diff options
| author | Mostafa Hagog <mustafa@il.ibm.com> | 2004-08-08 21:35:53 +0000 |
|---|---|---|
| committer | Mostafa Hagog <hagog@gcc.gnu.org> | 2004-08-08 21:35:53 +0000 |
| commit | d72372e46ac06b4db61143f91367a777089cc6f6 (patch) | |
| tree | d4a4dc1d2fd739eff16c75b3bbf87780df96f926 /gcc/doc | |
| parent | 94538bd12ae8ffa02164399a616ef806e77f797b (diff) | |
| download | gcc-d72372e46ac06b4db61143f91367a777089cc6f6.zip gcc-d72372e46ac06b4db61143f91367a777089cc6f6.tar.gz gcc-d72372e46ac06b4db61143f91367a777089cc6f6.tar.bz2 | |
common.opt (freschedule-modulo-scheduled-loops): New flag.
2004-08-08 Mostafa Hagog <mustafa@il.ibm.com>
Ayal Zaks <zaks@il.ibm.com>
* common.opt (freschedule-modulo-scheduled-loops): New flag.
* final.c (final_scan_insn): Handle NOTE_DISABLE_SCHED_OF_BLOCK.
* modulo-sched.c (sms_schedule): Emit a note to disable scheduling
when -freschedule-modulo-scheduled-loops flag is not specified.
(sms_schedule_by_order, ps_insn_advance_column, add_node_to_ps,
add_node_to_ps, ps_has_conflicts, ps_add_node_check_conflicts):
More accurate placing of insn in row of partial schedule.
(ps_insn_find_column): New function.
* rtl.h (NOTE_DISABLE_SCHED_OF_BLOCK): New note.
* sched-rgn.c (sched_is_disabled_for_current_region_p): New.
(schedule_region): Use sched_is_disabled_for_current_region_p.
* docs/invoke.texi: Document -freschedule-modulo-scheduled-loops.
Co-Authored-By: Ayal Zaks <zaks@il.ibm.com>
From-SVN: r85696
Diffstat (limited to 'gcc/doc')
| -rw-r--r-- | gcc/doc/invoke.texi | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index f1988b0..0018fc6 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -308,8 +308,8 @@ in the following sections. -fsched-spec-load-dangerous @gol -fsched-stalled-insns=@var{n} -sched-stalled-insns-dep=@var{n} @gol -fsched2-use-superblocks @gol --fsched2-use-traces -fsignaling-nans @gol --fsingle-precision-constant @gol +-fsched2-use-traces -freschedule-modulo-scheduled-loops @gol +-fsignaling-nans -fsingle-precision-constant @gol -fstrength-reduce -fstrict-aliasing -ftracer -fthread-jumps @gol -funroll-all-loops -funroll-loops -fpeel-loops @gol -funswitch-loops -fold-unroll-loops -fold-unroll-all-loops @gol @@ -4393,6 +4393,12 @@ reality and hurt the performance. This only makes sense when scheduling after register allocation, i.e.@: with @option{-fschedule-insns2} or at @option{-O2} or higher. +@item -freschedule-modulo-scheduled-loops +@opindex fscheduling-in-modulo-scheduled-loops +The modulo scheduling comes before the traditional scheduling, if a loop was modulo scheduled +we may want to prevent the later scheduling passes from changing its schedule, we use this +option to control that. + @item -fcaller-saves @opindex fcaller-saves Enable values to be allocated in registers that will be clobbered by |