aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/CGRecordLayoutBuilder.cpp
AgeCommit message (Collapse)Author
2013-04-16Standardize accesses to the TargetInfo in IR-gen.John McCall
Patch by Stephen Lin! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@179638 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-25First pass at abstracting out a class for the target C++ ABI.John McCall
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@173514 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-15Fix signed/unsigned CompareDavid Greene
Do some casting to avoid a signed/unsigned compare. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@172571 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-02Rewrite #includes for llvm/Foo.h to llvm/IR/Foo.h as appropriate toChandler Carruth
reflect the migration in r171366. Re-sort the #include lines to reflect the new paths. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@171369 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18Fix a compilation warning with gcc 4.6Eli Bendersky
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@170440 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-09Cleanup and fix an assert that was mis-firing.Chandler Carruth
Note that there is no test suite update. This was found by a couple of tests failing when the test suite was run on a powerpc64 host (thanks Roman!). The tests don't specify a triple, which might seem surprising for a codegen test. But in fact, these tests don't even inspect their output. Not at all. I could add a bunch of triples to these tests so that we'd get the test coverage for normal builds, but really someone needs to go through and add actual *tests* to these tests. =[ The ones in question are: test/CodeGen/bitfield-init.c test/CodeGen/union.c git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@169694 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-09Fix the bitfield record layout in codegen for big endian targets.Chandler Carruth
This was an egregious bug due to the several iterations of refactorings that took place. Size no longer meant what it original did by the time I finished, but this line of code never got updated. Unfortunately we had essentially zero tests for this in the regression test suite. =[ I've added a PPC64 run over the bitfield test case I've been primarily using. I'm still looking at adding more tests and making sure this is the *correct* bitfield access code on PPC64 linux, but it looks pretty close to me, and it is *worlds* better than before this patch as it no longer asserts! =] More commits to follow with at least additional tests and maybe more fixes. Sorry for the long breakage due to this.... git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@169691 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-06Rework the bitfield access IR generation to address PR13619 andChandler Carruth
generally support the C++11 memory model requirements for bitfield accesses by relying more heavily on LLVM's memory model. The primary change this introduces is to move from a manually aligned and strided access pattern across the bits of the bitfield to a much simpler lump access of all bits in the bitfield followed by math to extract the bits relevant for the particular field. This simplifies the code significantly, but relies on LLVM to intelligently lowering these integers. I have tested LLVM's lowering both synthetically and in benchmarks. The lowering appears to be functional, and there are no really significant performance regressions. Different code patterns accessing bitfields will vary in how this impacts them. The only real regressions I'm seeing are a few patterns where the LLVM code generation for loads that feed directly into a mask operation don't take advantage of the x86 ability to do a smaller load and a cheap zero-extension. This doesn't regress any benchmark in the nightly test suite on my box past the noise threshold, but my box is quite noisy. I'll be watching the LNT numbers, and will look into further improvements to the LLVM lowering as needed. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@169489 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-04Sort all of Clang's files under 'lib', and fix up the broken headersChandler Carruth
uncovered. This required manually correcting all of the incorrect main-module headers I could find, and running the new llvm/utils/sort_includes.py script over the files. I also manually added quite a few missing headers that were uncovered by shuffling the order or moving headers up to be main-module-headers. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@169237 91177308-0d34-0410-b5e6-96231b3b80d8
2012-10-12Make -mms-bitfields behave consistently.Eli Friedman
Patch by Jeremiah Zanin. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@165849 91177308-0d34-0410-b5e6-96231b3b80d8
2012-10-08Move TargetData to DataLayout.Micah Villmow
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@165395 91177308-0d34-0410-b5e6-96231b3b80d8
2012-06-27Propagate lvalue alignment into bitfields. Per report on cfe-dev.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@159295 91177308-0d34-0410-b5e6-96231b3b80d8
2012-06-06Revert Decl's iterators back to pointer value_type rather than reference ↵David Blaikie
value_type In addition, I've made the pointer and reference typedef 'void' rather than T* just so they can't get misused. I would've omitted them entirely but std::distance likes them to be there even if it doesn't use them. This rolls back r155808 and r155869. Review by Doug Gregor incorporating feedback from Chandler Carruth. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@158104 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-01Add support for laying out vtordisps according to our currentJohn McCall
working knowledge of the Microsoft ABI. Based on a patch by Dmitry Sokolov. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@155905 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-30Remove the ref/value inconsistency in filter_decl_iterator.David Blaikie
filter_decl_iterator had a weird mismatch where both op* and op-> returned T* making it difficult to generalize this filtering behavior into a reusable library of any kind. This change errs on the side of value, making op-> return T* and op* return T&. (reviewed by Richard Smith) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@155808 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-27Add a missing check in CodeGen of packed classes with vtables. ↵Eli Friedman
<rdar://problem/11324125>. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@155689 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
2012-03-04Silence a GCC warning about a set-but-not-used variable in release builds.Chandler Carruth
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@152005 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-13A base subobject type doesn't make sense for unions; don't try to compute ↵Eli Friedman
it. Based on patch by Yin Ma. Fixes PR11751. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@148093 91177308-0d34-0410-b5e6-96231b3b80d8
2011-12-12Make CGRecordLayoutBuilder correctly switch over to a packed class when a ↵Eli Friedman
class has a base whose alignment will break the class layout. <rdar://problem/10551376>. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@146443 91177308-0d34-0410-b5e6-96231b3b80d8
2011-12-07Make sure we correctly zero-initialize unions containing a pointer to data ↵Eli Friedman
member as the first field. PR11487. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@146009 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-08Fix the layout of vb-tables and vf-tables in the MS C++ ABI.John McCall
Based on work by Dmitry Sokolov! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@144072 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-18Rewrite parts of MS ABI C++ layout. Based on work by r4start; I ended up ↵Eli Friedman
doing this while I was trying to review his patch. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@142325 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-10Constant expression evaluation refactoring:Richard Smith
- Remodel Expr::EvaluateAsInt to behave like the other EvaluateAs* functions, and add Expr::EvaluateKnownConstInt to capture the current fold-or-assert behaviour. - Factor out evaluation of bitfield bit widths. - Fix a few places which would evaluate an expression twice: once to determine whether it is a constant expression, then again to get the value. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@141561 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-02Extend the ASTContext constructor to delay the initialization ofDouglas Gregor
builtin types (When requested). This is another step toward making ASTUnit build the ASTContext as needed when loading an AST file, rather than doing so after the fact. No actual functionality change (yet). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@138985 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-12switch clang to use the new-new way of creating llvm::StructType's.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@137472 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-23remove unneeded llvm:: namespace qualifiers on some core types now that ↵Chris Lattner
LLVM.h imports them into the clang namespace. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135852 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-18de-constify llvm::Type, patch by David Blaikie!Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135370 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-09clang side to match the LLVM IR type system rewrite patch.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@134831 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-21IRgen: Add a -fuse-register-sized-bitfield-access option, for testing.Daniel Dunbar
- Changes bit-field access policy to try to use (aligned) register sized accesses. The idea here is that by using larger accesses we expose more coalescing potential to the backend when we have situations like adjacent bit-fields in the same structure (which is common), and that the backend should be smart enough to narrow the accesses down when no coalescing is done or when it is shown not to be profitable. -- $ clang -m32 -O3 -S -o - t.c _f0: ## @f0 pushl %ebp movl %esp, %ebp movl 8(%ebp), %eax movb (%eax), %cl andb $-128, %cl orb $1, %cl movb %cl, (%eax) movb 1(%eax), %cl andb $-128, %cl orb $1, %cl movb %cl, 1(%eax) movb 2(%eax), %cl andb $-128, %cl orb $1, %cl movb %cl, 2(%eax) movb 3(%eax), %cl andb $-128, %cl orb $1, %cl movb %cl, 3(%eax) popl %ebp ret $ clang -m32 -O3 -S -o - t.c -Xclang -fuse-register-sized-bitfield-access _f0: ## @f0 pushl %ebp movl %esp, %ebp movl 8(%ebp), %eax movl $-2139062144, %ecx ## imm = 0xFFFFFFFF80808080 andl (%eax), %ecx orl $16843009, %ecx ## imm = 0x1010101 movl %ecx, (%eax) popl %ebp ret -- git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@133532 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-20Introduce Type::isSignedIntegerOrEnumerationType() andDouglas Gregor
Type::isUnsignedIntegerOrEnumerationType(), which are like Type::isSignedIntegerType() and Type::isUnsignedIntegerType() but also consider the underlying type of a C++0x scoped enumeration type. Audited all callers to the existing functions, switching those that need to also handle scoped enumeration types (e.g., those that deal with constant values) over to the new functions. Fixes PR9923 / <rdar://problem/9447851>. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@131735 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-03Finish off rules for z-length bitfields in ms_structFariborz Jahanian
structs. // rdar://8823265 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130783 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-02More rule enforcement of zero bitfields for ms_struct.Fariborz Jahanian
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130696 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-27Some refactoring of my ms_struct patch.Fariborz Jahanian
// rdar://8823265 related. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130311 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-26With ms_struct attribut, Zero-length bitfields following Fariborz Jahanian
non-bitfield members are ignore. // rdar://8823265 wip git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130257 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-24Simplify AppendTailPadding() by converting its parameter to CharUnits. NoKen Dyck
change in functionality intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130108 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-24Eliminate literal 8s from LayoutUnionField by converting NumBytesToAppend toKen Dyck
CharUnits. No change in functionality intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130107 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-24Eliminate literal 8s from LayoutBitField(), converting variables toKen Dyck
CharUnits or replacing the 8s with char align. No change in functionality intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130106 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-24Convert AccessInfo::AccessAlignment to CharUnits. No change in functionalityKen Dyck
intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130087 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-24Convert CGBitFieldInfo::FieldByteOffset to CharUnits. No change inKen Dyck
functionality intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130085 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17When laying out bases in, always try the "base subobject" LLVM type. If itAnders Carlsson
turns out that a field or base needs to be laid out in the tail padding of the base, CGRecordLayoutBuilder::ResizeLastBaseFieldIfNecessary will convert it to an array of i8. I've audited the new test results to make sure that they are still valid. I've also verified that we pass a self-host with this change. This (finally) fixes PR5589! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129673 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17Use a SmallVector for field types in CGRecordLayoutBuilder now that ↵Anders Carlsson
llvm::StructType::get takes an ArrayRef. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129667 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-22Fix typo in comment.Nick Lewycky
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128088 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-26A much-delayed response to Ken Dyck's review: make this somewhat moreJohn McCall
char-width agnostic. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@126542 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-17improve support for big endian targets, fixing PR8171, patchChris Lattner
by Heikki Kultala! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125784 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-15Refactor CGRecordLayoutBuilder to use CharUnits more consistently.John McCall
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125605 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-15Perform zero-initialization of virtual base classes when emitting John McCall
a zero constant for a complete class. rdar://problem/8424975 To make this happen, track the field indexes for virtual bases in the complete object. I'm curious whether we might be better off making CGRecordLayoutBuilder *much* more reliant on ASTRecordLayout; we're currently duplicating an awful lot of the ABI layout logic. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125555 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-15Convert RecordLayout::Alignment to CharUnits from bit units. No change inKen Dyck
functionality intended. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125549 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-11Add a helper function, ASTContext::toBits(), that converts sizes inKen Dyck
CharUnits to sizes in bits, and use it to tidy up the places where the conversion was done explicitly. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125332 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-10Drop the 'InBits' part from the name of RecordSizeInBits as the value is inKen Dyck
character units. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125281 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-04-29 22:13:01 +0000