| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2012-12-09 | Remove vax-specific configuration constructs. |   miod | 1 | -7/+1 |
| 2012-12-08 | Work-in-progress gcc 3.3.6 configuration for OpenBSD/vax. | miod | 4 | -2/+22 |
| 2012-12-08 | Bring in a couple of bugfixes from old (Aug 2000), but more recent than this | miod | 1 | -9/+13 |
| 2012-12-02 | m88k_layout_frame(): return early if the frame computations have already | miod | 1 | -11/+12 |
| 2012-12-01 | reload_cse_noop_set_p(): invoking rtx_equal_for_cselib_p() is not enough; if | miod | 1 | -1/+13 |
| 2012-11-28 | merge conflicts | miod | 23 | -268/+377 |
| 2012-11-28 | gcc 3.3.6 | miod | 80 | -3580/+28776 |
| 2012-11-21 | Disable the stack protector in OVERRIDE_OPTIONS for the time being. The stack | miod | 1 | -0/+9 |
| 2012-11-21 | Remove an md global which value is a duplicate of MI current_function_args_info. | miod | 1 | -10/+7 |
| 2012-11-21 | Turn OVERRIDE_OPTIONS into a function, will make it easier to extend on a | miod | 3 | -55/+59 |
| 2012-11-20 | Convert to new world order by defining TARGET_OS_CPP_BUILTINS instead of | miod | 2 | -16/+30 |
| 2012-11-20 | Make the mips and powerpc optimizer bug workarounds permanent (i.e. apply | miod | 4 | -9/+1 |
| 2012-11-19 | Make sure to invoke end_sequence() before passing it to any emit_insn() | miod | 1 | -13/+7 |
| 2012-11-19 | move_by_pieces_1(), store_by_pieces_2(): sync insn skipping logic in the | miod | 1 | -6/+14 |
| 2012-11-19 | Sprinkle TREE_SIDE_EFFECTS annotations around all the computations done in | miod | 1 | -0/+4 |
| 2012-11-17 | Revert the va_start() and va_arg() optimization for functions which have no | miod | 1 | -11/+8 |
| 2012-11-17 | In the expansion of __builtin_saveregs(), use double store instructions to | miod | 1 | -3/+29 |
| 2012-11-16 | In SETUP_INCOMING_VARARGS(), make sure to return in pretend_size the amount | miod | 1 | -0/+6 |
| 2012-11-16 | Rewrite the varags implementation again (third time's the charm), the upstream | miod | 2 | -85/+195 |
| 2012-11-15 | Only allow the "return" insn construct (which simply yields a "jmp r1" line) | miod | 1 | -4/+4 |
| 2012-11-15 | Make this cross-compile friendly by defining ATTRIBUTE_UNUSED to an empty | miod | 1 | -0/+3 |
| 2012-11-14 | Do not bother reloading r1 from the stack in the epilogue, if it has not been | miod | 1 | -1/+3 |
| 2012-11-14 | Generously sprinkle RTX_FRAME_RELATED_P() in the prologue and epilogue code, | miod | 1 | -5/+33 |
| 2012-11-14 | Define EPILOGUE_USES(). And suddenly the life analysis pass doesn't decide | miod | 1 | -0/+3 |
| 2012-11-14 | Revert the `sequence' part of 1.3 - it is actually not necessary and hides | miod | 1 | -12/+0 |
| 2012-11-11 | The next step towards a working gcc3/m88k: remove the argument area (from gcc2), | miod | 3 | -88/+60 |
| 2012-11-11 | Explicitely include <nlist.h> if !__ELF__ for this file to compile. | miod | 1 | -0/+3 |
| 2012-11-11 | Recognize "uahalf" and "uaword" statement, as synonyms for "half" and "word" | miod | 1 | -1/+3 |
| 2012-11-09 | Doh, in gcc3 the various *_ASM_OP need to output explicit whitespace before | miod | 1 | -2/+2 |
| 2012-11-09 | Move the actual operand contraints in define_split constructs from the lisp | miod | 1 | -101/+153 |
| 2012-11-09 | Put prologue and epilogue instructions into their own sequences, bounded by | miod | 1 | -6/+32 |
| 2012-11-09 | (subreg) in gcc 3 takes the subreg position in bytes, not in BYTES_PER_WORD | miod | 1 | -1/+1 |
| 2012-11-09 | Port the gcc 2.95 m88k fixes to the gcc 3 codebase, except for the function | miod | 6 | -86/+157 |
| 2012-11-09 | Allow OBSD_HAS_CORRECT_ASM_OPS to prevent <openbsd.h> from redefining the | miod | 1 | -0/+6 |
| 2012-11-05 | Backport a powerpc --relax fix present in binutils 2.17. |   landry | 2 | -31/+25 |
| 2012-11-04 | Less pessimistic OPT value for m68k. | miod | 1 | -2/+2 |
| 2012-11-02 | Remove mips64 -fno-stack-protector addition; at the moment gcc4 used on mips64 | miod | 1 | -5/+1 |
| 2012-10-26 | Fix for CVE-2012-5195; ok miod@ |   millert | 2 | -0/+4 |
| 2012-10-23 | Define __PIE__ and __pie__ when compiling PIE code, just like newer versions |   kettenis | 1 | -0/+5 |
| 2012-10-18 | Add support for stac/clac instructions to manipulate the flag |   jsg | 8 | -5/+35 |
| 2012-10-13 | fix typo, keyword is exists, not exist |   espie | 1 | -2/+2 |
| 2012-10-06 | Our system headers are supposed to be C++-aware these days so stop GCC from | kettenis | 2 | -3/+4 |
| 2012-10-02 | fix weird depend on -liberty, okay miod@. | espie | 1 | -1/+2 |
| 2012-09-27 | last stage of rfc changes, using consistent Rs/Re blocks, and moving the |   jmc | 1 | -9/+27 |
| 2012-09-16 | Build gcc as non-PIE. PIE breaks precompiled headers and causes a serious | kettenis | 1 | -1/+3 |
| 2012-09-09 | Set use_collect2 on m88k. Fixes the lib/csu/init_priority regress test. | miod | 1 | -0/+1 |
| 2012-09-07 | disassemble the rdrand instruction on i386/amd64 |   naddy | 2 | -8/+16 |
| 2012-09-03 | Teach as(1) about rdrand on i386/amd64. | kettenis | 4 | -2/+10 |
| 2012-09-02 | Provide ios::pos_type and ios::off_type on gcc 2.95 platforms, as some ports | miod | 1 | -0/+2 |
| 2012-09-01 | Bounds check before addressing call_used_regs[], avoids segfaults in | miod | 1 | -1/+1 |
