aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/CGBuiltin.cpp
AgeCommit message (Collapse)Author
2011-09-09Bring llvm.annotation* intrinsics support back to where it was in llvm-gcc: canJulien Lerouge
annotate global, local variables, struct fields, or arbitrary statements (using the __builtin_annotation), rdar://8037476. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@139423 91177308-0d34-0410-b5e6-96231b3b80d8
2011-09-07Switch clang over to using fence/atomicrmw/cmpxchg instead of the intrinsics ↵Eli Friedman
(which will go away). LLVM CodeGen does almost exactly the same thing with these and the old intrinsics, so I'm reasonably confident this will not break anything. There are still a few issues which need to be resolved with code generation for atomic load and store, so I'm not converting the places which need those for now. I'm not entirely sure what to do about __builtin_llvm_memory_barrier: the fence instruction doesn't expose all the possibilities which can be expressed by __builtin_llvm_memory_barrier. I would appreciate hearing from anyone who is using this intrinsic. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@139216 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-17Fix a handful of dead stores found by Clang's static analyzer. There's a ↵Ted Kremenek
bunch of others I haven't touched. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@137867 91177308-0d34-0410-b5e6-96231b3b80d8
2011-08-13Move handling of vget_lane/vset_lane before the code that checks the type.Bob Wilson
Unlike most of the other Neon intrinsics, these are not overloaded and do not have the extra argument that specifies the vector type. This has not been fatal because the lane number operand is supposed to be an ICE and so that value has harmlessly been used as the type identifier. Radar 9901281. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@137550 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-29Remove some unnecessary single element array temporaries.Jay Foad
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@136461 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-25Shorten some expressions by using ArrayRef::slice().Frits van Bommel
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135910 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
2011-07-18Migrate LLVM and Clang to use the new makeArrayRef(...) functions where ↵Frits van Bommel
previously explicit non-default constructors were used. Mostly mechanical with some manual reformatting. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135390 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
2011-07-15Convert CallInst and InvokeInst APIs to use ArrayRef.Jay Foad
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135265 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-14Change intrinsic getter to take an ArrayRef, now that the underlying ↵Benjamin Kramer
function in LLVM does. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@135155 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
2011-07-08Introduce __builtin_expect() intrinsic support.Jakub Staszak
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@134761 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-08Add codegen support for the fma/fmal/fmaf builtins.Cameron Zwarich
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@134743 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-24Revert "Shorten some ARM builtin names by removing unnecessary "neon" prefix."Bob Wilson
Sorry, this was a bad idea. Within clang these builtins are in a separate "ARM" namespace, but the actual builtin names should clearly distinguish tha they are target specific. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@133833 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-24Shorten some ARM builtin names by removing unnecessary "neon" prefix.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@133826 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-18update for api change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@133365 91177308-0d34-0410-b5e6-96231b3b80d8
2011-06-14Update the prefetch intrinsic usage. Now the last argument tells codegenBruno Cardoso Lopes
whether it's a data or instruction cache access. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@132977 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-28Eliminate temporary argument vectors.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@132260 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-28Add support for ARM ldrexd/strexd builtinsBruno Cardoso Lopes
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@132249 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-13Remove the 'unaligned load' builtins now that they're no longer used in the ↵Bill Wendling
*mmintrin.h files. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@131300 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-12LLVM doesn't always optimize away the four loads from this:Bill Wendling
(__m128){ p[0], p[1], p[2], p[3] } which produces really bad code. This could be done in instcombine, but it's probably better to do it in the front-end instead. <rdar://problem/9424836> git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@131237 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-04Simplification noticed by Chris.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130864 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-04Convert the non-temporal store builtins to LLVM-native IR.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130830 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-25Generalize case for built-in expressions havingFariborz Jahanian
side-effect to generate their ir. Not just for __builtin_expect. // rdar://9330105 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130172 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-25Ir-gen the side-effect(s) when __builtin_expect isFariborz Jahanian
constant-folded. // rdar://9330105 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@130163 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-20fix a crash on code that uses the result value of __builtin___memcpy_chk.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129892 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-17fold memcpy/set/move_chk to llvm.memcpy/set/move when the sizesChris Lattner
are trivial. This exposes opportunities earlier, and allows fastisel to do good things with these at -O0. This addresses rdar://9289468 - clang doesn't fold memset_chk at -O0 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129651 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-15Add 3DNow! Intrinsics.Michael J. Spencer
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129570 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13Removing the unaligned load tests from builtins-x86.c since they're ↵Bill Wendling
generated by a regular 'load' now. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129464 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13It looks like the FreeBSD buildbot needs this for the builtins-x86.c test.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129433 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13Just use a native "load" instead of translating the builtin later. Clang canBill Wendling
take it! I wasn't able to get __builtin_ia32_loaddqu to transform into an unaligned load...I'll have to look into it further. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129427 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13Convert the unaligned load builtins to the first-class versions.Bill Wendling
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129420 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-09add a __sync_swap builtin to fill out the rest of the __sync builtins. Chris Lattner
Patch by Dave Zarzycki! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129189 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-31Oops, prefer C-style cast hereMatt Beaumont-Gay
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128607 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-31Silence GCC warning about differing types on the branches of a conditional ↵Matt Beaumont-Gay
expression git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128605 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-31Use intrinsics for Neon vmull operations. Radar 9208957.Bob Wilson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128590 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-30Remove PHINode::reserveOperandSpace(). Instead, add a parameter toJay Foad
PHINode::Create() giving the (known or expected) number of operands. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128538 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-30(Almost) always call reserveOperandSpace() on newly created PHINodes.Jay Foad
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128534 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-24Make sure we aggressively attach nounwind (etc.) to calls to libraryEli Friedman
functions of the form __builtin_XXX. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@128198 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-14__clear_cache() is varargs and people will occasionally write it withoutEric Christopher
arguments. Process only the arguments that people write, but process all of them. Fixes rdar://8900346 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@127616 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-15update for ConstantVector API change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125538 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-14revert my ConstantVector patch, it seems to have made the llvm-gccChris Lattner
builders unhappy. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125505 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-14update for ConstantVector::get API change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125488 91177308-0d34-0410-b5e6-96231b3b80d8
2011-02-08Reorganize CodeGen{Function,Module} to eliminate the unfortunateJohn McCall
Block{Function,Module} base class. Minor other refactorings. Fixed a few address-space bugs while I was there. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@125085 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-23Null initialize a few variables flagged byTed Kremenek
clang's -Wuninitialized-experimental warning. While these don't look like real bugs, clang's -Wuninitialized-experimental analysis is stricter than GCC's, and these fixes have the benefit of being general nice cleanups. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@124072 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-12Fix a latent bug where, after emitting an expression statement, we wouldJohn McCall
delete the block we began emitting into if it had no predecessors. We never want to do this, because there are several valid cases during statement emission where an existing block has no known predecessors but will acquire some later. The case in my test case doesn't inherently fall into this category, because we could safely emit the case-range code before the statement body, but there are examples with labels that can't be fallen into that would also demonstrate this bug. rdar://problem/8837067 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@123303 91177308-0d34-0410-b5e6-96231b3b80d8
2011-01-09Make a helper function static.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@123118 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-30Simplify mem{cpy, move, set} creation with IRBuilder.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@122634 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-15Implement builtins for Neon half-precision float conversions.Bob Wilson
Also tweak the VCVT_F32_F16 entry in arm_neon.td to be more consistent with the other floating-point conversion builtins. Radar 8068427. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@121916 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-07-28 21:21:23 +0000