summaryrefslogtreecommitdiffstats
path: root/llvm/lib/CodeGen
Commit message (Collapse)AuthorAgeFilesLines
* Revert r128045 and r128051, debug info enhancements.Andrew Trick2011-03-223-77/+49
| | | | | | Temporarily reverting these to see if we can get llvm-objdump to link. Hopefully this is not the problem. llvm-svn: 128097
* Clear map after use.Jakob Stoklund Olesen2011-03-221-0/+1
| | | | | | This is likely to fix the segfault in llvm-gcc-x86_64-darwin10-cross-mingw32. llvm-svn: 128051
* Dont emit 'DBG_VALUE %noreg, ...' to terminate user variable ranges.Jakob Stoklund Olesen2011-03-223-49/+76
| | | | | | | | | | These ranges get completely jumbled by the post-ra scheduler, and it is not really reasonable to expect it to make sense of them. Instead, teach DwarfDebug to notice when user variables in registers are clobbered, and terminate the ranges there. llvm-svn: 128045
* Grammar-o.Eric Christopher2011-03-211-1/+1
| | | | llvm-svn: 128004
* We need to pass the TargetMachine object to the InstPrinter if we are printingBill Wendling2011-03-211-1/+1
| | | | | | | | | the alias of an InstAlias instead of the thing being aliased. Because we need to know the features that are valid for an InstAlias. This is part of a work-in-progress. llvm-svn: 127986
* Process all dead defs after rematerializing during splitting.Jakob Stoklund Olesen2011-03-201-22/+15
| | | | llvm-svn: 127973
* Also eliminate redundant spills downstream of inserted reloads.Jakob Stoklund Olesen2011-03-201-9/+17
| | | | | | | This can happen when multiple sibling registers are spilled after live range splitting. llvm-svn: 127965
* Change an argument to a LiveInterval instead of a register number to save ↵Jakob Stoklund Olesen2011-03-201-13/+13
| | | | | | some redundant lookups. llvm-svn: 127964
* Replace a broken LiveInterval::MergeValueInAsValue() with something simpler.Jakob Stoklund Olesen2011-03-191-46/+5
| | | | llvm-svn: 127960
* Add debug output.Jakob Stoklund Olesen2011-03-192-0/+5
| | | | llvm-svn: 127959
* Minor code re-structuring.Evan Cheng2011-03-191-1/+1
| | | | llvm-svn: 127952
* Add support for legalizing UINT_TO_FP of vectors on platforms which doNadav Rotem2011-03-191-1/+48
| | | | | | | | not have native support for this operation (such as X86). The legalized code uses two vector INT_TO_FP operations and is faster than scalarizing. llvm-svn: 127951
* Reapply 127939 since Daniel fixed the breakage. <rdar://problem/9012638>Stuart Hastings2011-03-191-7/+4
| | | | llvm-svn: 127944
* Revert 127939. <rdar://problem/9012638>Stuart Hastings2011-03-191-4/+7
| | | | llvm-svn: 127943
* Revise r126127 to address Daniel's comments. <rdar://problem/9012638>Stuart Hastings2011-03-191-7/+4
| | | | llvm-svn: 127939
* Beginnings of MC-JIT code generation.Jim Grosbach2011-03-181-0/+26
| | | | | | | | | | | Proof-of-concept code that code-gens a module to an in-memory MachO object. This will be hooked up to a run-time dynamic linker library (see: llvm-rtdyld for similarly conceptual work for that part) which will take the compiled object and link it together with the rest of the system, providing back to the JIT a table of available symbols which will be used to respond to the getPointerTo*() queries. llvm-svn: 127916
* Extend live debug values down the dominator tree by following copies.Jakob Stoklund Olesen2011-03-181-24/+136
| | | | | | | | | | | | | | | | | The llvm.dbg.value intrinsic refers to SSA values, not virtual registers, so we should be able to extend the range of a value by tracking that value through register copies. This greatly improves the debug value tracking for function arguments that for some reason are copied to a second virtual register at the end of the entry block. We only extend the debug value range where its register is killed. All original llvm.dbg.value locations are still respected. Copies from physical registers are ignored. That should not be a problem since the entry block already adds DBG_VALUE instructions for the virtual registers holding the function arguments. llvm-svn: 127912
* Hoist spills when the same value is known to be in less loopy sibling registers.Jakob Stoklund Olesen2011-03-181-22/+148
| | | | | | | | | | | | | | Stack slot real estate is virtually free compared to registers, so it is advantageous to spill earlier even though the same value is now kept in both a register and a stack slot. Also eliminate redundant spills by extending the stack slot live range underneath reloaded registers. This can trigger a dead code elimination, removing copies and even reloads that were only feeding spills. llvm-svn: 127868
* Accept instructions that read undefined values.Jakob Stoklund Olesen2011-03-181-1/+9
| | | | | | | This is not supposed to happen, but I have seen the x86 rematter getting confused when rematerializing partial redefs. llvm-svn: 127857
* Be more accurate about the slot index reading a register when dealing with defsJakob Stoklund Olesen2011-03-181-4/+20
| | | | | | | | and early clobbers. Assert when trying to find an undefined value. llvm-svn: 127856
* BuildUDIV: If the divisor is even we can simplify the fixup of the ↵Benjamin Kramer2011-03-171-9/+22
| | | | | | | | | | | | | | | | | | | | | | | multiplied value by introducing an early shift. This allows us to compile "unsigned foo(unsigned x) { return x/28; }" into shrl $2, %edi imulq $613566757, %rdi, %rax shrq $32, %rax ret instead of movl %edi, %eax imulq $613566757, %rax, %rcx shrq $32, %rcx subl %ecx, %eax shrl %eax addl %ecx, %eax shrl $4, %eax on x86_64 llvm-svn: 127829
* Dead code elimination may separate the live interval into multiple connected ↵Jakob Stoklund Olesen2011-03-175-22/+44
| | | | | | | | | components. I have convinced myself that it can only happen when a phi value dies. When it happens, allocate new virtual registers for the components. llvm-svn: 127827
* Move more logic into getTypeForExtArgOrReturn.Cameron Zwarich2011-03-171-6/+2
| | | | llvm-svn: 127809
* Rename getTypeForExtendedInteger() to getTypeForExtArgOrReturn().Cameron Zwarich2011-03-171-1/+1
| | | | llvm-svn: 127807
* Rewrite instructions as part of ConnectedVNInfoEqClasses::Distribute.Jakob Stoklund Olesen2011-03-173-47/+34
| | | | llvm-svn: 127779
* Add a LiveRangeEdit delegate callback before shrinking a live range.Jakob Stoklund Olesen2011-03-163-1/+19
| | | | | | The register allocator needs to adjust its live interval unions when that happens. llvm-svn: 127774
* Erase virtual registers that are unused after DCE.Jakob Stoklund Olesen2011-03-161-4/+12
| | | | llvm-svn: 127773
* Tag cached interference with a user-provided tag instead of the virtual ↵Jakob Stoklund Olesen2011-03-163-7/+16
| | | | | | | | | register number. The live range of a virtual register may change which invalidates the cached interference information. llvm-svn: 127772
* Clarify debugging output.Jakob Stoklund Olesen2011-03-163-7/+17
| | | | llvm-svn: 127771
* The x86-64 ABI says that a bool is only guaranteed to be sign-extended to a byteCameron Zwarich2011-03-161-5/+2
| | | | | | | | | | | rather than an int. Thankfully, this only causes LLVM to miss optimizations, not generate incorrect code. This just fixes the zext at the return. We still insert an i32 ZextAssert when reading a function's arguments, but it is followed by a truncate and another i8 ZextAssert so it is not optimized. llvm-svn: 127766
* Don't recompute something that we already have in a local variable.Cameron Zwarich2011-03-161-2/+2
| | | | llvm-svn: 127764
* Revert r127757, "Patch to a fix dwarf relocation problem on ARM. One-line fixDaniel Dunbar2011-03-161-5/+2
| | | | | | | plus the test where it used to break.", which broke Clang self-host of a Debug+Asserts compiler, on OS X. llvm-svn: 127763
* Patch to a fix dwarf relocation problem on ARM. One-line fix plus the test ↵Renato Golin2011-03-161-2/+5
| | | | | | where it used to break. llvm-svn: 127757
* Trace back through sibling copies to hoist spills and find rematerializable ↵Jakob Stoklund Olesen2011-03-151-10/+208
| | | | | | | | | | | | | defs. After live range splitting, an original value may be available in multiple registers. Tracing back through the registers containing the same value, find the best place to insert a spill, determine if the value has already been spilled, or discover a reaching def that may be rematerialized. This is only the analysis part. The information is not used for anything yet. llvm-svn: 127698
* Preserve both isPHIDef and isDefByCopy bits when copying parent values.Jakob Stoklund Olesen2011-03-151-5/+4
| | | | llvm-svn: 127697
* Add a peephole optimization to optimize pairs of bitcasts. e.g.Evan Cheng2011-03-151-4/+97
| | | | | | | | | | | | | | | | | | | | | | v2 = bitcast v1 ... v3 = bitcast v2 ... = v3 => v2 = bitcast v1 ... = v1 if v1 and v3 are of in the same register class. bitcast between i32 and fp (and others) are often not nops since they are in different register classes. These bitcast instructions are often left because they are in different basic blocks and cannot be eliminated by dag combine. rdar://9104514 llvm-svn: 127668
* sext(undef) = 0, because the top bits will all be the same.Evan Cheng2011-03-151-1/+5
| | | | | | zext(undef) = 0, because the top bits will be zero. llvm-svn: 127649
* There are some situations which can cause the URoR hack to infinitely recurseBill Wendling2011-03-151-6/+8
| | | | | | | | | | | | | and then go kablooie. The problem was that it was tracking the PHI nodes anew each time into this function. But it didn't need to. And because the recursion didn't know that a PHINode was visited before, it would go ahead and call itself. There is a testcase, but unfortunately it's too big to add. This problem will go away with the EH rewrite. <rdar://problem/8856298> llvm-svn: 127640
* Place context in member variables instead of passing around pointers.Jakob Stoklund Olesen2011-03-141-21/+18
| | | | | | Use the opportunity to get rid of the trailing underscore variable names. llvm-svn: 127618
* Rename members to match LLVM naming conventions more closely.Jakob Stoklund Olesen2011-03-141-111/+109
| | | | | | | | Remove the unused reserved_ bit vector, no functional change intended. This doesn't break 'svn blame', this file really is all my fault. llvm-svn: 127607
* BIT_CONVERT has been renamed to BITCAST.Evan Cheng2011-03-141-1/+1
| | | | llvm-svn: 127600
* Minor optimization. sign-ext/anyext of undef is still undef.Evan Cheng2011-03-141-0/+4
| | | | llvm-svn: 127598
* Now that we are deleting unused live intervals during allocation, pointers ↵Jakob Stoklund Olesen2011-03-131-2/+4
| | | | | | | | may be reused. Use the virtual register number as a cache tag instead. They are not reused. llvm-svn: 127561
* Tell the register allocator about new unused virtual registers.Jakob Stoklund Olesen2011-03-134-1/+26
| | | | | | | This allows the allocator to free any resources used by the virtual register, including physical register assignments. llvm-svn: 127560
* Speculatively revert commit 127478 (jsjodin) in an attempt to fix theDuncan Sands2011-03-121-0/+7
| | | | | | | | | | llvm-gcc-i386-linux-selfhost and llvm-x86_64-linux-checks buildbots. The original log entry: Remove optimization emitting a reference insted of label difference, since it can create more relocations. Removed isBaseAddressKnownZero method, because it is no longer used. llvm-svn: 127540
* Include snippets in the live stack interval.Jakob Stoklund Olesen2011-03-121-1/+3
| | | | llvm-svn: 127530
* Spill multiple registers at once.Jakob Stoklund Olesen2011-03-122-48/+210
| | | | | | | | | | | Live range splitting can create a number of small live ranges containing only a single real use. Spill these small live ranges along with the large range they are connected to with copies. This enables memory operand folding and maximizes the spill to fill distance. Work in progress with known bugs. llvm-svn: 127529
* That's it, I am declaring this a failure of the C++03 STL.Jakob Stoklund Olesen2011-03-121-119/+15
| | | | | | | | | | | | | | There are too many compatibility problems with using mixed types in std::upper_bound, and I don't want to spend 110 lines of boilerplate setting up a call to a 10-line function. Binary search is not /that/ hard to implement correctly. I tried terminating the binary search with a linear search, but that actually made the algorithm slower against my expectation. Most live intervals have less than 4 segments. The early test against endIndex() does pay, and this version is 25% faster than plain std::upper_bound(). llvm-svn: 127522
* Fix the GCC test suite issue exposed by r127477, which was caused by stackCameron Zwarich2011-03-111-3/+3
| | | | | | | protector insertion not working correctly with unreachable code. Since that revision was rolled out, this test doesn't actual fail before this fix. llvm-svn: 127497
* Teach FastISel to support register-immediate-immediate instructions.Owen Anderson2011-03-111-0/+23
| | | | llvm-svn: 127496
OpenPOWER on IntegriCloud