blob: 5a8b2b844721fa20e131f981dc3f23a9efcb3263 [file] [log] [blame]
; Copyright (C) 2005-2018 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 3, 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 COPYING3. If not see
; <http://www.gnu.org/licenses/>.
HeaderInclude
config/ia64/ia64-opts.h
; Which cpu are we scheduling for.
Variable
enum processor_type ia64_tune = PROCESSOR_ITANIUM2
mbig-endian
Target Report RejectNegative Mask(BIG_ENDIAN)
Generate big endian code.
mlittle-endian
Target Report RejectNegative InverseMask(BIG_ENDIAN)
Generate little endian code.
mgnu-as
Target Report Mask(GNU_AS)
Generate code for GNU as.
mgnu-ld
Target Report Mask(GNU_LD)
Generate code for GNU ld.
mvolatile-asm-stop
Target Report Mask(VOL_ASM_STOP)
Emit stop bits before and after volatile extended asms.
mregister-names
Target Mask(REG_NAMES)
Use in/loc/out register names.
mno-sdata
Target Report RejectNegative Mask(NO_SDATA)
msdata
Target Report RejectNegative InverseMask(NO_SDATA)
Enable use of sdata/scommon/sbss.
mno-pic
Target Report RejectNegative Mask(NO_PIC)
Generate code without GP reg.
mconstant-gp
Target Report RejectNegative Mask(CONST_GP)
gp is constant (but save/restore gp on indirect calls).
mauto-pic
Target Report RejectNegative Mask(AUTO_PIC)
Generate self-relocatable code.
minline-float-divide-min-latency
Target Report RejectNegative Var(TARGET_INLINE_FLOAT_DIV, 1)
Generate inline floating point division, optimize for latency.
minline-float-divide-max-throughput
Target Report RejectNegative Var(TARGET_INLINE_FLOAT_DIV, 2) Init(2)
Generate inline floating point division, optimize for throughput.
mno-inline-float-divide
Target Report RejectNegative Var(TARGET_INLINE_FLOAT_DIV, 0)
minline-int-divide-min-latency
Target Report RejectNegative Var(TARGET_INLINE_INT_DIV, 1)
Generate inline integer division, optimize for latency.
minline-int-divide-max-throughput
Target Report RejectNegative Var(TARGET_INLINE_INT_DIV, 2)
Generate inline integer division, optimize for throughput.
mno-inline-int-divide
Target Report RejectNegative Var(TARGET_INLINE_INT_DIV, 0)
Do not inline integer division.
minline-sqrt-min-latency
Target Report RejectNegative Var(TARGET_INLINE_SQRT, 1)
Generate inline square root, optimize for latency.
minline-sqrt-max-throughput
Target Report RejectNegative Var(TARGET_INLINE_SQRT, 2)
Generate inline square root, optimize for throughput.
mno-inline-sqrt
Target Report RejectNegative Var(TARGET_INLINE_SQRT, 0)
Do not inline square root.
mdwarf2-asm
Target Report Mask(DWARF2_ASM)
Enable DWARF line debug info via GNU as.
mearly-stop-bits
Target Report Mask(EARLY_STOP_BITS)
Enable earlier placing stop bits for better scheduling.
mfixed-range=
Target RejectNegative Joined Var(ia64_deferred_options) Defer
Specify range of registers to make fixed.
mtls-size=
Target RejectNegative Joined UInteger Var(ia64_tls_size) Init(22)
Specify bit size of immediate TLS offsets.
mtune=
Target RejectNegative Joined Enum(ia64_tune) Var(ia64_tune)
Schedule code for given CPU.
Enum
Name(ia64_tune) Type(enum processor_type)
Known Itanium CPUs (for use with the -mtune= option):
EnumValue
Enum(ia64_tune) String(itanium2) Value(PROCESSOR_ITANIUM2)
EnumValue
Enum(ia64_tune) String(mckinley) Value(PROCESSOR_ITANIUM2)
msched-br-data-spec
Target Report Var(mflag_sched_br_data_spec) Init(0)
Use data speculation before reload.
msched-ar-data-spec
Target Report Var(mflag_sched_ar_data_spec) Init(1)
Use data speculation after reload.
msched-control-spec
Target Report Var(mflag_sched_control_spec) Init(2)
Use control speculation.
msched-br-in-data-spec
Target Report Var(mflag_sched_br_in_data_spec) Init(1)
Use in block data speculation before reload.
msched-ar-in-data-spec
Target Report Var(mflag_sched_ar_in_data_spec) Init(1)
Use in block data speculation after reload.
msched-in-control-spec
Target Report Var(mflag_sched_in_control_spec) Init(1)
Use in block control speculation.
msched-spec-ldc
Target Report Var(mflag_sched_spec_ldc) Init(1)
Use simple data speculation check.
msched-spec-control-ldc
Target Report Var(mflag_sched_spec_control_ldc) Init(0)
Use simple data speculation check for control speculation.
msched-prefer-non-data-spec-insns
Target Ignore Warn(switch %qs is no longer supported)
msched-prefer-non-control-spec-insns
Target Ignore Warn(switch %qs is no longer supported)
msched-count-spec-in-critical-path
Target Report Var(mflag_sched_count_spec_in_critical_path) Init(0)
Count speculative dependencies while calculating priority of instructions.
msched-stop-bits-after-every-cycle
Target Report Var(mflag_sched_stop_bits_after_every_cycle) Init(1)
Place a stop bit after every cycle when scheduling.
msched-fp-mem-deps-zero-cost
Target Report Var(mflag_sched_fp_mem_deps_zero_cost) Init(0)
Assume that floating-point stores and loads are not likely to cause conflict when placed into one instruction group.
msched-max-memory-insns=
Target RejectNegative Joined UInteger Var(ia64_max_memory_insns) Init(1)
Soft limit on number of memory insns per instruction group, giving lower priority to subsequent memory insns attempting to schedule in the same insn group. Frequently useful to prevent cache bank conflicts. Default value is 1.
msched-max-memory-insns-hard-limit
Target Report Var(mflag_sched_mem_insns_hard_limit) Init(0)
Disallow more than 'msched-max-memory-insns' in instruction group. Otherwise, limit is 'soft' (prefer non-memory operations when limit is reached).
msel-sched-dont-check-control-spec
Target Report Var(mflag_sel_sched_dont_check_control_spec) Init(0)
Don't generate checks for control speculation in selective scheduling.
; This comment is to ensure we retain the blank line above.