| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 1998-02-17 | fold-const.c: Include "system.h" to get stdlib.h and stdio.h. | Kaveh R. Ghazi | 1 | -7/+8 |
| 1998-02-15 | tree.c (first_rtl_op): New fn. | Jason Merrill | 1 | -3/+5 |
| 1998-02-13 | varasm.c (output_constant_pool): Bring back 'done' label inside an appropriat... | Jeffrey A Law | 1 | -1/+0 |
| 1998-02-01 | expr.c (get_inner_reference): Use sbitsizetype for type sizes. | J"orn Rennecke | 1 | -15/+17 |
| 1997-12-20 | bitmap.c (bitmap_clear): Ensure `inline' is at the beginning of the declaration. | Kaveh R. Ghazi | 1 | -1/+3 |
| 1997-12-18 | tree.c (build_range_type): Allow creation of ranges with no maximum. | Richard Henderson | 1 | -10/+35 |
| 1997-12-11 | Another handful of gcc-2.8 changes (will it ever go out the door?). | Jeff Law | 1 | -5/+7 |
| 1997-12-06 | Merge from gcc-2.8 | Jeff Law | 1 | -3/+3 |
| 1997-11-14 | * fold-const.c (fold_truthop): Fix bug in last change. | J"orn Rennecke | 1 | -2/+2 |
| 1997-11-13 | patch from paul eggert to fix -0.0/0.0 confusion that breaks glibc | Paul Eggert | 1 | -2/+2 |
| 1997-11-13 | fold-const.c (fold_truthop): When changing a one-bit comparison against zero ... | J"orn Rennecke | 1 | -2/+19 |
| 1997-11-02 | Update mainline egcs to gcc2 snapshot 971021. | Jeff Law | 1 | -175/+265 |
| 1997-10-28 | fold-const.c (fold): For ((a * C1) / C3) or (((a * C1) + C2) / C3) optimizati... | Toon Moene | 1 | -0/+31 |
| 1997-10-26 | fold-const (fold): Also simplify FLOOR_DIV_EXPR to EXACT_DIV_EXPR if... | Toon Moene | 1 | -5/+4 |
| 1997-09-23 | fold-const.c (make_range): Correctly handle cases of converting from unsigned... | Jeff Law | 1 | -23/+81 |
| 1997-08-29 | pa.md (reload_peepholes): Make sure operand is a REG before examining REGNO. | Jeffrey A Law | 1 | -179/+155 |
| 1997-08-27 | c-decl.c (grokdeclarator): If array index or size calculations overflow, issu... | Doug Evans | 1 | -155/+179 |
| 1997-08-25 | version.c: Bump for new snapshot. | Jeff Law | 1 | -0/+113 |
| 1997-07-27 | (fold_range_test, fold): Avoid making SAVE_EXPR if at top level. | Richard Kenner | 1 | -11/+17 |
| 1997-06-26 | Document force_fit_type as an entry point in the file. | Doug Evans | 1 | -3/+7 |
| 1997-06-09 | x | Jason Merrill | 1 | -6/+13 |
| 1997-05-08 | (fold, truth_andor): Disable optimization for side effects on RHS. | Richard Kenner | 1 | -2/+4 |
| 1997-04-20 | (operand_equal_p): Constants are not equal if there has been an overflow. | Richard Kenner | 1 | -2/+7 |
| 1997-02-15 | (fold): Don't assume a nonexplicit constant cannot equal an explicit one. | Richard Kenner | 1 | -1/+2 |
| 1997-02-07 | (const_binop): Don't call size_int if low < 0. | Richard Kenner | 1 | -1/+1 |
| 1997-01-27 | (fold): Don't make COND_EXPR when both expr and one part are comparisons. | Richard Kenner | 1 | -2/+4 |
| 1997-01-06 | (operand_equal_p): Rework to consider two expressions that have embedded iden... | Richard Kenner | 1 | -33/+57 |
| 1997-01-04 | (const_binop): Rework to only make constants in one place; always use size_in... | Richard Kenner | 1 | -68/+62 |
| 1996-12-07 | (merge_ranges): In (+,-) case, if lower bounds are the same the result | Richard Kenner | 1 | -4/+3 |
| 1996-11-16 | (optimize_bit_field_compare, decode_field_reference): Pass new arg to | Richard Kenner | 1 | -4/+6 |
| 1996-11-15 | (unextend): Rewrite type conversions to avoid overflow. | Jim Wilson | 1 | -3/+13 |
| 1996-11-15 | (operand_equal_p): Do real comparison with REAL_VALUES_EQUAL. | Richard Kenner | 1 | -7/+29 |
| 1996-11-15 | (range_binop): Set SNG1 to zero if ARG1 is a not a lower or upper | Richard Kenner | 1 | -1/+1 |
| 1996-11-12 | (merge_ranges): In (+,-) case, treat subset specially | Jim Wilson | 1 | -4/+8 |
| 1996-11-07 | (const_binop, case *_DIV_EXPR): Don't do special sizetype case if a | Richard Kenner | 1 | -9/+7 |
| 1996-11-01 | (make_range): Don't merge unsigned case with range containing upper bound; | Richard Kenner | 1 | -15/+24 |
| 1996-10-31 | fold-const.c (make_range, [...]): Correct normalization of an unsigned range ... | Jeff Law | 1 | -3/+9 |
| 1996-10-30 | (range_binop): Strip NOPs from result and verify it's an INTEGER_CST. | Richard Kenner | 1 | -9/+12 |
| 1996-10-27 | * fold-const.c (merge_ranges): Fix thinko/typo. | Jeff Law | 1 | -1/+1 |
| 1996-10-24 | (make_range, case PLUS_EXPR): Normalize an unsigned | Jim Wilson | 1 | -1/+9 |
| 1996-10-24 | (make_range, comparison cases): When making range for unsigned to | Richard Kenner | 1 | -6/+11 |
| 1996-10-23 | (exact_real_inverse): Likewise, if no REAL_ARITHMETIC. | Richard Kenner | 1 | -5/+106 |
| 1996-10-12 | fold-const.c (make_range): Handle NULL operand 0 (like in BIND_EXPRs). | Doug Evans | 1 | -1/+1 |
| 1996-10-09 | (range_test): Function deleted. | Richard Kenner | 1 | -156/+461 |
| 1996-09-29 | (const_binop): Add new arg, TYPE, to call to build_complex. | Richard Kenner | 1 | -25/+30 |
| 1996-09-22 | Fix error in comments. | Richard Kenner | 1 | -1/+1 |
| 1996-07-03 | formatting tweaks | Mike Stump | 1 | -1/+1 |
| 1996-06-06 | (fold): Don't do anything with evaluated SAVE_EXPR. | Richard Kenner | 1 | -2/+3 |
| 1996-05-16 | (fold, case EQ_EXPR): When folding VAR++ = CONST, mask the new const | Richard Kenner | 1 | -0/+35 |
| 1996-04-30 | (fold): Don't call convert to recreate tree nodes | Jason Merrill | 1 | -1/+1 |
