX) 1
/* Compute extra cost of moving data between one register class
- and another. */
+ and another.
+ ??? v9: We ignore FPCC_REGS on the assumption they'll never be seen. */
#define REGISTER_MOVE_COST(CLASS1, CLASS2) \
- (((CLASS1 == FP_REGS && CLASS2 == GENERAL_REGS) \
- || (CLASS1 == GENERAL_REGS && CLASS2 == FP_REGS)) ? 6 : 2)
+ (((FP_REG_CLASS_P (CLASS1) && (CLASS2) == GENERAL_REGS) \
+ || ((CLASS1) == GENERAL_REGS && FP_REG_CLASS_P (CLASS2))) ? 6 : 2)
/* Provide the costs of a rtl expression. This is in the body of a
switch on CODE. The purpose for the cost of MULT is to encourage
#define RTX_COSTS(X,CODE,OUTER_CODE) \
case MULT: \
- return TARGET_V8 ? COSTS_N_INSNS (5) : COSTS_N_INSNS (25); \
+ return (TARGET_V8 || TARGET_SPARCLITE || TARGET_V9) \
+ ? COSTS_N_INSNS (5) : COSTS_N_INSNS (25); \
case DIV: \
case UDIV: \
case MOD: \
case FIX: \
return 19;
+/* Adjust the cost of dependencies. */
+#define ADJUST_COST(INSN,LINK,DEP,COST) \
+ if (sparc_cpu == PROCESSOR_SUPERSPARC) \
+ (COST) = supersparc_adjust_cost (INSN, LINK, DEP, COST)
+
/* Conditional branches with empty delay slots have a length of two. */
#define ADJUST_INSN_LENGTH(INSN, LENGTH) \
if (GET_CODE (INSN) == CALL_INSN \
#define ASM_APP_OFF ""
+/* ??? Try to make the style consistent here (_OP?). */
+
+#define ASM_LONGLONG ".xword"
#define ASM_LONG ".word"
#define ASM_SHORT ".half"
#define ASM_BYTE_OP ".byte"
+#define ASM_FLOAT ".single"
+#define ASM_DOUBLE ".double"
+#define ASM_LONGDOUBLE ".xxx" /* ??? Not known (or used yet). */
/* Output before read-only data. */
"%f0", "%f1", "%f2", "%f3", "%f4", "%f5", "%f6", "%f7", \
"%f8", "%f9", "%f10", "%f11", "%f12", "%f13", "%f14", "%f15", \
"%f16", "%f17", "%f18", "%f19", "%f20", "%f21", "%f22", "%f23", \
- "%f24", "%f25", "%f26", "%f27", "%f28", "%f29", "%f30", "%f31"}
-
-/* Define additional names for use in asm clobbers and asm declarations.
+ "%f24", "%f25", "%f26", "%f27", "%f28", "%f29", "%f30", "%f31", \
+ "%f32", "%f33", "%f34", "%f35", "%f36", "%f37", "%f38", "%f39", \
+ "%f40", "%f41", "%f42", "%f43", "%f44", "%f45", "%f46", "%f47", \
+ "%f48", "%f49", "%f50", "%f51", "%f52", "%f53", "%f54", "%f55", \
+ "%f56", "%f57", "%f58", "%f59", "%f60", "%f61", "%f62", "%f63", \
+ "%fcc0", "%fcc1", "%fcc2", "%fcc3", "%icc"}
- We define the fake Condition Code register as an alias for reg 0 (which
- is our `condition code' register), so that condition codes can easily
- be clobbered by an asm. No such register actually exists. Condition
- codes are partly stored in the PSR and partly in the FSR. */
+/* Define additional names for use in asm clobbers and asm declarations. */
-#define ADDITIONAL_REGISTER_NAMES {"ccr", 0, "cc", 0}
+#define ADDITIONAL_REGISTER_NAMES \
+{{"ccr", SPARC_ICC_REG}, {"cc", SPARC_ICC_REG}}
/* How to renumber registers for dbx and gdb. */
#define DBX_REGISTER_NUMBER(REGNO) (REGNO)
-/* On Sun 4, this limit is 2048. We use 1500 to be safe,
- since the length can run past this up to a continuation point. */
-#define DBX_CONTIN_LENGTH 1500
+/* On Sun 4, this limit is 2048. We use 1000 to be safe, since the length
+ can run past this up to a continuation point. Once we used 1500, but
+ a single entry in C++ can run more than 500 bytes, due to the length of
+ mangled symbol names. dbxout.c should really be fixed to do
+ continuations when they are actually needed instead of trying to
+ guess... */
+#define DBX_CONTIN_LENGTH 1000
/* This is how to output a note to DBX telling it the line number
to which the following sequence of instructions corresponds.
#define ASM_GLOBALIZE_LABEL(FILE,NAME) \
do { fputs ("\t.global ", FILE); assemble_name (FILE, NAME); fputs ("\n", FILE);} while (0)
-/* This is how to output a reference to a user-level label named NAME.
- `assemble_name' uses this. */
+/* The prefix to add to user-visible assembler symbols. */
-#define ASM_OUTPUT_LABELREF(FILE,NAME) \
- fprintf (FILE, "_%s", NAME)
+#define USER_LABEL_PREFIX "_"
/* This is how to output a definition of an internal numbered label where
PREFIX is the class of label and NUM is the number within the class. */
#define ASM_OUTPUT_INTERNAL_LABEL(FILE,PREFIX,NUM) \
fprintf (FILE, "%s%d:\n", PREFIX, NUM)
-/* This is how to output a reference to an internal numbered label where
- PREFIX is the class of label and NUM is the number within the class. */
-/* FIXME: This should be used throughout gcc, and documented in the texinfo
- files. There is no reason you should have to allocate a buffer and
- `sprintf' to reference an internal label (as opposed to defining it). */
-
-#define ASM_OUTPUT_INTERNAL_LABELREF(FILE,PREFIX,NUM) \
- fprintf (FILE, "%s%d", PREFIX, NUM)
-
/* This is how to store into the string LABEL
the symbol_ref name of an internal numbered label where
PREFIX is the class of label and NUM is the number within the class.
do { \
char label[30]; \
ASM_GENERATE_INTERNAL_LABEL (label, "L", VALUE); \
- fprintf (FILE, "\t.word\t"); \
+ if (Pmode == SImode) \
+ fprintf (FILE, "\t.word\t"); \
+ else if (TARGET_MEDLOW) \
+ fprintf (FILE, "\t.word\t0\n\t.word\t"); \
+ else \
+ fprintf (FILE, "\t.xword\t"); \
assemble_name (FILE, label); \
fprintf (FILE, "\n"); \
} while (0)
do { \
char label[30]; \
ASM_GENERATE_INTERNAL_LABEL (label, "L", VALUE); \
- fprintf (FILE, "\t.word\t"); \
+ if (Pmode == SImode) \
+ fprintf (FILE, "\t.word\t"); \
+ else if (TARGET_MEDLOW) \
+ fprintf (FILE, "\t.word\t0\n\t.word\t"); \
+ else \
+ fprintf (FILE, "\t.xword\t"); \
assemble_name (FILE, label); \
fprintf (FILE, "-1b\n"); \
} while (0)
to define a global common symbol. */
#define ASM_OUTPUT_COMMON(FILE, NAME, SIZE, ROUNDED) \
-( fputs ("\t.global ", (FILE)), \
- assemble_name ((FILE), (NAME)), \
- fputs ("\n\t.common ", (FILE)), \
+( fputs ("\t.common ", (FILE)), \
assemble_name ((FILE), (NAME)), \
- fprintf ((FILE), ",%u,\"bss\"\n", (ROUNDED)))
+ fprintf ((FILE), ",%u,\"bss\"\n", (SIZE)))
-/* This says how to output an assembler line
- to define a local common symbol. */
+/* This says how to output an assembler line to define a local common
+ symbol. */
-#define ASM_OUTPUT_LOCAL(FILE, NAME, SIZE, ROUNDED) \
-( fputs ("\n\t.reserve ", (FILE)), \
- assemble_name ((FILE), (NAME)), \
- fprintf ((FILE), ",%u,\"bss\"\n", (ROUNDED)))
+#define ASM_OUTPUT_ALIGNED_LOCAL(FILE, NAME, SIZE, ALIGNED) \
+( fputs ("\t.reserve ", (FILE)), \
+ assemble_name ((FILE), (NAME)), \
+ fprintf ((FILE), ",%u,\"bss\",%u\n", \
+ (SIZE), ((ALIGNED) / BITS_PER_UNIT)))
/* Store in OUTPUT a string (made with alloca) containing
an assembler-name for a local static variable named NAME.
#define ASM_OUTPUT_IDENT(FILE, NAME) \
fprintf (FILE, "\t%s\t\"%s\"\n", IDENT_ASM_OP, NAME);
+/* Output code to add DELTA to the first argument, and then jump to FUNCTION.
+ Used for C++ multiple inheritance. */
+#define ASM_OUTPUT_MI_THUNK(FILE, THUNK_FNDECL, DELTA, FUNCTION) \
+do { \
+ int big_delta = (DELTA) >= 4096 || (DELTA) < -4096; \
+ if (big_delta) \
+ fprintf (FILE, "\tset %d,%%g1\n\tadd %%o0,%%g1,%%o0\n", (DELTA)); \
+ if (TARGET_MEDANY || TARGET_FULLANY) \
+ { \
+ fprintf (FILE, "\tsetx "); \
+ assemble_name \
+ (FILE, IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (FUNCTION))); \
+ fprintf (FILE, ",%%g5,%%g1\n\tjmp %%g1\n"); \
+ } \
+ else \
+ { \
+ fprintf (FILE, "\tsethi %%hi("); \
+ assemble_name \
+ (FILE, IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (FUNCTION))); \
+ fprintf (FILE, "),%%g1\n\tjmp %%g1+%%lo("); \
+ assemble_name \
+ (FILE, IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (FUNCTION))); \
+ fprintf (FILE, ")\n"); \
+ } \
+ if (big_delta) fprintf (FILE, "\tnop\n"); \
+ else fprintf (FILE, "\tadd %%o0,%d,%%o0\n", DELTA); \
+} while (0)
+
/* Define the parentheses used to group arithmetic operations
in assembler code. */
fprintf (FILE, "%+d", offset); \
else if (GET_CODE (index) == REG) \
fprintf (FILE, "+%s", reg_names[REGNO (index)]); \
- else if (GET_CODE (index) == SYMBOL_REF) \
+ else if (GET_CODE (index) == SYMBOL_REF \
+ || GET_CODE (index) == CONST) \
fputc ('+', FILE), output_addr_const (FILE, index); \
else abort (); \
} \
} \
}
+/* The number of Pmode words for the setjmp buffer. */
+#define JMP_BUF_SIZE 12
+
/* Declare functions defined in sparc.c and used in templates. */
extern char *singlemove_string ();
extern char *output_block_move ();
extern char *output_scc_insn ();
extern char *output_cbranch ();
+extern char *output_v9branch ();
extern char *output_return ();
/* Defined in flags.h, but insn-emit.c does not include flags.h. */
OSDN Git Service
