bd52d17906
too large to list, but see: http://gcc.gnu.org/gcc-3.4/changes.html http://gcc.gnu.org/gcc-4.0/changes.html http://gcc.gnu.org/gcc-4.1/changes.html for the details.
192 lines
6.4 KiB
Plaintext
192 lines
6.4 KiB
Plaintext
; Options for the CRIS port of the compiler.
|
|
|
|
; Copyright (C) 2005 Free Software Foundation, Inc.
|
|
;
|
|
; This file is part of GCC.
|
|
;
|
|
; GCC is free software; you can redistribute it and/or modify it under
|
|
; the terms of the GNU General Public License as published by the Free
|
|
; Software Foundation; either version 2, or (at your option) any later
|
|
; version.
|
|
;
|
|
; GCC is distributed in the hope that it will be useful, but WITHOUT ANY
|
|
; WARRANTY; without even the implied warranty of MERCHANTABILITY or
|
|
; FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
|
|
; for more details.
|
|
;
|
|
; You should have received a copy of the GNU General Public License
|
|
; along with GCC; see the file COPYING. If not, write to the Free
|
|
; Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA
|
|
; 02110-1301, USA.
|
|
|
|
; TARGET_MUL_BUG: Whether or not to work around multiplication
|
|
; instruction hardware bug when generating code for models where
|
|
; it may be present. From the trouble report for Etrax 100 LX:
|
|
; "A multiply operation may cause incorrect cache behaviour
|
|
; under some specific circumstances. The problem can occur if
|
|
; the instruction following the multiply instruction causes a
|
|
; cache miss, and multiply operand 1 (source operand) bits
|
|
; [31:27] matches the logical mapping of the mode register
|
|
; address (0xb0....), and bits [9:2] of operand 1 matches the
|
|
; TLB register address (0x258-0x25f). There is such a mapping
|
|
; in kernel mode or when the MMU is off. Normally there is no
|
|
; such mapping in user mode, and the problem will therefore
|
|
; probably not occur in Linux user mode programs."
|
|
;
|
|
; We have no sure-fire way to know from within GCC that we're
|
|
; compiling a user program. For example, -fpic/PIC is used in
|
|
; libgcc which is linked into the kernel. However, the
|
|
; workaround option -mno-mul-bug can be safely used per-package
|
|
; when compiling programs. The same goes for general user-only
|
|
; libraries such as glibc, since there's no user-space
|
|
; driver-like program that gets a mapping of I/O registers (all
|
|
; on the same page, including the TLB registers).
|
|
mmul-bug-workaround
|
|
Target Report Mask(MUL_BUG)
|
|
Work around bug in multiplication instruction
|
|
|
|
; TARGET_ETRAX4_ADD: Instruction-set additions from Etrax 4 and up.
|
|
; (Just "lz", which we don't really generate from GCC -- yet).
|
|
metrax4
|
|
Target Report Mask(ETRAX4_ADD)
|
|
Compile for ETRAX 4 (CRIS v3)
|
|
|
|
; See cris_handle_option.
|
|
metrax100
|
|
Target Report RejectNegative
|
|
Compile for ETRAX 100 (CRIS v8)
|
|
|
|
; See cris_handle_option.
|
|
mno-etrax100
|
|
Target Report RejectNegative Undocumented
|
|
|
|
mpdebug
|
|
Target Report Mask(PDEBUG)
|
|
Emit verbose debug information in assembly code
|
|
|
|
; TARGET_CCINIT: Whether to use condition-codes generated by
|
|
; insns other than the immediately preceding compare/test insn.
|
|
; Used to check for errors in notice_update_cc.
|
|
mcc-init
|
|
Target Report Mask(CCINIT)
|
|
Do not use condition codes from normal instructions
|
|
|
|
; TARGET_SIDE_EFFECT_PREFIXES: Whether to use side-effect
|
|
; patterns. Used to debug the [rx=ry+i] type patterns.
|
|
mside-effects
|
|
Target Report RejectNegative Mask(SIDE_EFFECT_PREFIXES) Undocumented
|
|
|
|
mno-side-effects
|
|
Target Report RejectNegative InverseMask(SIDE_EFFECT_PREFIXES)
|
|
Do not emit addressing modes with side-effect assignment
|
|
|
|
; TARGET_STACK_ALIGN: Whether to *keep* (not force) alignment of
|
|
; stack at 16 (or 32, depending on TARGET_ALIGN_BY_32) bits.
|
|
mstack-align
|
|
Target Report RejectNegative Mask(STACK_ALIGN) Undocumented
|
|
|
|
mno-stack-align
|
|
Target Report RejectNegative InverseMask(STACK_ALIGN)
|
|
Do not tune stack alignment
|
|
|
|
; TARGET_DATA_ALIGN: Whether to do alignment on individual
|
|
; modifiable objects.
|
|
mdata-align
|
|
Target Report RejectNegative Mask(DATA_ALIGN) Undocumented
|
|
|
|
mno-data-align
|
|
Target Report RejectNegative InverseMask(DATA_ALIGN)
|
|
Do not tune writable data alignment
|
|
|
|
; TARGET_CONST_ALIGN: Whether to do alignment on individual
|
|
; non-modifiable objects.
|
|
mconst-align
|
|
Target Report RejectNegative Mask(CONST_ALIGN) Undocumented
|
|
|
|
mno-const-align
|
|
Target Report RejectNegative InverseMask(CONST_ALIGN)
|
|
Do not tune code and read-only data alignment
|
|
|
|
; See cris_handle_option.
|
|
m32-bit
|
|
Target Report RejectNegative Undocumented
|
|
|
|
; See cris_handle_option.
|
|
m32bit
|
|
Target Report RejectNegative
|
|
Align code and data to 32 bits
|
|
|
|
; See cris_handle_option.
|
|
m16-bit
|
|
Target Report RejectNegative Undocumented
|
|
|
|
; See cris_handle_option.
|
|
m16bit
|
|
Target Report RejectNegative Undocumented
|
|
|
|
; See cris_handle_option.
|
|
m8-bit
|
|
Target Report RejectNegative Undocumented
|
|
|
|
; See cris_handle_option.
|
|
m8bit
|
|
Target Report RejectNegative
|
|
Don't align items in code or data
|
|
|
|
; TARGET_PROLOGUE_EPILOGUE: Whether or not to omit function
|
|
; prologue and epilogue.
|
|
mprologue-epilogue
|
|
Target Report RejectNegative Mask(PROLOGUE_EPILOGUE) Undocumented
|
|
|
|
mno-prologue-epilogue
|
|
Target Report RejectNegative InverseMask(PROLOGUE_EPILOGUE)
|
|
Do not emit function prologue or epilogue
|
|
|
|
; We have to handle this m-option here since we can't wash it
|
|
; off in both CC1_SPEC and CC1PLUS_SPEC.
|
|
|
|
mbest-lib-options
|
|
Target Report RejectNegative
|
|
Use the most feature-enabling options allowed by other options
|
|
|
|
; FIXME: The following comment relates to gcc before cris.opt.
|
|
; Check it it's still valid:
|
|
; We must call it "override-" since calling it "no-" will cause
|
|
; gcc.c to forget it, if there's a "later" -mbest-lib-options.
|
|
; Kludgy, but needed for some multilibbed files.
|
|
moverride-best-lib-options
|
|
Target Report RejectNegative
|
|
Override -mbest-lib-options
|
|
|
|
mcpu=
|
|
Target Report RejectNegative Joined Undocumented Var(cris_cpu_str)
|
|
|
|
march=
|
|
Target Report RejectNegative Joined Var(cris_cpu_str) VarExists
|
|
-march=ARCH Generate code for the specified chip or CPU version
|
|
|
|
mtune=
|
|
Target Report RejectNegative Joined Var(cris_tune_str)
|
|
-mtune=ARCH Tune alignment for the specified chip or CPU version
|
|
|
|
mmax-stackframe=
|
|
Target Report RejectNegative Joined Var(cris_max_stackframe_str)
|
|
-mmax-stackframe=SIZE Warn when a stackframe is larger than the specified size
|
|
|
|
max-stackframe=
|
|
Target Report RejectNegative Joined Undocumented Var(cris_max_stackframe_str) VarExists
|
|
|
|
; TARGET_SVINTO: Currently this just affects alignment. FIXME:
|
|
; Redundant with TARGET_ALIGN_BY_32, or put machine stuff here?
|
|
; This and the others below could just as well be variables and
|
|
; TARGET_* defines in cris.h.
|
|
Mask(SVINTO)
|
|
|
|
; TARGET_ALIGN_BY_32: Say that all alignment specifications say
|
|
; to prefer 32 rather than 16 bits.
|
|
Mask(ALIGN_BY_32)
|
|
|
|
; TARGET_AVOID_GOTPLT is referred to in the .c and the .md so we
|
|
; need to allocate the flag and macros here.
|
|
Mask(AVOID_GOTPLT)
|