aboutsummaryrefslogtreecommitdiff
path: root/gcc/sched.c
AgeCommit message (Expand)Author
1997-02-19(schedule_insns): If there was no first scheduling pass,wilson
1996-12-17(update_flow_info): When add REG_DEAD notes for dest ofwilson
1996-10-08(schedule_block): Before scheduling, add code to make allwilson
1996-07-23install EH codemrs
1996-07-08(true_dependence, anti_dependence, output_dependence):wilson
1996-07-03formatting tweaksmrs
1996-03-31(schedule_block): CALL_INSNs don't affect fixed regs.kenner
1996-02-23(sched_analyze): Record NOTE_INSN_SETJMP if no CALL_INSN as prev;kenner
1996-02-10(flush_pending_lists): Add new arg, ONLY_WRITE.kenner
1996-02-10(add_dependence): Add test for next != CODE_LABEL.dje
1996-01-15 * sched.c (reemit_notes): Add prototype.dje
1996-01-14(schedule_block): Copy RTX_INTEGRATE_P bit when create a new note.wilson
1996-01-12 * sched.c (schedule_block): Maintain a valid chain sodje
1995-12-31 * sched.c (adjust_priority): Use ADJUST_PRIORITY if its defined.law
1995-12-18 * sched.c (canon_rtx): Recursively look for equivalences;law
1995-06-15Update FSF address.kenner
1995-06-05(attach_deaths): In last change, use find_reg_note instead ofkenner
1995-06-04(attach_deaths): Don't add a REG_DEAD note if a REG_UNUSED note iskenner
1995-05-25(reemit_notes): New function.wilson
1995-05-16Fix typos in comments.kenner
1995-01-25Update comments.kenner
1994-09-20(sched_analyze_insn): Parameter loop_note type changedwilson
1994-09-15(sched_analyze_insn): Don't set SCHED_GROUP_P for previous INSN a USEkenner
1994-09-08(attach_deaths): Correct typo, use GET_MODE not GET_CODE.wilson
1994-08-20* sched.c (sched_analyze_insn): Add new parameter loop_note.wilson
1994-08-14(attach_deaths): When adding REG_DEAD notes, check forwilson
1994-06-14Cast pointer operands to bzero, bcopy, and bcmp to (char *).kenner
1994-05-11(sched_analyze_insn): Fix typo in last change.kenner
1994-05-07(sched_analyze_insn, attach_deaths_insn): Look atkenner
1994-04-20(attach_deaths): Create partial register dead notes usingdje
1994-03-25(clear_units): Remove decl of unused variable UNIT.kenner
1994-03-09(rtx_equal_for_tmemref_p): Commutative operations are identical if thekenner
1994-02-06(reg_pending_sets, reg_pending_sets_all): New variables.wilson
1994-02-01(sched_analyze_insn): Add use dependencies for registerswilson
1994-02-01(priority): Call insn_cost() for anti- and output-wilson
1994-01-26(attach_deaths_insn): Add REG_DEAD notes to CLOBBER ifwilson
1994-01-10(new_insn_dead_notes): Compare registers with rtx_equal_p.wilson
1993-12-22(priority): Handle INSN_DELETED_P insns, in addition towilson
1993-12-21(sched_analyze): For CALL_INSN followed bywilson
1993-10-05(sched_analyze_2): Make volatile asms depend on allwilson
1993-09-29Add prototypes for static function.wilson
1993-09-28(schedule_insns): Don't zero reg_n_calls_crossed forwilson
1993-09-24* sched.c: (memrefs_conflict_p): We can work out whether referencesdje
1993-09-07(schedule_block): Correct type of BIT.kenner
1993-08-12(sched_analyze_insn): Handle USE before a JUMP_INSN justwilson
1993-08-09(schedule_insns): Don't emit a NOTE at the end of the lastwilson
1993-07-28(schedule_block): Do not emit line number notes before other notes.wilson
1993-06-12(add_dependence): Reject CODE_LABELs when searching forwilson
1993-06-10(sched_analyze_2): Don't ignore USE insns when creating dependencies.wilson
1993-05-14(split_hard_reg_notes): Handle 0th sub register, andwilson