| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Add a new option. | Chris Lattner | 2007-05-03 | 2 | -4/+33 |
| | | | | | llvm-svn: 36657 | ||||
| * | Non-algorithmic change. Moved definitions around into separate sections | Bill Wendling | 2007-05-02 | 1 | -1004/+1251 |
| | | | | | | | for SSE1, SSE2, SSE3, and SSSE3. llvm-svn: 36656 | ||||
| * | Properly set arguments bitwidth of EHSELECT node | Anton Korobeynikov | 2007-05-02 | 1 | -2/+2 |
| | | | | | llvm-svn: 36654 | ||||
| * | Update. | Bill Wendling | 2007-05-02 | 1 | -0/+27 |
| | | | | | llvm-svn: 36653 | ||||
| * | Use 'static const char' instead of 'static const int'. | Devang Patel | 2007-05-02 | 150 | -313/+312 |
| | | | | | | | | Due to darwin gcc bug, one version of darwin linker coalesces static const int, which defauts PassID based pass identification. llvm-svn: 36652 | ||||
| * | Disable RTTI handling until we're ready. | Reid Spencer | 2007-05-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 36651 | ||||
| * | Add a rule to get the footprint of binaries and libraries. | Reid Spencer | 2007-05-02 | 1 | -0/+14 |
| | | | | | llvm-svn: 36650 | ||||
| * | Re-install patch to enable use of PassID. | Devang Patel | 2007-05-02 | 3 | -35/+34 |
| | | | | | | | | I am preparing another patch to address the failure that prompted Chris to revert this patch earlier. llvm-svn: 36649 | ||||
| * | Fix build error. | Lauro Ramos Venancio | 2007-05-02 | 8 | -11/+14 |
| | | | | | llvm-svn: 36648 | ||||
| * | Emit correct register move information in eh frames for X86. This allows ↵ | Anton Korobeynikov | 2007-05-02 | 2 | -8/+56 |
| | | | | | | | | | Shootout-C++/except to pass on x86/linux with non-llvm-compiled (e.g. "native") unwind runtime. llvm-svn: 36647 | ||||
| * | Emit correct DWARF reg # for RA (return address) register | Anton Korobeynikov | 2007-05-02 | 2 | -1/+7 |
| | | | | | llvm-svn: 36646 | ||||
| * | Test that TREE_CONSTANT is being set correctly. | Duncan Sands | 2007-05-02 | 2 | -0/+17 |
| | | | | | llvm-svn: 36645 | ||||
| * | Test that inline assembler compiles (llvm-gcc added an extra "uses" parameter | Duncan Sands | 2007-05-02 | 1 | -0/+6 |
| | | | | | | | to ASM_EXPR). llvm-svn: 36644 | ||||
| * | update to reflect reality | Chris Lattner | 2007-05-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 36643 | ||||
| * | add reader logic for terminator instrs. | Chris Lattner | 2007-05-02 | 3 | -10/+93 |
| | | | | | llvm-svn: 36642 | ||||
| * | add reader support for a bunch of new instructions | Chris Lattner | 2007-05-02 | 1 | -14/+90 |
| | | | | | llvm-svn: 36641 | ||||
| * | read a few instructions, fix some bugs. This is enough to be able to round | Chris Lattner | 2007-05-02 | 1 | -15/+38 |
| | | | | | | | | | | | | | | | trip function bodies like this: define <2 x i64> @foo(<2 x i64> %x, <2 x i64> %y) { %tmp4 = bitcast <2 x i64> %y to <8 x i16> ; <<8 x i16>> [#uses=1] %tmp5 = bitcast <2 x i64> %x to <8 x i16> ; <<8 x i16>> [#uses=1] %tmp = add <8 x i16> %tmp5, %tmp4 ; <<8 x i16>> [#uses=1] %tmp6 = bitcast <8 x i16> %tmp to <2 x i64> ; <<2 x i64>> [#uses=1] ret <2 x i64> %tmp6 } llvm-svn: 36640 | ||||
| * | use the correct code for binop instrs | Chris Lattner | 2007-05-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 36639 | ||||
| * | revert enough of devang's recent patches to get the tree basically working again | Chris Lattner | 2007-05-02 | 3 | -35/+36 |
| | | | | | llvm-svn: 36638 | ||||
| * | disable this assertion as a hack to get the build more unbroken :( | Chris Lattner | 2007-05-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 36637 | ||||
| * | Make sign extension work correctly for unusual bit widths. | Reid Spencer | 2007-05-02 | 1 | -28/+100 |
| | | | | | llvm-svn: 36635 | ||||
| * | Add some support for (Darwin) code-generating directives in getInlineAsmLength. | Dale Johannesen | 2007-05-02 | 2 | -18/+115 |
| | | | | | | | Support is incomplete, but more accurate than gcc's. llvm-svn: 36634 | ||||
| * | Fix couple of bugs connected with eh info: | Anton Korobeynikov | 2007-05-01 | 4 | -24/+46 |
| | | | | | | | | 1. Correct output offsets on Linux 2. Fix "style" of personality function. It shouldn't be indirect. llvm-svn: 36633 | ||||
| * | Do not use typeinfo to identify pass in pass manager. | Devang Patel | 2007-05-01 | 154 | -129/+687 |
| | | | | | llvm-svn: 36632 | ||||
| * | Update doc to reflect changes I am about to install to fix PR 888. | Devang Patel | 2007-05-01 | 1 | -0/+19 |
| | | | | | llvm-svn: 36631 | ||||
| * | Doh. PC displacement is between the constantpool and the add instruction. | Evan Cheng | 2007-05-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 36630 | ||||
| * | Test handling of TRY_CATCH_EXPRs for which the handler is a sequence of | Duncan Sands | 2007-05-01 | 1 | -0/+12 |
| | | | | | | | ordinary statements, rather than a list of CATCH_EXPRs or an EH_FILTER_EXPR. llvm-svn: 36629 | ||||
| * | Use correct PC symbol | Anton Korobeynikov | 2007-05-01 | 1 | -0/+2 |
| | | | | | llvm-svn: 36628 | ||||
| * | Adjust correct EH-related sections | Anton Korobeynikov | 2007-05-01 | 1 | -0/+4 |
| | | | | | llvm-svn: 36627 | ||||
| * | eliminateFrameIndex() change. | Evan Cheng | 2007-05-01 | 12 | -14/+25 |
| | | | | | llvm-svn: 36626 | ||||
| * | If call frame is not part of stack frame and no dynamic alloc, ↵ | Evan Cheng | 2007-05-01 | 2 | -18/+49 |
| | | | | | | | eliminateFrameIndex() must adjust SP offset with size of call frames. llvm-svn: 36625 | ||||
| * | Pass call frame setup SP adjustment along to eliminateFrameIndex(). | Evan Cheng | 2007-05-01 | 1 | -5/+6 |
| | | | | | llvm-svn: 36624 | ||||
| * | Add SPAdj parameter to account for call frame setup SP adjustment. | Evan Cheng | 2007-05-01 | 1 | -3/+4 |
| | | | | | llvm-svn: 36623 | ||||
| * | Forgot about chain result; also UNDEF cannot have multiple values. | Evan Cheng | 2007-05-01 | 1 | -12/+12 |
| | | | | | llvm-svn: 36622 | ||||
| * | fix build with non-buggy compilers | Chris Lattner | 2007-05-01 | 1 | -10/+11 |
| | | | | | llvm-svn: 36621 | ||||
| * | handle function-level forward references, read binops. | Chris Lattner | 2007-05-01 | 2 | -20/+145 |
| | | | | | llvm-svn: 36620 | ||||
| * | A bit of feedback from Chris that I missed; error rather than asserting. | Nate Begeman | 2007-05-01 | 1 | -2/+5 |
| | | | | | llvm-svn: 36619 | ||||
| * | llvm bug #1350, parts 1, 2, and 3. | Nate Begeman | 2007-05-01 | 6 | -27/+133 |
| | | | | | llvm-svn: 36618 | ||||
| * | implement materializeModule, force deallocation of vector memory when we | Chris Lattner | 2007-05-01 | 2 | -15/+107 |
| | | | | | | | are done with them, start implementing ParseFunctionBody llvm-svn: 36617 | ||||
| * | several bitfixes to JumpToBit | Chris Lattner | 2007-05-01 | 1 | -4/+7 |
| | | | | | llvm-svn: 36616 | ||||
| * | The stream to read from is now an ivar | Chris Lattner | 2007-05-01 | 2 | -20/+18 |
| | | | | | llvm-svn: 36615 | ||||
| * | implement scafolding for lazy deserialization of function bodies | Chris Lattner | 2007-05-01 | 2 | -7/+85 |
| | | | | | llvm-svn: 36614 | ||||
| * | add JumpToBit, an explicit init method, and a default ctor. | Chris Lattner | 2007-05-01 | 1 | -2/+28 |
| | | | | | llvm-svn: 36613 | ||||
| * | Split target dependent test portions to target-specific directories. | Reid Spencer | 2007-05-01 | 3 | -4/+314 |
| | | | | | llvm-svn: 36612 | ||||
| * | add phi | Chris Lattner | 2007-05-01 | 1 | -1/+5 |
| | | | | | llvm-svn: 36611 | ||||
| * | write the symbol table for function bodies | Chris Lattner | 2007-05-01 | 1 | -30/+33 |
| | | | | | llvm-svn: 36610 | ||||
| * | encode all of the instructions. | Chris Lattner | 2007-05-01 | 1 | -3/+166 |
| | | | | | llvm-svn: 36609 | ||||
| * | add missing opcode. | Chris Lattner | 2007-05-01 | 1 | -9/+10 |
| | | | | | llvm-svn: 36608 | ||||
| * | Under normal circumstances, when a frame pointer is not required, we reserve | Evan Cheng | 2007-05-01 | 3 | -20/+43 |
| | | | | | | | | | | | | | argument space for call sites in the function immediately on entry to the current function. This eliminates the need for add/sub sp brackets around call sites. However, this is not always a good idea. If the "call frame" is large and the target load / store instructions have small immediate field to encode sp offset, this can cause poor codegen. In the worst case, this can make it impossible to scavenge a register if the reserved spill slot is pushed too far apart from sp / fp. llvm-svn: 36607 | ||||
| * | Added hook hasReservedCallFrame(). It returns true if the call frame is | Evan Cheng | 2007-05-01 | 1 | -0/+9 |
| | | | | | | | included as part of the stack frame. llvm-svn: 36606 | ||||

