| Commit message (Expand) | Author | Age | Files | Lines |
* | Fix tests. | Evan Cheng | 2009-08-15 | 2 | -2/+2 |
* | Turn on if-conversion for thumb2. | Evan Cheng | 2009-08-15 | 3 | -0/+189 |
* | Leaf functions which do not save CSRs can be frameless even with -disable-fp-... | Evan Cheng | 2009-08-14 | 1 | -0/+6 |
* | Also shrink immediate branches; also more assembler workarounds. | Evan Cheng | 2009-08-14 | 2 | -6/+77 |
* | Shrink ADR and LDR from constantpool late during constantpool island pass. | Evan Cheng | 2009-08-14 | 1 | -1/+1 |
* | Remove an Darwin assembler workaround. | Evan Cheng | 2009-08-12 | 1 | -1/+1 |
* | Shrink ADDS, ADC, RSB, and SUBS. | Evan Cheng | 2009-08-12 | 9 | -20/+61 |
* | Shrinkify Thumb2 r = add sp, imm. | Evan Cheng | 2009-08-11 | 1 | -2/+4 |
* | Shrinkify Thumb2 load / store multiple instructions. | Evan Cheng | 2009-08-11 | 1 | -7/+7 |
* | Fix the previous accidental commit. Now shrinking common Thumb2 load / store ... | Evan Cheng | 2009-08-11 | 7 | -34/+73 |
* | Enable Thumb2 instruction shrinking (32-bit to 16-bit) pass. Convert a bunch ... | Evan Cheng | 2009-08-10 | 26 | -80/+240 |
* | Handle the constantfp created during post-legalization dag combiner phase. | Evan Cheng | 2009-08-10 | 1 | -0/+15 |
* | Simplify RegScavenger::forward a bit more. | Jakob Stoklund Olesen | 2009-08-08 | 1 | -0/+20 |
* | Thumb2 32-bit ldm / stm needs .w suffix if submode is ia. | Evan Cheng | 2009-08-07 | 1 | -4/+4 |
* | Fix support to use NEON for single precision fp math. | Evan Cheng | 2009-08-07 | 1 | -0/+80 |
* | Another coalescer bug. When a dead copy is eliminated, transfer the kill to a... | Evan Cheng | 2009-08-07 | 1 | -0/+16 |
* | It turns out most of the thumb2 instructions are not allowed to touch SP. The... | Evan Cheng | 2009-08-07 | 2 | -3/+27 |
* | Disable stack coloring with register for now. It's not able to set kill markers. | Evan Cheng | 2009-08-05 | 1 | -0/+508 |
* | Another nasty coalescer bug (is there another kind): | Evan Cheng | 2009-08-05 | 1 | -0/+153 |
* | One more. Transfer kill of the larger register when lowering an EXTRACT_SUBREG. | Evan Cheng | 2009-08-05 | 1 | -0/+54 |
* | One more place where subreg lowering forgot to transfer undefness. | Evan Cheng | 2009-08-05 | 1 | -0/+42 |
* | If the insert_subreg source is <undef>, insert an implicit_def instead of a c... | Evan Cheng | 2009-08-05 | 1 | -0/+34 |
* | Fix part 1 of pr4682. PICADD is a 16-bit instruction even in thumb2 mode. | Evan Cheng | 2009-08-04 | 2 | -2/+2 |
* | Fix test. | Evan Cheng | 2009-08-04 | 1 | -1/+1 |
* | Enable load / store multiple pass for Thumb2. It's not using ldrd / strd yet. | Evan Cheng | 2009-08-04 | 1 | -0/+40 |
* | In thumb mode, r7 is used as frame register. This fixes pr4681. | Evan Cheng | 2009-08-04 | 1 | -0/+29 |
* | Emit sub r, #c instead of transforming it to add r, #-c if c fits in 8-bit. T... | Evan Cheng | 2009-08-04 | 1 | -1/+19 |
* | Fix a coaelescer bug. If a copy val# is extended to eliminate a non-trivially... | Evan Cheng | 2009-08-03 | 1 | -0/+46 |
* | Use the i12 variant of load / store opcodes if offset is zero. Now we pass al... | Evan Cheng | 2009-08-03 | 1 | -0/+85 |
* | Test both darwin and linux. | Evan Cheng | 2009-08-02 | 1 | -5/+12 |
* | Hack to make this test work on platforms which aren't Macs. Fixing this | Eli Friedman | 2009-08-01 | 1 | -1/+1 |
* | Workaround a couple of Darwin assembler bugs. | Evan Cheng | 2009-08-01 | 1 | -1/+1 |
* | Split t2MOVCCs since some assemblers do not recognize mov shifted register al... | Evan Cheng | 2009-08-01 | 1 | -0/+35 |
* | Fix Thumb2 function call isel. Thumb1 and Thumb2 should share the same | Evan Cheng | 2009-08-01 | 2 | -1/+21 |
* | Until we have a "ALIGN" pseudo instruction, have asm printer emitted a .align | Evan Cheng | 2009-07-31 | 1 | -0/+1 |
* | When fp is not eliminated, instructions with T2_i12 modes will be changed to ... | Evan Cheng | 2009-07-30 | 1 | -0/+193 |
* | Darwin assembler now recognizes "orn", so remove workaround. | David Goodwin | 2009-07-30 | 2 | -2/+0 |
* | Darwin assembler now supports "rrx", so remove workaround. | David Goodwin | 2009-07-30 | 1 | -2/+2 |
* | Cleanup and include code selection for some frame index cases. | David Goodwin | 2009-07-30 | 1 | -1/+1 |
* | tbb / tbh instructions only branch forward, not backwards. | Evan Cheng | 2009-07-29 | 2 | -0/+124 |
* | Optimize Thumb2 jumptable to use tbb / tbh when all the offsets fit in byte /... | Evan Cheng | 2009-07-29 | 2 | -2/+87 |
* | In thumb2 mode, add pc is unpredictable. Use add + mov pc instead (that is un... | Evan Cheng | 2009-07-28 | 1 | -1/+1 |
* | Remove support for ORN to workaround <rdar://problem/7096522>. | David Goodwin | 2009-07-28 | 2 | -0/+2 |
* | Add workaround for <rdar://problem/7098328>. | David Goodwin | 2009-07-28 | 1 | -2/+2 |
* | Add Thumb-2 patterns for ARMsrl_flag and ARMsra_flag. | David Goodwin | 2009-07-28 | 1 | -0/+19 |
* | - More refactoring. This gets rid of all of the getOpcode calls. | Evan Cheng | 2009-07-28 | 1 | -0/+28 |
* | ORN does not require (and can not have) the ".w" suffix. "Orthogonality" is a... | David Goodwin | 2009-07-27 | 1 | -5/+5 |
* | Add ".w" suffix for wide thumb-2 instructions. | David Goodwin | 2009-07-27 | 47 | -121/+124 |
* | Change Thumb2 jumptable codegen to one that uses two level jumps: | Evan Cheng | 2009-07-25 | 1 | -0/+3 |
* | Remove a duplicated test. | Evan Cheng | 2009-07-25 | 1 | -26/+0 |