aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2010-11-20On X86, MEMBARRIER, MFENCE, SFENCE, LFENCE are not target memory intrinsics,Duncan Sands
so don't claim they are. They are allocated using DAG.getNode, so attempts to access MemSDNode fields results in reading off the end of the allocated memory. This fixes crashes with "llc -debug" due to debug code trying to print MemSDNode fields for these barrier nodes (since the crashes are not deterministic, use valgrind to see this). Add some nasty checking to try to catch this kind of thing in the future. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119901 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-20Removing the useless test that I added recently. It was meant as an example, ↵Andrew Trick
but not complicated enough to merit another test. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119898 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-20RABasic fix. Regalloc is responsible for updating block live ins.Andrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119896 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-20Whitespace.Andrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119895 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-20Check for _setjmp too, because it's also used.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119875 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-19Make isScalarToVector to return false if the node is a scalar. This will preventMon P Wang
DAGCombine from making an illegal transformation of bitcast of a scalar to a vector into a scalar_to_vector. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119819 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-19Don't attempt trivial coalescing for sub-register copies.Jakob Stoklund Olesen
Patch by Krister Wombell! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119791 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-19Change some methods in MCDwarf.cpp to be able to handle an arbitraryRafael Espindola
MCStreamer instead of just MCObjectStreamer. Address changes cannot be as efficient as we have to use DW_LNE_set_addres, but at least most of the logic is shared. This will be used so that, with CodeGen still using EmitDwarfLocDirective, llvm-gcc is able to produce debug_line sections without needing an assembler that supports .loc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119777 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Move getInitialFrameState() to TargetFrameInfoAnton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119754 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Move hasFP() and few related hooks to TargetFrameInfo.Anton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119740 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Fix thinko: we must turn select(anyext, sext) into sext(select)Duncan Sands
not anyext(select). Spotted by Frits van Bommel. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119739 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18The DAGCombiner was threading select over pairs of extending loads evenDuncan Sands
if the extension types were not the same. The result was that if you fed a select with sext and zext loads, as in the testcase, then it would get turned into a zext (or sext) of the select, which is wrong in the cases when it should have been an sext (resp. zext). Reported and diagnosed by Sebastien Deldon. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119728 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Oops, missed this file when remaing ExpandPseudos to ExpandISelPseudos.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119717 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Rename ExpandPseudos to ExpandISelPseudos to help clarify its role.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119716 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Fix typos.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119712 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18ExpandPseudos doesn't have any dependencies, so it can use theDan Gohman
simple form of INITIALIZE_PASS. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119707 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Change CodeGen to use .loc directives. This produces a lot more readable outputRafael Espindola
and testing is easier. A good example is the unknown-location.ll test that now can just look for ".loc 1 0 0". We also don't use a DW_LNE_set_address for every address change anymore. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119613 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-18Do not throw away alignment when generating the DAG forDale Johannesen
memset; we may need it to decide between MOVAPS and MOVUPS later. Adjust a test that was looking for wrong code. PR 3866 / 8675131. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119605 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17Bug 8621 fix - pointer cast stripped from inline asm constraint argument.John Thompson
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119590 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17Remove ARM isel hacks that fold large immediates into a pair of add, sub, and,Evan Cheng
and xor. The 32-bit move immediates can be hoisted out of loops by machine LICM but the isel hacks were preventing them. Instead, let peephole optimization pass recognize registers that are defined by immediates and the ARM target hook will fold the immediates in. Other changes include 1) do not fold and / xor into cmp to isel TST / TEQ instructions if there are multiple uses. This happens when the 'and' is live out, machine sink would have sinked the computation and that ends up pessimizing code. The peephole pass would recognize situations where the 'and' can be toggled to define CPSR and eliminate the comparison anyway. 2) Move peephole pass to after machine LICM, sink, and CSE to avoid blocking important optimizations. rdar://8663787, rdar://8241368 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119548 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17With the newly simplified SourceMgr interfaces and the generalizedChris Lattner
SrcMgrDiagHandler, we can improve clang diagnostics for inline asm: instead of reporting them on a source line of the original line, we can report it on the correct line wherever the string literal came from. For something like this: void foo() { asm("push %rax\n" ".code32\n"); } we used to get this: (note that the line in t.c isn't helpful) t.c:4:7: error: warning: ignoring directive for now asm("push %rax\n" ^ <inline asm>:2:1: note: instantiated into assembly here .code32 ^ now we get: t.c:5:8: error: warning: ignoring directive for now ".code32\n" ^ <inline asm>:2:1: note: instantiated into assembly here .code32 ^ Note that we're pointing to line 5 properly now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119488 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17now that AsmPrinter::EmitInlineAsm is factored right, we can eliminate theChris Lattner
cookie argument to the SourceMgr diagnostic stuff. This cleanly separates LLVMContext's inlineasm handler from the sourcemgr error handling definition, increasing type safety and cleaning things up. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119486 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17rearrange how the handler in SourceMgr is installed, eliminating the use of Chris Lattner
the cookie argument to setDiagHandler git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119483 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17refactor the interface to EmitInlineAsm a bit, no functionality change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119482 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-17Only avoid the check if we're the last operand before the variableEric Christopher
operands in a variadic instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119446 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Fix grammaro.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119386 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Add ExpandPseudos.cpp.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119385 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Split pseudo-instruction expansion into a separate pass, to make itDan Gohman
easier to debug, and to avoid complications when the CFG changes in the middle of the instruction selection process. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119382 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Fix emergency spilling in LiveIntervals::spillPhysRegAroundRegDefsUses.Jakob Stoklund Olesen
Always spill the full representative register at any point where any subregister is live. This fixes PR8620 which caused the old logic to get confused and not spill anything at all. The fundamental problem here is that the coalescer is too aggressive about physical register coalescing. It sometimes makes it impossible to allocate registers without these emergency spills. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119375 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Print out the register class of the current interval.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119374 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Make the verifier a little quieter on instructions that it's probablyEric Christopher
(and likely) wrong about anyhow. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119320 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Remember to resize SpillSlotToUsesMap when allocating an emergency spill slot.Jakob Stoklund Olesen
Use amazing new function call technology instead of writing identical code in multiple places. This fixes PR8604. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119306 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-16Fix PR8612 in the standard spiller, take two.Jakob Stoklund Olesen
The live range of a register defined by an early clobber starts at the use slot, not the def slot. Except when it is an early clobber tied to a use operand. Then it starts at the def slot like a standard def. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119305 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15Revert "Fix PR8612 in the standard spiller as well."Jakob Stoklund Olesen
This reverts r119183 which borke the buildbots. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119270 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15Code clean up. The peephole pass should be the one updating the instructionEvan Cheng
iterator, not TII->OptimizeCompareInstr. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119186 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15Fix PR8612 in the standard spiller as well.Jakob Stoklund Olesen
The live range of a register defined by an early clobber starts at the use slot, not the def slot. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119183 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15When spilling a register defined by an early clobber, make sure that the newJakob Stoklund Olesen
live ranges for the spill register are also defined at the use slot instead of the normal def slot. This fixes PR8612 for the inline spiller. A use was being allocated to the same register as a spilled early clobber def. This problem exists in all the spillers. A fix for the standard spiller is forthcoming. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119182 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-15First step of huge frame-related refactoring: move emit{Prologue,Epilogue} ↵Anton Korobeynikov
out of TargetRegisterInfo to TargetFrameInfo, which is definitely much better suitable place git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119097 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-14move the pic base symbol stuff up to MachineFunctionChris Lattner
since it is trivial and will be shared between ppc and x86. This substantially simplifies the X86 backend also. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@119089 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-12Remove DW_AT_start_scope support. It is incomplete and superseeded by ↵Devang Patel
location entries support. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118940 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-12typo (4th checkin for one fix)Andrew Trick
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118913 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-12Fixes PR8287: SD scheduling time. The fix is a failsafe that preventsAndrew Trick
catastrophic compilation time in the event of unreasonable LLVM IR. Code quality is a separate issue--someone upstream needs to do a better job of reducing to llvm.memcpy. If the situation can be reproduced with any supported frontend, then it will be a separate bug. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118904 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-12tidy up.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118896 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-12Fix some style issues in PBQP. Patch by David Blaikie.Lang Hames
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118883 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-11Add a FIXME comment.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118803 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-11Check TRI->getReservedRegs because other allocators do it. Even thoughAndrew Trick
it makes no sense for allocation_order iterators to visit reserved regs. The inline spiller depends on AliasAnalysis. Manage the Query state to avoid uninitialized or stale results. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118800 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-11Remove the memmove->memcpy optimization from CodeGen. MemCpyOpt does this.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118789 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-11Delete SplittingSpiller. It was not being used by anyone, and it is beingJakob Stoklund Olesen
superceded by SplitKit. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118754 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-11Insert two blank SlotIndexes between basic blocks instead of just one.Jakob Stoklund Olesen
This is the first small step towards using closed intervals for liveness instead of the half-open intervals we're using now. We want to be able to distinguish between a SlotIndex that represents a variable being live-out of a basic block, and an index representing a variable live-in to its successor. That requires two separate indexes between blocks. One for live-outs and one for live-ins. With this change, getMBBEndIdx(MBB).getPrevSlot() becomes stable so it stays greater than any instructions inserted at the end of MBB. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118747 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-10No need to add liveness that's already there.Jakob Stoklund Olesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118742 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-05-01 11:23:54 +0000