aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/CGExpr.cpp
AgeCommit message (Collapse)Author
2012-05-09update calls to objectsize intrinsic to match LLVM r156473Nuno Lopes
add a test for -fbounds-checking code generation git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@156474 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-08add -fbounds-checking option.Nuno Lopes
When enabled, clang generates bounds checks for array and pointers dereferences. Work to follow in LLVM's backend. OK'ed by Chad; thanks for the review. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@156431 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-07remove code to add bound checks for simple array accesses, since those are ↵Nuno Lopes
already covered by the check with the objectsize builtin remove the comparison of objectsize with -1. since it's an unsigned comparison, it will always succeed if objectsize returns -1, which is enough to have the check removed git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@156311 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-01Push variable declaration into nested scope (the only place where it is ↵Ted Kremenek
used). Found by static analyzer. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@155922 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-01Abstract the emission of global destructors into ABI-specific codeJohn McCall
and only consider using __cxa_atexit in the Itanium logic. The default logic is to use atexit(). Emit "guarded" initializers in Microsoft mode unconditionally. This is definitely not correct, but it's closer to correct than just not emitting the initializer. Based on a patch by Timur Iskhodzhanov! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@155894 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-16Adjust for LLVM name tweaks requested by Chandler.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154824 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-16Propagate alignment on lvalues through EmitLValueForField. PR12395.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154789 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-15Use MDBuilder to help with metadata creation.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154767 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-14Rename "fpaccuracy" metadata to the more generic "fpmath". That's because I'mDuncan Sands
thinking of generalizing it to be able to specify other freedoms beyond accuracy (such as that NaN's don't have to be respected). I'd like the 3.1 release (the first one with this metadata) to have the more generic name already rather than having to auto-upgrade it in 3.2. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154745 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-13Step forward with supporting of ARM homogenous aggregates:Anton Korobeynikov
- Handle unions - Handle C++ classes git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154664 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-13Implement __atomic_fetch_nand and __atomic_nand_fetch to complete our set ofRichard Smith
GNU __atomic builtins. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154659 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-13Implement the missing pieces needed to support libstdc++4.7's <atomic>:Richard Smith
__atomic_test_and_set, __atomic_clear, plus a pile of undocumented __GCC_* predefined macros. Implement library fallback for __atomic_is_lock_free and __c11_atomic_is_lock_free, and implement __atomic_always_lock_free. Contrary to their documentation, GCC's __atomic_fetch_add family don't multiply the operand by sizeof(T) when operating on a pointer type. libstdc++ relies on this quirk. Remove this handling for all but the __c11_atomic_fetch_add and __c11_atomic_fetch_sub builtins. Contrary to their documentation, __atomic_test_and_set and __atomic_clear take a first argument of type 'volatile void *', not 'void *' or 'bool *', and __atomic_is_lock_free and __atomic_always_lock_free have an argument of type 'const volatile void *', not 'void *'. With this change, libstdc++4.7's <atomic> passes libc++'s atomic test suite, except for a couple of libstdc++ bugs and some cases where libc++'s test suite tests for properties which implementations have latitude to vary. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154640 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-12Fix some i1/i8 confusion within _Atomic(bool) in IR generation, bothDouglas Gregor
in general (such an atomic has boolean representation) and specifically for IR generation of __c11_atomic_init. The latter also means actually using initialization semantics for this initialization, rather than just creating a store. On a related note, make sure we actually put in non-atomic-to-atomic conversions when performing an implicit conversion sequence. IR generation is far too kind here, but we still want the ASTs to make sense. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154612 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-12Implement support for 18 of the GNU-compatible __atomic builtins.Richard Smith
This is not quite sufficient for libstdc++'s <atomic>: we still need __atomic_test_and_set and __atomic_clear, and may need a more complete __atomic_is_lock_free implementation. We are also missing an implementation of __atomic_always_lock_free, __atomic_nand_fetch, and __atomic_fetch_nand, but those aren't needed for libstdc++. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154579 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-11Make __atomic_init() (soon to be __c11_atomic_init()) work with non-scalar ↵David Chisnall
types. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154507 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-10Express the number of ULPs in fpaccuracy metadata as a real rather than a Duncan Sands
rational number, eg as 2.5 rather than 5, 2. OK'd by Peter Collingbourne. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@154388 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-29Call out to GCC-compatible runtime functions for atomic ops that we can't useDavid Chisnall
LLVM intrinsics for. I have an implementation of these functions, which wants to go in a libgcc_s equivalent in compiler-rt. It's currently here: http://people.freebsd.org/~theraven/atomic.c It will be committed to compiler-rt as soon as I work out where would be a sensible place to put it... git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153666 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-29Revert r153613 as it's causing large compile-time regressions on the nightly ↵Chad Rosier
testers. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153660 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-28When we can't prove that the target of an aggregate copy isJohn McCall
a complete object, the memcpy needs to use the data size of the structure instead of its sizeof() value. Fixes PR12204. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153613 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-27Move the emission of strict enum range metadata behind a flag (the sameChandler Carruth
flag as GCC uses: -fstrict-enums). There is a *lot* of code making unwarranted assumptions about the underlying type of enums, and it doesn't seem entirely reasonable to eagerly break all of it. Much more importantly, the current state of affairs is *very* good at optimizing based upon this information, which causes failures that are very distant from the actual enum. Before we push for enabling this by default, I think we need to implement -fcatch-undefined-behavior support for instrumenting and trapping whenever we store or load a value outside of the range. That way we can track down the misbehaving code very quickly. I discussed this with Rafael, and currently the only important cases he is aware of are the bool range-based optimizations which are staying hard enabled. We've not seen any issue with those either, and they are much more important for performance. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153550 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-24Add back r153360 with a fix for enums that cover all the 32 bit values.Rafael Espindola
Thanks to NAKAMURA Takumi for finding it! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153383 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-24Revert r153360 (and r153380), "Second part of PR12251. Produce the range ↵NAKAMURA Takumi
metadata in clang for booleans and". For i686 targets (eg. cygwin), I saw "Range must not be empty!" in verifier. It produces (i32)[0x80000000:0x80000000) from (uint64_t)[0xFFFFFFFF80000000ULL:0x0000000080000000ULL), for signed i32 on MDNode::Range. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153382 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-24Second part of PR12251. Produce the range metadata in clang for booleans andRafael Espindola
c++ enums. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153360 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-22Make sure we correctly set the alignment for vector loads and stores ↵Eli Friedman
associated with vector element lvalues. Patch by Kevin Schoedel (with some minor modifications by me). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@153285 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-11Unify naming of LangOptions variable/get function across the Clang stack ↵David Blaikie
(Lex to AST). The member variable is always "LangOpts" and the member function is always "getLangOpts". Reviewed by Chris Lattner git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@152536 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-10Remove BlockDeclRefExpr and introduce a bit on DeclRefExpr toJohn McCall
track whether the referenced declaration comes from an enclosing local context. I'm amenable to suggestions about the exact meaning of this bit. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@152491 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-10Unify the BlockDeclRefExpr and DeclRefExpr paths so thatJohn McCall
we correctly emit loads of BlockDeclRefExprs even when they don't qualify as ODR-uses. I think I'm adequately convinced that BlockDeclRefExpr can die. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@152479 91177308-0d34-0410-b5e6-96231b3b80d8
2012-03-07AST representation for user-defined literals, plus just enough of semanticRichard Smith
analysis to make the AST representation testable. They are represented by a new UserDefinedLiteral AST node, which is a sugared CallExpr. All semantic properties, including full CodeGen support, are achieved for free by this representation. UserDefinedLiterals can never be dependent, so no custom instantiation behavior is required. They are mangled as if they were direct calls to the underlying literal operator. This matches g++'s apparent behavior (but not its actual mangling, which is broken for literal-operator-ids). User-defined *string* literals are now fully-operational, but the semantic analysis is quite hacky and needs more work. No other forms of user-defined literal are created yet, but the AST support for them is present. This patch committed after midnight because we had already hit the quota for new kinds of literal yesterday. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@152211 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-22Generate an AST for the conversion from a lambda closure type to aDouglas Gregor
block pointer that returns a block literal which captures (by copy) the lambda closure itself. Some aspects of the block literal are left unspecified, namely the capture variable (which doesn't actually exist) and the body (which will be filled in by IRgen because it can't be written as an AST). Because we're switching to this model, this patch also eliminates tracking the copy-initialization expression for the block capture of the conversion function, since that information is now embedded in the synthesized block literal. -1 side tables FTW. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@151131 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-18Fix wrong-code bug: __imag on a scalar lvalue should produce a zero rvalue,Richard Smith
rather than an lvalue referring to the scalar. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150889 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-17Whether an argument is required (in contrast with being anJohn McCall
optional argument passed through the variadic ellipsis) potentially affects how we need to lower it. Propagate this information down to the various getFunctionInfo(...) overloads on CodeGenTypes. Furthermore, rename those overloads to clarify their distinct purposes, and make sure we're calling the right one in the right place. This has a nice side-effect of making it easier to construct a function type, since the 'variadic' bit is no longer separable. This shouldn't really change anything for our existing platforms, with one minor exception --- we should now call variadic ObjC methods with the ... in the "right place" (see the test case), which I guess matters for anyone running GNUStep on MIPS. Mostly it's just a substantial clean-up. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150788 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-15Split reinterpret_casts of member pointers out from CK_BitCast; thisJohn McCall
is general goodness because representations of member pointers are not always equivalent across member pointer types on all ABIs (even though this isn't really standard-endorsed). Take advantage of the new information to teach IR-generation how to do these reinterprets in constant initializers. Make sure this works when intermingled with hierarchy conversions (although this is not part of our motivating use case). Doing this in the constant-evaluator would probably have been better, but that would require a *lot* of extra structure in the representation of constant member pointers: you'd really have to track an arbitrary chain of hierarchy conversions and reinterpretations in order to get this right. Ultimately, this seems less complex. I also wasn't quite sure how to extend the constant evaluator to handle foldings that we don't actually want to treat as extended constant expressions. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150551 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-14Use a simpler (and more efficient) pattern to pad vectors.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150475 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-11Basic support for referring to captured variables from lambdas. Some simple ↵Eli Friedman
examples seem to work. Tests coming up soon. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150293 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-09Refactor lambda IRGen so AggExprEmitter::VisitLambdaExpr does the right thing.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150146 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-09Basic IRGen for LambdaExprs with captures.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150141 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-08A little bit of lambda IRGen.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@150058 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-07simplify a bunch of code to use the well-known LLVM IR types computed by ↵Chris Lattner
CodeGenModule. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@149943 91177308-0d34-0410-b5e6-96231b3b80d8
2012-02-05Basic: import SmallString<> into clang namespaceDylan Noblesmith
(I was going to fix the TODO about DenseMap too, but that would break self-host right now. See PR11922.) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@149799 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-30Simplify code by using the new getAggregateElement method that got addedChris Lattner
recently. This also conveniently gets clang ready for a change about to land in mainline. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@149225 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-26Use function pointers, rather than references, to pass DestroyersPeter Collingbourne
around, in the process cleaning up the various gcc/msvc compiler workarounds. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@149036 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Fixup r148926, for -Asserts.NAKAMURA Takumi
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148943 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25reapply r148902:Chris Lattner
"use the new ConstantVector::getSplat method where it makes sense." Also simplify a bunch of code to use the Builder->getInt32 instead of doing it the hard and ugly way. Much more progress could be made here, but I don't plan to do it. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148926 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Fix a crash for an edge case of the GNU ?: extension.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148923 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25Revert 148902 which was part of 148901 which was reverted in r148906.Argyrios Kyrtzidis
Original log: use the new ConstantVector::getSplat method where it makes sense. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148907 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-25use the new ConstantVector::getSplat method where it makes sense.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148902 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-21Add an assertion that our use-marking actually covers all uses of a ↵Eli Friedman
variable. The assertion doesn't cover quite as much as it should, but it's a good start, at least. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148625 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-20More dead code removal (using -Wunreachable-code)David Blaikie
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148577 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-16Some improvements to the handling of C11 atomic types:David Chisnall
- Add atomic-to/from-nonatomic cast types - Emit atomic operations for arithmetic on atomic types - Emit non-atomic stores for initialisation of atomic types, but atomic stores and loads for every other store / load - Add a __atomic_init() intrinsic which does a non-atomic store to an _Atomic() type. This is needed for the corresponding C11 stdatomic.h function. - Enables the relevant __has_feature() checks. The feature isn't 100% complete yet, but it's done enough that we want people testing it. Still to do: - Make the arithmetic operations on atomic types (e.g. Atomic(int) foo = 1; foo++;) use the correct LLVM intrinsic if one exists, not a loop with a cmpxchg. - Add a signal fence builtin - Properly set the fenv state in atomic operations on floating point values - Correctly handle things like _Atomic(_Complex double) which are too large for an atomic cmpxchg on some platforms (this requires working out what 'correctly' means in this context) - Fix the many remaining corner cases git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148242 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-04implement rdar://10639962 by keeping track of increased alignment Chris Lattner
information even in subscripting operations. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@147557 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-08-03 15:41:05 +0000