aboutsummaryrefslogtreecommitdiff
path: root/gcc/flow.c
AgeCommit message (Expand)AuthorFilesLines
1996-02-02(jmp_uses_reg_or_mem): Renamed from uses_reg_or_mem.Richard Kenner1-13/+31
1995-08-28Fix spelling errors.Richard Kenner1-1/+1
1995-06-15Update FSF address.Richard Kenner1-1/+2
1995-05-16Fix typos in comments.Richard Kenner1-3/+3
1995-05-13(find_basic_blocks): Only perform n_basic_blocks sanity check on first pass...Doug Evans1-1/+9
1995-05-11Eliminate dead code caused by last patchMichael Meissner1-10/+5
1995-05-11Do not segfault if SUBREG uses something other than REG.Michael Meissner1-0/+7
1995-03-29(mark_used_regs, case SUBREG): Only fall through to REG caseJim Wilson1-0/+5
1995-03-18(mark_used_regs, case SUBREG): Set reg_changes_size even for integerRichard Kenner1-4/+9
1995-02-13(find_basic_blocks): Revise calculation of when to make another pass.Doug Evans1-2/+7
1995-01-20(find_basic_blocks): No longer need to check for deletedJim Wilson1-5/+2
1994-12-19(mark_set_1): For multi-reg hard register, clear allJim Wilson1-3/+8
1994-11-17(find_auto_inc): Insert missing "return" statement.Richard Kenner1-0/+2
1994-11-15(flow_delete_insn): New function.Doug Evans1-45/+119
1994-10-26(find_basic_blocks): Ensure each live basic block is actually reachable.Richard Kenner1-3/+17
1994-10-18(propagate_block): Call mark_used_regs to show global regs used byRichard Kenner1-3/+3
1994-09-08(mark_used_regs, case REG): Correct typo, use GET_MODE not GET_CODE.Jim Wilson1-1/+1
1994-08-14(mark_used_regs): When adding REG_DEAD notes...Jim Wilson1-0/+10
1994-08-12(mark_used_regs): Only show changes size if one mode is integral.Richard Kenner1-1/+3
1994-08-11(find_auto_inc): Ensure that all proposed changes to insns are validRichard Kenner1-44/+56
1994-08-07(reg_changes_size): New variable.Richard Kenner1-0/+21
1994-07-07(mark_set_1): Record explicit hard registers in their raw mode.Doug Evans1-2/+4
1994-06-14Cast pointer operands to bzero, bcopy, and bcmp to (char *).Richard Kenner1-24/+33
1994-05-16(try_pre_increment_1): Don't try to create a pre-increment expressionRichard Kenner1-0/+3
1994-05-06(flow_analysis, find_basic_blocks): Change from USE and CLOBBER insnsRichard Kenner1-21/+13
1994-04-20*** empty log message ***Steve Chamberlain1-13/+16
1994-03-11(libcall_dead_p): Fail not abort for libcall that returnsJim Wilson1-1/+4
1994-02-08(find_basic_blocks): Also look for REG_LABEL notes on firstJim Wilson1-1/+4
1994-02-03Add prototypes for static functions.Richard Kenner1-114/+173
1994-01-28(find_auto_inc): Check for increment using single_set.Richard Kenner1-7/+9
1993-12-08(mark_used_regs, case CLOBBER): Refine previous change to just look atRichard Kenner1-1/+8
1993-12-07(mark_used_regs): Look inside a CLOBBER...Richard Kenner1-1/+0
1993-11-14(find_basic_blocks): When starting new basic block for a CALL_INSN in the pre...Richard Kenner1-3/+14
1993-09-24flow.c: (life_analysis): Mark HARD_FRAME_POINTER_REGNUM as live at the end of...Doug Evans1-0/+23
1993-07-02(find_auto_inc): In *p=expr;q=p+size case, avoid clobbering q if referenced i...Doug Evans1-3/+6
1993-04-24(mark_used_regs): Use proper type for {all,some}_needed.Richard Kenner1-2/+2
1993-04-17(uid_block_number): Likewise.Richard Kenner1-4/+4
1993-04-08(mark_set_1): Clear reg_next_use at the proper place.Richard Stallman1-3/+3
1993-04-07(mark_set_1): Clear reg_next_use for all kinds of regs.Richard Stallman1-3/+5
1993-02-19(regno_uninitialized): Test global_regs only for hard regs.Richard Stallman1-1/+2
1993-01-23(find_basic_blocks): Don't mark labels in nonlocal_label_listRichard Stallman1-2/+5
1992-10-29(life_analysis): Mark FP live at end of function.Richard Kenner1-0/+14
1992-10-26(find_basic_blocks): Check for LABEL_REF_NONLOCAL_P in label_value_list.Richard Stallman1-2/+4
1992-10-20Change global-alloc.c to global.c.Jim Wilson1-1/+1
1992-08-14(life_analysis, propagate_block, dump_flow_info): Add more missingRichard Kenner1-6/+9
1992-08-07(mark_used_regs): Do recurse in setting of global hard reg.Richard Stallman1-2/+9
1992-08-02(find_use_as_address): Use HOST_WIDE_INT instead of int when comparing a poin...Richard Kenner1-4/+4
1992-07-23entered into RCSRichard Stallman1-4/+4
1992-07-06*** empty log message ***Charles Hannum1-52/+71
1992-06-29*** empty log message ***Richard Stallman1-2/+3