aboutsummaryrefslogtreecommitdiff
path: root/gcc/reload.c
AgeCommit message (Expand)AuthorFilesLines
1993-09-01(find_reloads): Bump REJECT for a hard register; just not for aRichard Kenner1-1/+3
1993-09-01(find_reloads_address_1): Handle SUBREGs same as REGs.Jim Wilson1-20/+34
1993-08-31(push_reload): Abort is secondary_reload wants sameJim Wilson1-0/+15
1993-08-07(find_reloads): Don't combine two reloads if their reload_reg_rtxRichard Kenner1-1/+2
1993-08-07(push_reload, find_reloads): Treat (subreg (pseudo)) and (subreg (mem)) the s...Richard Kenner1-24/+38
1993-08-05(push_reload, find_reloads): Refine when we reload inside a SUBREG by respect...Richard Kenner1-37/+55
1993-07-05(subst_reloads): Don't use gen_lowpart_common to get reload registerRichard Kenner1-2/+4
1993-06-28(find_reloads_address): Be selective about what innerRichard Stallman1-2/+8
1993-06-26(find_reloads): When merging reloads again after they have been changed...Richard Stallman1-1/+3
1993-06-14(make_memloc): Don't reuse a memloc.Richard Stallman1-0/+10
1993-05-24(push_reload): Fix NULL arg in last change.Richard Stallman1-3/+32
1993-05-10(find_equiv_reg): Set goal_mem_addr_variesRichard Stallman1-1/+3
1993-05-07(push_reload): Don't use IN as the reload reg in in-outRichard Stallman1-0/+12
1993-05-06(find_equiv_reg): Don't look inside a CONST_DOUBLE; useRichard Kenner1-4/+8
1993-05-02(push_reload): Allow non-paradoxical SUBREGs of MEMRichard Stallman1-2/+6
1993-03-30(form_sum): Use mode of inputs when making new PLUS.Richard Kenner1-2/+9
1993-03-29(secondary_memlocs_elim): Now one per mode per operand.Richard Kenner1-16/+13
1993-03-29Add prototypes for all static functions and clean up some definitions.Richard Kenner1-431/+614
1993-03-16(get_secondary_mem): Use macro SECONDARY_MEMORY_NEEDED_RTX.Jim Wilson1-2/+8
1993-02-16(find_reloads_address): Use strict_memory_address_pRichard Stallman1-19/+6
1993-01-29reload.c (find_reloads_toplev): When handling a (subreg (reg)) where (reg) wi...Jeff Law1-1/+3
1993-01-10# Fix misspellings in comments.Paul Eggert1-1/+1
1992-12-31(operands_match_p): For WORDS_BIG_ENDIAN machines,Jim Wilson1-1/+11
1992-12-30(push_reload): Move the NO_REGS abort after the error check for asms, to avoi...Richard Stallman1-3/+12
1992-12-16(get_secondary_mem): Don't save address if it isn't valid.Jim Wilson1-1/+6
1992-11-27(find_reloads) [SMALL_REGISTER_CLASSES]:Richard Stallman1-0/+7
1992-11-17(combine_reloads): Don't combine with a reload needed for multipleRichard Kenner1-0/+1
1992-10-28(find_reloads_address): When copying memref after we've made recursiveRichard Kenner1-0/+2
1992-10-18(find_reloads_address): Redo last change: copy the memRichard Stallman1-1/+7
1992-10-14(find_reloads): Don't let an operand have a numberedJim Wilson1-0/+11
1992-10-06(find_reloads_address): If tem != ad, then create a newRichard Stallman1-0/+6
1992-09-23(find_reloads_address): Properly check for constant address.Richard Kenner1-1/+1
1992-09-20(find_reloads, find_reloads_toplev): Treat BYTE_LOADS_SIGN_EXTEND justRichard Kenner1-6/+6
1992-09-20Add comment in find_reloads.Richard Stallman1-0/+4
1992-09-19(find_reloads): Always use >= with FIRST_PSEUDO_REGISTER.Richard Stallman1-2/+2
1992-09-02(find_reloads): If have any earlyclobbers, don't let any reloads overlap with...Richard Stallman1-0/+7
1992-08-21(find_dummy_reload): Restore original RTX instead of passed IN rtx.Richard Stallman1-1/+3
1992-08-14(find_reloads): Ignore `%' in the last operand afterTom Wood1-1/+3
1992-07-31entered into RCSRichard Kenner1-3/+4
1992-07-20*** empty log message ***Richard Stallman1-0/+6
1992-07-17*** empty log message ***Richard Kenner1-0/+128
1992-07-17*** empty log message ***Richard Kenner1-3/+22
1992-07-06*** empty log message ***Charles Hannum1-38/+49
1992-06-27*** empty log message ***Richard Kenner1-2/+2
1992-06-24*** empty log message ***Richard Stallman1-1/+1
1992-06-07*** empty log message ***Richard Stallman1-1/+1
1992-05-22*** empty log message ***Richard Kenner1-33/+53
1992-05-21*** empty log message ***Richard Stallman1-6/+10
1992-05-11*** empty log message ***Richard Kenner1-14/+82
1992-05-10*** empty log message ***Richard Stallman1-3/+3
generated by cgit v1.1 at 2025-02-23 01:23:11 +0000