| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | - Fix atomic operation JIT encoding. | Evan Cheng | 2008-04-18 | 3 | -54/+30 |
| | | | | | | | - Remove unused instructions. llvm-svn: 49921 | ||||
| * | Add some more FIXME's for indexed loads and stores. | Duncan Sands | 2008-04-18 | 3 | -0/+6 |
| | | | | | llvm-svn: 49916 | ||||
| * | Provide an explicit list of operands to MakeLibcall, | Duncan Sands | 2008-04-18 | 3 | -72/+81 |
| | | | | | | | | | | | rather than having it suck them out of a node. Add a bunch of new libcalls, and remove dead softfloat code (dead, because FloatToInt is used not Expand in this case). Note that indexed stores probably aren't handled properly, likewise for loads. llvm-svn: 49915 | ||||
| * | Correct comment. | Evan Cheng | 2008-04-18 | 1 | -2/+2 |
| | | | | | llvm-svn: 49913 | ||||
| * | Not safe to "kill" a register if its live range extends pass the end of ↵ | Evan Cheng | 2008-04-18 | 1 | -3/+29 |
| | | | | | | | block branch. llvm-svn: 49911 | ||||
| * | Also support Intel asm syntax. | Evan Cheng | 2008-04-17 | 1 | -20/+20 |
| | | | | | llvm-svn: 49878 | ||||
| * | Remove the implicit conversion from SDOperandPtr to SDOperand*; this | Dan Gohman | 2008-04-17 | 1 | -4/+4 |
| | | | | | | | may fix a build error on Visual Studio. llvm-svn: 49876 | ||||
| * | Fix assembly code for atomic operations. | Evan Cheng | 2008-04-17 | 1 | -18/+18 |
| | | | | | llvm-svn: 49869 | ||||
| * | Be more careful with insert_subreg and extract_subreg where either source or ↵ | Evan Cheng | 2008-04-17 | 1 | -27/+54 |
| | | | | | | | destination operand has already been coalesced with another register that's defined by a insert_subreg or extract_subreg. llvm-svn: 49843 | ||||
| * | Make GVN able to remove unnecessary calls to read-only functions again. | Owen Anderson | 2008-04-17 | 2 | -18/+31 |
| | | | | | llvm-svn: 49842 | ||||
| * | Use correct name for method in comment. | Bill Wendling | 2008-04-17 | 1 | -4/+4 |
| | | | | | llvm-svn: 49841 | ||||
| * | Correct the SrcValue information in the Expand code for va_copy. | Dan Gohman | 2008-04-17 | 1 | -2/+2 |
| | | | | | llvm-svn: 49839 | ||||
| * | Remove unused variable | Scott Michel | 2008-04-17 | 1 | -1/+0 |
| | | | | | llvm-svn: 49838 | ||||
| * | Fix a sub-register indice propagation bug. | Evan Cheng | 2008-04-17 | 1 | -5/+14 |
| | | | | | llvm-svn: 49832 | ||||
| * | Workaround for PR2207, in which pred_iterator assert gets triggered due to a | Scott Michel | 2008-04-16 | 1 | -2/+9 |
| | | | | | | | wee problem in Xcode 2.[45]/gcc 4.0.1. llvm-svn: 49831 | ||||
| * | Don't forget about sub-register indices when rematting instructions. | Evan Cheng | 2008-04-16 | 1 | -0/+13 |
| | | | | | llvm-svn: 49830 | ||||
| * | Unbreak build on x86-64. | Dale Johannesen | 2008-04-16 | 1 | -1/+1 |
| | | | | | llvm-svn: 49822 | ||||
| * | Correlate stubs with functions in JIT: when emitting a stub, the JIT tells ↵ | Nicolas Geoffray | 2008-04-16 | 12 | -45/+67 |
| | | | | | | | | | the memory manager which function the stub will resolve. llvm-svn: 49814 | ||||
| * | After reading memory that's already freed. | Evan Cheng | 2008-04-16 | 2 | -22/+24 |
| | | | | | llvm-svn: 49810 | ||||
| * | Infrastructure for getting the machine code size of a function and an ↵ | Nicolas Geoffray | 2008-04-16 | 13 | -208/+623 |
| | | | | | | | instruction. X86, PowerPC and ARM are implemented llvm-svn: 49809 | ||||
| * | Initialize X863DNowLevel. | Evan Cheng | 2008-04-16 | 1 | -0/+1 |
| | | | | | llvm-svn: 49808 | ||||
| * | Fix PR2226. Avoid using uninitialized variables. | Evan Cheng | 2008-04-16 | 1 | -13/+28 |
| | | | | | llvm-svn: 49807 | ||||
| * | Empty basic block should have an empty range. | Evan Cheng | 2008-04-16 | 1 | -1/+3 |
| | | | | | llvm-svn: 49800 | ||||
| * | Don't read off end of the input array. | Dale Johannesen | 2008-04-16 | 1 | -1/+1 |
| | | | | | llvm-svn: 49799 | ||||
| * | Ongoing work on improving the instruction selection infrastructure: | Roman Levenstein | 2008-04-16 | 7 | -85/+86 |
| | | | | | | | | | | | Rename SDOperandImpl back to SDOperand. Introduce the SDUse class that represents a use of the SDNode referred by an SDOperand. Now it is more similar to Use/Value classes. Patch is approved by Dan Gohman. llvm-svn: 49795 | ||||
| * | Rewrite LiveVariable liveness computation. The new implementation is much ↵ | Evan Cheng | 2008-04-16 | 1 | -219/+239 |
| | | | | | | | | | simplified. It eliminated the nasty recursive routines and removed the partial def / use bookkeeping. There is also potential for performance improvement by replacing the conservative handling of partial physical register definitions. The code is currently disabled until live interval analysis is taught of the name scheme. This patch also fixed a couple of nasty corner cases. llvm-svn: 49784 | ||||
| * | Code clean up. | Evan Cheng | 2008-04-16 | 1 | -70/+54 |
| | | | | | llvm-svn: 49783 | ||||
| * | Major repairs to the post-dominators implementation. Patch from Florian ↵ | Owen Anderson | 2008-04-16 | 2 | -1/+8 |
| | | | | | | | Brandner! llvm-svn: 49768 | ||||
| * | Add support for the form of the SSE41 extractps instruction that | Dan Gohman | 2008-04-16 | 2 | -5/+6 |
| | | | | | | | puts its result in a 32-bit GPR. llvm-svn: 49762 | ||||
| * | Recreate the size SDNode instead of reusing the old one in the x86 | Dan Gohman | 2008-04-16 | 1 | -2/+2 |
| | | | | | | | | | memcpy lowering code; this ensures that the size node has the desired result type. This fixes a regression from r49572 with @llvm.memcpy.i64 on x86-32. llvm-svn: 49761 | ||||
| * | Avoid read after free. | Evan Cheng | 2008-04-16 | 1 | -1/+1 |
| | | | | | llvm-svn: 49760 | ||||
| * | Remove X86_64SRet; it isn't used anymore. | Dan Gohman | 2008-04-16 | 1 | -11/+0 |
| | | | | | llvm-svn: 49759 | ||||
| * | Add movd instructions to move from MMX registers | Dan Gohman | 2008-04-15 | 1 | -0/+12 |
| | | | | | | | to 64-bit GPR registers on x86-64. llvm-svn: 49757 | ||||
| * | Fix the new scheduler assertion checks to work when | Dan Gohman | 2008-04-15 | 1 | -2/+10 |
| | | | | | | | | the scheduler has inserted no-ops. This fixes the 2006-07-03-schedulers.ll regression on ppc32. llvm-svn: 49747 | ||||
| * | VisualStudio project files updated. #include <algorithm> added to make ↵ | Chuck Rose III | 2008-04-15 | 1 | -0/+15 |
| | | | | | | | VisualStudio happy. Also had to undefine setjmp because of #include <csetjmp> turning setjmp into _setjmp in VisualStudio. llvm-svn: 49743 | ||||
| * | Don't assume a tail call can't reference a byval | Dale Johannesen | 2008-04-15 | 1 | -2/+4 |
| | | | | | | | argument to the outer function, this isn't correct. llvm-svn: 49731 | ||||
| * | Use gv's --spartan option, which trades away an extra row of UI buttons | Dan Gohman | 2008-04-15 | 1 | -0/+1 |
| | | | | | | | for more space for displaying the graph. llvm-svn: 49730 | ||||
| * | Change Divided flag to Split, as suggested by Evan | Nicolas Geoffray | 2008-04-15 | 2 | -4/+4 |
| | | | | | llvm-svn: 49715 | ||||
| * | Treat EntryToken nodes as "passive" so that they aren't added to the | Dan Gohman | 2008-04-15 | 3 | -35/+51 |
| | | | | | | | | | | | | | | | | | | ScheduleDAG; they don't correspond to any actual instructions so they don't need to be scheduled. This fixes a bug where the EntryToken was being scheduled multiple times in some cases, though it ended up not causing any trouble because EntryToken doesn't expand into anything. With this fixed the schedulers reliably schedule the expected number of units, so we can check this with an assertion. This requires a tweak to test/CodeGen/X86/loop-hoist.ll because it ends up getting scheduled differently in a trivial way, though it was enough to fool the prcontext+grep that the test does. llvm-svn: 49701 | ||||
| * | In -view-sunit-dags, display "special" chain dependencies as cyan | Dan Gohman | 2008-04-14 | 1 | -0/+2 |
| | | | | | | | instead of blue to distinguish them from regular dependencies. llvm-svn: 49696 | ||||
| * | Remove unnecessary <sstream> includes. | Dan Gohman | 2008-04-14 | 2 | -2/+0 |
| | | | | | llvm-svn: 49681 | ||||
| * | Avoid creating MERGE_VALUES nodes for single values. | Dan Gohman | 2008-04-14 | 1 | -1/+2 |
| | | | | | llvm-svn: 49676 | ||||
| * | Teach AliasSetTracker about VAArgInst. | Dan Gohman | 2008-04-14 | 1 | -0/+17 |
| | | | | | llvm-svn: 49674 | ||||
| * | Minor whitespace and comment cleanups. | Dan Gohman | 2008-04-14 | 1 | -3/+3 |
| | | | | | llvm-svn: 49671 | ||||
| * | In the special case, call the comparison function instead of | Dan Gohman | 2008-04-14 | 1 | -2/+2 |
| | | | | | | | | | manually performing the comparison. This allows the special case to work correctly even in the case where someone is experimenting with a different comparison function :-). llvm-svn: 49670 | ||||
| * | Fix const-correctness issues with the SrcValue handling in the | Dan Gohman | 2008-04-14 | 5 | -18/+18 |
| | | | | | | | memory intrinsic expansion code. llvm-svn: 49666 | ||||
| * | Reverse sense of unwind-tables option. This means | Dale Johannesen | 2008-04-14 | 4 | -9/+9 |
| | | | | | | | | stack tracebacks on Darwin x86-64 won't work by default; nevertheless, everybody but me thinks this is a good idea. llvm-svn: 49663 | ||||
| * | Revert r49614. As Dan pointed out, some of these aren't correct. | Owen Anderson | 2008-04-14 | 5 | -6/+6 |
| | | | | | llvm-svn: 49657 | ||||
| * | Fix /test/CodeGen/PowerPC/big-endian-actual-args.ll for linux/ppc32 | Nicolas Geoffray | 2008-04-14 | 1 | -4/+4 |
| | | | | | llvm-svn: 49652 | ||||
| * | VAArg may trap. | Dan Gohman | 2008-04-14 | 1 | -0/+1 |
| | | | | | llvm-svn: 49646 | ||||

