maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp

2817 lines
102 KiB
C++
Raw Normal View History

add a comment line that got dropped llvm-svn: 100638
2010-04-08 02:10:38 +08:00
//===-- AsmPrinter.cpp - Common AsmPrinter code ---------------------------===//
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
//
// The LLVM Compiler Infrastructure
//
Remove attribution from file headers, per discussion on llvmdev. llvm-svn: 45418
2007-12-30 04:36:04 +08:00
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
//
//===----------------------------------------------------------------------===//
//
// This file implements the AsmPrinter class.
//
//===----------------------------------------------------------------------===//
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "AsmPrinterHandler.h"
[NFC] Header cleanup Removed some unused headers, replaced some headers with forward class declarations. Found using simple scripts like this one: clear && ack --cpp -l '#include "llvm/ADT/IndexedMap.h"' | xargs grep -L 'IndexedMap[<]' | xargs grep -n --color=auto 'IndexedMap' Patch by Eugene Kosov <claprix@yandex.ru> Differential Revision: http://reviews.llvm.org/D19219 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 266595
2016-04-18 17:17:29 +08:00
#include "CodeViewDebug.h"
change AsmPrinter to use DwarfDebug/DwarfException directly instead of going through DwarfWriter. llvm-svn: 100405
2010-04-05 13:11:15 +08:00
#include "DwarfDebug.h"
#include "DwarfException.h"
Rename Win64Exception.(cpp|h) to WinException.(cpp|h) This is in preparation for reusing this for 32-bit x86 EH table emission. Also updates the type name for consistency. NFC llvm-svn: 238521
2015-05-29 06:47:01 +08:00
#include "WinException.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/ADT/APFloat.h"
#include "llvm/ADT/APInt.h"
#include "llvm/ADT/DenseMap.h"
#include "llvm/ADT/SmallPtrSet.h"
#include "llvm/ADT/SmallString.h"
#include "llvm/ADT/SmallVector.h"
Use the new script to sort the includes of every file under lib. Sooooo many of these had incorrect or strange main module includes. I have manually inspected all of these, and fixed the main module include to be the nearest plausible thing I could find. If you own or care about any of these source files, I encourage you to take some time and check that these edits were sensible. I can't have broken anything (I strictly added headers, and reordered them, never removed), but they may not be the headers you'd really like to identify as containing the API being implemented. Many forward declarations and missing includes were added to a header files to allow them to parse cleanly when included first. The main module rule does in fact have its merits. =] llvm-svn: 169131
2012-12-04 00:50:05 +08:00
#include "llvm/ADT/Statistic.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/ADT/STLExtras.h"
#include "llvm/ADT/StringRef.h"
#include "llvm/ADT/Triple.h"
#include "llvm/ADT/Twine.h"
Use the new script to sort the includes of every file under lib. Sooooo many of these had incorrect or strange main module includes. I have manually inspected all of these, and fixed the main module include to be the nearest plausible thing I could find. If you own or care about any of these source files, I encourage you to take some time and check that these edits were sensible. I can't have broken anything (I strictly added headers, and reordered them, never removed), but they may not be the headers you'd really like to identify as containing the API being implemented. Many forward declarations and missing includes were added to a header files to allow them to parse cleanly when included first. The main module rule does in fact have its merits. =] llvm-svn: 169131
2012-12-04 00:50:05 +08:00
#include "llvm/Analysis/ConstantFolding.h"
Move llvm::canBeOmittedFromSymbolTable() to Analysis. llvm-svn: 299182
2017-03-31 12:46:31 +08:00
#include "llvm/Analysis/ObjectUtils.h"
Refactor duplicated code. llvm-svn: 214328
2014-07-31 03:42:16 +08:00
#include "llvm/CodeGen/Analysis.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/CodeGen/AsmPrinter.h"
#include "llvm/CodeGen/GCMetadata.h"
Rename some GC classes so that their roll will hopefully be clearer. In particular, Collector was confusing to implementors. Several thought that this compile-time class was the place to implement their runtime GC heap. Of course, it doesn't even exist at runtime. Specifically, the renames are: Collector -> GCStrategy CollectorMetadata -> GCFunctionInfo CollectorModuleMetadata -> GCModuleInfo CollectorRegistry -> GCRegistry Function::getCollector -> getGC (setGC, hasGC, clearGC) Several accessors and nested types have also been renamed to be consistent. These changes should be obvious. llvm-svn: 54899
2008-08-18 02:44:35 +08:00
#include "llvm/CodeGen/GCMetadataPrinter.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/CodeGen/GCStrategy.h"
#include "llvm/CodeGen/MachineBasicBlock.h"
Allow target to customize directive used to switch to arbitrary section in SwitchSection, add generic constant pool emitter llvm-svn: 24464
2005-11-21 16:25:09 +08:00
#include "llvm/CodeGen/MachineConstantPool.h"
Move DebugInfo checks into EmitComments and remove them from target-specific AsmPrinters. Not all comments need DebugInfo. Re-enable the line numbers comment test. llvm-svn: 88697
2009-11-14 05:34:57 +08:00
#include "llvm/CodeGen/MachineFrameInfo.h"
Make various changes suggested by Chris. llvm-svn: 79358
2009-08-19 03:22:55 +08:00
#include "llvm/CodeGen/MachineFunction.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/CodeGen/MachineFunctionPass.h"
#include "llvm/CodeGen/MachineInstr.h"
Handle bundled terminators in isBlockOnlyReachableByFallthrough. Targets like SPARC and MIPS have delay slots and normally bundle the delay slot instruction with the corresponding terminator. Teach isBlockOnlyReachableByFallthrough to find any MBB operands on bundled terminators so SPARC doesn't need to specialize this function. llvm-svn: 199061
2014-01-13 03:24:08 +08:00
#include "llvm/CodeGen/MachineInstrBundle.h"
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
#include "llvm/CodeGen/MachineJumpTableInfo.h"
Add support for printing loop structure information in asm comments. This definitely slows down asm output so put it under an -asm-exuberant flag. This information is useful when doing static analysis of performance issues. llvm-svn: 78567
2009-08-11 00:38:07 +08:00
#include "llvm/CodeGen/MachineLoopInfo.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/CodeGen/MachineMemOperand.h"
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
#include "llvm/CodeGen/MachineModuleInfoImpls.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/CodeGen/MachineOperand.h"
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
#include "llvm/CodeGen/MachineOptimizationRemarkEmitter.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/IR/BasicBlock.h"
#include "llvm/IR/Constant.h"
#include "llvm/IR/Constants.h"
Move all of the header files which are involved in modelling the LLVM IR into their new header subdirectory: include/llvm/IR. This matches the directory structure of lib, and begins to correct a long standing point of file layout clutter in LLVM. There are still more header files to move here, but I wanted to handle them in separate commits to make tracking what files make sense at each layer easier. The only really questionable files here are the target intrinsic tablegen files. But that's a battle I'd rather not fight today. I've updated both CMake and Makefile build systems (I think, and my tests think, but I may have missed something). I've also re-sorted the includes throughout the project. I'll be committing updates to Clang, DragonEgg, and Polly momentarily. llvm-svn: 171366
2013-01-02 19:36:10 +08:00
#include "llvm/IR/DataLayout.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/IR/DebugInfoMetadata.h"
#include "llvm/IR/DerivedTypes.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/GlobalAlias.h"
#include "llvm/IR/GlobalIFunc.h"
#include "llvm/IR/GlobalIndirectSymbol.h"
#include "llvm/IR/GlobalObject.h"
#include "llvm/IR/GlobalValue.h"
#include "llvm/IR/GlobalVariable.h"
Move the llvm mangler to lib/IR. This makes it available to tools that don't link with target (like llvm-ar). llvm-svn: 198708
2014-01-08 05:19:40 +08:00
#include "llvm/IR/Mangler.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/IR/Metadata.h"
Move all of the header files which are involved in modelling the LLVM IR into their new header subdirectory: include/llvm/IR. This matches the directory structure of lib, and begins to correct a long standing point of file layout clutter in LLVM. There are still more header files to move here, but I wanted to handle them in separate commits to make tracking what files make sense at each layer easier. The only really questionable files here are the target intrinsic tablegen files. But that's a battle I'd rather not fight today. I've updated both CMake and Makefile build systems (I think, and my tests think, but I may have missed something). I've also re-sorted the includes throughout the project. I'll be committing updates to Clang, DragonEgg, and Polly momentarily. llvm-svn: 171366
2013-01-02 19:36:10 +08:00
#include "llvm/IR/Module.h"
#include "llvm/IR/Operator.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/IR/Value.h"
split inline asm support out to its own .cpp file. llvm-svn: 100340
2010-04-05 02:34:07 +08:00
#include "llvm/MC/MCAsmInfo.h"
hoist MCContext/MCStreamer up to AsmPrinter since we're going to start creating MCSections soon instead of Section for all targets, and we need something to own them. llvm-svn: 77252
2009-07-28 05:28:04 +08:00
#include "llvm/MC/MCContext.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/MC/MCDirectives.h"
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
#include "llvm/MC/MCExpr.h"
Emit line numbers in asm comments when available. llvm-svn: 76117
2009-07-17 06:24:20 +08:00
#include "llvm/MC/MCInst.h"
switch off of 'Section' onto MCSection. We're not properly using MCSection subclasses yet, but this is a step in the right direction. llvm-svn: 77708
2009-08-01 02:48:30 +08:00
#include "llvm/MC/MCSection.h"
[XRay] Merge instrumentation point table emission code into AsmPrinter. Summary: No need to have this per-architecture. While there, unify 32-bit ARM's behaviour with what changed elsewhere and start function names lowercase as per the coding standards. Individual entry emission code goes to the entry's own class. Fully tested on amd64, cross-builds on both ARMs and PowerPC. Reviewers: dberris Subscribers: aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D28209 llvm-svn: 290858
2017-01-03 12:30:21 +08:00
#include "llvm/MC/MCSectionELF.h"
#include "llvm/MC/MCSectionMachO.h"
switch off of 'Section' onto MCSection. We're not properly using MCSection subclasses yet, but this is a step in the right direction. llvm-svn: 77708
2009-08-01 02:48:30 +08:00
#include "llvm/MC/MCStreamer.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/MC/MCSubtargetInfo.h"
#include "llvm/MC/MCSymbol.h"
#include "llvm/MC/MCTargetOptions.h"
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
#include "llvm/MC/MCValue.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/MC/SectionKind.h"
#include "llvm/Pass.h"
#include "llvm/Support/Casting.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/Dwarf.h"
#include "llvm/Support/ELF.h"
Use the new script to sort the includes of every file under lib. Sooooo many of these had incorrect or strange main module includes. I have manually inspected all of these, and fixed the main module include to be the nearest plausible thing I could find. If you own or care about any of these source files, I encourage you to take some time and check that these edits were sensible. I can't have broken anything (I strictly added headers, and reordered them, never removed), but they may not be the headers you'd really like to identify as containing the API being implemented. Many forward declarations and missing includes were added to a header files to allow them to parse cleanly when included first. The main module rule does in fact have its merits. =] llvm-svn: 169131
2012-12-04 00:50:05 +08:00
#include "llvm/Support/ErrorHandling.h"
#include "llvm/Support/Format.h"
#include "llvm/Support/MathExtras.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/Support/raw_ostream.h"
AsmPrinter::doFinalization is at the module level and so doesn't have access to a target specific subtarget info. Grab the module level MCSubtargetInfo for the JumpInstrTable output stubs. llvm-svn: 229974
2015-02-20 14:59:48 +08:00
#include "llvm/Support/TargetRegistry.h"
Use the new script to sort the includes of every file under lib. Sooooo many of these had incorrect or strange main module includes. I have manually inspected all of these, and fixed the main module include to be the nearest plausible thing I could find. If you own or care about any of these source files, I encourage you to take some time and check that these edits were sensible. I can't have broken anything (I strictly added headers, and reordered them, never removed), but they may not be the headers you'd really like to identify as containing the API being implemented. Many forward declarations and missing includes were added to a header files to allow them to parse cleanly when included first. The main module rule does in fact have its merits. =] llvm-svn: 169131
2012-12-04 00:50:05 +08:00
#include "llvm/Support/Timer.h"
Debug Info: Simplify Frame Index handling in DBG_VALUE Machine Instructions Rather than using the full power of target-specific addressing modes in DBG_VALUEs with Frame Indicies, simply use Frame Index + Offset. This reduces the complexity of debug info handling down to two representations of values (reg+offset and frame index+offset) rather than three or four. Ideally we could ensure that frame indicies had been eliminated by the time we reached an assembly or dwarf generation, but I haven't spent the time to figure out where the FIs are leaking through into that & whether there's a good place to convert them. Some FI+offset=>reg+offset conversion is done (see PrologEpilogInserter, for example) which is necessary for some SelectionDAG assumptions about registers, I believe, but it might be possible to make this a more thorough conversion & ensure there are no remaining FIs no matter how instruction selection is performed. llvm-svn: 184066
2013-06-17 04:34:15 +08:00
#include "llvm/Target/TargetFrameLowering.h"
Move DebugInfo checks into EmitComments and remove them from target-specific AsmPrinters. Not all comments need DebugInfo. Re-enable the line numbers comment test. llvm-svn: 88697
2009-11-14 05:34:57 +08:00
#include "llvm/Target/TargetInstrInfo.h"
If a target uses a GOT, put it in the jt data section, not the text section. This will fix alpha when Andrew implements AlphaTargetMachine::getTargetLowering(). llvm-svn: 30779
2006-10-07 06:50:56 +08:00
#include "llvm/Target/TargetLowering.h"
Rip all of the global variable lowering logic out of TargetAsmInfo. Since it is highly specific to the object file that will be generated in the end, this introduces a new TargetLoweringObjectFile interface that is implemented for each of ELF/MachO/COFF/Alpha/PIC16 and XCore. Though still is still a brutal and ugly refactoring, this is a major step towards goodness. This patch also: 1. fixes a bunch of dangling pointer problems in the PIC16 backend. 2. disables the TargetLowering copy ctor which PIC16 was accidentally using. 3. gets us closer to xcore having its own crazy target section flags and pic16 not having to shadow sections with its own objects. 4. fixes wierdness where ELF targets would set CStringSection but not CStringSection_. Factor the code better. 5. fixes some bugs in string lowering on ELF targets. llvm-svn: 77294
2009-07-28 11:13:23 +08:00
#include "llvm/Target/TargetLoweringObjectFile.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include "llvm/Target/TargetMachine.h"
Replace all target specific implicit def instructions with a target independent one: TargetInstrInfo::IMPLICIT_DEF. llvm-svn: 48380
2008-03-15 08:03:38 +08:00
#include "llvm/Target/TargetRegisterInfo.h"
Change MCStreamer EmitInstruction interface to take subtarget info llvm-svn: 200345
2014-01-29 07:12:42 +08:00
#include "llvm/Target/TargetSubtargetInfo.h"
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
#include <algorithm>
#include <cassert>
#include <cinttypes>
#include <cstdint>
#include <limits>
#include <memory>
#include <string>
#include <utility>
#include <vector>
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
using namespace llvm;
[Modules] Remove potential ODR violations by sinking the DEBUG_TYPE define below all header includes in the lib/CodeGen/... tree. While the current modules implementation doesn't check for this kind of ODR violation yet, it is likely to grow support for it in the future. It also removes one layer of macro pollution across all the included headers. Other sub-trees will follow. llvm-svn: 206837
2014-04-22 10:02:50 +08:00
#define DEBUG_TYPE "asm-printer"
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
static const char *const DWARFGroupName = "dwarf";
static const char *const DWARFGroupDescription = "DWARF Emission";
static const char *const DbgTimerName = "emit";
static const char *const DbgTimerDescription = "Debug Info Emission";
static const char *const EHTimerName = "write_exception";
static const char *const EHTimerDescription = "DWARF Exception Writer";
static const char *const CodeViewLineTablesGroupName = "linetables";
static const char *const CodeViewLineTablesGroupDescription =
"CodeView Line Tables";
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
STATISTIC(EmittedInsts, "Number of machine instrs printed");
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
static cl::opt<bool>
PrintSchedule("print-schedule", cl::Hidden, cl::init(false),
cl::desc("Print 'sched: [latency:throughput]' in .s output"));
Drop 'const' llvm-svn: 36662
2007-05-03 09:11:54 +08:00
char AsmPrinter::ID = 0;
rearrange MCContext ownership. Before LLVMTargetMachine created it and passing off ownership to AsmPrinter. Now MachineModuleInfo creates it and owns it by value. This allows us to use MCSymbols more consistently throughout the rest of the code generator, and simplifies a bit of code. This also allows MachineFunction to keep an MCContext reference handy, and cleans up the TargetRegistry interfaces for AsmPrinters. llvm-svn: 98450
2010-03-14 04:55:24 +08:00
Use unique_ptr for the result of Registry entries. llvm-svn: 206248
2014-04-15 13:53:26 +08:00
typedef DenseMap<GCStrategy*, std::unique_ptr<GCMetadataPrinter>> gcp_map_type;
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
static gcp_map_type &getGCMap(void *&P) {
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (!P)
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
P = new gcp_map_type();
return *(gcp_map_type*)P;
}
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
/// getGVAlignmentLog2 - Return the alignment to use for the specified global
/// value in log2 form. This rounds up to the preferred alignment if possible
/// and legal.
DataLayout is mandatory, update the API to reflect it with references. Summary: Now that the DataLayout is a mandatory part of the module, let's start cleaning the codebase. This patch is a first attempt at doing that. This patch is not exactly NFC as for instance some places were passing a nullptr instead of the DataLayout, possibly just because there was a default value on the DataLayout argument to many functions in the API. Even though it is not purely NFC, there is no change in the validation. I turned as many pointer to DataLayout to references, this helped figuring out all the places where a nullptr could come up. I had initially a local version of this patch broken into over 30 independant, commits but some later commit were cleaning the API and touching part of the code modified in the previous commits, so it seemed cleaner without the intermediate state. Test Plan: Reviewers: echristo Subscribers: llvm-commits From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 231740
2015-03-10 10:37:25 +08:00
static unsigned getGVAlignmentLog2(const GlobalValue *GV, const DataLayout &DL,
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
unsigned InBits = 0) {
unsigned NumBits = 0;
if (const GlobalVariable *GVar = dyn_cast<GlobalVariable>(GV))
DataLayout is mandatory, update the API to reflect it with references. Summary: Now that the DataLayout is a mandatory part of the module, let's start cleaning the codebase. This patch is a first attempt at doing that. This patch is not exactly NFC as for instance some places were passing a nullptr instead of the DataLayout, possibly just because there was a default value on the DataLayout argument to many functions in the API. Even though it is not purely NFC, there is no change in the validation. I turned as many pointer to DataLayout to references, this helped figuring out all the places where a nullptr could come up. I had initially a local version of this patch broken into over 30 independant, commits but some later commit were cleaning the API and touching part of the code modified in the previous commits, so it seemed cleaner without the intermediate state. Test Plan: Reviewers: echristo Subscribers: llvm-commits From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 231740
2015-03-10 10:37:25 +08:00
NumBits = DL.getPreferredAlignmentLog(GVar);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
// If InBits is specified, round it to it.
if (InBits > NumBits)
NumBits = InBits;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
// If the GV has a specified alignment, take it into account.
if (GV->getAlignment() == 0)
return NumBits;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
unsigned GVAlign = Log2_32(GV->getAlignment());
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
// If the GVAlign is larger than NumBits, or if we are required to obey
// NumBits because the GV has an assigned section, obey it.
if (GVAlign > NumBits || GV->hasSection())
NumBits = GVAlign;
return NumBits;
}
std::unique_ptrify the MCStreamer argument to createAsmPrinter llvm-svn: 226414
2015-01-19 04:29:04 +08:00
AsmPrinter::AsmPrinter(TargetMachine &tm, std::unique_ptr<MCStreamer> Streamer)
Remove the TargetMachine forwards for TargetSubtargetInfo based information and update all callers. No functional change. llvm-svn: 214781
2014-08-05 05:25:23 +08:00
: MachineFunctionPass(ID), TM(tm), MAI(tm.getMCAsmInfo()),
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
OutContext(Streamer->getContext()), OutStreamer(std::move(Streamer)) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
VerboseAsm = OutStreamer->isVerboseAsm();
CodeGen still defaults to non-verbose asm, but llc now overrides it and default to verbose. llvm-svn: 67668
2009-03-25 09:47:28 +08:00
}
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
Factor GC metadata table assembly generation out of Collector in preparation for splitting AsmPrinter into its own library. llvm-svn: 54880
2008-08-17 20:08:44 +08:00
AsmPrinter::~AsmPrinter() {
[C++] Use 'nullptr'. llvm-svn: 207394
2014-04-28 12:05:08 +08:00
assert(!DD && Handlers.empty() && "Debug/EH info didn't get finalized");
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (GCMetadataPrinters) {
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
gcp_map_type &GCMap = getGCMap(GCMetadataPrinters);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
delete &GCMap;
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
GCMetadataPrinters = nullptr;
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
}
Factor GC metadata table assembly generation out of Collector in preparation for splitting AsmPrinter into its own library. llvm-svn: 54880
2008-08-17 20:08:44 +08:00
}
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
Move isPositionIndependent up to AsmPrinter. Use it in ppc too. llvm-svn: 273877
2016-06-27 22:19:45 +08:00
bool AsmPrinter::isPositionIndependent() const {
Use isPositionIndependent in a few more places. I think this converts all the simple cases that really just care about the generated code being position independent or not. The remaining uses are a bit more complicated and are checking things like "is this a library or executable" or "can this symbol be preempted". llvm-svn: 274055
2016-06-29 04:13:36 +08:00
return TM.isPositionIndependent();
Move isPositionIndependent up to AsmPrinter. Use it in ppc too. llvm-svn: 273877
2016-06-27 22:19:45 +08:00
}
make MachineFunction keep track of its ID and make MachineFunctionAnalysis dole them out, instead of having AsmPrinter do both. Have the AsmPrinter::SetupMachineFunction method set the 'AsmPrinter::MF' variable. llvm-svn: 94509
2010-01-26 12:35:26 +08:00
/// getFunctionNumber - Return a unique ID for the current function.
///
unsigned AsmPrinter::getFunctionNumber() const {
return MF->getFunctionNumber();
}
Add const qualifiers to TargetLoweringObjectFile usage. llvm-svn: 101640
2010-04-18 00:44:48 +08:00
const TargetLoweringObjectFile &AsmPrinter::getObjFileLowering() const {
Only access TLOF via the TargetMachine, not TargetLowering. llvm-svn: 227949
2015-02-03 15:22:52 +08:00
return *TM.getObjFileLowering();
Rip all of the global variable lowering logic out of TargetAsmInfo. Since it is highly specific to the object file that will be generated in the end, this introduces a new TargetLoweringObjectFile interface that is implemented for each of ELF/MachO/COFF/Alpha/PIC16 and XCore. Though still is still a brutal and ugly refactoring, this is a major step towards goodness. This patch also: 1. fixes a bunch of dangling pointer problems in the PIC16 backend. 2. disables the TargetLowering copy ctor which PIC16 was accidentally using. 3. gets us closer to xcore having its own crazy target section flags and pic16 not having to shadow sections with its own objects. 4. fixes wierdness where ELF targets would set CStringSection but not CStringSection_. Factor the code better. 5. fixes some bugs in string lowering on ELF targets. llvm-svn: 77294
2009-07-28 11:13:23 +08:00
}
Move TargetData to DataLayout. llvm-svn: 165402
2012-10-09 00:38:25 +08:00
const DataLayout &AsmPrinter::getDataLayout() const {
Remove DataLayout from TargetLoweringObjectFile, redirect to Module Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11079 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242385
2015-07-16 14:04:17 +08:00
return MMI->getModule()->getDataLayout();
1) make DIE take AsmPrinter instead of DwarfPrinter. 2) change DwarfDebug to not inherit from DwarfPrinter. llvm-svn: 100372
2010-04-05 08:13:49 +08:00
}
Remove access to the DataLayout in the TargetMachine Summary: Replace getDataLayout() with a createDataLayout() method to make explicit that it is intended to create a DataLayout only and not accessing it for other purpose. This change is the last of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: jholewinski, llvm-commits, rafael, yaron.keren Differential Revision: http://reviews.llvm.org/D11103 (cherry picked from commit 5609fc56bca971e5a7efeaa6ca4676638eaec5ea) From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 243114
2015-07-25 00:04:22 +08:00
// Do not use the cached DataLayout because some client use it without a Module
Remove getTargetTriple and update all uses to use the Triple off of the TargetMachine. NFC. llvm-svn: 283002
2016-10-01 09:50:33 +08:00
// (llvm-dsymutil, llvm-dwarfdump).
Remove access to the DataLayout in the TargetMachine Summary: Replace getDataLayout() with a createDataLayout() method to make explicit that it is intended to create a DataLayout only and not accessing it for other purpose. This change is the last of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: jholewinski, llvm-commits, rafael, yaron.keren Differential Revision: http://reviews.llvm.org/D11103 (cherry picked from commit 5609fc56bca971e5a7efeaa6ca4676638eaec5ea) From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 243114
2015-07-25 00:04:22 +08:00
unsigned AsmPrinter::getPointerSize() const { return TM.getPointerSize(); }
Redirect pointerSize query to the TargetMachine in ASMPrinter Summary: Because llvm-dsymutil is using ASMPrinter without any MachineFunction of Module available. This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits Differential Revision: http://reviews.llvm.org/D11078 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242384
2015-07-16 13:59:25 +08:00
Change MCStreamer EmitInstruction interface to take subtarget info llvm-svn: 200345
2014-01-29 07:12:42 +08:00
const MCSubtargetInfo &AsmPrinter::getSubtargetInfo() const {
Remove more uses of TargetMachine::getSubtargetImpl from the AsmPrinter. getSubtargetInfo now asserts that the MachineFunction exists. Debug printing of register naming now uses the register info from MCAsmInfo as that's unchanging. llvm-svn: 229978
2015-02-20 15:16:19 +08:00
assert(MF && "getSubtargetInfo requires a valid MachineFunction!");
return MF->getSubtarget<MCSubtargetInfo>();
Change MCStreamer EmitInstruction interface to take subtarget info llvm-svn: 200345
2014-01-29 07:12:42 +08:00
}
void AsmPrinter::EmitToStreamer(MCStreamer &S, const MCInst &Inst) {
S.EmitInstruction(Inst, getSubtargetInfo());
}
Make AsmStreamer maintain a notion of the current section, pushing it up from the MCAsmStreamer. Based on this, eliminate the current section from AsmPrinter. While I'm at it, clean up the last of the horrible "switch to null section" stuff and add an assert. This change is in preparation for completely eliminating asmprinter::switchtosection. llvm-svn: 79324
2009-08-18 14:15:16 +08:00
/// getCurrentSection() - Return the current section we are emitting to.
const MCSection *AsmPrinter::getCurrentSection() const {
Tidy the calls to getCurrentSection().first -> getCurrentSectionOnly to help readability a bit. llvm-svn: 284202
2016-10-14 13:47:37 +08:00
return OutStreamer->getCurrentSectionOnly();
Provide direct function to switch to Section llvm-svn: 56571
2008-09-25 06:12:10 +08:00
}
Split SwitchSection into SwitchTo{Text|Data}Section methods. llvm-svn: 28184
2006-05-09 12:59:56 +08:00
Enabling the target-independent garbage collection infrastructure by hooking it up to the various compiler pipelines. This doesn't actually add support for any GC algorithms, which means it temporarily breaks a few tests. To be fixed shortly. llvm-svn: 45669
2008-01-07 09:30:38 +08:00
void AsmPrinter::getAnalysisUsage(AnalysisUsage &AU) const {
Use setPreservesAll and setPreservesCFG in CodeGen passes. llvm-svn: 77754
2009-08-01 07:37:33 +08:00
AU.setPreservesAll();
Enabling the target-independent garbage collection infrastructure by hooking it up to the various compiler pipelines. This doesn't actually add support for any GC algorithms, which means it temporarily breaks a few tests. To be fixed shortly. llvm-svn: 45669
2008-01-07 09:30:38 +08:00
MachineFunctionPass::getAnalysisUsage(AU);
rearrange MCContext ownership. Before LLVMTargetMachine created it and passing off ownership to AsmPrinter. Now MachineModuleInfo creates it and owns it by value. This allows us to use MCSymbols more consistently throughout the rest of the code generator, and simplifies a bit of code. This also allows MachineFunction to keep an MCContext reference handy, and cleans up the TargetRegistry interfaces for AsmPrinters. llvm-svn: 98450
2010-03-14 04:55:24 +08:00
AU.addRequired<MachineModuleInfo>();
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
AU.addRequired<MachineOptimizationRemarkEmitterPass>();
Rename some GC classes so that their roll will hopefully be clearer. In particular, Collector was confusing to implementors. Several thought that this compile-time class was the place to implement their runtime GC heap. Of course, it doesn't even exist at runtime. Specifically, the renames are: Collector -> GCStrategy CollectorMetadata -> GCFunctionInfo CollectorModuleMetadata -> GCModuleInfo CollectorRegistry -> GCRegistry Function::getCollector -> getGC (setGC, hasGC, clearGC) Several accessors and nested types have also been renamed to be consistent. These changes should be obvious. llvm-svn: 54899
2008-08-18 02:44:35 +08:00
AU.addRequired<GCModuleInfo>();
clean up the asmprinter header and privatize some stuff. llvm-svn: 100342
2010-04-05 02:52:31 +08:00
if (isVerbose())
Add support for printing loop structure information in asm comments. This definitely slows down asm output so put it under an -asm-exuberant flag. This information is useful when doing static analysis of performance issues. llvm-svn: 78567
2009-08-11 00:38:07 +08:00
AU.addRequired<MachineLoopInfo>();
Enabling the target-independent garbage collection infrastructure by hooking it up to the various compiler pipelines. This doesn't actually add support for any GC algorithms, which means it temporarily breaks a few tests. To be fixed shortly. llvm-svn: 45669
2008-01-07 09:30:38 +08:00
}
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
bool AsmPrinter::doInitialization(Module &M) {
change this back too llvm-svn: 100310
2010-04-04 13:09:10 +08:00
MMI = getAnalysisIfAvailable<MachineModuleInfo>();
rearrange MCContext ownership. Before LLVMTargetMachine created it and passing off ownership to AsmPrinter. Now MachineModuleInfo creates it and owns it by value. This allows us to use MCSymbols more consistently throughout the rest of the code generator, and simplifies a bit of code. This also allows MachineFunction to keep an MCContext reference handy, and cleans up the TargetRegistry interfaces for AsmPrinters. llvm-svn: 98450
2010-03-14 04:55:24 +08:00
refactor section construction in TLOF to be through an explicit initialize method, which can be called when an MCContext is available. llvm-svn: 77687
2009-08-01 01:42:42 +08:00
// Initialize TargetLoweringObjectFile.
const_cast<TargetLoweringObjectFile&>(getObjFileLowering())
.Initialize(OutContext, TM);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->InitSections(false);
Initialize object file info before output streamer r179494 switched to using the object file info to retrieve the default text section for some MC streamers. It is possible that initializing an MC streamer can request sections before the object file info is initialized when the AutoInitSections flag is set on the streamer. llvm-svn: 185670
2013-07-05 05:37:26 +08:00
MachO: Emit a version-min load command when possible. When deployment target version information is available, emit it to the target streamer for inclusion in the object file. rdar://11337778 llvm-svn: 204191
2014-03-19 06:09:08 +08:00
// Emit the version-min deplyment target directive if needed.
//
// FIXME: If we end up with a collection of these sorts of Darwin-specific
// or ELF-specific things, it may make sense to have a platform helper class
// that will work with the target helper class. For now keep it here, as the
// alternative is duplicated code in each of the target asm printers that
// use the directive, where it would need the same conditionalization
// anyway.
Remove getTargetTriple and update all uses to use the Triple off of the TargetMachine. NFC. llvm-svn: 283002
2016-10-01 09:50:33 +08:00
const Triple &TT = TM.getTargetTriple();
AsmPrinter: Fix wrong OS X versions being emitted for darwin triples The version numbers of the darwin kernel are different from the version numbers of OS X, so we need adjustments if we had "*-*-darwin" triples. Use the existing utility functions in TargetTriple for this. Fixes rdar://22056966 Differential Revision: http://reviews.llvm.org/D14601 llvm-svn: 257555
2016-01-13 09:18:13 +08:00
// If there is a version specified, Major will be non-zero.
if (TT.isOSDarwin() && TT.getOSMajorVersion() != 0) {
MachO: Emit a version-min load command when possible. When deployment target version information is available, emit it to the target streamer for inclusion in the object file. rdar://11337778 llvm-svn: 204191
2014-03-19 06:09:08 +08:00
unsigned Major, Minor, Update;
AsmPrinter: Fix wrong OS X versions being emitted for darwin triples The version numbers of the darwin kernel are different from the version numbers of OS X, so we need adjustments if we had "*-*-darwin" triples. Use the existing utility functions in TargetTriple for this. Fixes rdar://22056966 Differential Revision: http://reviews.llvm.org/D14601 llvm-svn: 257555
2016-01-13 09:18:13 +08:00
MCVersionMinType VersionType;
if (TT.isWatchOS()) {
VersionType = MCVM_WatchOSVersionMin;
TT.getWatchOSVersion(Major, Minor, Update);
} else if (TT.isTvOS()) {
VersionType = MCVM_TvOSVersionMin;
TT.getiOSVersion(Major, Minor, Update);
} else if (TT.isMacOSX()) {
VersionType = MCVM_OSXVersionMin;
if (!TT.getMacOSXVersion(Major, Minor, Update))
Major = 0;
} else {
VersionType = MCVM_IOSVersionMin;
TT.getiOSVersion(Major, Minor, Update);
ARM: support .watchos_version_min and .tvos_version_min. These MachO file directives are used by linkers and other tools to provide compatibility information, much like the existing .ios_version_min and .macosx_version_min. llvm-svn: 251569
2015-10-29 06:36:05 +08:00
}
AsmPrinter: Fix wrong OS X versions being emitted for darwin triples The version numbers of the darwin kernel are different from the version numbers of OS X, so we need adjustments if we had "*-*-darwin" triples. Use the existing utility functions in TargetTriple for this. Fixes rdar://22056966 Differential Revision: http://reviews.llvm.org/D14601 llvm-svn: 257555
2016-01-13 09:18:13 +08:00
if (Major != 0)
OutStreamer->EmitVersionMin(VersionType, Major, Minor, Update);
MachO: Emit a version-min load command when possible. When deployment target version information is available, emit it to the target streamer for inclusion in the object file. rdar://11337778 llvm-svn: 204191
2014-03-19 06:09:08 +08:00
}
Add a new virtual EmitStartOfAsmFile method to the AsmPrinter and use this to emit target-specific things at the beginning of the asm output. This fixes a problem for PPC, where the text sections are not being kept together as expected. The base class doInitialization code calls DW->BeginModule() which emits a bunch of DWARF section directives. The PPC doInitialization code then emits all the TEXT section directives, with the intention that they will be kept together. But as I understand it, the Darwin assembler treats the default TEXT section as a special case and moves it to the beginning of the file, which means that all those DWARF sections are in the middle of the text. With this change, the EmitStartOfAsmFile hook is called before the DWARF section directives are emitted, so that all the PPC text section directives come out right at the beginning of the file. llvm-svn: 83176
2009-10-01 06:06:26 +08:00
// Allow the target to emit any magic that it wants at the start of the file.
EmitStartOfAsmFile(M);
Fix bug 3140. Print a single parameter .file directive if we have an ELF target. llvm-svn: 60480
2008-12-03 19:01:37 +08:00
mcstreamerize .file and .file. This also fixes an issue where the normal form of .file would fail if the filename had a weird character in it. llvm-svn: 94437
2010-01-26 02:58:59 +08:00
// Very minimal debug info. It is ignored if we emit actual debug info. If we
// don't, this at least helps the user find where a global came from.
rename TAI -> MAI, being careful not to make MAILJMP instructions :) llvm-svn: 79777
2009-08-23 05:43:10 +08:00
if (MAI->hasSingleParameterDotFile()) {
mcstreamerize .file and .file. This also fixes an issue where the normal form of .file would fail if the filename had a weird character in it. llvm-svn: 94437
2010-01-26 02:58:59 +08:00
// .file "foo.c"
CodeGen: Use the source filename as the argument to .file, rather than the module ID. Using the module ID here is wrong for a couple of reasons: 1) The module ID is not persisted, so we can end up with different object file contents given the same input file (for example if the same file is accessed via different paths). 2) With ThinLTO the module ID field may contain the path to a bitcode file, which is incorrect, as the .file argument is supposed to contain the path to a source file. Differential Revision: https://reviews.llvm.org/D30584 llvm-svn: 297853
2017-03-16 00:24:52 +08:00
OutStreamer->EmitFileDirective(M.getSourceFileName());
Fix bug 3140. Print a single parameter .file directive if we have an ELF target. llvm-svn: 60480
2008-12-03 19:01:37 +08:00
}
Add a new virtual EmitStartOfAsmFile method to the AsmPrinter and use this to emit target-specific things at the beginning of the asm output. This fixes a problem for PPC, where the text sections are not being kept together as expected. The base class doInitialization code calls DW->BeginModule() which emits a bunch of DWARF section directives. The PPC doInitialization code then emits all the TEXT section directives, with the intention that they will be kept together. But as I understand it, the Darwin assembler treats the default TEXT section as a special case and moves it to the beginning of the file, which means that all those DWARF sections are in the middle of the text. With this change, the EmitStartOfAsmFile hook is called before the DWARF section directives are emitted, so that all the PPC text section directives come out right at the beginning of the file. llvm-svn: 83176
2009-10-01 06:06:26 +08:00
GCModuleInfo *MI = getAnalysisIfAvailable<GCModuleInfo>();
assert(MI && "AsmPrinter didn't require GCModuleInfo?");
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (auto &I : *MI)
if (GCMetadataPrinter *MP = GetOrCreateGCPrinter(*I))
GCStrategy should not own GCFunctionInfo This change moves the ownership and access of GCFunctionInfo (the object which describes the safepoints associated with a safepoint under GCRoot) to GCModuleInfo. Previously, this was owned by GCStrategy which was in turned owned by GCModuleInfo. This made GCStrategy module specific which is 'surprising' given it's name and other purposes. There's a few more changes needed, but we're getting towards the point we can reuse GCStrategy for gc.statepoint as well. p.s. The style of this code ends up being a mess. I was trying to move code around without otherwise changing much. Once I get the ownership structure rearranged, I will go through and fixup spacing, naming, comments etc. Differential Revision: http://reviews.llvm.org/D6587 llvm-svn: 223994
2014-12-11 09:47:23 +08:00
MP->beginAssembly(M, *MI, *this);
add a new EmitInlineAsm function to asmprinter to handle inline asm. If we have an MCAsmStreamer, we continue to emit asm textually, otherwise we (currently) emit an error to errs and ignore it. llvm-svn: 100289
2010-04-04 05:35:55 +08:00
// Emit module-level inline asm if it exists.
mc'ize comment printing around file scope inline asm. llvm-svn: 100288
2010-04-04 05:13:18 +08:00
if (!M.getModuleInlineAsm().empty()) {
[AsmPrinter] Use the per-function subtarget to emit inline asm instructions that are not at the file level. Previously, the default subtarget created from the target triple was used to emit inline asm instructions. Compilation would fail in cases where the feature bits necessary to assemble an inline asm instruction in a function weren't set. llvm-svn: 232392
2015-03-17 02:02:16 +08:00
// We're at the module level. Construct MCSubtarget from the default CPU
// and target triple.
std::unique_ptr<MCSubtargetInfo> STI(TM.getTarget().createMCSubtargetInfo(
Replace string GNU Triples with llvm::Triple in TargetMachine::getTargetTriple(). NFC. Summary: This continues the patch series to eliminate StringRef forms of GNU triples from the internals of LLVM that began in r239036. Reviewers: rengolin Reviewed By: rengolin Subscribers: llvm-commits, rengolin Differential Revision: http://reviews.llvm.org/D10381 llvm-svn: 239815
2015-06-16 21:15:50 +08:00
TM.getTargetTriple().str(), TM.getTargetCPU(),
TM.getTargetFeatureString()));
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("Start of file scope inline assembly");
OutStreamer->AddBlankLine();
Reduce the size of MCRelaxableFragment. MCRelaxableFragment previously kept a copy of MCSubtargetInfo and MCInst to enable re-encoding the MCInst later during relaxation. A copy of MCSubtargetInfo (instead of a reference or pointer) was needed because the feature bits could be modified by the parser. This commit replaces the MCSubtargetInfo copy in MCRelaxableFragment with a constant reference to MCSubtargetInfo. The copies of MCSubtargetInfo are kept in MCContext, and the target parsers are now responsible for asking MCContext to provide a copy whenever the feature bits of MCSubtargetInfo have to be toggled. With this patch, I saw a 4% reduction in peak memory usage when I compiled verify-uselistorder.lto.bc using llc. rdar://problem/21736951 Differential Revision: http://reviews.llvm.org/D14346 llvm-svn: 253127
2015-11-14 14:35:56 +08:00
EmitInlineAsm(M.getModuleInlineAsm()+"\n",
OutContext.getSubtargetCopy(*STI), TM.Options.MCOptions);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("End of file scope inline assembly");
OutStreamer->AddBlankLine();
mc'ize comment printing around file scope inline asm. llvm-svn: 100288
2010-04-04 05:13:18 +08:00
}
Print file-scope inline asm blocks at the start of the output file. llvm-svn: 25565
2006-01-24 07:47:53 +08:00
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
if (MAI->doesSupportDebugInformation()) {
If the "CodeView" module flag is set, emit codeview instead of DWARF Summary: Emit both DWARF and CodeView if "CodeView" and "Dwarf Version" module flags are set. Reviewers: majnemer Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D11756 llvm-svn: 244158
2015-08-06 06:26:20 +08:00
bool EmitCodeView = MMI->getModule()->getCodeViewFlag();
[AsmPrinter] Enable codeview for windows-itanium Enable codeview emission for windows-itanium targets. Co-opt an existing test (which is derived from a C source file and should therefore be identical across the Itanium and MS ABIs). Differential Revision: https://reviews.llvm.org/D26693 llvm-svn: 287567
2016-11-22 04:13:32 +08:00
if (EmitCodeView && (TM.getTargetTriple().isKnownWindowsMSVCEnvironment() ||
TM.getTargetTriple().isWindowsItaniumEnvironment())) {
Rename WinCodeViewLineTables to CodeViewDebug, similar to DwarfDebug Soon it will be responsible for more than line tables. Reviewers: majnemer Differential Revision: http://reviews.llvm.org/D16199 llvm-svn: 257792
2016-01-15 03:25:04 +08:00
Handlers.push_back(HandlerInfo(new CodeViewDebug(this),
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
DbgTimerName, DbgTimerDescription,
CodeViewLineTablesGroupName,
CodeViewLineTablesGroupDescription));
Band-aid fix for PR22032: don't emit DWARF debug info if AddressSanitizer is enabled on Windows llvm-svn: 224860
2014-12-27 01:00:51 +08:00
}
If the "CodeView" module flag is set, emit codeview instead of DWARF Summary: Emit both DWARF and CodeView if "CodeView" and "Dwarf Version" module flags are set. Reviewers: majnemer Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D11756 llvm-svn: 244158
2015-08-06 06:26:20 +08:00
if (!EmitCodeView || MMI->getModule()->getDwarfVersion()) {
Band-aid fix for PR22032: don't emit DWARF debug info if AddressSanitizer is enabled on Windows llvm-svn: 224860
2014-12-27 01:00:51 +08:00
DD = new DwarfDebug(this, &M);
CodeGen: Move the call to DwarfDebug::beginModule() out of the constructor. This gives AsmPrinter a chance to initialize its DD field before we call beginModule(), which is about to start using it. Differential Revision: http://reviews.llvm.org/D20413 llvm-svn: 270258
2016-05-21 03:35:35 +08:00
DD->beginModule();
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
Handlers.push_back(HandlerInfo(DD, DbgTimerName, DbgTimerDescription,
DWARFGroupName, DWARFGroupDescription));
Band-aid fix for PR22032: don't emit DWARF debug info if AddressSanitizer is enabled on Windows llvm-svn: 224860
2014-12-27 01:00:51 +08:00
}
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
}
Split stuff as a preparation for CFI directives-based frame information emission llvm-svn: 123473
2011-01-15 05:57:45 +08:00
PR 31534: When emitting both DWARF unwind tables and debug information, do not use .cfi_sections. This requires checking if any non-declaration function in the module needs an unwind table. llvm-svn: 291172
2017-01-06 04:55:28 +08:00
switch (MAI->getExceptionHandlingType()) {
case ExceptionHandling::SjLj:
case ExceptionHandling::DwarfCFI:
case ExceptionHandling::ARM:
isCFIMoveForDebugging = true;
if (MAI->getExceptionHandlingType() != ExceptionHandling::DwarfCFI)
break;
for (auto &F: M.getFunctionList()) {
// If the module contains any function with unwind data,
// .eh_frame has to be emitted.
// Ignore functions that won't get emitted.
if (!F.isDeclarationForLinker() && F.needsUnwindTableEntry()) {
isCFIMoveForDebugging = false;
break;
}
}
break;
default:
isCFIMoveForDebugging = false;
break;
}
CodeGen: refactor DwarfException DwarfException served as a base class for exception handling directive emission. However, this is also used by other exception models (e.g. Win64EH). Rename this class to EHStreamer and split it out of DwarfException.h. NFC. Use the opportunity to fix up some of the documentation comments to match current LLVM style. Also rename some functions to conform better with current LLVM coding style. llvm-svn: 210622
2014-06-11 09:19:03 +08:00
EHStreamer *ES = nullptr;
List all exception types in a switch. llvm-svn: 130944
2011-05-06 03:48:34 +08:00
switch (MAI->getExceptionHandlingType()) {
case ExceptionHandling::None:
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
break;
List all exception types in a switch. llvm-svn: 130944
2011-05-06 03:48:34 +08:00
case ExceptionHandling::SjLj:
case ExceptionHandling::DwarfCFI:
CodeGen: refactor DwarfException DwarfException served as a base class for exception handling directive emission. However, this is also used by other exception models (e.g. Win64EH). Rename this class to EHStreamer and split it out of DwarfException.h. NFC. Use the opportunity to fix up some of the documentation comments to match current LLVM style. Also rename some functions to conform better with current LLVM coding style. llvm-svn: 210622
2014-06-11 09:19:03 +08:00
ES = new DwarfCFIException(this);
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
break;
List all exception types in a switch. llvm-svn: 130944
2011-05-06 03:48:34 +08:00
case ExceptionHandling::ARM:
CodeGen: refactor DwarfException DwarfException served as a base class for exception handling directive emission. However, this is also used by other exception models (e.g. Win64EH). Rename this class to EHStreamer and split it out of DwarfException.h. NFC. Use the opportunity to fix up some of the documentation comments to match current LLVM style. Also rename some functions to conform better with current LLVM coding style. llvm-svn: 210622
2014-06-11 09:19:03 +08:00
ES = new ARMException(this);
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
break;
Classify functions by EH personality type rather than using the triple This mostly reverts commit r222062 and replaces it with a new enum. At some point this enum will grow at least for other MSVC EH personalities. Also beefs up the way we were sniffing the personality function. Previously we would emit the Itanium LSDA despite using __C_specific_handler. Reviewers: majnemer Differential Revision: http://reviews.llvm.org/D6987 llvm-svn: 226920
2015-01-24 02:49:01 +08:00
case ExceptionHandling::WinEH:
CodeGen: indicate Windows unwind data format The structures for Windows unwinding are shared across multiple platforms. Indicate the encoding to be used for the particular target. Use this to switch the unwind emitter instantiated by the AsmPrinter. llvm-svn: 216895
2014-09-02 07:48:39 +08:00
switch (MAI->getWinEHEncodingType()) {
default: llvm_unreachable("unsupported unwinding information encoding");
Re-land "[WinEH] Add an EH registration and state insertion pass for 32-bit x86" This reverts commit r236360. This change exposed a bug in WinEHPrepare by opting win32 code into EH preparation. We already knew that WinEHPrepare has bugs, and is the status quo for x64, so I don't think that's a reason to hold off on this change. I disabled exceptions in the sanitizer tests in r236505 and an earlier revision. llvm-svn: 236508
2015-05-06 01:44:16 +08:00
case WinEH::EncodingType::Invalid:
break;
[WinEH] Emit EH tables for __CxxFrameHandler3 on 32-bit x86 Small (really small!) C++ exception handling examples work on 32-bit x86 now. This change disables the use of .seh_* directives in WinException when CFI is not in use. It also uses absolute symbol references in the tables instead of imagerel32 relocations. Also fixes a cache invalidation bug in MMI personality classification. llvm-svn: 238575
2015-05-30 01:00:57 +08:00
case WinEH::EncodingType::X86:
CodeGen: indicate Windows unwind data format The structures for Windows unwinding are shared across multiple platforms. Indicate the encoding to be used for the particular target. Use this to switch the unwind emitter instantiated by the AsmPrinter. llvm-svn: 216895
2014-09-02 07:48:39 +08:00
case WinEH::EncodingType::Itanium:
Rename Win64Exception.(cpp|h) to WinException.(cpp|h) This is in preparation for reusing this for 32-bit x86 EH table emission. Also updates the type name for consistency. NFC llvm-svn: 238521
2015-05-29 06:47:01 +08:00
ES = new WinException(this);
CodeGen: indicate Windows unwind data format The structures for Windows unwinding are shared across multiple platforms. Indicate the encoding to be used for the particular target. Use this to switch the unwind emitter instantiated by the AsmPrinter. llvm-svn: 216895
2014-09-02 07:48:39 +08:00
break;
}
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
break;
List all exception types in a switch. llvm-svn: 130944
2011-05-06 03:48:34 +08:00
}
CodeGen: refactor DwarfException DwarfException served as a base class for exception handling directive emission. However, this is also used by other exception models (e.g. Win64EH). Rename this class to EHStreamer and split it out of DwarfException.h. NFC. Use the opportunity to fix up some of the documentation comments to match current LLVM style. Also rename some functions to conform better with current LLVM coding style. llvm-svn: 210622
2014-06-11 09:19:03 +08:00
if (ES)
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
Handlers.push_back(HandlerInfo(ES, EHTimerName, EHTimerDescription,
DWARFGroupName, DWARFGroupDescription));
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
return false;
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
}
Refactor logic into a function predicate. No functionality change. llvm-svn: 200976
2014-02-08 00:07:11 +08:00
static bool canBeHidden(const GlobalValue *GV, const MCAsmInfo &MAI) {
if (!MAI.hasWeakDefCanBeHiddenDirective())
return false;
Refactor duplicated code. llvm-svn: 214328
2014-07-31 03:42:16 +08:00
return canBeOmittedFromSymbolTable(GV);
Refactor logic into a function predicate. No functionality change. llvm-svn: 200976
2014-02-08 00:07:11 +08:00
}
Produce .weak_def_can_be_hidden for some linkonce_odr values With this patch llvm produces a weak_def_can_be_hidden for linkonce_odr if they are also unnamed_addr or don't have their address taken. There is not a lot of documentation about .weak_def_can_be_hidden, but from the old discussion about linkonce_odr_auto_hide and the name of the directive this looks correct: these symbols can be hidden. Testing this with the ld64 in Xcode 5 linking clang reduces the number of exported symbols from 21053 to 19049. llvm-svn: 193718
2013-10-31 06:08:11 +08:00
void AsmPrinter::EmitLinkage(const GlobalValue *GV, MCSymbol *GVSym) const {
GlobalValue::LinkageTypes Linkage = GV->getLinkage();
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
switch (Linkage) {
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
case GlobalValue::CommonLinkage:
case GlobalValue::LinkOnceAnyLinkage:
case GlobalValue::LinkOnceODRLinkage:
case GlobalValue::WeakAnyLinkage:
case GlobalValue::WeakODRLinkage:
Convert two char* that are only ever used as booleans to bool. llvm-svn: 196168
2013-12-03 07:04:51 +08:00
if (MAI->hasWeakDefDirective()) {
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
// .globl _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_Global);
Create the new linker type "linker_private_weak_def_auto". It's similar to "linker_private_weak", but it's known that the address of the object is not taken. For instance, functions that had an inline definition, but the compiler decided not to inline it. Note, unlike linker_private and linker_private_weak, linker_private_weak_def_auto may have only default visibility. The symbols are removed by the linker from the final linked image (executable or dynamic library). llvm-svn: 111684
2010-08-21 06:05:50 +08:00
Refactor logic into a function predicate. No functionality change. llvm-svn: 200976
2014-02-08 00:07:11 +08:00
if (!canBeHidden(GV, *MAI))
Create the new linker type "linker_private_weak_def_auto". It's similar to "linker_private_weak", but it's known that the address of the object is not taken. For instance, functions that had an inline definition, but the compiler decided not to inline it. Note, unlike linker_private and linker_private_weak, linker_private_weak_def_auto may have only default visibility. The symbols are removed by the linker from the final linked image (executable or dynamic library). llvm-svn: 111684
2010-08-21 06:05:50 +08:00
// .weak_definition _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_WeakDefinition);
Create the new linker type "linker_private_weak_def_auto". It's similar to "linker_private_weak", but it's known that the address of the object is not taken. For instance, functions that had an inline definition, but the compiler decided not to inline it. Note, unlike linker_private and linker_private_weak, linker_private_weak_def_auto may have only default visibility. The symbols are removed by the linker from the final linked image (executable or dynamic library). llvm-svn: 111684
2010-08-21 06:05:50 +08:00
else
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_WeakDefAutoPrivate);
Convert two char* that are only ever used as booleans to bool. llvm-svn: 196168
2013-12-03 07:04:51 +08:00
} else if (MAI->hasLinkOnceDirective()) {
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
// .globl _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_Global);
Remove unused variable. Tweak a comment while there. llvm-svn: 103586
2010-05-12 15:11:33 +08:00
//NOTE: linkonce is handled by the section the symbol was assigned to.
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
} else {
// .weak _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_Weak);
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
}
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
return;
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
case GlobalValue::ExternalLinkage:
Fail early on unknown appending linkage variables. In practice only a few well known appending linkage variables work. Currently if codegen sees an unknown appending linkage variable it will just print it as a regular global. That is wrong as the symbol in the produced object file has different semantics as the one provided by the appending linkage. This just errors early instead of producing a broken .o. llvm-svn: 269706
2016-05-17 05:14:24 +08:00
// If external, declare as a global symbol: .globl _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_Global);
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
return;
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
case GlobalValue::PrivateLinkage:
case GlobalValue::InternalLinkage:
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
return;
Fail early on unknown appending linkage variables. In practice only a few well known appending linkage variables work. Currently if codegen sees an unknown appending linkage variable it will just print it as a regular global. That is wrong as the symbol in the produced object file has different semantics as the one provided by the appending linkage. This just errors early instead of producing a broken .o. llvm-svn: 269706
2016-05-17 05:14:24 +08:00
case GlobalValue::AppendingLinkage:
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
case GlobalValue::AvailableExternallyLinkage:
case GlobalValue::ExternalWeakLinkage:
Merge two unreachable cases. llvm-svn: 269189
2016-05-11 22:41:30 +08:00
llvm_unreachable("Should never emit this");
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
}
Reduce casting and use a fully covered switch. llvm-svn: 193272
2013-10-24 05:24:34 +08:00
llvm_unreachable("Unknown linkage type!");
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
}
Add back r201608, r201622, r201624 and r201625 r201608 made llvm corretly handle private globals with MachO. r201622 fixed a bug in it and r201624 and r201625 were changes for using private linkage, assuming that llvm would do the right thing. They all got reverted because r201608 introduced a crash in LTO. This patch includes a fix for that. The issue was that TargetLoweringObjectFile now has to be initialized before we can mangle names of private globals. This is trivially true during the normal codegen pipeline (the asm printer does it), but LTO has to do it manually. llvm-svn: 201700
2014-02-20 01:23:20 +08:00
void AsmPrinter::getNameWithPrefix(SmallVectorImpl<char> &Name,
const GlobalValue *GV) const {
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
TM.getNameWithPrefix(Name, GV, getObjFileLowering().getMangler());
Add back r201608, r201622, r201624 and r201625 r201608 made llvm corretly handle private globals with MachO. r201622 fixed a bug in it and r201624 and r201625 were changes for using private linkage, assuming that llvm would do the right thing. They all got reverted because r201608 introduced a crash in LTO. This patch includes a fix for that. The issue was that TargetLoweringObjectFile now has to be initialized before we can mangle names of private globals. This is trivially true during the normal codegen pipeline (the asm printer does it), but LTO has to do it manually. llvm-svn: 201700
2014-02-20 01:23:20 +08:00
}
Add a helper getSymbol to AsmPrinter. llvm-svn: 193627
2013-10-30 01:07:16 +08:00
MCSymbol *AsmPrinter::getSymbol(const GlobalValue *GV) const {
CodeGen: simplify TargetMachine::getSymbol interface. NFC. No-one actually had a mangler handy when calling this function, and getSymbol itself went most of the way towards getting its own mangler (with a local TLOF variable) so forcing all callers to supply one was just extra complication. llvm-svn: 287645
2016-11-23 00:17:20 +08:00
return TM.getSymbol(GV);
Add a helper getSymbol to AsmPrinter. llvm-svn: 193627
2013-10-30 01:07:16 +08:00
}
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
hoist handling of external globals and special globals up to common code. This makes a similar code dead in all the other targets, I'll clean it up in a bit. This also moves handling of lcomm up before acquisition of a section, since lcomm never needs a section. llvm-svn: 93851
2010-01-19 12:39:15 +08:00
/// EmitGlobalVariable - Emit the specified global variable to the .s file.
void AsmPrinter::EmitGlobalVariable(const GlobalVariable *GV) {
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
bool IsEmuTLSVar = TM.Options.EmulatedTLS && GV->isThreadLocal();
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
assert(!(IsEmuTLSVar && GV->hasCommonLinkage()) &&
"No emulated TLS variables in the common section");
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
// Never emit TLS variable xyz in emulated TLS model.
// The initialization value is in __emutls_t.xyz instead of xyz.
if (IsEmuTLSVar)
return;
Print visibility info for external variables. llvm-svn: 128887
2011-04-05 23:51:32 +08:00
if (GV->hasInitializer()) {
// Check to see if this is a special global used by LLVM, if so, emit it.
if (EmitSpecialLLVMGlobal(GV))
return;
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Skip the emission of global equivalents. The symbol can be emitted later
// on by emitGlobalGOTEquivs in case it turns out to be needed.
if (GlobalGOTEquivs.count(getSymbol(GV)))
return;
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
if (isVerbose()) {
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
// When printing the control variable __emutls_v.*,
// we don't need to print the original TLS variable name.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
GV->printAsOperand(OutStreamer->GetCommentOS(),
Print visibility info for external variables. llvm-svn: 128887
2011-04-05 23:51:32 +08:00
/*PrintType=*/false, GV->getParent());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->GetCommentOS() << '\n';
Print visibility info for external variables. llvm-svn: 128887
2011-04-05 23:51:32 +08:00
}
Move the verbose asm output up a bit so it can be used in the special cases as well. llvm-svn: 104642
2010-05-26 05:49:43 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add a helper getSymbol to AsmPrinter. llvm-svn: 193627
2013-10-30 01:07:16 +08:00
MCSymbol *GVSym = getSymbol(GV);
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
MCSymbol *EmittedSym = GVSym;
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
80-column and comment fixups. llvm-svn: 274335
2016-07-01 14:07:31 +08:00
// getOrCreateEmuTLSControlSym only creates the symbol with name and default
// attributes.
// GV's or GVSym's attributes will be used for the EmittedSym.
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
EmitVisibility(EmittedSym, GV->getVisibility(), !GV->isDeclaration());
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
Print visibility info for external variables. llvm-svn: 128887
2011-04-05 23:51:32 +08:00
if (!GV->hasInitializer()) // External globals require no extra code.
return;
CodeGen: Allow aliases to be overridden by variables llvm-svn: 224827
2014-12-25 06:44:29 +08:00
GVSym->redefineIfPossible();
CodeGen: Error on redefinitions instead of asserting It's possible to have a prior definition of a symbol in module asm. Raise an error instead of crashing. llvm-svn: 224828
2014-12-25 07:06:55 +08:00
if (GVSym->isDefined() || GVSym->isVariable())
report_fatal_error("symbol '" + Twine(GVSym->getName()) +
"' is already defined");
CodeGen: Allow aliases to be overridden by variables llvm-svn: 224827
2014-12-25 06:44:29 +08:00
emit ELF .type directives through MCStreamer instead of doing it textually. llvm-svn: 94436
2010-01-26 02:33:40 +08:00
if (MAI->hasDotTypeDotSizeDirective())
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
OutStreamer->EmitSymbolAttribute(EmittedSym, MCSA_ELF_TypeObject);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
SectionKind GVKind = TargetLoweringObjectFile::getKindForGlobal(GV, TM);
Remove DataLayout from TargetLoweringObjectFile, redirect to Module Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11079 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242385
2015-07-16 14:04:17 +08:00
const DataLayout &DL = GV->getParent()->getDataLayout();
uint64_t Size = DL.getTypeAllocSize(GV->getType()->getElementType());
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
fix PR6921 a different way. Intead of increasing the alignment of globals with a specified alignment, we fix common variables to obey their alignment. Add a comment explaining why this behavior is important. llvm-svn: 102365
2010-04-27 02:46:46 +08:00
// If the alignment is specified, we *must* obey it. Overaligning a global
// with a specified alignment is a prompt way to break globals emitted to
// sections and expected to be contiguous (e.g. ObjC metadata).
Remove DataLayout from TargetLoweringObjectFile, redirect to Module Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11079 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242385
2015-07-16 14:04:17 +08:00
unsigned AlignLog = getGVAlignmentLog2(GV, DL);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription,
HI.TimerGroupName, HI.TimerGroupDescription,
TimePassesIsEnabled);
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
HI.Handler->setSymbolSize(GVSym, Size);
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
}
Fixed debug_aranges handling for common symbols. The size of common symbols is now tracked correctly, so they can be listed in the arange section without needing knowledge of other following symbols. .comm (and .lcomm) do not indicate to the system assembler any particular section to use, so we have to treat them as having no section. Test case update to account for this. llvm-svn: 191210
2013-09-24 01:56:20 +08:00
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
// Handle common symbols
if (GVKind.isCommon()) {
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
if (Size == 0) Size = 1; // .comm Foo, 0 is undefined, avoid it.
Don't drop alignment info on local common symbols. - On COFF the .lcomm directive has an alignment argument. - On ELF we fall back to .local + .comm Based on a patch by NAKAMURA Takumi. Fixes PR9337, PR9483 and PR10128. llvm-svn: 138976
2011-09-02 07:04:27 +08:00
unsigned Align = 1 << AlignLog;
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
if (!getObjFileLowering().getCommDirectiveSupportsAlignment())
Align = 0;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
// .comm _foo, 42, 4
OutStreamer->EmitCommonSymbol(GVSym, Size, Align);
return;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
// Determine to which section this global should be emitted.
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
MCSection *TheSection = getObjFileLowering().SectionForGlobal(GV, GVKind, TM);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
// If we have a bss global going to a section that supports the
// zerofill directive, do so here.
if (GVKind.isBSS() && MAI->hasMachoZeroFillDirective() &&
TheSection->isVirtualSection()) {
if (Size == 0)
Size = 1; // zerofill of 0 bytes is undefined.
unsigned Align = 1 << AlignLog;
EmitLinkage(GV, GVSym);
// .zerofill __DATA, __bss, _foo, 400, 5
OutStreamer->EmitZerofill(TheSection, GVSym, Size, Align);
return;
}
// If this is a BSS local symbol and we are emitting in the BSS
// section use .lcomm/.comm directive.
if (GVKind.isBSSLocal() &&
getObjFileLowering().getBSSSection() == TheSection) {
if (Size == 0)
Size = 1; // .comm Foo, 0 is undefined, avoid it.
unsigned Align = 1 << AlignLog;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Never use .lcomm on platforms where it does not accept an alignment argument. Instead, use a pair of .local and .comm directives. This avoids spurious differences between binaries built by the integrated assembler vs. those built by the external assembler, since the external assembler may impose alignment requirements on .lcomm symbols where the integrated assembler does not. llvm-svn: 168704
2012-11-28 00:11:16 +08:00
// Use .lcomm only if it supports user-specified alignment.
// Otherwise, while it would still be correct to use .lcomm in some
// cases (e.g. when Align == 1), the external assembler might enfore
// some -unknown- default alignment behavior, which could cause
// spurious differences between external and integrated assembler.
// Prefer to simply fall back to .local / .comm in this case.
if (MAI->getLCOMMDirectiveAlignmentType() != LCOMM::NoAlignment) {
stop using the .lcomm pseudoop on darwin, instead, directly use the .zerofill directive. Streamerize its generation. llvm-svn: 93868
2010-01-19 14:25:51 +08:00
// .lcomm _foo, 42
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLocalCommonSymbol(GVSym, Size, Align);
stop using the .lcomm pseudoop on darwin, instead, directly use the .zerofill directive. Streamerize its generation. llvm-svn: 93868
2010-01-19 14:25:51 +08:00
return;
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
}
the latest assembler that runs on powerpc 10.4 machines doesn't support aligned comm. Detect when compiling for 10.4 and don't emit an alignment for comm. THis will hopefully fix PR8198. llvm-svn: 114817
2010-09-27 14:44:54 +08:00
if (!getObjFileLowering().getCommDirectiveSupportsAlignment())
Align = 0;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
stop using the .lcomm pseudoop on darwin, instead, directly use the .zerofill directive. Streamerize its generation. llvm-svn: 93868
2010-01-19 14:25:51 +08:00
// .local _foo
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(GVSym, MCSA_Local);
stop using the .lcomm pseudoop on darwin, instead, directly use the .zerofill directive. Streamerize its generation. llvm-svn: 93868
2010-01-19 14:25:51 +08:00
// .comm _foo, 42, 4
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitCommonSymbol(GVSym, Size, Align);
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
return;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add support for initialized global data for darwin tls. Update comments and testcases accordingly. llvm-svn: 104635
2010-05-26 05:28:50 +08:00
// Handle thread local data for mach-o which requires us to output an
// additional structure of data and mangle the original symbol so that we
// can reference it later.
add a comment about where this should eventually move. llvm-svn: 113117
2010-09-06 04:33:40 +08:00
//
// TODO: This should become an "emit thread local global" method on TLOF.
// All of this macho specific stuff should be sunk down into TLOFMachO and
// stuff like "TLSExtraDataSection" should no longer be part of the parent
// TLOF class. This will also make it more obvious that stuff like
// MCStreamer::EmitTBSSSymbol is macho specific and only called from macho
// specific code.
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
if (GVKind.isThreadLocal() && MAI->hasMachoTBSSDirective()) {
Add full bss data support for darwin tls variables. llvm-svn: 104414
2010-05-22 08:10:22 +08:00
// Emit the .tbss symbol
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
MCSymbol *MangSym =
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
OutContext.getOrCreateSymbol(GVSym->getName() + Twine("$tlv$init"));
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Using the integrated assembler we'd fail to change section to the .tbss section for zerofill thread locals. Make sure we do this before emitting the zerofills. Fixes PR15972. llvm-svn: 187913
2013-08-08 05:13:06 +08:00
if (GVKind.isThreadBSS()) {
TheSection = getObjFileLowering().getTLSBSSSection();
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitTBSSSymbol(TheSection, MangSym, Size, 1 << AlignLog);
Using the integrated assembler we'd fail to change section to the .tbss section for zerofill thread locals. Make sure we do this before emitting the zerofills. Fixes PR15972. llvm-svn: 187913
2013-08-08 05:13:06 +08:00
} else if (GVKind.isThreadData()) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(TheSection);
Add support for initialized global data for darwin tls. Update comments and testcases accordingly. llvm-svn: 104635
2010-05-26 05:28:50 +08:00
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
EmitAlignment(AlignLog, GV);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(MangSym);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitGlobalConstant(GV->getParent()->getDataLayout(),
GV->getInitializer());
Add support for initialized global data for darwin tls. Update comments and testcases accordingly. llvm-svn: 104635
2010-05-26 05:28:50 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add full bss data support for darwin tls variables. llvm-svn: 104414
2010-05-22 08:10:22 +08:00
// Emit the variable struct for the runtime.
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
MCSection *TLVSect = getObjFileLowering().getTLSExtraDataSection();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(TLVSect);
Add full bss data support for darwin tls variables. llvm-svn: 104414
2010-05-22 08:10:22 +08:00
// Emit the linkage here.
Produce .weak_def_can_be_hidden for some linkonce_odr values With this patch llvm produces a weak_def_can_be_hidden for linkonce_odr if they are also unnamed_addr or don't have their address taken. There is not a lot of documentation about .weak_def_can_be_hidden, but from the old discussion about linkonce_odr_auto_hide and the name of the directive this looks correct: these symbols can be hidden. Testing this with the ld64 in Xcode 5 linking clang reduces the number of exported symbols from 21053 to 19049. llvm-svn: 193718
2013-10-31 06:08:11 +08:00
EmitLinkage(GV, GVSym);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(GVSym);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add full bss data support for darwin tls variables. llvm-svn: 104414
2010-05-22 08:10:22 +08:00
// Three pointers in size:
// - __tlv_bootstrap - used to make sure support exists
// - spare pointer, used when mapped by the runtime
// - pointer to mangled symbol above with initializer
Remove DataLayout from TargetLoweringObjectFile, redirect to Module Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11079 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242385
2015-07-16 14:04:17 +08:00
unsigned PtrSize = DL.getPointerTypeSize(GV->getType());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolValue(GetExternalSymbolSymbol("_tlv_bootstrap"),
80-column and tab character fixes. llvm-svn: 184792
2013-06-25 07:20:02 +08:00
PtrSize);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIntValue(0, PtrSize);
OutStreamer->EmitSymbolValue(MangSym, PtrSize);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
Partial code for emitting thread local bss data. llvm-svn: 104197
2010-05-20 08:49:07 +08:00
return;
}
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
Fix BSS global handling in AsmPrinter Change EmitGlobalVariable to check final assembler section is in BSS before using .lcomm/.comm directive. This prevents globals from being put into .bss erroneously when -data-sections is used. This fixes PR26570. Reviewers: echristo, rafael Subscribers: llvm-commits, mehdi_amini Differential Revision: http://reviews.llvm.org/D21146 llvm-svn: 272674
2016-06-14 23:09:30 +08:00
MCSymbol *EmittedInitSym = GVSym;
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(TheSection);
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
[TLS] New lower emutls pass, fix linkage bugs. Previous implementation in http://reviews.llvm.org/D10522 created external references to __emutls_v.* variables. Such references are inaccurate and cannot be handled by all linkers, e.g. Android dynamic and gold linkers for aarch64. Now a new LowerEmuTLS pass to go through all global variables, and add emutls_v.* and emutls_t.* variables. These __emutls* variables have the same linkage and visibility as the associated user defined TLS variable. Also removed old code that dump __emutls* variables in AsmPrinter.cpp, and updated TLS unit tests. Differential Revision: http://reviews.llvm.org/D15300 llvm-svn: 257718
2016-01-14 07:56:37 +08:00
EmitLinkage(GV, EmittedInitSym);
Revert r102300/102301, which serious broke objc apps. llvm-svn: 102359
2010-04-27 02:30:45 +08:00
EmitAlignment(AlignLog, GV);
pull linkage emission code out to a new EmitLinkage function. llvm-svn: 94621
2010-01-27 07:47:12 +08:00
Implement target independent TLS compatible with glibc's emutls.c. The 'common' section TLS is not implemented. Current C/C++ TLS variables are not placed in common section. DWARF debug info to get the address of TLS variables is not generated yet. clang and driver changes in http://reviews.llvm.org/D10524 Added -femulated-tls flag to select the emulated TLS model, which will be used for old targets like Android that do not support ELF TLS models. Added TargetLowering::LowerToTLSEmulatedModel as a target-independent function to convert a SDNode of TLS variable address to a function call to __emutls_get_address. Added into lib/Target/*/*ISelLowering.cpp to call LowerToTLSEmulatedModel for TLSModel::Emulated. Although all targets supporting ELF TLS models are enhanced, emulated TLS model has been tested only for Android ELF targets. Modified AsmPrinter.cpp to print the emutls_v.* and emutls_t.* variables for emulated TLS variables. Modified DwarfCompileUnit.cpp to skip some DIE for emulated TLS variabls. TODO: Add proper DIE for emulated TLS variables. Added new unit tests with emulated TLS. Differential Revision: http://reviews.llvm.org/D10522 llvm-svn: 243438
2015-07-29 00:24:05 +08:00
OutStreamer->EmitLabel(EmittedInitSym);
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitGlobalConstant(GV->getParent()->getDataLayout(), GV->getInitializer());
Now that we have everything nicely factored (e.g. asmprinter is not doing global variable classification anymore) and hookized, sink almost all target targets global variable emission code into AsmPrinter and out of each target. Some notes: 1. PIC16 does completely custom and crazy stuff, so it is not changed. 2. XCore has some custom handling for extra directives. I'll look at it next. 3. This switches linux/ppc to use .globl instead of .global. If .globl is actually wrong, let me know and I'll fix it. 4. This makes linux/ppc get a lot of random cases right which were obviously wrong before, it is probably now a bit healthier. 5. Blackfin will probably start getting .comm and other things that it didn't before. If this is undesirable, it should explicitly opt out of these things by clearing the relevant fields of MCAsmInfo. This leads to a nice diffstat: 14 files changed, 127 insertions(+), 830 deletions(-) llvm-svn: 93858
2010-01-19 13:38:33 +08:00
if (MAI->hasDotTypeDotSizeDirective())
add a comment llvm-svn: 94417
2010-01-25 15:53:05 +08:00
// .size foo, 42
[MC] Refactor emitELFSize to make usage more consistent. NFC. Move the cast<MCSymbolELF> inside emitELFSize, so that: - it's done in one place instead of at each call - it's more consistent with similar functions like EmitCOFFSafeSEH - ambiguity between cast<> and dyn_cast<> is avoided (which also eliminates an unnecessary dyn_cast call) This also makes it easier to experiment with using ".size" directives on non-ELF targets. llvm-svn: 288437
2016-12-02 07:39:08 +08:00
OutStreamer->emitELFSize(EmittedInitSym,
Create a MCSymbolELF. This create a MCSymbolELF class and moves SymbolSize since only ELF needs a size expression. This reduces the size of MCSymbol from 56 to 48 bytes. llvm-svn: 238801
2015-06-02 08:25:12 +08:00
MCConstantExpr::create(Size, OutContext));
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
hoist handling of external globals and special globals up to common code. This makes a similar code dead in all the other targets, I'll clean it up in a bit. This also moves handling of lcomm up before acquisition of a section, since lcomm never needs a section. llvm-svn: 93851
2010-01-19 12:39:15 +08:00
}
[mips] Fix debug information for __thread variable This patch fixes debug information for __thread variable on Mips using .dtprelword and .dtpreldword directives. Patch by Aleksandar Beserminji. Differential Revision: http://reviews.llvm.org/D28770 llvm-svn: 292624
2017-01-21 01:53:30 +08:00
/// Emit the directive and value for debug thread local expression
///
/// \p Value - The value to emit.
/// \p Size - The size of the integer (in bytes) to emit.
[DebugInfo] Rename EmitDebugValue to EmitDebugThreadLocal (NFC) As pointed out by David Blaikie in the post commit review of r292624, EmitDebugValue should be called EmitDebugThreadLocal. llvm-svn: 294500
2017-02-09 03:03:46 +08:00
void AsmPrinter::EmitDebugThreadLocal(const MCExpr *Value,
[mips] Fix debug information for __thread variable This patch fixes debug information for __thread variable on Mips using .dtprelword and .dtpreldword directives. Patch by Aleksandar Beserminji. Differential Revision: http://reviews.llvm.org/D28770 llvm-svn: 292624
2017-01-21 01:53:30 +08:00
unsigned Size) const {
OutStreamer->EmitValue(Value, Size);
}
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
/// EmitFunctionHeader - This method emits the header for the current
/// function.
void AsmPrinter::EmitFunctionHeader() {
// Print out constants referenced by the function
Drop the argument to AsmPrinter::EmitConstantPool and make it virtual. Overload it in the ARM backend to do nothing, since is does insane constant pool emission. llvm-svn: 94708
2010-01-28 08:19:24 +08:00
EmitConstantPool();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
// Print the 'header' of function.
const Function *F = MF->getFunction();
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
OutStreamer->SwitchSection(getObjFileLowering().SectionForGlobal(F, TM));
rename printVisibility to EmitVisibility and make it private, constify EmitLinkage. llvm-svn: 94705
2010-01-28 08:05:10 +08:00
EmitVisibility(CurrentFnSym, F->getVisibility());
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
Produce .weak_def_can_be_hidden for some linkonce_odr values With this patch llvm produces a weak_def_can_be_hidden for linkonce_odr if they are also unnamed_addr or don't have their address taken. There is not a lot of documentation about .weak_def_can_be_hidden, but from the old discussion about linkonce_odr_auto_hide and the name of the directive this looks correct: these symbols can be hidden. Testing this with the ld64 in Xcode 5 linking clang reduces the number of exported symbols from 21053 to 19049. llvm-svn: 193718
2013-10-31 06:08:11 +08:00
EmitLinkage(F, CurrentFnSym);
[NVPTXAsmPrinter] do not print .align on function headers Summary: PTX does not allow .align directives on function headers. Fixes PR21551. Test Plan: test/Codegen/NVPTX/function-align.ll Reviewers: eliben, jholewinski Reviewed By: eliben, jholewinski Subscribers: llvm-commits, eliben, jpienaar, jholewinski Differential Revision: http://reviews.llvm.org/D8274 llvm-svn: 232004
2015-03-12 09:50:30 +08:00
if (MAI->hasFunctionAlignment())
EmitAlignment(MF->getAlignment(), F);
rearrange some directives, no functionality change. llvm-svn: 94620
2010-01-27 07:41:48 +08:00
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
if (MAI->hasDotTypeDotSizeDirective())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(CurrentFnSym, MCSA_ELF_TypeFunction);
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
clean up the asmprinter header and privatize some stuff. llvm-svn: 100342
2010-04-05 02:52:31 +08:00
if (isVerbose()) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
F->printAsOperand(OutStreamer->GetCommentOS(),
mcize label emission for functions. llvm-svn: 94624
2010-01-27 07:53:39 +08:00
/*PrintType=*/false, F->getParent());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->GetCommentOS() << '\n';
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
}
Prologue support Patch by Ben Gamari! This redefines the `prefix` attribute introduced previously and introduces a `prologue` attribute. There are a two primary usecases that these attributes aim to serve, 1. Function prologue sigils 2. Function hot-patching: Enable the user to insert `nop` operations at the beginning of the function which can later be safely replaced with a call to some instrumentation facility 3. Runtime metadata: Allow a compiler to insert data for use by the runtime during execution. GHC is one example of a compiler that needs this functionality for its tables-next-to-code functionality. Previously `prefix` served cases (1) and (2) quite well by allowing the user to introduce arbitrary data at the entrypoint but before the function body. Case (3), however, was poorly handled by this approach as it required that prefix data was valid executable code. Here we redefine the notion of prefix data to instead be data which occurs immediately before the function entrypoint (i.e. the symbol address). Since prefix data now occurs before the function entrypoint, there is no need for the data to be valid code. The previous notion of prefix data now goes under the name "prologue data" to emphasize its duality with the function epilogue. The intention here is to handle cases (1) and (2) with prologue data and case (3) with prefix data. References ---------- This idea arose out of discussions[1] with Reid Kleckner in response to a proposal to introduce the notion of symbol offsets to enable handling of case (3). [1] http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-May/073235.html Test Plan: testsuite Differential Revision: http://reviews.llvm.org/D6454 llvm-svn: 223189
2014-12-03 10:08:38 +08:00
// Emit the prefix data.
Ensure that prefix data is preserved with subsections-via-symbols On MachO platforms that use subsections-via-symbols dead code stripping will drop prefix data. Unfortunately there is no great way to convey the relationship between a function and its prefix data to the linker. We are forced to use a bit of a hack: we give the prefix data it’s own symbol, and mark the actual function entry an .alt_entry. Patch by Moritz Angermann! Differential Revision: https://reviews.llvm.org/D30770 llvm-svn: 297804
2017-03-15 12:18:16 +08:00
if (F->hasPrefixData()) {
if (MAI->hasSubsectionsViaSymbols()) {
// Preserving prefix data on platforms which use subsections-via-symbols
// is a bit tricky. Here we introduce a symbol for the prefix data
// and use the .alt_entry attribute to mark the function's real entry point
// as an alternative entry point to the prefix-data symbol.
MCSymbol *PrefixSym = OutContext.createLinkerPrivateTempSymbol();
OutStreamer->EmitLabel(PrefixSym);
EmitGlobalConstant(F->getParent()->getDataLayout(), F->getPrefixData());
// Emit an .alt_entry directive for the actual function symbol.
OutStreamer->EmitSymbolAttribute(CurrentFnSym, MCSA_AltEntry);
} else {
EmitGlobalConstant(F->getParent()->getDataLayout(), F->getPrefixData());
}
}
Prologue support Patch by Ben Gamari! This redefines the `prefix` attribute introduced previously and introduces a `prologue` attribute. There are a two primary usecases that these attributes aim to serve, 1. Function prologue sigils 2. Function hot-patching: Enable the user to insert `nop` operations at the beginning of the function which can later be safely replaced with a call to some instrumentation facility 3. Runtime metadata: Allow a compiler to insert data for use by the runtime during execution. GHC is one example of a compiler that needs this functionality for its tables-next-to-code functionality. Previously `prefix` served cases (1) and (2) quite well by allowing the user to introduce arbitrary data at the entrypoint but before the function body. Case (3), however, was poorly handled by this approach as it required that prefix data was valid executable code. Here we redefine the notion of prefix data to instead be data which occurs immediately before the function entrypoint (i.e. the symbol address). Since prefix data now occurs before the function entrypoint, there is no need for the data to be valid code. The previous notion of prefix data now goes under the name "prologue data" to emphasize its duality with the function epilogue. The intention here is to handle cases (1) and (2) with prologue data and case (3) with prefix data. References ---------- This idea arose out of discussions[1] with Reid Kleckner in response to a proposal to introduce the notion of symbol offsets to enable handling of case (3). [1] http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-May/073235.html Test Plan: testsuite Differential Revision: http://reviews.llvm.org/D6454 llvm-svn: 223189
2014-12-03 10:08:38 +08:00
Fix "the the" and similar typos. llvm-svn: 95781
2010-02-11 00:03:48 +08:00
// Emit the CurrentFnSym. This is a virtual function to allow targets to
add a new AsmPrinter::EmitFunctionEntryLabel virtual function, which allows targets to override function entry label emission. Use it to convert linux/ppc to use EmitFunctionHeader(). llvm-svn: 94667
2010-01-27 15:21:55 +08:00
// do their wild and crazy things as required.
EmitFunctionEntryLabel();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Implement support for the case when a reference to a addr-of-bb label is generated, but then the block is deleted. Since the value is undefined, we just emit the label right after the entry label of the function. It might matter that the label is in the same section as the function was afterall. llvm-svn: 98579
2010-03-16 04:39:00 +08:00
// If the function had address-taken blocks that got deleted, then we have
// references to the dangling symbols. Emit them at the start of the function
// so that we don't get references to undefined symbols.
std::vector<MCSymbol*> DeadBlockSyms;
MMI->takeDeletedSymbolsForFunction(F, DeadBlockSyms);
for (unsigned i = 0, e = DeadBlockSyms.size(); i != e; ++i) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("Address taken block that was later removed");
OutStreamer->EmitLabel(DeadBlockSyms[i]);
Implement support for the case when a reference to a addr-of-bb label is generated, but then the block is deleted. Since the value is undefined, we just emit the label right after the entry label of the function. It might matter that the label is in the same section as the function was afterall. llvm-svn: 98579
2010-03-16 04:39:00 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Use the correct func begin symbol in all places in ppc. I missed an occurrence of the old symbol in my previous patch. llvm-svn: 231398
2015-03-06 03:47:50 +08:00
if (CurrentFnBegin) {
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
if (MAI->useAssignmentForEHBegin()) {
MC: Clean up method names in MCContext. The naming was a mish-mash of old and new style. Update to be consistent with the new. NFC. llvm-svn: 237594
2015-05-19 02:43:14 +08:00
MCSymbol *CurPos = OutContext.createTempSymbol();
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(CurPos);
OutStreamer->EmitAssignment(CurrentFnBegin,
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
MCSymbolRefExpr::create(CurPos, OutContext));
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
} else {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(CurrentFnBegin);
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
}
}
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
// Emit pre-function debug and/or EH information.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription, HI.TimerGroupName,
HI.TimerGroupDescription, TimePassesIsEnabled);
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
HI.Handler->beginFunction(MF);
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
}
Emit prefix data after debug and EH directives. This ensures that the prefix data is treated as part of the function for the purpose of debug info. This provides a better debugging experience, among other things by allowing a debug info client to correctly look up a function in debug info given a function pointer. llvm-svn: 193042
2013-10-20 10:16:21 +08:00
Prologue support Patch by Ben Gamari! This redefines the `prefix` attribute introduced previously and introduces a `prologue` attribute. There are a two primary usecases that these attributes aim to serve, 1. Function prologue sigils 2. Function hot-patching: Enable the user to insert `nop` operations at the beginning of the function which can later be safely replaced with a call to some instrumentation facility 3. Runtime metadata: Allow a compiler to insert data for use by the runtime during execution. GHC is one example of a compiler that needs this functionality for its tables-next-to-code functionality. Previously `prefix` served cases (1) and (2) quite well by allowing the user to introduce arbitrary data at the entrypoint but before the function body. Case (3), however, was poorly handled by this approach as it required that prefix data was valid executable code. Here we redefine the notion of prefix data to instead be data which occurs immediately before the function entrypoint (i.e. the symbol address). Since prefix data now occurs before the function entrypoint, there is no need for the data to be valid code. The previous notion of prefix data now goes under the name "prologue data" to emphasize its duality with the function epilogue. The intention here is to handle cases (1) and (2) with prologue data and case (3) with prefix data. References ---------- This idea arose out of discussions[1] with Reid Kleckner in response to a proposal to introduce the notion of symbol offsets to enable handling of case (3). [1] http://lists.cs.uiuc.edu/pipermail/llvmdev/2014-May/073235.html Test Plan: testsuite Differential Revision: http://reviews.llvm.org/D6454 llvm-svn: 223189
2014-12-03 10:08:38 +08:00
// Emit the prologue data.
if (F->hasPrologueData())
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitGlobalConstant(F->getParent()->getDataLayout(), F->getPrologueData());
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
}
add a new AsmPrinter::EmitFunctionEntryLabel virtual function, which allows targets to override function entry label emission. Use it to convert linux/ppc to use EmitFunctionHeader(). llvm-svn: 94667
2010-01-27 15:21:55 +08:00
/// EmitFunctionEntryLabel - Emit the label that is the entrypoint for the
/// function. This can be overridden by targets as required to do custom stuff.
void AsmPrinter::EmitFunctionEntryLabel() {
MC: Label definitions are permitted after .set directives .set directives may be overridden by other .set directives as well as label definitions. This fixes PR22019. llvm-svn: 224811
2014-12-24 18:27:50 +08:00
CurrentFnSym->redefineIfPossible();
Fix PR7054 - Assertion `Symbol->isUndefined() && "Cannot define a symbol twice!"' failed. Users can write broken code that emits the same label twice with asm renaming, detect this and emit a fatal backend error instead of aborting. llvm-svn: 103140
2010-05-06 08:05:37 +08:00
// The function label could have already been emitted if two symbols end up
// conflicting due to asm renaming. Detect this and emit an error.
CodeGen: Error on redefinitions instead of asserting It's possible to have a prior definition of a symbol in module asm. Raise an error instead of crashing. llvm-svn: 224828
2014-12-25 07:06:55 +08:00
if (CurrentFnSym->isVariable())
report_fatal_error("'" + Twine(CurrentFnSym->getName()) +
"' is a protected alias");
if (CurrentFnSym->isDefined())
report_fatal_error("'" + Twine(CurrentFnSym->getName()) +
"' label emitted multiple times to assembly file");
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
return OutStreamer->EmitLabel(CurrentFnSym);
add a new AsmPrinter::EmitFunctionEntryLabel virtual function, which allows targets to override function entry label emission. Use it to convert linux/ppc to use EmitFunctionHeader(). llvm-svn: 94667
2010-01-27 15:21:55 +08:00
}
now that enough stuff is constified, move function header printing logic up from X86 into the common code. The other targets will hopefully start using this soon. llvm-svn: 94614
2010-01-27 07:18:44 +08:00
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
/// emitComments - Pretty-print comments for instructions.
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
static void emitComments(const MachineInstr &MI, raw_ostream &CommentOS,
AsmPrinter *AP) {
Print InlinedAt location. llvm-svn: 107214
2010-06-30 06:29:15 +08:00
const MachineFunction *MF = MI.getParent()->getParent();
Remove the MCInstrInfo cached variable as it was only used in a single place and replace calls to getSubtargetImpl with calls to get the subtarget from the MachineFunction where valid. llvm-svn: 229971
2015-02-20 14:35:21 +08:00
const TargetInstrInfo *TII = MF->getSubtarget().getInstrInfo();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
// Check for spills and reloads
int FI;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
MachineFunction: Return reference for getFrameInfo(); NFC getFrameInfo() never returns nullptr so we should use a reference instead of a pointer. llvm-svn: 277017
2016-07-29 02:40:00 +08:00
const MachineFrameInfo &MFI = MF->getFrameInfo();
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
bool Commented = false;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
// We assume a single instruction only has a spill or reload, not
// both.
const MachineMemOperand *MMO;
CodeGen: Use MachineInstr& in TargetInstrInfo, NFC This is mostly a mechanical change to make TargetInstrInfo API take MachineInstr& (instead of MachineInstr* or MachineBasicBlock::iterator) when the argument is expected to be a valid MachineInstr. This is a general API improvement. Although it would be possible to do this one function at a time, that would demand a quadratic amount of churn since many of these functions call each other. Instead I've done everything as a block and just updated what was necessary. This is mostly mechanical fixes: adding and removing `*` and `&` operators. The only non-mechanical change is to split ARMBaseInstrInfo::getOperandLatencyImpl out from ARMBaseInstrInfo::getOperandLatency. Previously, the latter took a `MachineInstr*` which it updated to the instruction bundle leader; now, the latter calls the former either with the same `MachineInstr&` or the bundle leader. As a side effect, this removes a bunch of MachineInstr* to MachineBasicBlock::iterator implicit conversions, a necessary step toward fixing PR26753. Note: I updated WebAssembly, Lanai, and AVR (despite being off-by-default) since it turned out to be easy. I couldn't run tests for AVR since llc doesn't link with it turned on. llvm-svn: 274189
2016-06-30 08:01:54 +08:00
if (TII->isLoadFromStackSlotPostFE(MI, FI)) {
MachineFunction: Return reference for getFrameInfo(); NFC getFrameInfo() never returns nullptr so we should use a reference instead of a pointer. llvm-svn: 277017
2016-07-29 02:40:00 +08:00
if (MFI.isSpillSlotObjectIndex(FI)) {
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
MMO = *MI.memoperands_begin();
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
CommentOS << MMO->getSize() << "-byte Reload";
Commented = true;
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
}
CodeGen: Use MachineInstr& in TargetInstrInfo, NFC This is mostly a mechanical change to make TargetInstrInfo API take MachineInstr& (instead of MachineInstr* or MachineBasicBlock::iterator) when the argument is expected to be a valid MachineInstr. This is a general API improvement. Although it would be possible to do this one function at a time, that would demand a quadratic amount of churn since many of these functions call each other. Instead I've done everything as a block and just updated what was necessary. This is mostly mechanical fixes: adding and removing `*` and `&` operators. The only non-mechanical change is to split ARMBaseInstrInfo::getOperandLatencyImpl out from ARMBaseInstrInfo::getOperandLatency. Previously, the latter took a `MachineInstr*` which it updated to the instruction bundle leader; now, the latter calls the former either with the same `MachineInstr&` or the bundle leader. As a side effect, this removes a bunch of MachineInstr* to MachineBasicBlock::iterator implicit conversions, a necessary step toward fixing PR26753. Note: I updated WebAssembly, Lanai, and AVR (despite being off-by-default) since it turned out to be easy. I couldn't run tests for AVR since llc doesn't link with it turned on. llvm-svn: 274189
2016-06-30 08:01:54 +08:00
} else if (TII->hasLoadFromStackSlot(MI, MMO, FI)) {
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
if (MFI.isSpillSlotObjectIndex(FI)) {
CommentOS << MMO->getSize() << "-byte Folded Reload";
Commented = true;
}
CodeGen: Use MachineInstr& in TargetInstrInfo, NFC This is mostly a mechanical change to make TargetInstrInfo API take MachineInstr& (instead of MachineInstr* or MachineBasicBlock::iterator) when the argument is expected to be a valid MachineInstr. This is a general API improvement. Although it would be possible to do this one function at a time, that would demand a quadratic amount of churn since many of these functions call each other. Instead I've done everything as a block and just updated what was necessary. This is mostly mechanical fixes: adding and removing `*` and `&` operators. The only non-mechanical change is to split ARMBaseInstrInfo::getOperandLatencyImpl out from ARMBaseInstrInfo::getOperandLatency. Previously, the latter took a `MachineInstr*` which it updated to the instruction bundle leader; now, the latter calls the former either with the same `MachineInstr&` or the bundle leader. As a side effect, this removes a bunch of MachineInstr* to MachineBasicBlock::iterator implicit conversions, a necessary step toward fixing PR26753. Note: I updated WebAssembly, Lanai, and AVR (despite being off-by-default) since it turned out to be easy. I couldn't run tests for AVR since llc doesn't link with it turned on. llvm-svn: 274189
2016-06-30 08:01:54 +08:00
} else if (TII->isStoreToStackSlotPostFE(MI, FI)) {
MachineFunction: Return reference for getFrameInfo(); NFC getFrameInfo() never returns nullptr so we should use a reference instead of a pointer. llvm-svn: 277017
2016-07-29 02:40:00 +08:00
if (MFI.isSpillSlotObjectIndex(FI)) {
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
MMO = *MI.memoperands_begin();
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
CommentOS << MMO->getSize() << "-byte Spill";
Commented = true;
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
}
CodeGen: Use MachineInstr& in TargetInstrInfo, NFC This is mostly a mechanical change to make TargetInstrInfo API take MachineInstr& (instead of MachineInstr* or MachineBasicBlock::iterator) when the argument is expected to be a valid MachineInstr. This is a general API improvement. Although it would be possible to do this one function at a time, that would demand a quadratic amount of churn since many of these functions call each other. Instead I've done everything as a block and just updated what was necessary. This is mostly mechanical fixes: adding and removing `*` and `&` operators. The only non-mechanical change is to split ARMBaseInstrInfo::getOperandLatencyImpl out from ARMBaseInstrInfo::getOperandLatency. Previously, the latter took a `MachineInstr*` which it updated to the instruction bundle leader; now, the latter calls the former either with the same `MachineInstr&` or the bundle leader. As a side effect, this removes a bunch of MachineInstr* to MachineBasicBlock::iterator implicit conversions, a necessary step toward fixing PR26753. Note: I updated WebAssembly, Lanai, and AVR (despite being off-by-default) since it turned out to be easy. I couldn't run tests for AVR since llc doesn't link with it turned on. llvm-svn: 274189
2016-06-30 08:01:54 +08:00
} else if (TII->hasStoreToStackSlot(MI, MMO, FI)) {
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
if (MFI.isSpillSlotObjectIndex(FI)) {
CommentOS << MMO->getSize() << "-byte Folded Spill";
Commented = true;
}
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
// Check for spill-induced copies
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
if (MI.getAsmPrinterFlag(MachineInstr::ReloadReuse)) {
Commented = true;
CommentOS << " Reload Reuse";
}
if (Commented && AP->EnablePrintSchedInfo)
// If any comment was added above and we need sched info comment then
// add this new comment just after the above comment w/o "\n" between them.
CommentOS << " " << MF->getSubtarget().getSchedInfoStr(MI) << "\n";
else if (Commented)
CommentOS << "\n";
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
}
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
/// emitImplicitDef - This method emits the specified machine instruction
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
/// that is an implicit def.
Make AsmPrinter::emitImplicitDef a virtual method so targets can emit custom comments for implicit defs For NVPTX, this fixes a crash where the emitImplicitDef implementation was expecting physical registers, while NVPTX uses virtual registers (with a couple of exceptions). Now, the implicit def comment will be emitted as a true PTX register name. Other targets can use this to customize the output of implicit def comments. Fixes PR17519 llvm-svn: 192444
2013-10-11 20:39:36 +08:00
void AsmPrinter::emitImplicitDef(const MachineInstr *MI) const {
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
unsigned RegNo = MI->getOperand(0).getReg();
Use TargetRegisterInfo for printing MachineOperand register comments Several places in AsmPrinter.cpp print comments describing MachineOperand registers using MCRegisterInfo, which uses MCOperand-oriented names. This doesn't work for targets that use virtual registers exclusively, as WebAssembly does, since virtual registers are represented and printed differently. This patch preserves what seems to be the spirit of r229978, avoiding the use of TM.getSubtargetImpl(), while still using MachineOperand-oriented printing for MachineOperands. Differential Revision: http://reviews.llvm.org/D14709 llvm-svn: 253338
2015-11-18 00:01:28 +08:00
SmallString<128> Str;
raw_svector_ostream OS(Str);
OS << "implicit-def: "
<< PrintReg(RegNo, MF->getSubtarget().getRegisterInfo());
OutStreamer->AddComment(OS.str());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
}
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
static void emitKill(const MachineInstr *MI, AsmPrinter &AP) {
Use TargetRegisterInfo for printing MachineOperand register comments Several places in AsmPrinter.cpp print comments describing MachineOperand registers using MCRegisterInfo, which uses MCOperand-oriented names. This doesn't work for targets that use virtual registers exclusively, as WebAssembly does, since virtual registers are represented and printed differently. This patch preserves what seems to be the spirit of r229978, avoiding the use of TM.getSubtargetImpl(), while still using MachineOperand-oriented printing for MachineOperands. Differential Revision: http://reviews.llvm.org/D14709 llvm-svn: 253338
2015-11-18 00:01:28 +08:00
std::string Str;
raw_string_ostream OS(Str);
OS << "kill:";
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
for (unsigned i = 0, e = MI->getNumOperands(); i != e; ++i) {
const MachineOperand &Op = MI->getOperand(i);
assert(Op.isReg() && "KILL instruction must have only register operands");
Use TargetRegisterInfo for printing MachineOperand register comments Several places in AsmPrinter.cpp print comments describing MachineOperand registers using MCRegisterInfo, which uses MCOperand-oriented names. This doesn't work for targets that use virtual registers exclusively, as WebAssembly does, since virtual registers are represented and printed differently. This patch preserves what seems to be the spirit of r229978, avoiding the use of TM.getSubtargetImpl(), while still using MachineOperand-oriented printing for MachineOperands. Differential Revision: http://reviews.llvm.org/D14709 llvm-svn: 253338
2015-11-18 00:01:28 +08:00
OS << ' '
<< PrintReg(Op.getReg(),
AP.MF->getSubtarget().getRegisterInfo())
<< (Op.isDef() ? "<def>" : "<kill>");
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
}
AsmPrinter: Fix emitKill() not flush()ing a raw_string_ostream llvm-svn: 274951
2016-07-09 08:18:43 +08:00
AP.OutStreamer->AddComment(OS.str());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->AddBlankLine();
more interface cleanup make some helpers static functions. llvm-svn: 100343
2010-04-05 02:58:53 +08:00
}
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
/// emitDebugValueComment - This method handles the target-independent form
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
/// of DBG_VALUE, returning true if it was able to do so. A false return
/// means the target will need to handle MI in EmitInstruction.
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
static bool emitDebugValueComment(const MachineInstr *MI, AsmPrinter &AP) {
Move the complex address expression out of DIVariable and into an extra argument of the llvm.dbg.declare/llvm.dbg.value intrinsics. Previously, DIVariable was a variable-length field that has an optional reference to a Metadata array consisting of a variable number of complex address expressions. In the case of OpPiece expressions this is wasting a lot of storage in IR, because when an aggregate type is, e.g., SROA'd into all of its n individual members, the IR will contain n copies of the DIVariable, all alike, only differing in the complex address reference at the end. By making the complex address into an extra argument of the dbg.value/dbg.declare intrinsics, all of the pieces can reference the same variable and the complex address expressions can be uniqued across the CU, too. Down the road, this will allow us to move other flags, such as "indirection" out of the DIVariable, too. The new intrinsics look like this: declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr) declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr) This patch adds a new LLVM-local tag to DIExpressions, so we can detect and pretty-print DIExpression metadata nodes. What this patch doesn't do: This patch does not touch the "Indirect" field in DIVariable; but moving that into the expression would be a natural next step. http://reviews.llvm.org/D4919 rdar://problem/17994491 Thanks to dblaikie and dexonsmith for reviewing this patch! Note: I accidentally committed a bogus older version of this patch previously. llvm-svn: 218787
2014-10-02 02:55:02 +08:00
// This code handles only the 4-operand target-independent form.
if (MI->getNumOperands() != 4)
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
return false;
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
SmallString<128> Str;
raw_svector_ostream OS(Str);
Add an emitRawComment function and use it to simplify some uses of EmitRawText. llvm-svn: 199397
2014-01-17 00:28:37 +08:00
OS << "DEBUG_VALUE: ";
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
const DILocalVariable *V = MI->getDebugVariable();
if (auto *SP = dyn_cast<DISubprogram>(V->getScope())) {
DebugInfo: Gut DISubprogram and DILexicalBlock* Gut the `DIDescriptor` wrappers around `MDLocalScope` subclasses. Note that `DILexicalBlock` wraps `MDLexicalBlockBase`, not `MDLexicalBlock`. llvm-svn: 234850
2015-04-14 11:40:37 +08:00
StringRef Name = SP->getDisplayName();
Debug Info: Don't print the display name and colon prefix for DEBUG_VALUE comments if the display name is empty llvm-svn: 184026
2013-06-15 08:33:47 +08:00
if (!Name.empty())
OS << Name << ":";
}
DebugInfo: Gut DIVariable and DIGlobalVariable Gut all the non-pointer API from the variable wrappers, except an implicit conversion from `DIGlobalVariable` to `DIDescriptor`. Note that if you're updating out-of-tree code, `DIVariable` wraps `MDLocalVariable` (`MDVariable` is a common base class shared with `MDGlobalVariable`). llvm-svn: 234840
2015-04-14 10:22:36 +08:00
OS << V->getName();
Move the complex address expression out of DIVariable and into an extra argument of the llvm.dbg.declare/llvm.dbg.value intrinsics. Previously, DIVariable was a variable-length field that has an optional reference to a Metadata array consisting of a variable number of complex address expressions. In the case of OpPiece expressions this is wasting a lot of storage in IR, because when an aggregate type is, e.g., SROA'd into all of its n individual members, the IR will contain n copies of the DIVariable, all alike, only differing in the complex address reference at the end. By making the complex address into an extra argument of the dbg.value/dbg.declare intrinsics, all of the pieces can reference the same variable and the complex address expressions can be uniqued across the CU, too. Down the road, this will allow us to move other flags, such as "indirection" out of the DIVariable, too. The new intrinsics look like this: declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr) declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr) This patch adds a new LLVM-local tag to DIExpressions, so we can detect and pretty-print DIExpression metadata nodes. What this patch doesn't do: This patch does not touch the "Indirect" field in DIVariable; but moving that into the expression would be a natural next step. http://reviews.llvm.org/D4919 rdar://problem/17994491 Thanks to dblaikie and dexonsmith for reviewing this patch! Note: I accidentally committed a bogus older version of this patch previously. llvm-svn: 218787
2014-10-02 02:55:02 +08:00
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
const DIExpression *Expr = MI->getDebugExpression();
Refactor the DIExpression fragment query interface (NFC) ... so it becomes available to DIExpressionCursor. llvm-svn: 290322
2016-12-22 13:27:12 +08:00
auto Fragment = Expr->getFragmentInfo();
if (Fragment)
OS << " [fragment offset=" << Fragment->OffsetInBits
<< " size=" << Fragment->SizeInBits << "]";
Debug info: Infrastructure to support debug locations for fragmented variables (for example, by-value struct arguments passed in registers, or large integer values split across several smaller registers). On the IR level, this adds a new type of complex address operation OpPiece to DIVariable that describes size and offset of a variable fragment. On the DWARF emitter level, all pieces describing the same variable are collected, sorted and emitted as DWARF expressions using the DW_OP_piece and DW_OP_bit_piece operators. http://reviews.llvm.org/D3373 rdar://problem/15928306 What this patch doesn't do / Future work: - This patch only adds the backend machinery to make this work, patches that change SROA and SelectionDAG's type legalizer to actually create such debug info will follow. (http://reviews.llvm.org/D2680) - Making the DIVariable complex expressions into an argument of dbg.value will reduce the memory footprint of the debug metadata. - The sorting/uniquing of pieces should be moved into DebugLocEntry, to facilitate the merging of multi-piece entries. llvm-svn: 214576
2014-08-02 06:11:58 +08:00
OS << " <- ";
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
Reapply an improved version of r180816/180817. Change the informal convention of DBG_VALUE machine instructions so that we can express a register-indirect address with an offset of 0. The old convention was that a DBG_VALUE is a register-indirect value if the offset (operand 1) is nonzero. The new convention is that a DBG_VALUE is register-indirect if the first operand is a register and the second operand is an immediate. For plain register values the combination reg, reg is used. MachineInstrBuilder::BuildMI knows how to build the new DBG_VALUES. rdar://problem/13658587 llvm-svn: 185966
2013-07-10 04:28:37 +08:00
// The second operand is only an offset if it's an immediate.
PR32382: Fix emitting complex DWARF expressions. The DWARF specification knows 3 kinds of non-empty simple location descriptions: 1. Register location descriptions - describe a variable in a register - consist of only a DW_OP_reg 2. Memory location descriptions - describe the address of a variable 3. Implicit location descriptions - describe the value of a variable - end with DW_OP_stack_value & friends The existing DwarfExpression code is pretty much ignorant of these restrictions. This used to not matter because we only emitted very short expressions that we happened to get right by accident. This patch makes DwarfExpression aware of the rules defined by the DWARF standard and now chooses the right kind of location description for each expression being emitted. This would have been an NFC commit (for the existing testsuite) if not for the way that clang describes captured block variables. Based on how the previous code in LLVM emitted locations, DW_OP_deref operations that should have come at the end of the expression are put at its beginning. Fixing this means changing the semantics of DIExpression, so this patch bumps the version number of DIExpression and implements a bitcode upgrade. There are two major changes in this patch: I had to fix the semantics of dbg.declare for describing function arguments. After this patch a dbg.declare always takes the *address* of a variable as the first argument, even if the argument is not an alloca. When lowering a DBG_VALUE, the decision of whether to emit a register location description or a memory location description depends on the MachineLocation — register machine locations may get promoted to memory locations based on their DIExpression. (Future) optimization passes that want to salvage implicit debug location for variables may do so by appending a DW_OP_stack_value. For example: DBG_VALUE, [RBP-8] --> DW_OP_fbreg -8 DBG_VALUE, RAX --> DW_OP_reg0 +0 DBG_VALUE, RAX, DIExpression(DW_OP_deref) --> DW_OP_reg0 +0 All testcases that were modified were regenerated from clang. I also added source-based testcases for each of these to the debuginfo-tests repository over the last week to make sure that no synchronized bugs slip in. The debuginfo-tests compile from source and run the debugger. https://bugs.llvm.org/show_bug.cgi?id=32382 <rdar://problem/31205000> Differential Revision: https://reviews.llvm.org/D31439 llvm-svn: 300522
2017-04-18 09:21:53 +08:00
bool Deref = false;
bool MemLoc = MI->getOperand(0).isReg() && MI->getOperand(1).isImm();
int64_t Offset = MemLoc ? MI->getOperand(1).getImm() : 0;
Clean up the processing of dbg.value in various places Summary: First up is instcombine, where in the dbg.declare -> dbg.value conversion, the llvm.dbg.value needs to be called on the actual loaded value, rather than the address (since the whole point of this transformation is to be able to get rid of the alloca). Further, now that that's cleaned up, we can remove a hack in the backend, that would add an implicit OP_deref if the argument to dbg.value was an alloca. This stems from before the existence of DIExpression and is no longer necessary since the deref can be expressed explicitly. Now, in order to make sure that the tests pass with this change, we need to correct the printing of DEBUG_VALUE comments to take into account the expression, which wasn't taken into account before. Unfortunately, for both these changes, there were a number of incorrect test cases (mostly the wrong number of DW_OP_derefs, but also a couple where the test itself was broken more badly). aprantl and I have gone through and adjusted these test case in order to make them pass with these fixes and in some cases to make sure they're actually testing what they are meant to test. Reviewers: aprantl Subscribers: dsanders Differential Revision: http://reviews.llvm.org/D14186 llvm-svn: 256077
2015-12-19 10:02:44 +08:00
for (unsigned i = 0; i < Expr->getNumElements(); ++i) {
Improve DEBUG_VALUE assembly comments for spilled bitpieces Previously we would give up when we saw the bitpiece DWARF expression and print "[complex expression]" when actually we handled bitpiece expressions outside the loop. llvm-svn: 283355
2016-10-06 02:36:02 +08:00
uint64_t Op = Expr->getElement(i);
[DIExpression] Introduce a dedicated DW_OP_LLVM_fragment operation so we can stop using DW_OP_bit_piece with the wrong semantics. The entire back story can be found here: http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20161114/405934.html The gist is that in LLVM we've been misinterpreting DW_OP_bit_piece's offset field to mean the offset into the source variable rather than the offset into the location at the top the DWARF expression stack. In order to be able to fix this in a subsequent patch, this patch introduces a dedicated DW_OP_LLVM_fragment operation with the semantics that we used to apply to DW_OP_bit_piece, which is what we actually need while inside of LLVM. This patch is complete with a bitcode upgrade for expressions using the old format. It does not yet fix the DWARF backend to use DW_OP_bit_piece correctly. Implementation note: We discussed several options for implementing this, including reserving a dedicated field in DIExpression for the fragment size and offset, but using an custom operator at the end of the expression works just fine and is more efficient because we then only pay for it when we need it. Differential Revision: https://reviews.llvm.org/D27361 rdar://problem/29335809 llvm-svn: 288683
2016-12-06 02:04:47 +08:00
if (Op == dwarf::DW_OP_LLVM_fragment) {
Improve DEBUG_VALUE assembly comments for spilled bitpieces Previously we would give up when we saw the bitpiece DWARF expression and print "[complex expression]" when actually we handled bitpiece expressions outside the loop. llvm-svn: 283355
2016-10-06 02:36:02 +08:00
// There can't be any operands after this in a valid expression
break;
} else if (Deref) {
Clean up the processing of dbg.value in various places Summary: First up is instcombine, where in the dbg.declare -> dbg.value conversion, the llvm.dbg.value needs to be called on the actual loaded value, rather than the address (since the whole point of this transformation is to be able to get rid of the alloca). Further, now that that's cleaned up, we can remove a hack in the backend, that would add an implicit OP_deref if the argument to dbg.value was an alloca. This stems from before the existence of DIExpression and is no longer necessary since the deref can be expressed explicitly. Now, in order to make sure that the tests pass with this change, we need to correct the printing of DEBUG_VALUE comments to take into account the expression, which wasn't taken into account before. Unfortunately, for both these changes, there were a number of incorrect test cases (mostly the wrong number of DW_OP_derefs, but also a couple where the test itself was broken more badly). aprantl and I have gone through and adjusted these test case in order to make them pass with these fixes and in some cases to make sure they're actually testing what they are meant to test. Reviewers: aprantl Subscribers: dsanders Differential Revision: http://reviews.llvm.org/D14186 llvm-svn: 256077
2015-12-19 10:02:44 +08:00
// We currently don't support extra Offsets or derefs after the first
PR32382: Fix emitting complex DWARF expressions. The DWARF specification knows 3 kinds of non-empty simple location descriptions: 1. Register location descriptions - describe a variable in a register - consist of only a DW_OP_reg 2. Memory location descriptions - describe the address of a variable 3. Implicit location descriptions - describe the value of a variable - end with DW_OP_stack_value & friends The existing DwarfExpression code is pretty much ignorant of these restrictions. This used to not matter because we only emitted very short expressions that we happened to get right by accident. This patch makes DwarfExpression aware of the rules defined by the DWARF standard and now chooses the right kind of location description for each expression being emitted. This would have been an NFC commit (for the existing testsuite) if not for the way that clang describes captured block variables. Based on how the previous code in LLVM emitted locations, DW_OP_deref operations that should have come at the end of the expression are put at its beginning. Fixing this means changing the semantics of DIExpression, so this patch bumps the version number of DIExpression and implements a bitcode upgrade. There are two major changes in this patch: I had to fix the semantics of dbg.declare for describing function arguments. After this patch a dbg.declare always takes the *address* of a variable as the first argument, even if the argument is not an alloca. When lowering a DBG_VALUE, the decision of whether to emit a register location description or a memory location description depends on the MachineLocation — register machine locations may get promoted to memory locations based on their DIExpression. (Future) optimization passes that want to salvage implicit debug location for variables may do so by appending a DW_OP_stack_value. For example: DBG_VALUE, [RBP-8] --> DW_OP_fbreg -8 DBG_VALUE, RAX --> DW_OP_reg0 +0 DBG_VALUE, RAX, DIExpression(DW_OP_deref) --> DW_OP_reg0 +0 All testcases that were modified were regenerated from clang. I also added source-based testcases for each of these to the debuginfo-tests repository over the last week to make sure that no synchronized bugs slip in. The debuginfo-tests compile from source and run the debugger. https://bugs.llvm.org/show_bug.cgi?id=32382 <rdar://problem/31205000> Differential Revision: https://reviews.llvm.org/D31439 llvm-svn: 300522
2017-04-18 09:21:53 +08:00
// one. Bail out early instead of emitting an incorrect comment.
Clean up the processing of dbg.value in various places Summary: First up is instcombine, where in the dbg.declare -> dbg.value conversion, the llvm.dbg.value needs to be called on the actual loaded value, rather than the address (since the whole point of this transformation is to be able to get rid of the alloca). Further, now that that's cleaned up, we can remove a hack in the backend, that would add an implicit OP_deref if the argument to dbg.value was an alloca. This stems from before the existence of DIExpression and is no longer necessary since the deref can be expressed explicitly. Now, in order to make sure that the tests pass with this change, we need to correct the printing of DEBUG_VALUE comments to take into account the expression, which wasn't taken into account before. Unfortunately, for both these changes, there were a number of incorrect test cases (mostly the wrong number of DW_OP_derefs, but also a couple where the test itself was broken more badly). aprantl and I have gone through and adjusted these test case in order to make them pass with these fixes and in some cases to make sure they're actually testing what they are meant to test. Reviewers: aprantl Subscribers: dsanders Differential Revision: http://reviews.llvm.org/D14186 llvm-svn: 256077
2015-12-19 10:02:44 +08:00
OS << " [complex expression]";
AP.OutStreamer->emitRawComment(OS.str());
return true;
Improve DEBUG_VALUE assembly comments for spilled bitpieces Previously we would give up when we saw the bitpiece DWARF expression and print "[complex expression]" when actually we handled bitpiece expressions outside the loop. llvm-svn: 283355
2016-10-06 02:36:02 +08:00
} else if (Op == dwarf::DW_OP_deref) {
Clean up the processing of dbg.value in various places Summary: First up is instcombine, where in the dbg.declare -> dbg.value conversion, the llvm.dbg.value needs to be called on the actual loaded value, rather than the address (since the whole point of this transformation is to be able to get rid of the alloca). Further, now that that's cleaned up, we can remove a hack in the backend, that would add an implicit OP_deref if the argument to dbg.value was an alloca. This stems from before the existence of DIExpression and is no longer necessary since the deref can be expressed explicitly. Now, in order to make sure that the tests pass with this change, we need to correct the printing of DEBUG_VALUE comments to take into account the expression, which wasn't taken into account before. Unfortunately, for both these changes, there were a number of incorrect test cases (mostly the wrong number of DW_OP_derefs, but also a couple where the test itself was broken more badly). aprantl and I have gone through and adjusted these test case in order to make them pass with these fixes and in some cases to make sure they're actually testing what they are meant to test. Reviewers: aprantl Subscribers: dsanders Differential Revision: http://reviews.llvm.org/D14186 llvm-svn: 256077
2015-12-19 10:02:44 +08:00
Deref = true;
continue;
}
Improve DEBUG_VALUE assembly comments for spilled bitpieces Previously we would give up when we saw the bitpiece DWARF expression and print "[complex expression]" when actually we handled bitpiece expressions outside the loop. llvm-svn: 283355
2016-10-06 02:36:02 +08:00
Clean up the processing of dbg.value in various places Summary: First up is instcombine, where in the dbg.declare -> dbg.value conversion, the llvm.dbg.value needs to be called on the actual loaded value, rather than the address (since the whole point of this transformation is to be able to get rid of the alloca). Further, now that that's cleaned up, we can remove a hack in the backend, that would add an implicit OP_deref if the argument to dbg.value was an alloca. This stems from before the existence of DIExpression and is no longer necessary since the deref can be expressed explicitly. Now, in order to make sure that the tests pass with this change, we need to correct the printing of DEBUG_VALUE comments to take into account the expression, which wasn't taken into account before. Unfortunately, for both these changes, there were a number of incorrect test cases (mostly the wrong number of DW_OP_derefs, but also a couple where the test itself was broken more badly). aprantl and I have gone through and adjusted these test case in order to make them pass with these fixes and in some cases to make sure they're actually testing what they are meant to test. Reviewers: aprantl Subscribers: dsanders Differential Revision: http://reviews.llvm.org/D14186 llvm-svn: 256077
2015-12-19 10:02:44 +08:00
uint64_t ExtraOffset = Expr->getElement(i++);
if (Op == dwarf::DW_OP_plus)
Offset += ExtraOffset;
else {
assert(Op == dwarf::DW_OP_minus);
Offset -= ExtraOffset;
}
}
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
// Register or immediate value. Register 0 means undef.
if (MI->getOperand(0).isFPImm()) {
APFloat APF = APFloat(MI->getOperand(0).getFPImm()->getValueAPF());
if (MI->getOperand(0).getFPImm()->getType()->isFloatTy()) {
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
OS << (double)APF.convertToFloat();
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
} else if (MI->getOperand(0).getFPImm()->getType()->isDoubleTy()) {
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
OS << APF.convertToDouble();
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
} else {
// There is no good way to print long double. Convert a copy to
// double. Ah well, it's only a comment.
bool ignored;
Replace APFloatBase static fltSemantics data members with getter functions At least the plugin used by the LibreOffice build (<https://wiki.documentfoundation.org/Development/Clang_plugins>) indirectly uses those members (through inline functions in LLVM/Clang include files in turn using them), but they are not exported by utils/extract_symbols.py on Windows, and accessing data across DLL/EXE boundaries on Windows is generally problematic. Differential Revision: https://reviews.llvm.org/D26671 llvm-svn: 289647
2016-12-14 19:57:17 +08:00
APF.convert(APFloat::IEEEdouble(), APFloat::rmNearestTiesToEven,
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
&ignored);
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
OS << "(long double) " << APF.convertToDouble();
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
}
} else if (MI->getOperand(0).isImm()) {
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
OS << MI->getOperand(0).getImm();
Handle debug info for i128 constants. llvm-svn: 133821
2011-06-25 04:46:11 +08:00
} else if (MI->getOperand(0).isCImm()) {
MI->getOperand(0).getCImm()->getValue().print(OS, false /*isSigned*/);
tidy up llvm-svn: 100700
2010-04-08 06:29:10 +08:00
} else {
Debug Info: Simplify Frame Index handling in DBG_VALUE Machine Instructions Rather than using the full power of target-specific addressing modes in DBG_VALUEs with Frame Indicies, simply use Frame Index + Offset. This reduces the complexity of debug info handling down to two representations of values (reg+offset and frame index+offset) rather than three or four. Ideally we could ensure that frame indicies had been eliminated by the time we reached an assembly or dwarf generation, but I haven't spent the time to figure out where the FIs are leaking through into that & whether there's a good place to convert them. Some FI+offset=>reg+offset conversion is done (see PrologEpilogInserter, for example) which is necessary for some SelectionDAG assumptions about registers, I believe, but it might be possible to make this a more thorough conversion & ensure there are no remaining FIs no matter how instruction selection is performed. llvm-svn: 184066
2013-06-17 04:34:15 +08:00
unsigned Reg;
if (MI->getOperand(0).isReg()) {
Reg = MI->getOperand(0).getReg();
} else {
assert(MI->getOperand(0).isFI() && "Unknown operand type");
Remove more uses of TargetMachine::getSubtargetImpl from the AsmPrinter. getSubtargetInfo now asserts that the MachineFunction exists. Debug printing of register naming now uses the register info from MCAsmInfo as that's unchanging. llvm-svn: 229978
2015-02-20 15:16:19 +08:00
const TargetFrameLowering *TFI = AP.MF->getSubtarget().getFrameLowering();
80-column and tab character fixes. llvm-svn: 184792
2013-06-25 07:20:02 +08:00
Offset += TFI->getFrameIndexReference(*AP.MF,
MI->getOperand(0).getIndex(), Reg);
Debug Info: Simplify Frame Index handling in DBG_VALUE Machine Instructions Rather than using the full power of target-specific addressing modes in DBG_VALUEs with Frame Indicies, simply use Frame Index + Offset. This reduces the complexity of debug info handling down to two representations of values (reg+offset and frame index+offset) rather than three or four. Ideally we could ensure that frame indicies had been eliminated by the time we reached an assembly or dwarf generation, but I haven't spent the time to figure out where the FIs are leaking through into that & whether there's a good place to convert them. Some FI+offset=>reg+offset conversion is done (see PrologEpilogInserter, for example) which is necessary for some SelectionDAG assumptions about registers, I believe, but it might be possible to make this a more thorough conversion & ensure there are no remaining FIs no matter how instruction selection is performed. llvm-svn: 184066
2013-06-17 04:34:15 +08:00
Deref = true;
}
DebugInfo: print DBG_VALUE MachineInstrs with [] for deref and drop the offset when it's zero llvm-svn: 184045
2013-06-15 23:52:58 +08:00
if (Reg == 0) {
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
// Suppress offset, it is not meaningful here.
Use raw_ostream. llvm-svn: 100615
2010-04-07 17:26:51 +08:00
OS << "undef";
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
// NOTE: Want this comment at start of line, don't emit with AddComment.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->emitRawComment(OS.str());
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
return true;
}
PR32382: Fix emitting complex DWARF expressions. The DWARF specification knows 3 kinds of non-empty simple location descriptions: 1. Register location descriptions - describe a variable in a register - consist of only a DW_OP_reg 2. Memory location descriptions - describe the address of a variable 3. Implicit location descriptions - describe the value of a variable - end with DW_OP_stack_value & friends The existing DwarfExpression code is pretty much ignorant of these restrictions. This used to not matter because we only emitted very short expressions that we happened to get right by accident. This patch makes DwarfExpression aware of the rules defined by the DWARF standard and now chooses the right kind of location description for each expression being emitted. This would have been an NFC commit (for the existing testsuite) if not for the way that clang describes captured block variables. Based on how the previous code in LLVM emitted locations, DW_OP_deref operations that should have come at the end of the expression are put at its beginning. Fixing this means changing the semantics of DIExpression, so this patch bumps the version number of DIExpression and implements a bitcode upgrade. There are two major changes in this patch: I had to fix the semantics of dbg.declare for describing function arguments. After this patch a dbg.declare always takes the *address* of a variable as the first argument, even if the argument is not an alloca. When lowering a DBG_VALUE, the decision of whether to emit a register location description or a memory location description depends on the MachineLocation — register machine locations may get promoted to memory locations based on their DIExpression. (Future) optimization passes that want to salvage implicit debug location for variables may do so by appending a DW_OP_stack_value. For example: DBG_VALUE, [RBP-8] --> DW_OP_fbreg -8 DBG_VALUE, RAX --> DW_OP_reg0 +0 DBG_VALUE, RAX, DIExpression(DW_OP_deref) --> DW_OP_reg0 +0 All testcases that were modified were regenerated from clang. I also added source-based testcases for each of these to the debuginfo-tests repository over the last week to make sure that no synchronized bugs slip in. The debuginfo-tests compile from source and run the debugger. https://bugs.llvm.org/show_bug.cgi?id=32382 <rdar://problem/31205000> Differential Revision: https://reviews.llvm.org/D31439 llvm-svn: 300522
2017-04-18 09:21:53 +08:00
if (MemLoc || Deref)
DebugInfo: print DBG_VALUE MachineInstrs with [] for deref and drop the offset when it's zero llvm-svn: 184045
2013-06-15 23:52:58 +08:00
OS << '[';
Use TargetRegisterInfo for printing MachineOperand register comments Several places in AsmPrinter.cpp print comments describing MachineOperand registers using MCRegisterInfo, which uses MCOperand-oriented names. This doesn't work for targets that use virtual registers exclusively, as WebAssembly does, since virtual registers are represented and printed differently. This patch preserves what seems to be the spirit of r229978, avoiding the use of TM.getSubtargetImpl(), while still using MachineOperand-oriented printing for MachineOperands. Differential Revision: http://reviews.llvm.org/D14709 llvm-svn: 253338
2015-11-18 00:01:28 +08:00
OS << PrintReg(Reg, AP.MF->getSubtarget().getRegisterInfo());
tidy up llvm-svn: 100700
2010-04-08 06:29:10 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
PR32382: Fix emitting complex DWARF expressions. The DWARF specification knows 3 kinds of non-empty simple location descriptions: 1. Register location descriptions - describe a variable in a register - consist of only a DW_OP_reg 2. Memory location descriptions - describe the address of a variable 3. Implicit location descriptions - describe the value of a variable - end with DW_OP_stack_value & friends The existing DwarfExpression code is pretty much ignorant of these restrictions. This used to not matter because we only emitted very short expressions that we happened to get right by accident. This patch makes DwarfExpression aware of the rules defined by the DWARF standard and now chooses the right kind of location description for each expression being emitted. This would have been an NFC commit (for the existing testsuite) if not for the way that clang describes captured block variables. Based on how the previous code in LLVM emitted locations, DW_OP_deref operations that should have come at the end of the expression are put at its beginning. Fixing this means changing the semantics of DIExpression, so this patch bumps the version number of DIExpression and implements a bitcode upgrade. There are two major changes in this patch: I had to fix the semantics of dbg.declare for describing function arguments. After this patch a dbg.declare always takes the *address* of a variable as the first argument, even if the argument is not an alloca. When lowering a DBG_VALUE, the decision of whether to emit a register location description or a memory location description depends on the MachineLocation — register machine locations may get promoted to memory locations based on their DIExpression. (Future) optimization passes that want to salvage implicit debug location for variables may do so by appending a DW_OP_stack_value. For example: DBG_VALUE, [RBP-8] --> DW_OP_fbreg -8 DBG_VALUE, RAX --> DW_OP_reg0 +0 DBG_VALUE, RAX, DIExpression(DW_OP_deref) --> DW_OP_reg0 +0 All testcases that were modified were regenerated from clang. I also added source-based testcases for each of these to the debuginfo-tests repository over the last week to make sure that no synchronized bugs slip in. The debuginfo-tests compile from source and run the debugger. https://bugs.llvm.org/show_bug.cgi?id=32382 <rdar://problem/31205000> Differential Revision: https://reviews.llvm.org/D31439 llvm-svn: 300522
2017-04-18 09:21:53 +08:00
if (MemLoc || Deref)
Reapply an improved version of r180816/180817. Change the informal convention of DBG_VALUE machine instructions so that we can express a register-indirect address with an offset of 0. The old convention was that a DBG_VALUE is a register-indirect value if the offset (operand 1) is nonzero. The new convention is that a DBG_VALUE is register-indirect if the first operand is a register and the second operand is an immediate. For plain register values the combination reg, reg is used. MachineInstrBuilder::BuildMI knows how to build the new DBG_VALUES. rdar://problem/13658587 llvm-svn: 185966
2013-07-10 04:28:37 +08:00
OS << '+' << Offset << ']';
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
// NOTE: Want this comment at start of line, don't emit with AddComment.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->emitRawComment(OS.str());
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
return true;
}
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
Use .cfi_sections to put the unwind info in .debug_frame when possible. With this clang will use .debug_frame in, for example, clang -g -c -m32 test.c This matches gcc's behaviour. It looks like .debug_frame is a bit bigger than .eh_frame, but has the big advantage of not being allocated. llvm-svn: 131140
2011-05-11 02:39:09 +08:00
AsmPrinter::CFIMoveType AsmPrinter::needsCFIMoves() {
Replace the -unwind-tables option with a per function flag. This is more LTO friendly as we can now correctly merge files compiled with or without -fasynchronous-unwind-tables. llvm-svn: 132033
2011-05-25 11:44:17 +08:00
if (MAI->getExceptionHandlingType() == ExceptionHandling::DwarfCFI &&
MF->getFunction()->needsUnwindTableEntry())
return CFI_M_EH;
Factor some code to needsCFIMoves. Avoid printing moves when we don't have to. llvm-svn: 130501
2011-04-29 22:14:06 +08:00
Use .cfi_sections to put the unwind info in .debug_frame when possible. With this clang will use .debug_frame in, for example, clang -g -c -m32 test.c This matches gcc's behaviour. It looks like .debug_frame is a bit bigger than .eh_frame, but has the big advantage of not being allocated. llvm-svn: 131140
2011-05-11 02:39:09 +08:00
if (MMI->hasDebugInfo())
return CFI_M_Debug;
Factor some code to needsCFIMoves. Avoid printing moves when we don't have to. llvm-svn: 130501
2011-04-29 22:14:06 +08:00
Use .cfi_sections to put the unwind info in .debug_frame when possible. With this clang will use .debug_frame in, for example, clang -g -c -m32 test.c This matches gcc's behaviour. It looks like .debug_frame is a bit bigger than .eh_frame, but has the big advantage of not being allocated. llvm-svn: 131140
2011-05-11 02:39:09 +08:00
return CFI_M_None;
Factor some code to needsCFIMoves. Avoid printing moves when we don't have to. llvm-svn: 130501
2011-04-29 22:14:06 +08:00
}
When generating code for Win64 EH, emit StartProc and EndProc directives. llvm-svn: 132250
2011-05-28 12:21:04 +08:00
bool AsmPrinter::needsSEHMoves() {
Add the ExceptionHandling::MSVC enumeration It is intended to be used for a family of personality functions that have similar IR preparation requirements. Typically when interoperating with MSVC personality functions, bits of functionality need to be outlined from the main function into helper functions. There is also usually more than one landing pad per invoke, which does not match the LLVM IR landingpad representation. None of this is implemented yet. This change just adds a new enum that is active for *-windows-msvc and delegates to the EH removal preparation pass. No functionality change for other targets. llvm-svn: 224625
2014-12-20 06:19:48 +08:00
return MAI->usesWindowsCFI() && MF->getFunction()->needsUnwindTableEntry();
When generating code for Win64 EH, emit StartProc and EndProc directives. llvm-svn: 132250
2011-05-28 12:21:04 +08:00
}
Replace PROLOG_LABEL with a new CFI_INSTRUCTION. The old system was fairly convoluted: * A temporary label was created. * A single PROLOG_LABEL was created with it. * A few MCCFIInstructions were created with the same label. The semantics were that the cfi instructions were mapped to the PROLOG_LABEL via the temporary label. The output position was that of the PROLOG_LABEL. The temporary label itself was used only for doing the mapping. The new CFI_INSTRUCTION has a 1:1 mapping to MCCFIInstructions and points to one by holding an index into the CFI instructions of this function. I did consider removing MMI.getFrameInstructions completelly and having CFI_INSTRUCTION own a MCCFIInstruction, but MCCFIInstructions have non trivial constructors and destructors and are somewhat big, so the this setup is probably better. The net result is that we don't create temporary labels that are never used. llvm-svn: 203204
2014-03-07 14:08:31 +08:00
void AsmPrinter::emitCFIInstruction(const MachineInstr &MI) {
CodeGen: rename Win64 ExceptionHandling to WinEH This exception format is not specific to Windows x64. A similar approach is taken on nearly all architectures. Generalise the name to reflect reality. This will eventually be used for Windows on ARM data emission as well. Switch the enum and namespace into an enum class. llvm-svn: 212000
2014-06-30 05:43:47 +08:00
ExceptionHandling ExceptionHandlingType = MAI->getExceptionHandlingType();
Generate the DWARF stack frame decode operations in the function prologue for ARM/Thumb functions. Patch by Keith Walker! llvm-svn: 201423
2014-02-15 01:19:07 +08:00
if (ExceptionHandlingType != ExceptionHandling::DwarfCFI &&
ExceptionHandlingType != ExceptionHandling::ARM)
Add 129518 back with a fix for when we are producing eh just because of debug info. Change ELF systems to use CFI for producing the EH tables. This reduces the size of the clang binary in Debug builds from 690MB to 679MB. llvm-svn: 129571
2011-04-15 23:11:06 +08:00
return;
Use .cfi_sections to put the unwind info in .debug_frame when possible. With this clang will use .debug_frame in, for example, clang -g -c -m32 test.c This matches gcc's behaviour. It looks like .debug_frame is a bit bigger than .eh_frame, but has the big advantage of not being allocated. llvm-svn: 131140
2011-05-11 02:39:09 +08:00
if (needsCFIMoves() == CFI_M_None)
Factor some code to needsCFIMoves. Avoid printing moves when we don't have to. llvm-svn: 130501
2011-04-29 22:14:06 +08:00
return;
Move FrameInstructions from MachineModuleInfo to MachineFunction This is per function data so it is better kept at the function instead of the module. This is a necessary step to have machine module passes work properly. Differential Revision: https://reviews.llvm.org/D27185 llvm-svn: 288291
2016-12-01 07:48:42 +08:00
const std::vector<MCCFIInstruction> &Instrs = MF->getFrameInstructions();
Replace PROLOG_LABEL with a new CFI_INSTRUCTION. The old system was fairly convoluted: * A temporary label was created. * A single PROLOG_LABEL was created with it. * A few MCCFIInstructions were created with the same label. The semantics were that the cfi instructions were mapped to the PROLOG_LABEL via the temporary label. The output position was that of the PROLOG_LABEL. The temporary label itself was used only for doing the mapping. The new CFI_INSTRUCTION has a 1:1 mapping to MCCFIInstructions and points to one by holding an index into the CFI instructions of this function. I did consider removing MMI.getFrameInstructions completelly and having CFI_INSTRUCTION own a MCCFIInstruction, but MCCFIInstructions have non trivial constructors and destructors and are somewhat big, so the this setup is probably better. The net result is that we don't create temporary labels that are never used. llvm-svn: 203204
2014-03-07 14:08:31 +08:00
unsigned CFIIndex = MI.getOperand(0).getCFIIndex();
const MCCFIInstruction &CFI = Instrs[CFIIndex];
emitCFIInstruction(CFI);
Add 129518 back with a fix for when we are producing eh just because of debug info. Change ELF systems to use CFI for producing the EH tables. This reduces the size of the clang binary in Debug builds from 690MB to 679MB. llvm-svn: 129571
2011-04-15 23:11:06 +08:00
}
Add the llvm.frameallocate and llvm.recoverframeallocation intrinsics These intrinsics allow multiple functions to share a single stack allocation from one function's call frame. The function with the allocation may only perform one allocation, and it must be in the entry block. Functions accessing the allocation call llvm.recoverframeallocation with the function whose frame they are accessing and a frame pointer from an active call frame of that function. These intrinsics are very difficult to inline correctly, so the intention is that they be introduced rarely, or at least very late during EH preparation. Reviewers: echristo, andrew.w.kaylor Differential Revision: http://reviews.llvm.org/D6493 llvm-svn: 225746
2015-01-13 08:48:10 +08:00
void AsmPrinter::emitFrameAlloc(const MachineInstr &MI) {
// The operands are the MCSymbol and the frame offset of the allocation.
MCSymbol *FrameAllocSym = MI.getOperand(0).getMCSymbol();
int FrameOffset = MI.getOperand(1).getImm();
// Emit a symbol assignment.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitAssignment(FrameAllocSym,
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
MCConstantExpr::create(FrameOffset, OutContext));
Add the llvm.frameallocate and llvm.recoverframeallocation intrinsics These intrinsics allow multiple functions to share a single stack allocation from one function's call frame. The function with the allocation may only perform one allocation, and it must be in the entry block. Functions accessing the allocation call llvm.recoverframeallocation with the function whose frame they are accessing and a frame pointer from an active call frame of that function. These intrinsics are very difficult to inline correctly, so the intention is that they be introduced rarely, or at least very late during EH preparation. Reviewers: echristo, andrew.w.kaylor Differential Revision: http://reviews.llvm.org/D6493 llvm-svn: 225746
2015-01-13 08:48:10 +08:00
}
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
/// EmitFunctionBody - This method emits the body and trailer for a
/// function.
void AsmPrinter::EmitFunctionBody() {
Make EmitFunctionHeader a private helper. llvm-svn: 232481
2015-03-17 22:38:30 +08:00
EmitFunctionHeader();
add target hooks for emitting random gunk before and after the function body. llvm-svn: 94732
2010-01-28 09:58:58 +08:00
// Emit target-specific gunk before the function body.
EmitFunctionBodyStart();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
bool ShouldPrintDebugScopes = MMI->hasDebugInfo();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
// Print out code for the function.
bool HasAnyRealCode = false;
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
int NumInstsInFunction = 0;
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (auto &MBB : *MF) {
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
// Print a label for the basic block.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
EmitBasicBlockStart(MBB);
for (auto &MI : MBB) {
Consider this function: void foo() { __builtin_unreachable(); } It will output the following on Darwin X86: _func1: Leh_func_begin0: pushq %rbp Ltmp0: movq %rsp, %rbp Ltmp1: Leh_func_end0: This prolog adds a new Call Frame Information (CFI) row to the FDE with an address that is not within the address range of the code it describes -- part is equal to the end of the function -- and therefore results in an invalid EH frame. If we emit a nop in this situation, then the CFI row is now within the address range. llvm-svn: 108568
2010-07-17 06:51:10 +08:00
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
// Print the assembly for the instruction.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (!MI.isPosition() && !MI.isImplicitDef() && !MI.isKill() &&
!MI.isDebugValue()) {
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
HasAnyRealCode = true;
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
++NumInstsInFunction;
Do not count kill, implicit_def instructions as printed instructions. llvm-svn: 102453
2010-04-28 03:38:45 +08:00
}
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
if (ShouldPrintDebugScopes) {
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription,
HI.TimerGroupName, HI.TimerGroupDescription,
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
TimePassesIsEnabled);
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
HI.Handler->beginInstruction(&MI);
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
}
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
clean up the asmprinter header and privatize some stuff. llvm-svn: 100342
2010-04-05 02:52:31 +08:00
if (isVerbose())
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
emitComments(MI, OutStreamer->GetCommentOS(), this);
Move verbose asm instruction comments to using MCStreamer. The major win of this is that the code is simpler and they print on the same line as the instruction again: movl %eax, 96(%esp) ## 4-byte Spill movl 96(%esp), %eax ## 4-byte Reload cmpl 92(%esp), %eax ## 4-byte Folded Reload jl LBB7_86 llvm-svn: 95738
2010-02-10 08:47:53 +08:00
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
switch (MI.getOpcode()) {
Replace PROLOG_LABEL with a new CFI_INSTRUCTION. The old system was fairly convoluted: * A temporary label was created. * A single PROLOG_LABEL was created with it. * A few MCCFIInstructions were created with the same label. The semantics were that the cfi instructions were mapped to the PROLOG_LABEL via the temporary label. The output position was that of the PROLOG_LABEL. The temporary label itself was used only for doing the mapping. The new CFI_INSTRUCTION has a 1:1 mapping to MCCFIInstructions and points to one by holding an index into the CFI instructions of this function. I did consider removing MMI.getFrameInstructions completelly and having CFI_INSTRUCTION own a MCCFIInstruction, but MCCFIInstructions have non trivial constructors and destructors and are somewhat big, so the this setup is probably better. The net result is that we don't create temporary labels that are never used. llvm-svn: 203204
2014-03-07 14:08:31 +08:00
case TargetOpcode::CFI_INSTRUCTION:
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
emitCFIInstruction(MI);
Add 129518 back with a fix for when we are producing eh just because of debug info. Change ELF systems to use CFI for producing the EH tables. This reduces the size of the clang binary in Debug builds from 690MB to 679MB. llvm-svn: 129571
2011-04-15 23:11:06 +08:00
break;
Rename llvm.frameescape and llvm.framerecover to localescape and localrecover Summary: Initially, these intrinsics seemed like part of a family of "frame" related intrinsics, but now I think that's more confusing than helpful. Initially, the LangRef specified that this would create a new kind of allocation that would be allocated at a fixed offset from the frame pointer (EBP/RBP). We ended up dropping that design, and leaving the stack frame layout alone. These intrinsics are really about sharing local stack allocations, not frame pointers. I intend to go further and add an `llvm.localaddress()` intrinsic that returns whatever register (EBP, ESI, ESP, RBX) is being used to address locals, which should not be confused with the frame pointer. Naming suggestions at this point are welcome, I'm happy to re-run sed. Reviewers: majnemer, nicholas Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D11011 llvm-svn: 241633
2015-07-08 06:25:32 +08:00
case TargetOpcode::LOCAL_ESCAPE:
Add the llvm.frameallocate and llvm.recoverframeallocation intrinsics These intrinsics allow multiple functions to share a single stack allocation from one function's call frame. The function with the allocation may only perform one allocation, and it must be in the entry block. Functions accessing the allocation call llvm.recoverframeallocation with the function whose frame they are accessing and a frame pointer from an active call frame of that function. These intrinsics are very difficult to inline correctly, so the intention is that they be introduced rarely, or at least very late during EH preparation. Reviewers: echristo, andrew.w.kaylor Differential Revision: http://reviews.llvm.org/D6493 llvm-svn: 225746
2015-01-13 08:48:10 +08:00
emitFrameAlloc(MI);
break;
move target-independent opcodes out of TargetInstrInfo into TargetOpcodes.h. #include the new TargetOpcodes.h into MachineInstr. Add new inline accessors (like isPHI()) to MachineInstr, and start using them throughout the codebase. llvm-svn: 95687
2010-02-10 03:54:29 +08:00
case TargetOpcode::EH_LABEL:
case TargetOpcode::GC_LABEL:
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(MI.getOperand(0).getMCSymbol());
sink handling of target-independent machine instrs (other than DEBUG_VALUE :( ) into the target indep AsmPrinter.cpp file. This allows elimination of the NO_ASM_WRITER_BOILERPLATE hack among other things. llvm-svn: 95177
2010-02-03 09:00:52 +08:00
break;
move target-independent opcodes out of TargetInstrInfo into TargetOpcodes.h. #include the new TargetOpcodes.h into MachineInstr. Add new inline accessors (like isPHI()) to MachineInstr, and start using them throughout the codebase. llvm-svn: 95687
2010-02-10 03:54:29 +08:00
case TargetOpcode::INLINEASM:
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
EmitInlineAsm(&MI);
sink handling of target-independent machine instrs (other than DEBUG_VALUE :( ) into the target indep AsmPrinter.cpp file. This allows elimination of the NO_ASM_WRITER_BOILERPLATE hack among other things. llvm-svn: 95177
2010-02-03 09:00:52 +08:00
break;
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
case TargetOpcode::DBG_VALUE:
if (isVerbose()) {
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (!emitDebugValueComment(&MI, *this))
EmitInstruction(&MI);
Move printing of target-indepedent DEBUG_VALUE comments into AsmPrinter. Target-dependent form is still generated by FastISel and still handled in X86 code. llvm-svn: 100596
2010-04-07 09:15:14 +08:00
}
break;
move target-independent opcodes out of TargetInstrInfo into TargetOpcodes.h. #include the new TargetOpcodes.h into MachineInstr. Add new inline accessors (like isPHI()) to MachineInstr, and start using them throughout the codebase. llvm-svn: 95687
2010-02-10 03:54:29 +08:00
case TargetOpcode::IMPLICIT_DEF:
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (isVerbose()) emitImplicitDef(&MI);
sink handling of target-independent machine instrs (other than DEBUG_VALUE :( ) into the target indep AsmPrinter.cpp file. This allows elimination of the NO_ASM_WRITER_BOILERPLATE hack among other things. llvm-svn: 95177
2010-02-03 09:00:52 +08:00
break;
move target-independent opcodes out of TargetInstrInfo into TargetOpcodes.h. #include the new TargetOpcodes.h into MachineInstr. Add new inline accessors (like isPHI()) to MachineInstr, and start using them throughout the codebase. llvm-svn: 95687
2010-02-10 03:54:29 +08:00
case TargetOpcode::KILL:
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (isVerbose()) emitKill(&MI, *this);
sink handling of target-independent machine instrs (other than DEBUG_VALUE :( ) into the target indep AsmPrinter.cpp file. This allows elimination of the NO_ASM_WRITER_BOILERPLATE hack among other things. llvm-svn: 95177
2010-02-03 09:00:52 +08:00
break;
default:
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
EmitInstruction(&MI);
sink handling of target-independent machine instrs (other than DEBUG_VALUE :( ) into the target indep AsmPrinter.cpp file. This allows elimination of the NO_ASM_WRITER_BOILERPLATE hack among other things. llvm-svn: 95177
2010-02-03 09:00:52 +08:00
break;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
if (ShouldPrintDebugScopes) {
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription,
HI.TimerGroupName, HI.TimerGroupDescription,
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
TimePassesIsEnabled);
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
HI.Handler->endInstruction();
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
}
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
}
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
}
[X86] Optimize stackmap shadows on X86. This patch minimizes the number of nops that must be emitted on X86 to satisfy stackmap shadow constraints. To minimize the number of nops inserted, the X86AsmPrinter now records the size of the most recent stackmap's shadow in the StackMapShadowTracker class, and tracks the number of instruction bytes emitted since the that stackmap instruction was encountered. Padding is emitted (if it is required at all) immediately before the next stackmap/patchpoint instruction, or at the end of the basic block. This optimization should reduce code-size and improve performance for people using the llvm stackmap intrinsic on X86. <rdar://problem/14959522> llvm-svn: 213892
2014-07-25 04:40:55 +08:00
EmitBasicBlockEnd(MBB);
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
}
Consider this function: void foo() { __builtin_unreachable(); } It will output the following on Darwin X86: _func1: Leh_func_begin0: pushq %rbp Ltmp0: movq %rsp, %rbp Ltmp1: Leh_func_end0: This prolog adds a new Call Frame Information (CFI) row to the FDE with an address that is not within the address range of the code it describes -- part is equal to the end of the function -- and therefore results in an invalid EH frame. If we emit a nop in this situation, then the CFI row is now within the address range. llvm-svn: 108568
2010-07-17 06:51:10 +08:00
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
EmittedInsts += NumInstsInFunction;
MachineOptimizationRemarkAnalysis R(DEBUG_TYPE, "InstructionCount",
MF->getFunction()->getSubprogram(),
&MF->front());
R << ore::NV("NumInstructions", NumInstsInFunction)
<< " instructions in function";
ORE->emit(R);
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
// If the function is empty and the object file uses .subsections_via_symbols,
Give AsmPrinter the most common expected implementation of runOnMachineFunction, and switch PPC to use EmitFunctionBody. The two ppc asmprinters now don't heave to define runOnMachineFunction. llvm-svn: 94722
2010-01-28 09:28:58 +08:00
// then we need to emit *something* to the function body to prevent the
on darwin empty functions need to codegen into something of non-zero length, otherwise labels get incorrectly merged. We handled this by emitting a ".byte 0", but this isn't correct on thumb/arm targets where the text segment needs to be a multiple of 2/4 bytes. Handle this by emitting a noop. This is more gross than it should be because arm/ppc are not fully mc'ized yet. This fixes rdar://7908505 llvm-svn: 102400
2010-04-27 07:37:21 +08:00
// labels from collapsing together. Just emit a noop.
Revert r301040 "X86: Don't emit zero-byte functions on Windows" This broke almost all bots. Reverting while fixing. llvm-svn: 301041
2017-04-22 05:10:37 +08:00
if ((MAI->hasSubsectionsViaSymbols() && !HasAnyRealCode)) {
on darwin empty functions need to codegen into something of non-zero length, otherwise labels get incorrectly merged. We handled this by emitting a ".byte 0", but this isn't correct on thumb/arm targets where the text segment needs to be a multiple of 2/4 bytes. Handle this by emitting a noop. This is more gross than it should be because arm/ppc are not fully mc'ized yet. This fixes rdar://7908505 llvm-svn: 102400
2010-04-27 07:37:21 +08:00
MCInst Noop;
Revert r301040 "X86: Don't emit zero-byte functions on Windows" This broke almost all bots. Reverting while fixing. llvm-svn: 301041
2017-04-22 05:10:37 +08:00
MF->getSubtarget().getInstrInfo()->getNoopForMachoTarget(Noop);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("avoids zero-length function");
Revert r301040 "X86: Don't emit zero-byte functions on Windows" This broke almost all bots. Reverting while fixing. llvm-svn: 301041
2017-04-22 05:10:37 +08:00
// Targets can opt-out of emitting the noop here by leaving the opcode
// unspecified.
if (Noop.getOpcode())
OutStreamer->EmitInstruction(Noop, getSubtargetInfo());
on darwin empty functions need to codegen into something of non-zero length, otherwise labels get incorrectly merged. We handled this by emitting a ".byte 0", but this isn't correct on thumb/arm targets where the text segment needs to be a multiple of 2/4 bytes. Handle this by emitting a noop. This is more gross than it should be because arm/ppc are not fully mc'ized yet. This fixes rdar://7908505 llvm-svn: 102400
2010-04-27 07:37:21 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
We currently use a callback to handle an IL pass deleting a BB that still has a reference to it. Unfortunately, that doesn't work for codegen passes since we don't get notified of MBB's being deleted (the original BB stays). Use that fact to our advantage and after printing a function, check if any of the IL BBs corresponds to a symbol that was not printed. This fixes pr11202. llvm-svn: 144674
2011-11-16 03:08:46 +08:00
const Function *F = MF->getFunction();
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const auto &BB : *F) {
if (!BB.hasAddressTaken())
We currently use a callback to handle an IL pass deleting a BB that still has a reference to it. Unfortunately, that doesn't work for codegen passes since we don't get notified of MBB's being deleted (the original BB stays). Use that fact to our advantage and after printing a function, check if any of the IL BBs corresponds to a symbol that was not printed. This fixes pr11202. llvm-svn: 144674
2011-11-16 03:08:46 +08:00
continue;
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
MCSymbol *Sym = GetBlockAddressSymbol(&BB);
We currently use a callback to handle an IL pass deleting a BB that still has a reference to it. Unfortunately, that doesn't work for codegen passes since we don't get notified of MBB's being deleted (the original BB stays). Use that fact to our advantage and after printing a function, check if any of the IL BBs corresponds to a symbol that was not printed. This fixes pr11202. llvm-svn: 144674
2011-11-16 03:08:46 +08:00
if (Sym->isDefined())
continue;
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("Address of block that was removed by CodeGen");
OutStreamer->EmitLabel(Sym);
We currently use a callback to handle an IL pass deleting a BB that still has a reference to it. Unfortunately, that doesn't work for codegen passes since we don't get notified of MBB's being deleted (the original BB stays). Use that fact to our advantage and after printing a function, check if any of the IL BBs corresponds to a symbol that was not printed. This fixes pr11202. llvm-svn: 144674
2011-11-16 03:08:46 +08:00
}
add target hooks for emitting random gunk before and after the function body. llvm-svn: 94732
2010-01-28 09:58:58 +08:00
// Emit target-specific gunk after the function body.
EmitFunctionBodyEnd();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move most EH from MachineModuleInfo to MachineFunction Recommitting r288293 with some extra fixes for GlobalISel code. Most of the exception handling members in MachineModuleInfo is actually per function data (talks about the "current function") so it is better to keep it at the function instead of the module. This is a necessary step to have machine module passes work properly. Also: - Rename TidyLandingPads() to tidyLandingPads() - Use doxygen member groups instead of "//===- EH ---"... so it is clear where a group ends. - I had to add an ugly const_cast at two places in the AsmPrinter because the available MachineFunction pointers are const, but the code wants to call tidyLandingPads() in between (markFunctionEnd()/endFunction()). Differential Revision: https://reviews.llvm.org/D27227 llvm-svn: 288405
2016-12-02 03:32:15 +08:00
if (!MF->getLandingPads().empty() || MMI->hasDebugInfo() ||
MF->hasEHFunclets() || MAI->hasDotTypeDotSizeDirective()) {
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
// Create a symbol for the end of function.
Centralize the handling of unique ids for temporary labels. Before this patch code wanting to create temporary labels for a given entity (function, cu, exception range, etc) had to keep its own counter to have stable symbol names. createTempSymbol would still add a suffix to make sure a new symbol was always returned, but it kept a single counter. Because of that, if we were to use just createTempSymbol("cu_begin"), the label could change from cu_begin42 to cu_begin43 because some other code started using temporary labels. Simplify this by just keeping one counter per prefix and removing the various specialized counters. llvm-svn: 232535
2015-03-18 04:07:06 +08:00
CurrentFnEnd = createTempSymbol("func_end");
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(CurrentFnEnd);
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
}
emit the cygwin stub thing through mcstreamer. llvm-svn: 100295
2010-04-04 06:19:41 +08:00
// If the target wants a .size directive for the size of the function, emit
// it.
if (MAI->hasDotTypeDotSizeDirective()) {
Use the vanilla func_end symbol for .size. No need to create yet another temp symbol. llvm-svn: 231198
2015-03-04 09:35:23 +08:00
// We can get the size as difference between the function label and the
// temp label.
Reformat. llvm-svn: 239691
2015-06-14 08:23:33 +08:00
const MCExpr *SizeExp = MCBinaryExpr::createSub(
MCSymbolRefExpr::create(CurrentFnEnd, OutContext),
MCSymbolRefExpr::create(CurrentFnSymForSize, OutContext), OutContext);
[MC] Refactor emitELFSize to make usage more consistent. NFC. Move the cast<MCSymbolELF> inside emitELFSize, so that: - it's done in one place instead of at each call - it's more consistent with similar functions like EmitCOFFSafeSEH - ambiguity between cast<> and dyn_cast<> is avoided (which also eliminates an unnecessary dyn_cast call) This also makes it easier to experiment with using ".size" directives on non-ELF targets. llvm-svn: 288437
2016-12-02 07:39:08 +08:00
OutStreamer->emitELFSize(CurrentFnSym, SizeExp);
emit the cygwin stub thing through mcstreamer. llvm-svn: 100295
2010-04-04 06:19:41 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription, HI.TimerGroupName,
HI.TimerGroupDescription, TimePassesIsEnabled);
Print jump tables before exception tables. In the case where just tables are part of the function section, this produces more readable assembly by avoiding switching to the eh section and back to .text. This would also break with non unique section names, as trying to switch to a unique section actually creates a new one. llvm-svn: 231677
2015-03-10 02:29:12 +08:00
HI.Handler->markFunctionEnd();
Workaround the breakage in r100616 by guarding all timers with TimePassesIsEnabled. This should allow make check to pass. llvm-svn: 100618
2010-04-07 18:44:46 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
// Print out jump tables referenced by the function.
EmitJumpTableInfo();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Print jump tables before exception tables. In the case where just tables are part of the function section, this produces more readable assembly by avoiding switching to the eh section and back to .text. This would also break with non unique section names, as trying to switch to a unique section actually creates a new one. llvm-svn: 231677
2015-03-10 02:29:12 +08:00
// Emit post-function debug and/or EH information.
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription, HI.TimerGroupName,
HI.TimerGroupDescription, TimePassesIsEnabled);
Print jump tables before exception tables. In the case where just tables are part of the function section, this produces more readable assembly by avoiding switching to the eh section and back to .text. This would also break with non unique section names, as trying to switch to a unique section actually creates a new one. llvm-svn: 231677
2015-03-10 02:29:12 +08:00
HI.Handler->endFunction(MF);
}
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
}
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
/// \brief Compute the number of Global Variables that uses a Constant.
static unsigned getNumGlobalVariableUses(const Constant *C) {
if (!C)
return 0;
if (isa<GlobalVariable>(C))
return 1;
unsigned NumUses = 0;
for (auto *CU : C->users())
NumUses += getNumGlobalVariableUses(dyn_cast<Constant>(CU));
return NumUses;
}
/// \brief Only consider global GOT equivalents if at least one user is a
/// cstexpr inside an initializer of another global variables. Also, don't
/// handle cstexpr inside instructions. During global variable emission,
/// candidates are skipped and are emitted later in case at least one cstexpr
/// isn't replaced by a PC relative GOT entry access.
static bool isGOTEquivalentCandidate(const GlobalVariable *GV,
unsigned &NumGOTEquivUsers) {
// Global GOT equivalents are unnamed private globals with a constant
// pointer initializer to another global symbol. They must point to a
// GlobalVariable or Function, i.e., as GlobalValue.
80-column and comment fixups. llvm-svn: 274335
2016-07-01 14:07:31 +08:00
if (!GV->hasGlobalUnnamedAddr() || !GV->hasInitializer() ||
!GV->isConstant() || !GV->isDiscardableIfUnused() ||
!dyn_cast<GlobalValue>(GV->getOperand(0)))
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
return false;
// To be a got equivalent, at least one of its users need to be a constant
// expression used by another global variable.
for (auto *U : GV->users())
[AsmPrinter] Don't assert on GOT equivalent non-constant users. We used to dyn_cast<Constant> in the recursive call, but cast<> in the initial one, and there can be non-Constant initial users. llvm-svn: 233346
2015-03-27 09:40:54 +08:00
NumGOTEquivUsers += getNumGlobalVariableUses(dyn_cast<Constant>(U));
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
return NumGOTEquivUsers > 0;
}
/// \brief Unnamed constant global variables solely contaning a pointer to
/// another globals variable is equivalent to a GOT table entry; it contains the
/// the address of another symbol. Optimize it and replace accesses to these
/// "GOT equivalents" by using the GOT entry for the final global instead.
/// Compute GOT equivalent candidates among all global variables to avoid
/// emitting them if possible later on, after it use is replaced by a GOT entry
/// access.
void AsmPrinter::computeGlobalGOTEquivs(Module &M) {
if (!getObjFileLowering().supportIndirectSymViaGOTPCRel())
return;
for (const auto &G : M.globals()) {
unsigned NumGOTEquivUsers = 0;
if (!isGOTEquivalentCandidate(&G, NumGOTEquivUsers))
continue;
const MCSymbol *GOTEquivSym = getSymbol(&G);
GlobalGOTEquivs[GOTEquivSym] = std::make_pair(&G, NumGOTEquivUsers);
}
}
/// \brief Constant expressions using GOT equivalent globals may not be eligible
/// for PC relative GOT entry conversion, in such cases we need to emit such
/// globals we previously omitted in EmitGlobalVariable.
void AsmPrinter::emitGlobalGOTEquivs() {
if (!getObjFileLowering().supportIndirectSymViaGOTPCRel())
return;
[AsmPrinter][TLOF] Reintroduce AArch64 test Follow up from r231505. Fix the non-determinism by using a MapVector and reintroduce the AArch64 testcase. Defer deleting the got candidates up to the end and remove them in a bulk, avoiding linear time removal of each element. Thanks to Renato Golin for trying it out on other platforms. llvm-svn: 231830
2015-03-11 04:05:23 +08:00
SmallVector<const GlobalVariable *, 8> FailedCandidates;
for (auto &I : GlobalGOTEquivs) {
const GlobalVariable *GV = I.second.first;
unsigned Cnt = I.second.second;
if (Cnt)
FailedCandidates.push_back(GV);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
}
[AsmPrinter][TLOF] Reintroduce AArch64 test Follow up from r231505. Fix the non-determinism by using a MapVector and reintroduce the AArch64 testcase. Defer deleting the got candidates up to the end and remove them in a bulk, avoiding linear time removal of each element. Thanks to Renato Golin for trying it out on other platforms. llvm-svn: 231830
2015-03-11 04:05:23 +08:00
GlobalGOTEquivs.clear();
for (auto *GV : FailedCandidates)
EmitGlobalVariable(GV);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
}
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
void AsmPrinter::emitGlobalIndirectSymbol(Module &M,
const GlobalIndirectSymbol& GIS) {
MCSymbol *Name = getSymbol(&GIS);
if (GIS.hasExternalLinkage() || !MAI->getWeakRefDirective())
OutStreamer->EmitSymbolAttribute(Name, MCSA_Global);
else if (GIS.hasWeakLinkage() || GIS.hasLinkOnceLinkage())
OutStreamer->EmitSymbolAttribute(Name, MCSA_WeakReference);
else
[GCC] Attribute ifunc support in llvm This patch add support for GCC attribute((ifunc("resolver"))) for targets that use ELF as object file format. In general ifunc is a special kind of function alias with type @gnu_indirect_function. Patch for Clang http://reviews.llvm.org/D15524 Differential Revision: http://reviews.llvm.org/D15525 llvm-svn: 265667
2016-04-07 20:32:19 +08:00
assert(GIS.hasLocalLinkage() && "Invalid alias or ifunc linkage");
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
// Set the symbol type to function if the alias has a function type.
// This affects codegen when the aliasee is not a function.
[GCC] Attribute ifunc support in llvm This patch add support for GCC attribute((ifunc("resolver"))) for targets that use ELF as object file format. In general ifunc is a special kind of function alias with type @gnu_indirect_function. Patch for Clang http://reviews.llvm.org/D15524 Differential Revision: http://reviews.llvm.org/D15525 llvm-svn: 265667
2016-04-07 20:32:19 +08:00
if (GIS.getType()->getPointerElementType()->isFunctionTy()) {
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
OutStreamer->EmitSymbolAttribute(Name, MCSA_ELF_TypeFunction);
[GCC] Attribute ifunc support in llvm This patch add support for GCC attribute((ifunc("resolver"))) for targets that use ELF as object file format. In general ifunc is a special kind of function alias with type @gnu_indirect_function. Patch for Clang http://reviews.llvm.org/D15524 Differential Revision: http://reviews.llvm.org/D15525 llvm-svn: 265667
2016-04-07 20:32:19 +08:00
if (isa<GlobalIFunc>(GIS))
OutStreamer->EmitSymbolAttribute(Name, MCSA_ELF_TypeIndFunction);
}
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
EmitVisibility(Name, GIS.getVisibility());
const MCExpr *Expr = lowerConstant(GIS.getIndirectSymbol());
if (isa<GlobalAlias>(&GIS) && MAI->hasAltEntry() && isa<MCBinaryExpr>(Expr))
OutStreamer->EmitSymbolAttribute(Name, MCSA_AltEntry);
// Emit the directives as assignments aka .set:
OutStreamer->EmitAssignment(Name, Expr);
if (auto *GA = dyn_cast<GlobalAlias>(&GIS)) {
// If the aliasee does not correspond to a symbol in the output, i.e. the
// alias is not of an object or the aliased object is private, then set the
// size of the alias symbol from the type of the alias. We don't do this in
// other situations as the alias and aliasee having differing types but same
// size may be intentional.
const GlobalObject *BaseObject = GA->getBaseObject();
if (MAI->hasDotTypeDotSizeDirective() && GA->getValueType()->isSized() &&
(!BaseObject || BaseObject->hasPrivateLinkage())) {
const DataLayout &DL = M.getDataLayout();
uint64_t Size = DL.getTypeAllocSize(GA->getValueType());
[MC] Refactor emitELFSize to make usage more consistent. NFC. Move the cast<MCSymbolELF> inside emitELFSize, so that: - it's done in one place instead of at each call - it's more consistent with similar functions like EmitCOFFSafeSEH - ambiguity between cast<> and dyn_cast<> is avoided (which also eliminates an unnecessary dyn_cast call) This also makes it easier to experiment with using ".size" directives on non-ELF targets. llvm-svn: 288437
2016-12-02 07:39:08 +08:00
OutStreamer->emitELFSize(Name, MCConstantExpr::create(Size, OutContext));
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
}
}
}
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
bool AsmPrinter::doFinalization(Module &M) {
At the beginning of doFinalization set the MachineFunction to nullptr so that users get an earlier dereferencing error and so that we can use it to conditionalize access to MachineFunction specific data. llvm-svn: 232820
2015-03-21 00:03:39 +08:00
// Set the MachineFunction to nullptr so that we can catch attempted
// accesses to MF specific features at the module level and so that
// we can conditionalize accesses based on whether or not it is nullptr.
MF = nullptr;
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Gather all GOT equivalent globals in the module. We really need two
// passes over the globals: one to compute and another to avoid its emission
// in EmitGlobalVariable, otherwise we would not be able to handle cases
// where the got equivalent shows up before its use.
computeGlobalGOTEquivs(M);
make AsmPrinter::doFinalization iterate over the global variables and call PrintGlobalVariable, allowing elimination and simplification of various targets. llvm-svn: 76604
2009-07-22 02:38:57 +08:00
// Emit global variables.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const auto &G : M.globals())
EmitGlobalVariable(&G);
Print the visibility of declarations. llvm-svn: 124468
2011-01-28 11:20:10 +08:00
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Emit remaining GOT equivalent globals.
emitGlobalGOTEquivs();
Print the visibility of declarations. llvm-svn: 124468
2011-01-28 11:20:10 +08:00
// Emit visibility info for declarations
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const Function &F : M) {
Print the visibility of available_externally functions. We were already printing it for declarations, but not available_externally. llvm-svn: 242027
2015-07-13 21:55:18 +08:00
if (!F.isDeclarationForLinker())
Print the visibility of declarations. llvm-svn: 124468
2011-01-28 11:20:10 +08:00
continue;
GlobalValue::VisibilityTypes V = F.getVisibility();
if (V == GlobalValue::DefaultVisibility)
continue;
Add a helper getSymbol to AsmPrinter. llvm-svn: 193627
2013-10-30 01:07:16 +08:00
MCSymbol *Name = getSymbol(&F);
Omit private_extern declarations of extern symbols; followup to r124468. Patch by Rafael Avila de Espindola! llvm-svn: 126297
2011-02-23 10:27:05 +08:00
EmitVisibility(Name, V, false);
Print the visibility of declarations. llvm-svn: 124468
2011-01-28 11:20:10 +08:00
}
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
const TargetLoweringObjectFile &TLOF = getObjFileLowering();
Add code to the target lowering object file module to handle module flags. The MachO back-end needs to emit the garbage collection flags specified in the module flags. This is a WIP, so the front-end hasn't been modified to emit these flags just yet. Documentation and front-end switching to occur soon. llvm-svn: 150507
2012-02-15 05:28:13 +08:00
// Emit module flags.
Modify the code that emits the module flags to use the new module flags accessor method. This allows the target lowering code to not have to deal with MDNodes. Also, avoid leaking memory like a sieve by not creating a global variable for the image info section, but just emitting the code directly. llvm-svn: 150624
2012-02-16 06:36:15 +08:00
SmallVector<Module::ModuleFlagEntry, 8> ModuleFlags;
M.getModuleFlagsMetadata(ModuleFlags);
if (!ModuleFlags.empty())
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
TLOF.emitModuleFlags(*OutStreamer, ModuleFlags, TM);
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
Replace string GNU Triples with llvm::Triple in TargetMachine::getTargetTriple(). NFC. Summary: This continues the patch series to eliminate StringRef forms of GNU triples from the internals of LLVM that began in r239036. Reviewers: rengolin Reviewed By: rengolin Subscribers: llvm-commits, rengolin Differential Revision: http://reviews.llvm.org/D10381 llvm-svn: 239815
2015-06-16 21:15:50 +08:00
if (TM.getTargetTriple().isOSBinFormatELF()) {
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
MachineModuleInfoELF &MMIELF = MMI->getObjFileInfo<MachineModuleInfoELF>();
// Output stubs for external and common global variables.
MachineModuleInfoELF::SymbolListTy Stubs = MMIELF.GetGVStubList();
if (!Stubs.empty()) {
Stop producing .data.rel sections. If a section is rw, it is irrelevant if the dynamic linker will write to it or not. It looks like llvm implemented this because gcc was doing it. It looks like gcc implemented this in the hope that it would put all the relocated items close together and speed up the dynamic linker. There are two problem with this: * It doesn't work. Both bfd and gold will map .data.rel to .data and concatenate the input sections in the order they are seen. * If we want a feature like that, it can be implemented directly in the linker since it knowns where the dynamic relocations are. llvm-svn: 253436
2015-11-18 14:02:15 +08:00
OutStreamer->SwitchSection(TLOF.getDataSection());
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = M.getDataLayout();
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
for (const auto &Stub : Stubs) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(Stub.first);
OutStreamer->EmitSymbolValue(Stub.second.getPointer(),
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
DL.getPointerSize());
Refactor a lot of duplicated code for stub output. This also moves it earlier so that it they are produced before we print an end symbol for the data section. llvm-svn: 234315
2015-04-07 21:42:44 +08:00
}
}
}
Add code to the target lowering object file module to handle module flags. The MachO back-end needs to emit the garbage collection flags specified in the module flags. This is a WIP, so the front-end hasn't been modified to emit these flags just yet. Documentation and front-end switching to occur soon. llvm-svn: 150507
2012-02-15 05:28:13 +08:00
change AsmPrinter to use DwarfDebug/DwarfException directly instead of going through DwarfWriter. llvm-svn: 100405
2010-04-05 13:11:15 +08:00
// Finalize debug and EH information.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const HandlerInfo &HI : Handlers) {
Timer: Track name and description. The previously used "names" are rather descriptions (they use multiple words and contain spaces), use short programming language identifier like strings for the "names" which should be used when exporting to machine parseable formats. Also removed a unused TimerGroup from Hexxagon. Differential Revision: https://reviews.llvm.org/D25583 llvm-svn: 287369
2016-11-19 03:43:18 +08:00
NamedRegionTimer T(HI.TimerName, HI.TimerDescription, HI.TimerGroupName,
HI.TimerGroupDescription, TimePassesIsEnabled);
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
HI.Handler->endModule();
delete HI.Handler;
change AsmPrinter to use DwarfDebug/DwarfException directly instead of going through DwarfWriter. llvm-svn: 100405
2010-04-05 13:11:15 +08:00
}
Reland 196270 "Generalize debug info / EH emission in AsmPrinter" Addressing the existense AMDGPUAsmPrinter and other subclasses of AsmPrinter llvm-svn: 196288
2013-12-03 23:10:23 +08:00
Handlers.clear();
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
DD = nullptr;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
eliminate the ExtWeakSymbols set from AsmPrinter. This eliminates a bunch of code from all the targets, and eliminates nondeterministic ordering of directives being emitted in the output. llvm-svn: 74096
2009-06-25 02:52:01 +08:00
// If the target wants to know about weak references, print them all.
rename TAI -> MAI, being careful not to make MAILJMP instructions :) llvm-svn: 79777
2009-08-23 05:43:10 +08:00
if (MAI->getWeakRefDirective()) {
eliminate the ExtWeakSymbols set from AsmPrinter. This eliminates a bunch of code from all the targets, and eliminates nondeterministic ordering of directives being emitted in the output. llvm-svn: 74096
2009-06-25 02:52:01 +08:00
// FIXME: This is not lazy, it would be nice to only print weak references
// to stuff that is actually used. Note that doing so would require targets
// to notice uses in operands (due to constant exprs etc). This should
// happen with the MC stuff eventually.
IR: Introduce Module::global_objects(). This is a convenience iterator that allows clients to enumerate the GlobalObjects within a Module. Also start using it in a few places where it is obviously the right thing to use. Differential Revision: http://reviews.llvm.org/D21580 llvm-svn: 273470
2016-06-23 04:29:42 +08:00
// Print out module-level global objects here.
for (const auto &GO : M.global_objects()) {
if (!GO.hasExternalWeakLinkage())
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
continue;
IR: Introduce Module::global_objects(). This is a convenience iterator that allows clients to enumerate the GlobalObjects within a Module. Also start using it in a few places where it is obviously the right thing to use. Differential Revision: http://reviews.llvm.org/D21580 llvm-svn: 273470
2016-06-23 04:29:42 +08:00
OutStreamer->EmitSymbolAttribute(getSymbol(&GO), MCSA_WeakReference);
eliminate the ExtWeakSymbols set from AsmPrinter. This eliminates a bunch of code from all the targets, and eliminates nondeterministic ordering of directives being emitted in the output. llvm-svn: 74096
2009-06-25 02:52:01 +08:00
}
move ExtWeakSymbols to AsmPrinter llvm-svn: 32648
2006-12-18 11:37:18 +08:00
}
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddBlankLine();
Honour aliases visibility during asm emission llvm-svn: 48249
2008-03-12 05:41:14 +08:00
[AsmPrinter] Print aliases in topological order Print aliases in topological order, that is, for any alias a = b, b must be printed before a. This is because on some targets (e.g. PowerPC) linker expects aliases in such an order to generate correct TOC information. GCC also prints aliases in topological order. llvm-svn: 265064
2016-04-01 06:08:19 +08:00
// Print aliases in topological order, that is, for each alias a = b,
// b must be printed before a.
// This is because on some targets (e.g. PowerPC) linker expects aliases in
// such an order to generate correct TOC information.
SmallVector<const GlobalAlias *, 16> AliasStack;
SmallPtrSet<const GlobalAlias *, 16> AliasVisited;
for (const auto &Alias : M.aliases()) {
for (const GlobalAlias *Cur = &Alias; Cur;
Cur = dyn_cast<GlobalAlias>(Cur->getAliasee())) {
if (!AliasVisited.insert(Cur).second)
break;
AliasStack.push_back(Cur);
}
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
for (const GlobalAlias *AncestorAlias : llvm::reverse(AliasStack))
[IFUNC] Use GlobalIndirectSymbol when aliases and ifuncs have something similar Second part extracted from http://reviews.llvm.org/D15525 Use GlobalIndirectSymbol in all cases when aliases and ifuncs have something in common. Differential Revision: http://reviews.llvm.org/D18754 llvm-svn: 265382
2016-04-05 16:47:51 +08:00
emitGlobalIndirectSymbol(M, *AncestorAlias);
[AsmPrinter] Print aliases in topological order Print aliases in topological order, that is, for any alias a = b, b must be printed before a. This is because on some targets (e.g. PowerPC) linker expects aliases in such an order to generate correct TOC information. GCC also prints aliases in topological order. llvm-svn: 265064
2016-04-01 06:08:19 +08:00
AliasStack.clear();
Implement aliases. This fixes PR1017 and it's dependent bugs. CFE part will follow. llvm-svn: 36435
2007-04-25 22:27:10 +08:00
}
[GCC] Attribute ifunc support in llvm This patch add support for GCC attribute((ifunc("resolver"))) for targets that use ELF as object file format. In general ifunc is a special kind of function alias with type @gnu_indirect_function. Patch for Clang http://reviews.llvm.org/D15524 Differential Revision: http://reviews.llvm.org/D15525 llvm-svn: 265667
2016-04-07 20:32:19 +08:00
for (const auto &IFunc : M.ifuncs())
emitGlobalIndirectSymbol(M, IFunc);
Implement aliases. This fixes PR1017 and it's dependent bugs. CFE part will follow. llvm-svn: 36435
2007-04-25 22:27:10 +08:00
Rename getAnalysisToUpdate to getAnalysisIfAvailable. llvm-svn: 63198
2009-01-28 21:14:17 +08:00
GCModuleInfo *MI = getAnalysisIfAvailable<GCModuleInfo>();
Rename some GC classes so that their roll will hopefully be clearer. In particular, Collector was confusing to implementors. Several thought that this compile-time class was the place to implement their runtime GC heap. Of course, it doesn't even exist at runtime. Specifically, the renames are: Collector -> GCStrategy CollectorMetadata -> GCFunctionInfo CollectorModuleMetadata -> GCModuleInfo CollectorRegistry -> GCRegistry Function::getCollector -> getGC (setGC, hasGC, clearGC) Several accessors and nested types have also been renamed to be consistent. These changes should be obvious. llvm-svn: 54899
2008-08-18 02:44:35 +08:00
assert(MI && "AsmPrinter didn't require GCModuleInfo?");
for (GCModuleInfo::iterator I = MI->end(), E = MI->begin(); I != E; )
Use unique_ptr to manage ownership of GCStrategy objects in GCMetadata llvm-svn: 206246
2014-04-15 13:34:49 +08:00
if (GCMetadataPrinter *MP = GetOrCreateGCPrinter(**--I))
GCStrategy should not own GCFunctionInfo This change moves the ownership and access of GCFunctionInfo (the object which describes the safepoints associated with a safepoint under GCRoot) to GCModuleInfo. Previously, this was owned by GCStrategy which was in turned owned by GCModuleInfo. This made GCStrategy module specific which is 'surprising' given it's name and other purposes. There's a few more changes needed, but we're getting towards the point we can reuse GCStrategy for gc.statepoint as well. p.s. The style of this code ends up being a mess. I was trying to move code around without otherwise changing much. Once I get the ownership structure rearranged, I will go through and fixup spacing, naming, comments etc. Differential Revision: http://reviews.llvm.org/D6587 llvm-svn: 223994
2014-12-11 09:47:23 +08:00
MP->finishAssembly(M, *MI, *this);
Enabling the target-independent garbage collection infrastructure by hooking it up to the various compiler pipelines. This doesn't actually add support for any GC algorithms, which means it temporarily breaks a few tests. To be fixed shortly. llvm-svn: 45669
2008-01-07 09:30:38 +08:00
Add support for metadata representing .ident directives. llvm-svn: 192764
2013-10-16 09:49:05 +08:00
// Emit llvm.ident metadata in an '.ident' directive.
EmitModuleIdents(M);
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
// Emit __morestack address if needed for indirect calls.
if (MMI->usesMorestackAddr()) {
[X86] Use the correct alignment for COMDAT constant pool entries COFF doesn't have sections with mergeable contents. Instead, each constant pool entry ends up in a COMDAT section. The linker, when choosing between COMDAT sections, doesn't choose the max alignment of the two sections. You just get whatever alignment was on the section. If one constant needed a higher alignment in one object file from another one, then we will get into trouble if the linker chooses the lower alignment one. Instead, lets promote the alignment of the constant pool entry to make sure we don't use an under aligned constant with an instruction which assumed otherwise. This fixes PR26680. llvm-svn: 261462
2016-02-21 09:30:30 +08:00
unsigned Align = 1;
Remove DataLayout from TargetLoweringObjectFile, redirect to Module Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11079 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242385
2015-07-16 14:04:17 +08:00
MCSection *ReadOnlySection = getObjFileLowering().getSectionForConstant(
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
getDataLayout(), SectionKind::getReadOnly(),
[X86] Use the correct alignment for COMDAT constant pool entries COFF doesn't have sections with mergeable contents. Instead, each constant pool entry ends up in a COMDAT section. The linker, when choosing between COMDAT sections, doesn't choose the max alignment of the two sections. You just get whatever alignment was on the section. If one constant needed a higher alignment in one object file from another one, then we will get into trouble if the linker chooses the lower alignment one. Instead, lets promote the alignment of the constant pool entry to make sure we don't use an under aligned constant with an instruction which assumed otherwise. This fixes PR26680. llvm-svn: 261462
2016-02-21 09:30:30 +08:00
/*C=*/nullptr, Align);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(ReadOnlySection);
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
MCSymbol *AddrSymbol =
MC: Clean up method names in MCContext. The naming was a mish-mash of old and new style. Update to be consistent with the new. NFC. llvm-svn: 237594
2015-05-19 02:43:14 +08:00
OutContext.getOrCreateSymbol(StringRef("__morestack_addr"));
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(AddrSymbol);
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
Distinguish between code pointer size and DataLayout::getPointerSize() in DWARF info generation llvm-svn: 300463
2017-04-18 01:41:25 +08:00
unsigned PtrSize = MAI->getCodePointerSize();
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolValue(GetExternalSymbolSymbol("__morestack"),
PtrSize);
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
}
Add AsmPrinter support for emitting a directive to declare that the code being generated does not require an executable stack. Also, add target-specific code to make use of this on Linux on x86. llvm-svn: 50634
2008-05-05 08:28:39 +08:00
// If we don't have any trampolines, then we don't require stack memory
// to be executable. Some targets have a directive to declare this.
eliminate the ExtWeakSymbols set from AsmPrinter. This eliminates a bunch of code from all the targets, and eliminates nondeterministic ordering of directives being emitted in the output. llvm-svn: 74096
2009-06-25 02:52:01 +08:00
Function *InitTrampolineIntrinsic = M.getFunction("llvm.init.trampoline");
Add AsmPrinter support for emitting a directive to declare that the code being generated does not require an executable stack. Also, add target-specific code to make use of this on Linux on x86. llvm-svn: 50634
2008-05-05 08:28:39 +08:00
if (!InitTrampolineIntrinsic || InitTrampolineIntrinsic->use_empty())
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
if (MCSection *S = MAI->getNonexecutableStackSection(OutContext))
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(S);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
add a new hook to allow targets to splat stuff at the end of the file. Overriding doFinalization is pretty lame. llvm-svn: 82268
2009-09-19 04:17:03 +08:00
// Allow the target to emit any magic that it wants at the end of the file,
// after everything else has gone out.
EmitEndOfAsmFile(M);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
MMI = nullptr;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->Finish();
OutStreamer->reset();
Make the MCStreamer have a reset method and call that after finalization of the asm printer, also changed MCContext to a single reset only method for simplicity as requested on the list llvm-svn: 170041
2012-12-13 06:59:46 +08:00
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
return false;
}
Move the EH symbol to the asm printer and use it for the SJLJ case too. llvm-svn: 232475
2015-03-17 21:57:48 +08:00
MCSymbol *AsmPrinter::getCurExceptionSym() {
if (!CurExceptionSym)
Centralize the handling of unique ids for temporary labels. Before this patch code wanting to create temporary labels for a given entity (function, cu, exception range, etc) had to keep its own counter to have stable symbol names. createTempSymbol would still add a suffix to make sure a new symbol was always returned, but it kept a single counter. Because of that, if we were to use just createTempSymbol("cu_begin"), the label could change from cu_begin42 to cu_begin43 because some other code started using temporary labels. Simplify this by just keeping one counter per prefix and removing the various specialized counters. llvm-svn: 232535
2015-03-18 04:07:06 +08:00
CurExceptionSym = createTempSymbol("exception");
Move the EH symbol to the asm printer and use it for the SJLJ case too. llvm-svn: 232475
2015-03-17 21:57:48 +08:00
return CurExceptionSym;
}
Adjust to capitalized asmprinter method names llvm-svn: 24457
2005-11-21 15:51:36 +08:00
void AsmPrinter::SetupMachineFunction(MachineFunction &MF) {
make MachineFunction keep track of its ID and make MachineFunctionAnalysis dole them out, instead of having AsmPrinter do both. Have the AsmPrinter::SetupMachineFunction method set the 'AsmPrinter::MF' variable. llvm-svn: 94509
2010-01-26 12:35:26 +08:00
this->MF = &MF;
CurrentFnName is now dead, remove it. llvm-svn: 93612
2010-01-16 09:24:10 +08:00
// Get the function symbol.
Add a helper getSymbol to AsmPrinter. llvm-svn: 193627
2013-10-30 01:07:16 +08:00
CurrentFnSym = getSymbol(MF.getFunction());
Allow the use of an alternate symbol for calculating a function's size. The standard function epilog includes a .size directive, but ppc64 uses an alternate local symbol to tag the actual start of each function. Until recently, binutils accepted the .size directive as: .size test1, .Ltmp0-test1 however, using this directive with recent binutils will result in the error: .size expression for XXX does not evaluate to a constant so we must use the label which actually tags the start of the function. llvm-svn: 151200
2012-02-23 05:11:47 +08:00
CurrentFnSymForSize = CurrentFnSym;
Use the correct func begin symbol in all places in ppc. I missed an occurrence of the old symbol in my previous patch. llvm-svn: 231398
2015-03-06 03:47:50 +08:00
CurrentFnBegin = nullptr;
Move the EH symbol to the asm printer and use it for the SJLJ case too. llvm-svn: 232475
2015-03-17 21:57:48 +08:00
CurExceptionSym = nullptr;
Use the correct func begin symbol in all places in ppc. I missed an occurrence of the old symbol in my previous patch. llvm-svn: 231398
2015-03-06 03:47:50 +08:00
bool NeedsLocalForSize = MAI->needsLocalForSize();
Move most EH from MachineModuleInfo to MachineFunction Recommitting r288293 with some extra fixes for GlobalISel code. Most of the exception handling members in MachineModuleInfo is actually per function data (talks about the "current function") so it is better to keep it at the function instead of the module. This is a necessary step to have machine module passes work properly. Also: - Rename TidyLandingPads() to tidyLandingPads() - Use doxygen member groups instead of "//===- EH ---"... so it is clear where a group ends. - I had to add an ugly const_cast at two places in the AsmPrinter because the available MachineFunction pointers are const, but the code wants to call tidyLandingPads() in between (markFunctionEnd()/endFunction()). Differential Revision: https://reviews.llvm.org/D27227 llvm-svn: 288405
2016-12-02 03:32:15 +08:00
if (!MF.getLandingPads().empty() || MMI->hasDebugInfo() ||
MF.hasEHFunclets() || NeedsLocalForSize) {
Centralize the handling of unique ids for temporary labels. Before this patch code wanting to create temporary labels for a given entity (function, cu, exception range, etc) had to keep its own counter to have stable symbol names. createTempSymbol would still add a suffix to make sure a new symbol was always returned, but it kept a single counter. Because of that, if we were to use just createTempSymbol("cu_begin"), the label could change from cu_begin42 to cu_begin43 because some other code started using temporary labels. Simplify this by just keeping one counter per prefix and removing the various specialized counters. llvm-svn: 232535
2015-03-18 04:07:06 +08:00
CurrentFnBegin = createTempSymbol("func_begin");
Use the correct func begin symbol in all places in ppc. I missed an occurrence of the old symbol in my previous patch. llvm-svn: 231398
2015-03-06 03:47:50 +08:00
if (NeedsLocalForSize)
CurrentFnSymForSize = CurrentFnBegin;
}
Add support for printing loop structure information in asm comments. This definitely slows down asm output so put it under an -asm-exuberant flag. This information is useful when doing static analysis of performance issues. llvm-svn: 78567
2009-08-11 00:38:07 +08:00
OptDiag: Summarize the instruction count in asm-printer Add an optimization remark to asm-printer that summarizes the number of instructions emitted per function. llvm-svn: 296053
2017-02-24 08:19:22 +08:00
ORE = &getAnalysis<MachineOptimizationRemarkEmitterPass>().getORE();
clean up the asmprinter header and privatize some stuff. llvm-svn: 100342
2010-04-05 02:52:31 +08:00
if (isVerbose())
Add support for printing loop structure information in asm comments. This definitely slows down asm output so put it under an -asm-exuberant flag. This information is useful when doing static analysis of performance issues. llvm-svn: 78567
2009-08-11 00:38:07 +08:00
LI = &getAnalysis<MachineLoopInfo>();
This patch closes PR#32216: Better testing of schedule model instruction latencies/throughputs. The details are here: https://reviews.llvm.org/D30941 llvm-svn: 300311
2017-04-14 15:44:23 +08:00
const TargetSubtargetInfo &STI = MF.getSubtarget();
EnablePrintSchedInfo = PrintSchedule.getNumOccurrences()
? PrintSchedule
: STI.supportPrintSchedInfo();
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
}
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
namespace {
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
// Keep track the alignment, constpool entries per Section.
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
struct SectionCPs {
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
MCSection *S;
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
unsigned Alignment;
SmallVector<unsigned, 4> CPEs;
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
SectionCPs(MCSection *s, unsigned a) : S(s), Alignment(a) {}
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
};
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
} // end anonymous namespace
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
Allow target to customize directive used to switch to arbitrary section in SwitchSection, add generic constant pool emitter llvm-svn: 24464
2005-11-21 16:25:09 +08:00
/// EmitConstantPool - Print to the current output stream assembly
/// representations of the constants in the constant pool MCP. This is
/// used to print out constants which have been "spilled to memory" by
/// the code generator.
///
Drop the argument to AsmPrinter::EmitConstantPool and make it virtual. Overload it in the ARM backend to do nothing, since is does insane constant pool emission. llvm-svn: 94708
2010-01-28 08:19:24 +08:00
void AsmPrinter::EmitConstantPool() {
const MachineConstantPool *MCP = MF->getConstantPool();
rename fields of constant pool entries llvm-svn: 26076
2006-02-09 12:22:52 +08:00
const std::vector<MachineConstantPoolEntry> &CP = MCP->getConstants();
Allow target to customize directive used to switch to arbitrary section in SwitchSection, add generic constant pool emitter llvm-svn: 24464
2005-11-21 16:25:09 +08:00
if (CP.empty()) return;
Add support to print 4-, 8-, and 16- byte constant literals in special sections. e.g. On Darwin that would be .literal4 and .literal8. llvm-svn: 28977
2006-06-29 08:26:09 +08:00
Use generic section-handling stuff to emit constant pool entries llvm-svn: 56579
2008-09-25 06:17:59 +08:00
// Calculate sections for constant pool entries. We collect entries to go into
// the same section together to reduce amount of section switch statements.
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
SmallVector<SectionCPs, 4> CPSections;
Add support to print 4-, 8-, and 16- byte constant literals in special sections. e.g. On Darwin that would be .literal4 and .literal8. llvm-svn: 28977
2006-06-29 08:26:09 +08:00
for (unsigned i = 0, e = CP.size(); i != e; ++i) {
remove the SelectSectionForMachineConst hook, replacing it with a new getSectionForMergableConstant hook. This removes one dependence of TAI on Type, and provides the hook with enough info to make the right decision based on whether the global has relocations etc. llvm-svn: 76705
2009-07-22 08:28:43 +08:00
const MachineConstantPoolEntry &CPE = CP[i];
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
unsigned Align = CPE.getAlignment();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
SectionKind Kind = CPE.getSectionKind(&getDataLayout());
simplify getSectionForMergableConstant to take a SectionKind. llvm-svn: 77134
2009-07-26 14:26:55 +08:00
CodeGen: Stick constant pool entries in COMDAT sections for WinCOFF COFF lacks a feature that other object file formats support: mergeable sections. To work around this, MSVC sticks constant pool entries in special COMDAT sections so that each constant is in it's own section. This permits unused constants to be dropped and it also allows duplicate constants in different translation units to get merged together. This fixes PR20262. Differential Revision: http://reviews.llvm.org/D4482 llvm-svn: 213006
2014-07-15 06:57:27 +08:00
const Constant *C = nullptr;
if (!CPE.isMachineConstantPoolEntry())
C = CPE.Val.ConstVal;
[X86] Use the correct alignment for COMDAT constant pool entries COFF doesn't have sections with mergeable contents. Instead, each constant pool entry ends up in a COMDAT section. The linker, when choosing between COMDAT sections, doesn't choose the max alignment of the two sections. You just get whatever alignment was on the section. If one constant needed a higher alignment in one object file from another one, then we will get into trouble if the linker chooses the lower alignment one. Instead, lets promote the alignment of the constant pool entry to make sure we don't use an under aligned constant with an instruction which assumed otherwise. This fixes PR26680. llvm-svn: 261462
2016-02-21 09:30:30 +08:00
MCSection *S = getObjFileLowering().getSectionForConstant(getDataLayout(),
Kind, C, Align);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
// The number of sections are small, just do a linear search from the
// last section to the first.
bool Found = false;
unsigned SecIdx = CPSections.size();
while (SecIdx != 0) {
if (CPSections[--SecIdx].S == S) {
Found = true;
break;
}
}
if (!Found) {
SecIdx = CPSections.size();
CPSections.push_back(SectionCPs(S, Align));
}
if (Align > CPSections[SecIdx].Alignment)
CPSections[SecIdx].Alignment = Align;
CPSections[SecIdx].CPEs.push_back(i);
Add support to print 4-, 8-, and 16- byte constant literals in special sections. e.g. On Darwin that would be .literal4 and .literal8. llvm-svn: 28977
2006-06-29 08:26:09 +08:00
}
Use generic section-handling stuff to emit constant pool entries llvm-svn: 56579
2008-09-25 06:17:59 +08:00
// Now print stuff into the calculated sections.
CodeGen: Stick constant pool entries in COMDAT sections for WinCOFF COFF lacks a feature that other object file formats support: mergeable sections. To work around this, MSVC sticks constant pool entries in special COMDAT sections so that each constant is in it's own section. This permits unused constants to be dropped and it also allows duplicate constants in different translation units to get merged together. This fixes PR20262. Differential Revision: http://reviews.llvm.org/D4482 llvm-svn: 213006
2014-07-15 06:57:27 +08:00
const MCSection *CurSection = nullptr;
unsigned Offset = 0;
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
for (unsigned i = 0, e = CPSections.size(); i != e; ++i) {
for (unsigned j = 0, ee = CPSections[i].CPEs.size(); j != ee; ++j) {
unsigned CPI = CPSections[i].CPEs[j];
CodeGen: Stick constant pool entries in COMDAT sections for WinCOFF COFF lacks a feature that other object file formats support: mergeable sections. To work around this, MSVC sticks constant pool entries in special COMDAT sections so that each constant is in it's own section. This permits unused constants to be dropped and it also allows duplicate constants in different translation units to get merged together. This fixes PR20262. Differential Revision: http://reviews.llvm.org/D4482 llvm-svn: 213006
2014-07-15 06:57:27 +08:00
MCSymbol *Sym = GetCPISymbol(CPI);
if (!Sym->isUndefined())
continue;
if (CurSection != CPSections[i].S) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(CPSections[i].S);
CodeGen: Stick constant pool entries in COMDAT sections for WinCOFF COFF lacks a feature that other object file formats support: mergeable sections. To work around this, MSVC sticks constant pool entries in special COMDAT sections so that each constant is in it's own section. This permits unused constants to be dropped and it also allows duplicate constants in different translation units to get merged together. This fixes PR20262. Differential Revision: http://reviews.llvm.org/D4482 llvm-svn: 213006
2014-07-15 06:57:27 +08:00
EmitAlignment(Log2_32(CPSections[i].Alignment));
CurSection = CPSections[i].S;
Offset = 0;
}
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
MachineConstantPoolEntry CPE = CP[CPI];
// Emit inter-object padding for alignment.
unsigned AlignMask = CPE.getAlignment() - 1;
unsigned NewOffset = (Offset + AlignMask) & ~AlignMask;
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitZeros(NewOffset - Offset);
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
land David Blaikie's patch to de-constify Type, with a few tweaks. llvm-svn: 135375
2011-07-18 12:54:35 +08:00
Type *Ty = CPE.getType();
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
Offset = NewOffset + getDataLayout().getTypeAllocSize(Ty);
mcize jump table and constant pool entry labels, .local on elf, and some .weak directives. llvm-svn: 94284
2010-01-23 13:19:23 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(Sym);
Fix some significant problems with constant pools that resulted in unnecessary paddings between constant pool entries, larger than necessary alignments (e.g. 8 byte alignment for .literal4 sections), and potentially other issues. 1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants. 2. MachineConstantPool alignment field is also a log2 value. 3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values. 4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries. 5. Asm printer uses expensive data structure multimap to track constant pool entries by sections. 6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic. Solutions: 1. ConstantPoolSDNode alignment field is changed to keep non-log2 value. 2. MachineConstantPool alignment field is also changed to keep non-log2 value. 3. Functions that create ConstantPool nodes are passing in non-log2 alignments. 4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT. 5. Asm printer uses cheaper data structure to group constant pool entries. 6. Asm printer compute entry offsets after grouping is done. 7. Change JIT code to compute entry offsets on the fly. llvm-svn: 66875
2009-03-13 15:51:59 +08:00
if (CPE.isMachineConstantPoolEntry())
EmitMachineConstantPoolValue(CPE.Val.MachineCPVal);
Use generic section-handling stuff to emit constant pool entries llvm-svn: 56579
2008-09-25 06:17:59 +08:00
else
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitGlobalConstant(getDataLayout(), CPE.Val.ConstVal);
Adjust to MachineConstantPool interface change: instead of keeping a value/alignment pair for each constant, keep a value/offset pair. llvm-svn: 26078
2006-02-09 12:46:04 +08:00
}
Allow target to customize directive used to switch to arbitrary section in SwitchSection, add generic constant pool emitter llvm-svn: 24464
2005-11-21 16:25:09 +08:00
}
}
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
/// EmitJumpTableInfo - Print assembly representations of the jump tables used
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
/// by the current function to the current output stream.
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
///
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
void AsmPrinter::EmitJumpTableInfo() {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = MF->getDataLayout();
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
const MachineJumpTableInfo *MJTI = MF->getJumpTableInfo();
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (!MJTI) return;
Add a new jump table encoding to indicate jump tables entries are inside the function by the target at the point of use. llvm-svn: 98255
2010-03-11 22:58:16 +08:00
if (MJTI->getEntryKind() == MachineJumpTableInfo::EK_Inline) return;
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
const std::vector<MachineJumpTableEntry> &JT = MJTI->getJumpTables();
if (JT.empty()) return;
Fix PIC jump table codegen on x86-32/linux. In fact, such thing should be applied to all targets uses GOT-relative offsets for PIC (Alpha?) llvm-svn: 44108
2007-11-14 17:18:41 +08:00
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
// Pick the directive to use to print the jump table entries, and switch to
Support jump tables when in PIC relocation model llvm-svn: 29318
2006-07-27 09:13:04 +08:00
// the appropriate section.
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
const Function *F = MF->getFunction();
Put each jump table in an independent section if the function is too. This allows the linker to GC both, fixing pr22557. llvm-svn: 228937
2015-02-13 01:16:46 +08:00
const TargetLoweringObjectFile &TLOF = getObjFileLowering();
Add r228939 back with a fix. The problem in the original patch was not switching back to .text after printing an eh table. Original message: On ELF, put PIC jump tables in a non executable section. Fixes PR22558. llvm-svn: 229586
2015-02-18 07:34:51 +08:00
bool JTInDiffSection = !TLOF.shouldPutJumpTableInFunctionSection(
MJTI->getEntryKind() == MachineJumpTableInfo::EK_LabelDifference32,
*F);
Print jump tables before exception tables. In the case where just tables are part of the function section, this produces more readable assembly by avoiding switching to the eh section and back to .text. This would also break with non unique section names, as trying to switch to a unique section actually creates a new one. llvm-svn: 231677
2015-03-10 02:29:12 +08:00
if (JTInDiffSection) {
// Drop it in the readonly section.
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
MCSection *ReadOnlySection = TLOF.getSectionForJumpTable(*F, TM);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(ReadOnlySection);
Support jump tables when in PIC relocation model llvm-svn: 29318
2006-07-27 09:13:04 +08:00
}
Rearrange handling of jump tables. Highlights: 1. MachineJumpTableInfo is now created lazily for a function the first time it actually makes a jump table instead of for every function. 2. The encoding of jump table entries is now described by the MachineJumpTableInfo::JTEntryKind enum. This enum is determined by the TLI::getJumpTableEncoding() hook, instead of by lots of code scattered throughout the compiler that "knows" that jump table entries are always 32-bits in pic mode (for example). 3. The size and alignment of jump table entries is now calculated based on their kind, instead of at machinefunction creation time. Future work includes using the EntryKind in more places in the compiler, eliminating other logic that "knows" the layout of jump tables in various situations. llvm-svn: 94470
2010-01-26 07:26:13 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitAlignment(Log2_32(MJTI->getEntryAlignment(DL)));
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Mark jump tables in code sections with DataRegion directives. Even out-of-line jump tables can be in the code section, so mark them as data-regions for those targets which support the directives. rdar://12362871&12362974 llvm-svn: 164571
2012-09-25 07:06:27 +08:00
// Jump tables in code sections are marked with a data_region directive
// where that's supported.
if (!JTInDiffSection)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitDataRegion(MCDR_DataRegionJT32);
Mark jump tables in code sections with DataRegion directives. Even out-of-line jump tables can be in the code section, so mark them as data-regions for those targets which support the directives. rdar://12362871&12362974 llvm-svn: 164571
2012-09-25 07:06:27 +08:00
Now that printPICJumpTableSetLabel is not overloaded, inline it into its only caller, allowing us to simplify it and hoist bits out of the loop. llvm-svn: 94528
2010-01-26 14:42:44 +08:00
for (unsigned JTI = 0, e = JT.size(); JTI != e; ++JTI) {
const std::vector<MachineBasicBlock*> &JTBBs = JT[JTI].MBBs;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
// If this jump table was deleted, ignore it.
don't print dead jump tables llvm-svn: 31266
2006-10-29 02:10:06 +08:00
if (JTBBs.empty()) continue;
Emit .set directives for jump table entries when possible, which reduces the number of relocations in object files, shrinkifying them. llvm-svn: 29650
2006-08-13 05:29:52 +08:00
Fix a bit of confusion about .set and produce more readable assembly. Every target we support has support for assembly that looks like a = b - c .long a What is special about MachO is that the above combination suppresses the production of a relocation. With this change we avoid producing the intermediary labels when they don't add any value. llvm-svn: 220256
2014-10-21 09:17:30 +08:00
// For the EK_LabelDifference32 entry, if using .set avoids a relocation,
/// emit a .set directive for each unique entry.
simplify asmprinter: only emit .set directives when entries have EK_LabelDifference32 kind and the target has .set support. Simplify X86AsmPrinter::printPICJumpTableSetLabel to make use of recent helpers. llvm-svn: 94518
2010-01-26 13:15:20 +08:00
if (MJTI->getEntryKind() == MachineJumpTableInfo::EK_LabelDifference32 &&
doesSetDirectiveSuppressesReloc -> doesSetDirectiveSuppressReloc, the former is grammatically incorrect. llvm-svn: 273100
2016-06-19 07:25:37 +08:00
MAI->doesSetDirectiveSuppressReloc()) {
Now that printPICJumpTableSetLabel is not overloaded, inline it into its only caller, allowing us to simplify it and hoist bits out of the loop. llvm-svn: 94528
2010-01-26 14:42:44 +08:00
SmallPtrSet<const MachineBasicBlock*, 16> EmittedSets;
Update and remove a few calls to TargetMachine::getSubtargetImpl out of the asm printer. llvm-svn: 229883
2015-02-20 02:46:23 +08:00
const TargetLowering *TLI = MF->getSubtarget().getTargetLowering();
Remove the argument from EmitJumpTableInfo, because it doesn't need it. Move the X86 implementation of function body emission up to AsmPrinter::EmitFunctionBody, which works by calling the virtual EmitInstruction method. llvm-svn: 94716
2010-01-28 09:02:27 +08:00
const MCExpr *Base = TLI->getPICJumpTableRelocBaseExpr(MF,JTI,OutContext);
Now that printPICJumpTableSetLabel is not overloaded, inline it into its only caller, allowing us to simplify it and hoist bits out of the loop. llvm-svn: 94528
2010-01-26 14:42:44 +08:00
for (unsigned ii = 0, ee = JTBBs.size(); ii != ee; ++ii) {
const MachineBasicBlock *MBB = JTBBs[ii];
Update SetVector to rely on the underlying set's insert to return a pair<iterator, bool> This is to be consistent with StringSet and ultimately with the standard library's associative container insert function. This lead to updating SmallSet::insert to return pair<iterator, bool>, and then to update SmallPtrSet::insert to return pair<iterator, bool>, and then to update all the existing users of those functions... llvm-svn: 222334
2014-11-19 15:49:26 +08:00
if (!EmittedSets.insert(MBB).second)
continue;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
emit jump table an alias ".set" directives through MCStreamer as assignments. .set x, a-b is the same as: x = a-b llvm-svn: 94596
2010-01-27 05:53:08 +08:00
// .set LJTSet, LBB32-base
const MCExpr *LHS =
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
MCSymbolRefExpr::create(MBB->getSymbol(), OutContext);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitAssignment(GetJTSetSymbol(JTI, MBB->getNumber()),
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
MCBinaryExpr::createSub(LHS, Base,
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutContext));
Now that printPICJumpTableSetLabel is not overloaded, inline it into its only caller, allowing us to simplify it and hoist bits out of the loop. llvm-svn: 94528
2010-01-26 14:42:44 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
}
Spelling fix: consequtive -> consecutive. llvm-svn: 125563
2011-02-15 17:23:02 +08:00
// On some targets (e.g. Darwin) we want to emit two consecutive labels
add new JumpTableSpecialLabelPrefix hook to asmprinter for jumptable emission. llvm-svn: 33314
2007-01-18 09:12:56 +08:00
// before each jump table. The first label is never referenced, but tells
// the assembler and linker the extents of the jump table object. The
// second label is actually referenced by the code.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
if (JTInDiffSection && DL.hasLinkerPrivateGlobalPrefix())
add a new MachineJumpTableInfo::getJTISymbol method, use it to implement the default TargetLowering::getPICJumpTableRelocBaseExpr llvm-svn: 94523
2010-01-26 13:58:28 +08:00
// FIXME: This doesn't have to have any specific name, just any randomly
// named and numbered 'l' label would work. Simplify GetJTISymbol.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(GetJTISymbol(JTI, true));
mcize jump table and constant pool entry labels, .local on elf, and some .weak directives. llvm-svn: 94284
2010-01-23 13:19:23 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(GetJTISymbol(JTI));
mcize jump table and constant pool entry labels, .local on elf, and some .weak directives. llvm-svn: 94284
2010-01-23 13:19:23 +08:00
rename printPICJumpTableEntry -> EmitJumpTableEntry, make it private and non-virtual. It handles the non-pic case too, so just use it, simplifying EmitJumpTableInfo. llvm-svn: 94517
2010-01-26 13:10:10 +08:00
for (unsigned ii = 0, ee = JTBBs.size(); ii != ee; ++ii)
Now that printPICJumpTableSetLabel is not overloaded, inline it into its only caller, allowing us to simplify it and hoist bits out of the loop. llvm-svn: 94528
2010-01-26 14:42:44 +08:00
EmitJumpTableEntry(MJTI, JTBBs[ii], JTI);
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
}
Mark jump tables in code sections with DataRegion directives. Even out-of-line jump tables can be in the code section, so mark them as data-regions for those targets which support the directives. rdar://12362871&12362974 llvm-svn: 164571
2012-09-25 07:06:27 +08:00
if (!JTInDiffSection)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitDataRegion(MCDR_DataRegionEnd);
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
}
rename printPICJumpTableEntry -> EmitJumpTableEntry, make it private and non-virtual. It handles the non-pic case too, so just use it, simplifying EmitJumpTableInfo. llvm-svn: 94517
2010-01-26 13:10:10 +08:00
/// EmitJumpTableEntry - Emit a jump table entry for the specified MBB to the
/// current stream.
void AsmPrinter::EmitJumpTableEntry(const MachineJumpTableInfo *MJTI,
const MachineBasicBlock *MBB,
unsigned UID) const {
Assert on bad jump tables. llvm-svn: 125225
2011-02-10 05:52:06 +08:00
assert(MBB && MBB->getNumber() >= 0 && "Invalid basic block");
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
const MCExpr *Value = nullptr;
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
switch (MJTI->getEntryKind()) {
Add a new jump table encoding to indicate jump tables entries are inside the function by the target at the point of use. llvm-svn: 98255
2010-03-11 22:58:16 +08:00
case MachineJumpTableInfo::EK_Inline:
More dead code removal (using -Wunreachable-code) llvm-svn: 148578
2012-01-21 05:51:11 +08:00
llvm_unreachable("Cannot emit EK_Inline jump table entry");
Add support for target-specific 32-bit custom-lowered jump table entries. llvm-svn: 94505
2010-01-26 12:05:28 +08:00
case MachineJumpTableInfo::EK_Custom32:
Update and remove a few calls to TargetMachine::getSubtargetImpl out of the asm printer. llvm-svn: 229883
2015-02-20 02:46:23 +08:00
Value = MF->getSubtarget().getTargetLowering()->LowerCustomJumpTableEntry(
MJTI, MBB, UID, OutContext);
Add support for target-specific 32-bit custom-lowered jump table entries. llvm-svn: 94505
2010-01-26 12:05:28 +08:00
break;
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
case MachineJumpTableInfo::EK_BlockAddress:
// EK_BlockAddress - Each entry is a plain address of block, e.g.:
// .word LBB123
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
Value = MCSymbolRefExpr::create(MBB->getSymbol(), OutContext);
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
break;
case MachineJumpTableInfo::EK_GPRel32BlockAddress: {
// EK_GPRel32BlockAddress - Each entry is an address of block, encoded
// with a relocation as gp-relative, e.g.:
// .gprel32 LBB123
eliminate the now-unneeded context argument of MBB::getSymbol() llvm-svn: 98451
2010-03-14 05:04:28 +08:00
MCSymbol *MBBSym = MBB->getSymbol();
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
OutStreamer->EmitGPRel32Value(MCSymbolRefExpr::create(MBBSym, OutContext));
rename MAI::PICJumpTableDirective to MAI::GPRel32Directive to make it clear what it is, instead of how it is used. llvm-svn: 94448
2010-01-26 05:10:10 +08:00
return;
Fix PIC jump table codegen on x86-32/linux. In fact, such thing should be applied to all targets uses GOT-relative offsets for PIC (Alpha?) llvm-svn: 44108
2007-11-14 17:18:41 +08:00
}
Add support for target-specific 32-bit custom-lowered jump table entries. llvm-svn: 94505
2010-01-26 12:05:28 +08:00
Add a new MachineJumpTableInfo entry type, EK_GPRel64BlockAddress, which is needed to emit a 64-bit gp-relative relocation entry. Make changes necessary for emitting jump tables which have entries with directive .gpdword. This patch does not implement the parts needed for direct object emission or JIT. llvm-svn: 149668
2012-02-03 12:33:00 +08:00
case MachineJumpTableInfo::EK_GPRel64BlockAddress: {
// EK_GPRel64BlockAddress - Each entry is an address of block, encoded
// with a relocation as gp-relative, e.g.:
// .gpdword LBB123
MCSymbol *MBBSym = MBB->getSymbol();
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
OutStreamer->EmitGPRel64Value(MCSymbolRefExpr::create(MBBSym, OutContext));
Add a new MachineJumpTableInfo entry type, EK_GPRel64BlockAddress, which is needed to emit a 64-bit gp-relative relocation entry. Make changes necessary for emitting jump tables which have entries with directive .gpdword. This patch does not implement the parts needed for direct object emission or JIT. llvm-svn: 149668
2012-02-03 12:33:00 +08:00
return;
}
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
case MachineJumpTableInfo::EK_LabelDifference32: {
Fix a bit of confusion about .set and produce more readable assembly. Every target we support has support for assembly that looks like a = b - c .long a What is special about MachO is that the above combination suppresses the production of a relocation. With this change we avoid producing the intermediary labels when they don't add any value. llvm-svn: 220256
2014-10-21 09:17:30 +08:00
// Each entry is the address of the block minus the address of the jump
// table. This is used for PIC jump tables where gprel32 is not supported.
// e.g.:
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
// .word LBB123 - LJTI1_2
Fix a bit of confusion about .set and produce more readable assembly. Every target we support has support for assembly that looks like a = b - c .long a What is special about MachO is that the above combination suppresses the production of a relocation. With this change we avoid producing the intermediary labels when they don't add any value. llvm-svn: 220256
2014-10-21 09:17:30 +08:00
// If the .set directive avoids relocations, this is emitted as:
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
// .set L4_5_set_123, LBB123 - LJTI1_2
// .word L4_5_set_123
doesSetDirectiveSuppressesReloc -> doesSetDirectiveSuppressReloc, the former is grammatically incorrect. llvm-svn: 273100
2016-06-19 07:25:37 +08:00
if (MAI->doesSetDirectiveSuppressReloc()) {
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
Value = MCSymbolRefExpr::create(GetJTSetSymbol(UID, MBB->getNumber()),
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
OutContext);
break;
}
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
Value = MCSymbolRefExpr::create(MBB->getSymbol(), OutContext);
Remove more uses of TargetMachine::getSubtargetImpl from the AsmPrinter. getSubtargetInfo now asserts that the MachineFunction exists. Debug printing of register naming now uses the register info from MCAsmInfo as that's unchanging. llvm-svn: 229978
2015-02-20 15:16:19 +08:00
const TargetLowering *TLI = MF->getSubtarget().getTargetLowering();
Compute the correct jump table entries on 32 bit windows. On 32 bit windows we use label differences and .set does not suppress rolocations, a combination that was not used before r220256. This fixes PR21497. llvm-svn: 221456
2014-11-06 22:39:49 +08:00
const MCExpr *Base = TLI->getPICJumpTableRelocBaseExpr(MF, UID, OutContext);
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
Value = MCBinaryExpr::createSub(Value, Base, OutContext);
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
break;
}
Fix PIC jump table codegen on x86-32/linux. In fact, such thing should be applied to all targets uses GOT-relative offsets for PIC (Alpha?) llvm-svn: 44108
2007-11-14 17:18:41 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
switch jump table entry emission to be based on EntryKind instead of magic variables. llvm-svn: 94500
2010-01-26 11:43:22 +08:00
assert(Value && "Unknown entry kind!");
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
unsigned EntrySize = MJTI->getEntrySize(getDataLayout());
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitValue(Value, EntrySize);
Fix PIC jump table codegen on x86-32/linux. In fact, such thing should be applied to all targets uses GOT-relative offsets for PIC (Alpha?) llvm-svn: 44108
2007-11-14 17:18:41 +08:00
}
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
/// EmitSpecialLLVMGlobal - Check to see if the specified global is a
/// special global used by LLVM. If so, emit it and return true, otherwise
/// do nothing and return false.
bool AsmPrinter::EmitSpecialLLVMGlobal(const GlobalVariable *GV) {
Remove Value::{isName, getNameRef}. Also, change MDString to use a StringRef. llvm-svn: 77098
2009-07-26 07:55:21 +08:00
if (GV->getName() == "llvm.used") {
mcstreamerize .no_dead_strip and .reference for static ctors/dtors. llvm-svn: 94290
2010-01-23 13:51:36 +08:00
if (MAI->hasNoDeadStrip()) // No need to emit this at all.
Clarify that llvm.used can contain aliases. Also add a check for llvm.used in the verifier and simplify clients now that they can assume they have a ConstantArray. llvm-svn: 180019
2013-04-22 22:58:02 +08:00
EmitLLVMUsedList(cast<ConstantArray>(GV->getInitializer()));
move this check. ppc outputs .no_dead_strip properly llvm-svn: 41286
2007-08-23 03:33:11 +08:00
return true;
}
implement a new magic global "llvm.compiler.used" which is like llvm.used, but doesn't cause ".no_dead_strip" to be emitted on darwin. llvm-svn: 76399
2009-07-20 14:14:25 +08:00
// Ignore debug and non-emitted data. This handles llvm.compiler.used.
Return a StringRef from getSection. This is similar to how getName is handled. llvm-svn: 269218
2016-05-12 02:21:59 +08:00
if (GV->getSection() == "llvm.metadata" ||
Add a new "available_externally" linkage type. This is intended to support C99 inline, GNU extern inline, etc. Related bugzilla's include PR3517, PR3100, & PR2933. Nothing uses this yet, but it appears to work. llvm-svn: 68940
2009-04-13 13:44:34 +08:00
GV->hasAvailableExternallyLinkage())
return true;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Use "llvm.metadata" section for debug globals. Filter out these globals in the asm printer. llvm-svn: 26599
2006-03-08 06:00:35 +08:00
if (!GV->hasAppendingLinkage()) return false;
assert(GV->hasInitializer() && "Not a special LLVM global!");
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Make the code generator rip of dead constant expr uses before deciding whether a global is dead or not. This should fix PR3749 - linker adds spurious use to appending globals. I can't reasonably add a testcase for this, because the bc writer/reader strip dead constant users. llvm-svn: 66404
2009-03-09 13:52:15 +08:00
if (GV->getName() == "llvm.global_ctors") {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitXXStructorList(GV->getParent()->getDataLayout(), GV->getInitializer(),
/* isCtor */ true);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
just remove the use_empty() check entirely, the only reason it existed was for llvm-gcc 3.4 (which used the __main hack) which is really really long dead. llvm-svn: 66417
2009-03-09 16:18:48 +08:00
return true;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
}
Make the code generator rip of dead constant expr uses before deciding whether a global is dead or not. This should fix PR3749 - linker adds spurious use to appending globals. I can't reasonably add a testcase for this, because the bc writer/reader strip dead constant users. llvm-svn: 66404
2009-03-09 13:52:15 +08:00
if (GV->getName() == "llvm.global_dtors") {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitXXStructorList(GV->getParent()->getDataLayout(), GV->getInitializer(),
/* isCtor */ false);
move production of .reference directives for static ctor/dtor list on darwin into common code. llvm-svn: 93849
2010-01-19 12:34:02 +08:00
just remove the use_empty() check entirely, the only reason it existed was for llvm-gcc 3.4 (which used the __main hack) which is really really long dead. llvm-svn: 66417
2009-03-09 16:18:48 +08:00
return true;
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fail early on unknown appending linkage variables. In practice only a few well known appending linkage variables work. Currently if codegen sees an unknown appending linkage variable it will just print it as a regular global. That is wrong as the symbol in the produced object file has different semantics as the one provided by the appending linkage. This just errors early instead of producing a broken .o. llvm-svn: 269706
2016-05-17 05:14:24 +08:00
report_fatal_error("unknown special variable");
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
}
rename TAI -> MAI, being careful not to make MAILJMP instructions :) llvm-svn: 79777
2009-08-23 05:43:10 +08:00
/// EmitLLVMUsedList - For targets that define a MAI::UsedDirective, mark each
Move the uglier parts of deciding not to emit a UsedDirective for some symbols in llvm.used into Darwin-specific code. I've decided LessPrivateGlobal is potentially a useful abstraction and left it in the target-independent area, with improved comment. llvm-svn: 56024
2008-09-10 06:29:13 +08:00
/// global in the specified llvm.used list for which emitUsedDirectiveFor
/// is true, as being used with this directive.
Clarify that llvm.used can contain aliases. Also add a check for llvm.used in the verifier and simplify clients now that they can assume they have a ConstantArray. llvm-svn: 180019
2013-04-22 22:58:02 +08:00
void AsmPrinter::EmitLLVMUsedList(const ConstantArray *InitList) {
Fix old-style type names in comments. llvm-svn: 73362
2009-06-15 07:30:43 +08:00
// Should be an array of 'i8*'.
Add support for targets that want to do something with the llvm.used list, because they have an aggressive linker that does dead code stripping. llvm-svn: 30604
2006-09-26 11:38:18 +08:00
for (unsigned i = 0, e = InitList->getNumOperands(); i != e; ++i) {
remove AsmPrinter::findGlobalValue, just use Value::stripPointerCasts instead. llvm-svn: 76246
2009-07-18 06:00:23 +08:00
const GlobalValue *GV =
dyn_cast<GlobalValue>(InitList->getOperand(i)->stripPointerCasts());
Remove shouldEmitUsedDirectiveFor. Clang now uses llvm.compiler.used for these cases. llvm-svn: 203174
2014-03-07 06:47:08 +08:00
if (GV)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(getSymbol(GV), MCSA_NoDeadStrip);
Add support for targets that want to do something with the llvm.used list, because they have an aggressive linker that does dead code stripping. llvm-svn: 30604
2006-09-26 11:38:18 +08:00
}
}
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
namespace {
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
struct Structor {
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
int Priority = 0;
Constant *Func = nullptr;
GlobalValue *ComdatKey = nullptr;
Structor() = default;
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
};
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
} // end anonymous namespace
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
/// EmitXXStructorList - Emit the ctor or dtor list taking into account the init
/// priority.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
void AsmPrinter::EmitXXStructorList(const DataLayout &DL, const Constant *List,
bool isCtor) {
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
// Should be an array of '{ int, void ()* }' structs. The first value is the
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
// init priority.
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
if (!isa<ConstantArray>(List)) return;
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
// Sanity check the structors list.
const ConstantArray *InitList = dyn_cast<ConstantArray>(List);
if (!InitList) return; // Not an array!
StructType *ETy = dyn_cast<StructType>(InitList->getType()->getElementType());
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
// FIXME: Only allow the 3-field form in LLVM 4.0.
if (!ETy || ETy->getNumElements() < 2 || ETy->getNumElements() > 3)
return; // Not an array of two or three elements!
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
if (!isa<IntegerType>(ETy->getTypeAtIndex(0U)) ||
!isa<PointerType>(ETy->getTypeAtIndex(1U))) return; // Not (int, ptr).
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
if (ETy->getNumElements() == 3 && !isa<PointerType>(ETy->getTypeAtIndex(2U)))
return; // Not (int, ptr, ptr).
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
// Gather the structors in a form that's convenient for sorting by priority.
SmallVector<Structor, 8> Structors;
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
for (Value *O : InitList->operands()) {
ConstantStruct *CS = dyn_cast<ConstantStruct>(O);
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
if (!CS) continue; // Malformed.
if (CS->getOperand(1)->isNullValue())
break; // Found a null terminator, skip the rest.
ConstantInt *Priority = dyn_cast<ConstantInt>(CS->getOperand(0));
if (!Priority) continue; // Malformed.
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
Structors.push_back(Structor());
Structor &S = Structors.back();
S.Priority = Priority->getLimitedValue(65535);
S.Func = CS->getOperand(1);
if (ETy->getNumElements() == 3 && !CS->getOperand(2)->isNullValue())
80-column and comment fixups. llvm-svn: 274335
2016-07-01 14:07:31 +08:00
S.ComdatKey =
dyn_cast<GlobalValue>(CS->getOperand(2)->stripPointerCasts());
Fix PR5329: pay attention to constructor/destructor priority when outputting them. With this, the entire LLVM testsuite passes when built with dragonegg. llvm-svn: 138724
2011-08-28 21:17:22 +08:00
}
Properly emit ctors / dtors with priorities into desired sections and let linker handle the rest. This finally fixes PR5329 llvm-svn: 148990
2012-01-26 06:24:19 +08:00
// Emit the function pointers in the target-specific order
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
unsigned Align = Log2_32(DL.getPointerPrefAlignment());
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
std::stable_sort(Structors.begin(), Structors.end(),
[](const Structor &L,
const Structor &R) { return L.Priority < R.Priority; });
for (Structor &S : Structors) {
const TargetLoweringObjectFile &Obj = getObjFileLowering();
const MCSymbol *KeySym = nullptr;
Don't emit structors for available_externally globals (PR19933) We would previously assert here when trying to figure out the section for the global. This makes us handle the situation more gracefully since the IR isn't malformed. Differential Revision: http://reviews.llvm.org/D4022 llvm-svn: 210215
2014-06-05 05:04:54 +08:00
if (GlobalValue *GV = S.ComdatKey) {
Don't create a comdat group for a dropped def with initializer Non-prevailing weak/linkonce odr symbols will be dropped by ThinLTO to available_externally when possible. If they had an initializer in the global_ctors list, a comdat group was being created. This code already had logic to skip available_externally defs, but now the EliminateAvailableExternally pass will drop these symbols to declarations earlier. Change the check to skip all declarations for linker (which includes available_externally along with declarations). Reviewers: mehdi_amini Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D28737 llvm-svn: 292408
2017-01-19 00:58:43 +08:00
if (GV->isDeclarationForLinker())
// If the associated variable is not defined in this module
// (it might be available_externally, or have been an
// available_externally definition that was dropped by the
// EliminateAvailableExternally pass), some other TU
Don't emit structors for available_externally globals (PR19933) We would previously assert here when trying to figure out the section for the global. This makes us handle the situation more gracefully since the IR isn't malformed. Differential Revision: http://reviews.llvm.org/D4022 llvm-svn: 210215
2014-06-05 05:04:54 +08:00
// will provide its dynamic initializer.
continue;
KeySym = getSymbol(GV);
Add comdat key field to llvm.global_ctors and llvm.global_dtors This allows us to put dynamic initializers for weak data into the same comdat group as the data being initialized. This is necessary for MSVC ABI compatibility. Once we have comdats for guard variables, we can use the combination to help GlobalOpt fire more often for weak data with guarded initialization on other platforms. Reviewers: nlewycky Differential Revision: http://reviews.llvm.org/D3499 llvm-svn: 209015
2014-05-17 04:39:27 +08:00
}
Move alignment from MCSectionData to MCSection. This starts merging MCSection and MCSectionData. There are a few issues with the current split between MCSection and MCSectionData. * It optimizes the the not as important case. We want the production of .o files to be really fast, but the split puts the information used for .o emission in a separate data structure. * The ELF/COFF/MachO hierarchy is not represented in MCSectionData, leading to some ad-hoc ways to represent the various flags. * It makes it harder to remember where each item is. The attached patch starts merging the two by moving the alignment from MCSectionData to MCSection. Most of the patch is actually just dropping 'const', since MCSectionData is mutable, but MCSection was not. llvm-svn: 237936
2015-05-22 03:20:38 +08:00
MCSection *OutputSection =
Fix a few issues with comdat handling on COFF. * Section association cannot use just the section name as many sections can have the same name. With this patch, the comdat symbol in an assoc section is interpreted to mean a symbol in the associated section and the mapping is discovered from it. * Comdat symbols were not being set correctly. Instead we were getting whatever was output first for that section. A consequence is that associative sections now must use .section to set the association. Using .linkonce would not work since it is not possible to change a sections comdat symbol (it is used to decide if we should create a new section or reuse an existing one). This includes r210298, which was reverted because it was asserting on an associated section having the same comdat as the associated section. llvm-svn: 210367
2014-06-07 03:26:12 +08:00
(isCtor ? Obj.getStaticCtorSection(S.Priority, KeySym)
: Obj.getStaticDtorSection(S.Priority, KeySym));
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->SwitchSection(OutputSection);
if (OutStreamer->getCurrentSection() != OutStreamer->getPreviousSection())
Properly emit ctors / dtors with priorities into desired sections and let linker handle the rest. This finally fixes PR5329 llvm-svn: 148990
2012-01-26 06:24:19 +08:00
EmitAlignment(Align);
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
EmitXXStructor(DL, S.Func);
Properly emit ctors / dtors with priorities into desired sections and let linker handle the rest. This finally fixes PR5329 llvm-svn: 148990
2012-01-26 06:24:19 +08:00
}
Add a couple more fields, move ctor init list to .cpp file, add support for emitting the ctor/dtor list for common targets. llvm-svn: 24694
2005-12-13 14:32:10 +08:00
}
Allow target to customize directive used to switch to arbitrary section in SwitchSection, add generic constant pool emitter llvm-svn: 24464
2005-11-21 16:25:09 +08:00
Add support for metadata representing .ident directives. llvm-svn: 192764
2013-10-16 09:49:05 +08:00
void AsmPrinter::EmitModuleIdents(Module &M) {
if (!MAI->hasIdentDirective())
return;
if (const NamedMDNode *NMD = M.getNamedMetadata("llvm.ident")) {
for (unsigned i = 0, e = NMD->getNumOperands(); i != e; ++i) {
Revert "IR: MDNode => Value" Instead, we're going to separate metadata from the Value hierarchy. See PR21532. This reverts commit r221375. This reverts commit r221373. This reverts commit r221359. This reverts commit r221167. This reverts commit r221027. This reverts commit r221024. This reverts commit r221023. This reverts commit r220995. This reverts commit r220994. llvm-svn: 221711
2014-11-12 05:30:22 +08:00
const MDNode *N = NMD->getOperand(i);
whitespace llvm-svn: 201181
2014-02-12 05:23:02 +08:00
assert(N->getNumOperands() == 1 &&
Add support for metadata representing .ident directives. llvm-svn: 192764
2013-10-16 09:49:05 +08:00
"llvm.ident metadata entry can have only one operand");
const MDString *S = cast<MDString>(N->getOperand(0));
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIdent(S->getString());
Add support for metadata representing .ident directives. llvm-svn: 192764
2013-10-16 09:49:05 +08:00
}
}
}
Migrate print routines to asm to be shared by exception handling. llvm-svn: 33498
2007-01-25 23:12:02 +08:00
//===--------------------------------------------------------------------===//
// Emission and print routines
//
/// EmitInt8 - Emit a byte directive and value.
///
void AsmPrinter::EmitInt8(int Value) const {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIntValue(Value, 1);
Migrate print routines to asm to be shared by exception handling. llvm-svn: 33498
2007-01-25 23:12:02 +08:00
}
/// EmitInt16 - Emit a short directive and value.
///
void AsmPrinter::EmitInt16(int Value) const {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIntValue(Value, 2);
Migrate print routines to asm to be shared by exception handling. llvm-svn: 33498
2007-01-25 23:12:02 +08:00
}
/// EmitInt32 - Emit a long directive and value.
///
void AsmPrinter::EmitInt32(int Value) const {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIntValue(Value, 4);
Migrate print routines to asm to be shared by exception handling. llvm-svn: 33498
2007-01-25 23:12:02 +08:00
}
Fix a bit of confusion about .set and produce more readable assembly. Every target we support has support for assembly that looks like a = b - c .long a What is special about MachO is that the above combination suppresses the production of a relocation. With this change we avoid producing the intermediary labels when they don't add any value. llvm-svn: 220256
2014-10-21 09:17:30 +08:00
/// Emit something like ".long Hi-Lo" where the size in bytes of the directive
/// is specified by Size and Hi/Lo specify the labels. This implicitly uses
/// .set if it avoids relocations.
move .set generation out of DwarfPrinter into AsmPrinter and MCize it. llvm-svn: 98010
2010-03-09 07:58:37 +08:00
void AsmPrinter::EmitLabelDifference(const MCSymbol *Hi, const MCSymbol *Lo,
unsigned Size) const {
Generalize emitAbsoluteSymbolDiff. This makes emitAbsoluteSymbolDiff always succeed and moves logic from the asm printer to it. The object one now also works on ELF. If two symbols are in the same fragment, we will never move them apart. llvm-svn: 239552
2015-06-12 02:58:08 +08:00
OutStreamer->emitAbsoluteSymbolDiff(Hi, Lo, Size);
move .set generation out of DwarfPrinter into AsmPrinter and MCize it. llvm-svn: 98010
2010-03-09 07:58:37 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
/// EmitLabelPlusOffset - Emit something like ".long Label+Offset"
Fix .debug_range for linux. Patch by Krister Wombell. llvm-svn: 112830
2010-09-03 00:43:44 +08:00
/// where the size in bytes of the directive is specified by Size and Label
/// specifies the label. This implicitly uses .set if it is available.
void AsmPrinter::EmitLabelPlusOffset(const MCSymbol *Label, uint64_t Offset,
Reformat const for readability. llvm-svn: 195522
2013-11-23 08:05:06 +08:00
unsigned Size,
bool IsSectionRelative) const {
Rename DataLayout variables TD -> DL llvm-svn: 191927
2013-10-04 03:50:01 +08:00
if (MAI->needsDwarfSectionOffsetDirective() && IsSectionRelative) {
Reapply "[CodeGen] Fix invalid DWARF info on Win64" This reapplies rL289013 (reverted in rL289014) with the fixes identified in D21731. Should hopefully pass the buildbots this time. llvm-svn: 290809
2017-01-02 11:00:19 +08:00
OutStreamer->EmitCOFFSecRel32(Label, Offset);
if (Size > 4)
OutStreamer->EmitZeros(Size - 4);
Bugfix for making the DWARF debug strings and labels to code emitted as secrel32 instead of long opcodes (only for coff). This makes them debuggable with GDB (with fix for 64bits msvc) llvm-svn: 187656
2013-08-03 00:14:15 +08:00
return;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Emit relocations for DW_AT_location entries on systems which need it. This is a recommit of r127757. Fixes PR9493. Patch by Paul Robinson! llvm-svn: 158957
2012-06-22 09:25:12 +08:00
// Emit Label+Offset (or just Label if Offset is zero)
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
const MCExpr *Expr = MCSymbolRefExpr::create(Label, OutContext);
Emit relocations for DW_AT_location entries on systems which need it. This is a recommit of r127757. Fixes PR9493. Patch by Paul Robinson! llvm-svn: 158957
2012-06-22 09:25:12 +08:00
if (Offset)
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
Expr = MCBinaryExpr::createAdd(
Expr, MCConstantExpr::create(Offset, OutContext), OutContext);
Emit relocations for DW_AT_location entries on systems which need it. This is a recommit of r127757. Fixes PR9493. Patch by Paul Robinson! llvm-svn: 158957
2012-06-22 09:25:12 +08:00
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitValue(Expr, Size);
Fix .debug_range for linux. Patch by Krister Wombell. llvm-svn: 112830
2010-09-03 00:43:44 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Migrate print routines to asm to be shared by exception handling. llvm-svn: 33498
2007-01-25 23:12:02 +08:00
//===----------------------------------------------------------------------===//
Fix the asmprinter so that a globalvalue can specify an explicit alignment smaller than the preferred alignment, but so that the target can actually specify a minimum alignment if needed. This fixes some objc protocol failures Devang tracked down. llvm-svn: 37373
2007-06-01 02:57:45 +08:00
// EmitAlignment - Emit an alignment directive to the specified power of
// two boundary. For example, if you pass in 3 here, you will get an 8
// byte alignment. If a global value is specified, and if that global has
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
// an explicit alignment requested, it will override the alignment request
// if required for correctness.
Fix the asmprinter so that a globalvalue can specify an explicit alignment smaller than the preferred alignment, but so that the target can actually specify a minimum alignment if needed. This fixes some objc protocol failures Devang tracked down. llvm-svn: 37373
2007-06-01 02:57:45 +08:00
//
Split GlobalValue into GlobalValue and GlobalObject. This allows code to statically accept a Function or a GlobalVariable, but not an alias. This is already a cleanup by itself IMHO, but the main reason for it is that it gives a lot more confidence that the refactoring to fix the design of GlobalAlias is correct. That will be a followup patch. llvm-svn: 208716
2014-05-14 02:45:48 +08:00
void AsmPrinter::EmitAlignment(unsigned NumBits, const GlobalObject *GV) const {
Remove the TargetMachine forwards for TargetSubtargetInfo based information and update all callers. No functional change. llvm-svn: 214781
2014-08-05 05:25:23 +08:00
if (GV)
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
NumBits = getGVAlignmentLog2(GV, GV->getParent()->getDataLayout(), NumBits);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Rework global alignment computation again. Now we do round up alignment of globals to the preferred alignment, but only when there is no section specified on the global (by far the common case). llvm-svn: 102515
2010-04-29 03:58:07 +08:00
if (NumBits == 0) return; // 1-byte aligned: no need to emit alignment.
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Fixing -Wsign-compare warnings; NFC. llvm-svn: 224337
2014-12-16 22:04:11 +08:00
assert(NumBits <
static_cast<unsigned>(std::numeric_limits<unsigned>::digits) &&
Silence more static analyzer warnings. Add in definedness checks for shift operators, null checks when pointers are assumed by the code to be non-null, and explicit unreachables. llvm-svn: 224255
2014-12-16 02:48:43 +08:00
"undefined behavior");
Make AsmStreamer maintain a notion of the current section, pushing it up from the MCAsmStreamer. Based on this, eliminate the current section from AsmPrinter. While I'm at it, clean up the last of the horrible "switch to null section" stuff and add an assert. This change is in preparation for completely eliminating asmprinter::switchtosection. llvm-svn: 79324
2009-08-18 14:15:16 +08:00
if (getCurrentSection()->getKind().isText())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitCodeAlignment(1u << NumBits);
use OutStreamer.EmitCodeAlignment for alignment in the text segment. llvm-svn: 96967
2010-02-24 02:46:22 +08:00
else
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitValueToAlignment(1u << NumBits);
Add support for alignment llvm-svn: 15888
2004-08-18 03:14:29 +08:00
}
Fix some column padding bugs, reorganize things as suggested by Chris and eliminate complexity. Yay! llvm-svn: 78243
2009-08-06 05:00:52 +08:00
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
//===----------------------------------------------------------------------===//
// Constant emission.
//===----------------------------------------------------------------------===//
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
const MCExpr *AsmPrinter::lowerConstant(const Constant *CV) {
MCContext &Ctx = OutContext;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
if (CV->isNullValue() || isa<UndefValue>(CV))
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
return MCConstantExpr::create(0, Ctx);
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
if (const ConstantInt *CI = dyn_cast<ConstantInt>(CV))
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
return MCConstantExpr::create(CI->getZExtValue(), Ctx);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
if (const GlobalValue *GV = dyn_cast<GlobalValue>(CV))
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
return MCSymbolRefExpr::create(getSymbol(GV), Ctx);
Improve whitespace. llvm-svn: 111384
2010-08-19 02:41:13 +08:00
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
if (const BlockAddress *BA = dyn_cast<BlockAddress>(CV))
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
return MCSymbolRefExpr::create(GetBlockAddressSymbol(BA), Ctx);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
const ConstantExpr *CE = dyn_cast<ConstantExpr>(CV);
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (!CE) {
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
llvm_unreachable("Unknown constant value to lower!");
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
switch (CE->getOpcode()) {
When CodeGen'ing unoptimized code, there may be unfolded constant expressions in global initializers. Instead of aborting, attempt to fold them on the spot. If folding succeeds, emit the folded expression instead. This fixes PR6255. llvm-svn: 95583
2010-02-09 06:02:38 +08:00
default:
// If the code isn't optimized, there may be outstanding folding
Move TargetData to DataLayout. llvm-svn: 165402
2012-10-09 00:38:25 +08:00
// opportunities. Attempt to fold the expression using DataLayout as a
When CodeGen'ing unoptimized code, there may be unfolded constant expressions in global initializers. Instead of aborting, attempt to fold them on the spot. If folding succeeds, emit the folded expression instead. This fixes PR6255. llvm-svn: 95583
2010-02-09 06:02:38 +08:00
// last resort before giving up.
[ConstnatFolding] Teach the folder how to fold ConstantVector A ConstantVector can have ConstantExpr operands and vice versa. However, the folder had no ability to fold ConstantVectors which, in some cases, was an optimization barrier. Instead, rephrase the folder in terms of Constants instead of ConstantExprs and teach callers how to deal with failure. llvm-svn: 277099
2016-07-29 11:27:26 +08:00
if (Constant *C = ConstantFoldConstant(CE, getDataLayout()))
ConstantFoldConstantExpression can theoretically return the original expression; don't go into an infinite loop if it does. llvm-svn: 95591
2010-02-09 06:19:11 +08:00
if (C != CE)
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
return lowerConstant(C);
Change this llvm_unreachable to report_fatal_error, since it can be triggered by valid, if dubious, IR. llvm-svn: 110240
2010-08-05 02:51:09 +08:00
// Otherwise report the problem to the user.
{
Revert "Introduce a string_ostream string builder facilty" Temporarily back out commits r211749, r211752 and r211754. llvm-svn: 211814
2014-06-27 06:52:05 +08:00
std::string S;
raw_string_ostream OS(S);
Change this llvm_unreachable to report_fatal_error, since it can be triggered by valid, if dubious, IR. llvm-svn: 110240
2010-08-05 02:51:09 +08:00
OS << "Unsupported expression in static initializer: ";
Put the functionality for printing a value to a raw_ostream as an operand into the Value interface just like the core print method is. That gives a more conistent organization to the IR printing interfaces -- they are all attached to the IR objects themselves. Also, update all the users. This removes the 'Writer.h' header which contained only a single function declaration. llvm-svn: 198836
2014-01-09 10:29:41 +08:00
CE->printAsOperand(OS, /*PrintType=*/false,
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
!MF ? nullptr : MF->getFunction()->getParent());
Change this llvm_unreachable to report_fatal_error, since it can be triggered by valid, if dubious, IR. llvm-svn: 110240
2010-08-05 02:51:09 +08:00
report_fatal_error(OS.str());
}
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::GetElementPtr: {
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
// Generate a symbolic expression for the byte address
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
APInt OffsetAI(getDataLayout().getPointerTypeSizeInBits(CE->getType()), 0);
cast<GEPOperator>(CE)->accumulateConstantOffset(getDataLayout(), OffsetAI);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
const MCExpr *Base = lowerConstant(CE->getOperand(0));
convert a bunch of callers from DataLayout::getIndexedOffset() to GEP::accumulateConstantOffset(). The later API is nicer than the former, and is correct regarding wrap-around offsets (if anyone cares). There are a few more places left with duplicated code, which I'll remove soon. llvm-svn: 171259
2012-12-31 00:25:48 +08:00
if (!OffsetAI)
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
return Base;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
convert a bunch of callers from DataLayout::getIndexedOffset() to GEP::accumulateConstantOffset(). The later API is nicer than the former, and is correct regarding wrap-around offsets (if anyone cares). There are a few more places left with duplicated code, which I'll remove soon. llvm-svn: 171259
2012-12-31 00:25:48 +08:00
int64_t Offset = OffsetAI.getSExtValue();
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
return MCBinaryExpr::createAdd(Base, MCConstantExpr::create(Offset, Ctx),
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
Ctx);
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
case Instruction::Trunc:
// We emit the value and depend on the assembler to truncate the generated
// expression properly. This is important for differences between
// blockaddress labels. Since the two labels are in the same function, it
// is reasonable to treat their delta as a 32-bit value.
Replace a few more "fall through" comments with LLVM_FALLTHROUGH Follow up to r278902. I had missed "fall through", with a space. llvm-svn: 278970
2016-08-18 04:30:52 +08:00
LLVM_FALLTHROUGH;
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::BitCast:
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
return lowerConstant(CE->getOperand(0));
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::IntToPtr: {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = getDataLayout();
Move DataLayout back to the TargetMachine from TargetSubtargetInfo derived classes. Since global data alignment, layout, and mangling is often based on the DataLayout, move it to the TargetMachine. This ensures that global data is going to be layed out and mangled consistently if the subtarget changes on a per function basis. Prior to this all targets(*) have had subtarget dependent code moved out and onto the TargetMachine. *One target hasn't been migrated as part of this change: R600. The R600 port has, as a subtarget feature, the size of pointers and this affects global data layout. I've currently hacked in a FIXME to enable progress, but the port needs to be updated to either pass the 64-bitness to the TargetMachine, or fix the DataLayout to avoid subtarget dependent features. llvm-svn: 227113
2015-01-27 03:03:15 +08:00
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
// Handle casts to pointers by changing them into casts to the appropriate
// integer type. This promotes constant folding and simplifies this code.
Constant *Op = CE->getOperand(0);
Fix CodeGen for different size address space GEPs llvm-svn: 193111
2013-10-22 04:03:54 +08:00
Op = ConstantExpr::getIntegerCast(Op, DL.getIntPtrType(CV->getType()),
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
false/*ZExt*/);
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
return lowerConstant(Op);
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::PtrToInt: {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = getDataLayout();
Move DataLayout back to the TargetMachine from TargetSubtargetInfo derived classes. Since global data alignment, layout, and mangling is often based on the DataLayout, move it to the TargetMachine. This ensures that global data is going to be layed out and mangled consistently if the subtarget changes on a per function basis. Prior to this all targets(*) have had subtarget dependent code moved out and onto the TargetMachine. *One target hasn't been migrated as part of this change: R600. The R600 port has, as a subtarget feature, the size of pointers and this affects global data layout. I've currently hacked in a FIXME to enable progress, but the port needs to be updated to either pass the 64-bitness to the TargetMachine, or fix the DataLayout to avoid subtarget dependent features. llvm-svn: 227113
2015-01-27 03:03:15 +08:00
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
// Support only foldable casts to/from pointers that can be eliminated by
// changing the pointer to the appropriately sized integer type.
Constant *Op = CE->getOperand(0);
land David Blaikie's patch to de-constify Type, with a few tweaks. llvm-svn: 135375
2011-07-18 12:54:35 +08:00
Type *Ty = CE->getType();
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
const MCExpr *OpExpr = lowerConstant(Op);
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
// We can emit the pointer value into this slot if the slot is an
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
// integer slot equal to the size of the pointer.
Rename DataLayout variables TD -> DL llvm-svn: 191927
2013-10-04 03:50:01 +08:00
if (DL.getTypeAllocSize(Ty) == DL.getTypeAllocSize(Op->getType()))
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
return OpExpr;
// Otherwise the pointer is smaller than the resultant integer, mask off
// the high bits so we are sure to get a proper truncation if the input is
// a constant expr.
Rename DataLayout variables TD -> DL llvm-svn: 191927
2013-10-04 03:50:01 +08:00
unsigned InBits = DL.getTypeAllocSizeInBits(Op->getType());
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
const MCExpr *MaskExpr = MCConstantExpr::create(~0ULL >> (64-InBits), Ctx);
return MCBinaryExpr::createAnd(OpExpr, MaskExpr, Ctx);
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
CodeGen: Use PLT relocations for relative references to unnamed_addr functions. The relative vtable ABI (PR26723) needs PLT relocations to refer to virtual functions defined in other DSOs. The unnamed_addr attribute means that the function's address is not significant, so we're allowed to substitute it with the address of a PLT entry. Also includes a bonus feature: addends for COFF image-relative references. Differential Revision: http://reviews.llvm.org/D17938 llvm-svn: 267211
2016-04-23 04:40:10 +08:00
case Instruction::Sub: {
GlobalValue *LHSGV;
APInt LHSOffset;
if (IsConstantOffsetFromGlobal(CE->getOperand(0), LHSGV, LHSOffset,
getDataLayout())) {
GlobalValue *RHSGV;
APInt RHSOffset;
if (IsConstantOffsetFromGlobal(CE->getOperand(1), RHSGV, RHSOffset,
getDataLayout())) {
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
const MCExpr *RelocExpr =
getObjFileLowering().lowerRelativeReference(LHSGV, RHSGV, TM);
CodeGen: Use PLT relocations for relative references to unnamed_addr functions. The relative vtable ABI (PR26723) needs PLT relocations to refer to virtual functions defined in other DSOs. The unnamed_addr attribute means that the function's address is not significant, so we're allowed to substitute it with the address of a PLT entry. Also includes a bonus feature: addends for COFF image-relative references. Differential Revision: http://reviews.llvm.org/D17938 llvm-svn: 267211
2016-04-23 04:40:10 +08:00
if (!RelocExpr)
RelocExpr = MCBinaryExpr::createSub(
MCSymbolRefExpr::create(getSymbol(LHSGV), Ctx),
MCSymbolRefExpr::create(getSymbol(RHSGV), Ctx), Ctx);
int64_t Addend = (LHSOffset - RHSOffset).getSExtValue();
if (Addend != 0)
RelocExpr = MCBinaryExpr::createAdd(
RelocExpr, MCConstantExpr::create(Addend, Ctx), Ctx);
return RelocExpr;
}
}
}
// else fallthrough
Implement AsmPrinter support for several more operators which have direct MCExpr equivalents. Don't use MCExpr::Shr because it isn't consistent between targets. llvm-svn: 95620
2010-02-09 08:02:37 +08:00
// The MC library also has a right-shift operator, but it isn't consistently
// signed or unsigned between different targets.
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::Add:
Implement AsmPrinter support for several more operators which have direct MCExpr equivalents. Don't use MCExpr::Shr because it isn't consistent between targets. llvm-svn: 95620
2010-02-09 08:02:37 +08:00
case Instruction::Mul:
case Instruction::SDiv:
case Instruction::SRem:
case Instruction::Shl:
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
case Instruction::And:
case Instruction::Or:
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
case Instruction::Xor: {
Move lowerConstant to AsmPrinter This was a static function before, and NVPTX duplicated it because it wasn't exposed. llvm-svn: 224354
2014-12-17 03:16:14 +08:00
const MCExpr *LHS = lowerConstant(CE->getOperand(0));
const MCExpr *RHS = lowerConstant(CE->getOperand(1));
reduce indentation and use early exits in AsmPrinter::EmitConstantValueOnly llvm-svn: 93290
2010-01-13 12:29:19 +08:00
switch (CE->getOpcode()) {
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
default: llvm_unreachable("Unknown binary operator constant cast expr");
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
case Instruction::Add: return MCBinaryExpr::createAdd(LHS, RHS, Ctx);
case Instruction::Sub: return MCBinaryExpr::createSub(LHS, RHS, Ctx);
case Instruction::Mul: return MCBinaryExpr::createMul(LHS, RHS, Ctx);
case Instruction::SDiv: return MCBinaryExpr::createDiv(LHS, RHS, Ctx);
case Instruction::SRem: return MCBinaryExpr::createMod(LHS, RHS, Ctx);
case Instruction::Shl: return MCBinaryExpr::createShl(LHS, RHS, Ctx);
case Instruction::And: return MCBinaryExpr::createAnd(LHS, RHS, Ctx);
case Instruction::Or: return MCBinaryExpr::createOr (LHS, RHS, Ctx);
case Instruction::Xor: return MCBinaryExpr::createXor(LHS, RHS, Ctx);
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
}
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
}
Initial implementation of the asmprinter base class llvm-svn: 15838
2004-08-17 07:15:22 +08:00
}
}
Implement emitGlobalConstant llvm-svn: 15868
2004-08-17 14:36:49 +08:00
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantImpl(const DataLayout &DL, const Constant *C,
AsmPrinter &AP,
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const Constant *BaseCV = nullptr,
uint64_t Offset = 0);
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
AsmPrinter: Use emitGlobalConstantFP to emit elements of constant data It's strange to duplicate the logic for emitting FP values into emitGlobalConstantDataSequential, and it's even stranger that we end up printing the verbose assembly comments differently between the two paths. Just call into emitGlobalConstantFP rather than crudely duplicating its logic. llvm-svn: 254988
2015-12-08 10:37:48 +08:00
static void emitGlobalConstantFP(const ConstantFP *CFP, AsmPrinter &AP);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
/// isRepeatedByteSequence - Determine whether the given value is
/// composed of a repeated sequence of identical bytes and return the
/// byte value. If it is not a repeated sequence, return -1.
static int isRepeatedByteSequence(const ConstantDataSequential *V) {
StringRef Data = V->getRawDataValues();
assert(!Data.empty() && "Empty aggregates should be CAZ node");
char C = Data[0];
for (unsigned i = 1, e = Data.size(); i != e; ++i)
if (Data[i] != C) return -1;
Chris's constant data sequence refactoring actually enabled printing vectors of all one bits to be printed more cleverly in the AsmPrinter. Unfortunately, the byte value for all one bits is the same with -fsigned-char as the error return of '-1'. Force this to be the unsigned byte value when returning it to avoid this problem, and update the test case for the shiny new behavior. Yay for building LLVM and Clang with -funsigned-char. Chris, please review, and let me know if there is any reason to not desire this change. It seems good on the surface, and certainly intended based on the code written. llvm-svn: 149299
2012-01-31 07:47:44 +08:00
return static_cast<uint8_t>(C); // Ensure 255 is not returned as -1.
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
}
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
/// isRepeatedByteSequence - Determine whether the given value is
/// composed of a repeated sequence of identical bytes and return the
/// byte value. If it is not a repeated sequence, return -1.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static int isRepeatedByteSequence(const Value *V, const DataLayout &DL) {
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
if (const ConstantInt *CI = dyn_cast<ConstantInt>(V)) {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
uint64_t Size = DL.getTypeAllocSizeInBits(V->getType());
[AsmPrinter] Make isRepeatedByteSequence smarter about odd integer types - zext the value to alloc size first, then check if the value repeats with zero padding included. If so we can still emit a .space - Do the checking with APInt.isSplat(8), which handles non-pow2 types - Also handle large constants (bit width > 64) - In a ConstantArray all elements have the same type, so it's sufficient to check the first constant recursively and then just compare if all following constants are the same by pointer compare llvm-svn: 239977
2015-06-18 07:55:17 +08:00
assert(Size % 8 == 0);
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
[AsmPrinter] Make isRepeatedByteSequence smarter about odd integer types - zext the value to alloc size first, then check if the value repeats with zero padding included. If so we can still emit a .space - Do the checking with APInt.isSplat(8), which handles non-pow2 types - Also handle large constants (bit width > 64) - In a ConstantArray all elements have the same type, so it's sufficient to check the first constant recursively and then just compare if all following constants are the same by pointer compare llvm-svn: 239977
2015-06-18 07:55:17 +08:00
// Extend the element to take zero padding into account.
APInt Value = CI->getValue().zextOrSelf(Size);
if (!Value.isSplat(8))
return -1;
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
[AsmPrinter] Make isRepeatedByteSequence smarter about odd integer types - zext the value to alloc size first, then check if the value repeats with zero padding included. If so we can still emit a .space - Do the checking with APInt.isSplat(8), which handles non-pow2 types - Also handle large constants (bit width > 64) - In a ConstantArray all elements have the same type, so it's sufficient to check the first constant recursively and then just compare if all following constants are the same by pointer compare llvm-svn: 239977
2015-06-18 07:55:17 +08:00
return Value.zextOrTrunc(8).getZExtValue();
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
}
if (const ConstantArray *CA = dyn_cast<ConstantArray>(V)) {
// Make sure all array elements are sequences of the same repeated
// byte.
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
assert(CA->getNumOperands() != 0 && "Should be a CAZ");
[AsmPrinter] Make isRepeatedByteSequence smarter about odd integer types - zext the value to alloc size first, then check if the value repeats with zero padding included. If so we can still emit a .space - Do the checking with APInt.isSplat(8), which handles non-pow2 types - Also handle large constants (bit width > 64) - In a ConstantArray all elements have the same type, so it's sufficient to check the first constant recursively and then just compare if all following constants are the same by pointer compare llvm-svn: 239977
2015-06-18 07:55:17 +08:00
Constant *Op0 = CA->getOperand(0);
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
int Byte = isRepeatedByteSequence(Op0, DL);
[AsmPrinter] Make isRepeatedByteSequence smarter about odd integer types - zext the value to alloc size first, then check if the value repeats with zero padding included. If so we can still emit a .space - Do the checking with APInt.isSplat(8), which handles non-pow2 types - Also handle large constants (bit width > 64) - In a ConstantArray all elements have the same type, so it's sufficient to check the first constant recursively and then just compare if all following constants are the same by pointer compare llvm-svn: 239977
2015-06-18 07:55:17 +08:00
if (Byte == -1)
return -1;
// All array elements must be equal.
for (unsigned i = 1, e = CA->getNumOperands(); i != e; ++i)
if (CA->getOperand(i) != Op0)
return -1;
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
return Byte;
}
Whitespace. llvm-svn: 168383
2012-11-21 04:34:44 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (const ConstantDataSequential *CDS = dyn_cast<ConstantDataSequential>(V))
return isRepeatedByteSequence(CDS);
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
return -1;
}
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantDataSequential(const DataLayout &DL,
const ConstantDataSequential *CDS,
AsmPrinter &AP) {
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
// See if we can aggregate this into a .fill, if so, emit it as such.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
int Value = isRepeatedByteSequence(CDS, DL);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (Value != -1) {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
uint64_t Bytes = DL.getTypeAllocSize(CDS->getType());
don't emit a 1-byte object as a .fill. This is silly and causes CodeGen/X86/global-sections.ll to fail with CDArray llvm-svn: 149343
2012-01-31 11:39:24 +08:00
// Don't emit a 1-byte object as a .fill.
if (Bytes > 1)
[MC] Rename EmitFill to emitFill This is to match the overloaded variants as well as the new style. Differential Revision: http://reviews.llvm.org/D20690 llvm-svn: 271359
2016-06-01 09:59:58 +08:00
return AP.OutStreamer->emitFill(Bytes, Value);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
}
Whitespace. llvm-svn: 168383
2012-11-21 04:34:44 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
// If this can be emitted with .ascii/.asciz, emit it as such.
if (CDS->isString())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
return AP.OutStreamer->EmitBytes(CDS->getAsString());
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
// Otherwise, emit the values in successive locations.
unsigned ElementByteSize = CDS->getElementByteSize();
if (isa<IntegerType>(CDS->getElementType())) {
Use the right method to get the # elements in a CDS. llvm-svn: 148897
2012-01-25 09:27:20 +08:00
for (unsigned i = 0, e = CDS->getNumElements(); i != e; ++i) {
when verbose asm is on, print integers in ConstantDataSequentials just like normal integers. llvm-svn: 149223
2012-01-30 13:55:11 +08:00
if (AP.isVerbose())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->GetCommentOS() << format("0x%" PRIx64 "\n",
CDS->getElementAsInteger(i));
AP.OutStreamer->EmitIntValue(CDS->getElementAsInteger(i),
ElementByteSize);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
}
don't lose tail padding on ConstantDataAggregate vec3's. llvm-svn: 149222
2012-01-30 13:49:43 +08:00
} else {
AsmPrinter: Use emitGlobalConstantFP to emit elements of constant data It's strange to duplicate the logic for emitting FP values into emitGlobalConstantDataSequential, and it's even stranger that we end up printing the verbose assembly comments differently between the two paths. Just call into emitGlobalConstantFP rather than crudely duplicating its logic. llvm-svn: 254988
2015-12-08 10:37:48 +08:00
for (unsigned I = 0, E = CDS->getNumElements(); I != E; ++I)
emitGlobalConstantFP(cast<ConstantFP>(CDS->getElementAsConstant(I)), AP);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
}
Rename DataLayout variables TD -> DL llvm-svn: 191927
2013-10-04 03:50:01 +08:00
unsigned Size = DL.getTypeAllocSize(CDS->getType());
unsigned EmittedSize = DL.getTypeAllocSize(CDS->getType()->getElementType()) *
don't lose tail padding on ConstantDataAggregate vec3's. llvm-svn: 149222
2012-01-30 13:49:43 +08:00
CDS->getNumElements();
if (unsigned Padding = Size - EmittedSize)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitZeros(Padding);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
}
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantArray(const DataLayout &DL,
const ConstantArray *CA, AsmPrinter &AP,
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const Constant *BaseCV, uint64_t Offset) {
reapply the patches reverted in r149470 that reenable ConstantDataArray, but with a critical fix to the SelectionDAG code that optimizes copies from strings into immediate stores: the previous code was stopping reading string data at the first nul. Address this by adding a new argument to llvm::getConstantStringInfo, preserving the behavior before the patch. llvm-svn: 149800
2012-02-05 10:29:43 +08:00
// See if we can aggregate some values. Make sure it can be
// represented as a series of bytes of the constant value.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
int Value = isRepeatedByteSequence(CA, DL);
Compress Repeated Byte Output Emit a repeated sequence of bytes using .zero. This saves an enormous amount of asm file space for certain programs. llvm-svn: 138864
2011-09-01 01:30:56 +08:00
reapply the patches reverted in r149470 that reenable ConstantDataArray, but with a critical fix to the SelectionDAG code that optimizes copies from strings into immediate stores: the previous code was stopping reading string data at the first nul. Address this by adding a new argument to llvm::getConstantStringInfo, preserving the behavior before the patch. llvm-svn: 149800
2012-02-05 10:29:43 +08:00
if (Value != -1) {
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
uint64_t Bytes = DL.getTypeAllocSize(CA->getType());
[MC] Rename EmitFill to emitFill This is to match the overloaded variants as well as the new style. Differential Revision: http://reviews.llvm.org/D20690 llvm-svn: 271359
2016-06-01 09:59:58 +08:00
AP.OutStreamer->emitFill(Bytes, Value);
reapply the patches reverted in r149470 that reenable ConstantDataArray, but with a critical fix to the SelectionDAG code that optimizes copies from strings into immediate stores: the previous code was stopping reading string data at the first nul. Address this by adding a new argument to llvm::getConstantStringInfo, preserving the behavior before the patch. llvm-svn: 149800
2012-02-05 10:29:43 +08:00
}
else {
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
for (unsigned i = 0, e = CA->getNumOperands(); i != e; ++i) {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
emitGlobalConstantImpl(DL, CA->getOperand(i), AP, BaseCV, Offset);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
Offset += DL.getTypeAllocSize(CA->getOperand(i)->getType());
}
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
}
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantVector(const DataLayout &DL,
const ConstantVector *CV, AsmPrinter &AP) {
eliminate some uses of AsmPrinter::EmitIntXXX llvm-svn: 93996
2010-01-20 15:41:15 +08:00
for (unsigned i = 0, e = CV->getType()->getNumElements(); i != e; ++i)
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
emitGlobalConstantImpl(DL, CV->getOperand(i), AP);
Emit trailing padding on constant vectors when TargetData says that the vector is larger than the sum of the elements (including per-element padding). llvm-svn: 133631
2011-06-23 02:55:03 +08:00
Rename DataLayout variables TD -> DL llvm-svn: 191927
2013-10-04 03:50:01 +08:00
unsigned Size = DL.getTypeAllocSize(CV->getType());
unsigned EmittedSize = DL.getTypeAllocSize(CV->getType()->getElementType()) *
Emit trailing padding on constant vectors when TargetData says that the vector is larger than the sum of the elements (including per-element padding). llvm-svn: 133631
2011-06-23 02:55:03 +08:00
CV->getType()->getNumElements();
if (unsigned Padding = Size - EmittedSize)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitZeros(Padding);
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantStruct(const DataLayout &DL,
const ConstantStruct *CS, AsmPrinter &AP,
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const Constant *BaseCV, uint64_t Offset) {
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
// Print the fields in successive locations. Pad to align if needed!
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
unsigned Size = DL.getTypeAllocSize(CS->getType());
const StructLayout *Layout = DL.getStructLayout(CS->getType());
refactor code to be static functions instead of methods on AsmPrinter. This fixes some bugs handling address spaces. llvm-svn: 93891
2010-01-20 03:10:44 +08:00
uint64_t SizeSoFar = 0;
for (unsigned i = 0, e = CS->getNumOperands(); i != e; ++i) {
eliminate some uses of AsmPrinter::EmitIntXXX llvm-svn: 93996
2010-01-20 15:41:15 +08:00
const Constant *Field = CS->getOperand(i);
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Print the actual field value.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
emitGlobalConstantImpl(DL, Field, AP, BaseCV, Offset + SizeSoFar);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
// Check if padding is needed and insert one or more 0s.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
uint64_t FieldSize = DL.getTypeAllocSize(Field->getType());
signficant cleanups to EmitGlobalConstant (including streamerization of int initializers), change some methods to be static functions, use raw_ostream::write_hex instead of a smallstring dance with APValue::toStringUnsigned(S, 16). llvm-svn: 93991
2010-01-20 15:11:32 +08:00
uint64_t PadSize = ((i == e-1 ? Size : Layout->getElementOffset(i+1))
- Layout->getElementOffset(i)) - FieldSize;
SizeSoFar += FieldSize + PadSize;
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
// Insert padding - this may include padding to increase the size of the
// current field up to the ABI size (if the struct is not packed) as well
// as padding to ensure that the next field starts at the right offset.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitZeros(PadSize);
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
signficant cleanups to EmitGlobalConstant (including streamerization of int initializers), change some methods to be static functions, use raw_ostream::write_hex instead of a smallstring dance with APValue::toStringUnsigned(S, 16). llvm-svn: 93991
2010-01-20 15:11:32 +08:00
assert(SizeSoFar == Layout->getSizeInBytes() &&
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
"Layout of constant struct may be incorrect!");
}
Remove address spaces from MC. This is dead code since PIC16 was removed in 2010. The result was an odd mix, where some parts would carefully pass it along and others would assert it was zero (most of the object streamer for example). llvm-svn: 185436
2013-07-02 23:49:13 +08:00
static void emitGlobalConstantFP(const ConstantFP *CFP, AsmPrinter &AP) {
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
APInt API = CFP->getValueAPF().bitcastToAPInt();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
// First print a comment with what we think the original floating-point value
// should have been.
if (AP.isVerbose()) {
SmallString<8> StrVal;
CFP->getValueAPF().toString(StrVal);
Add basic generic CodeGen support for half. llvm-svn: 146927
2011-12-20 08:02:33 +08:00
Add back functionality removed in r210497. Instead of asserting, output a message stating that a null pointer was found. llvm-svn: 211430
2014-06-21 10:43:02 +08:00
if (CFP->getType())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
CFP->getType()->print(AP.OutStreamer->GetCommentOS());
Add back functionality removed in r210497. Instead of asserting, output a message stating that a null pointer was found. llvm-svn: 211430
2014-06-21 10:43:02 +08:00
else
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->GetCommentOS() << "Printing <null> Type";
AP.OutStreamer->GetCommentOS() << ' ' << StrVal << '\n';
Add basic generic CodeGen support for half. llvm-svn: 146927
2011-12-20 08:02:33 +08:00
}
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
// Now iterate through the APInt chunks, emitting them in endian-correct
// order, possibly with a smaller chunk at beginning/end (e.g. for x87 80-bit
// floats).
unsigned NumBytes = API.getBitWidth() / 8;
unsigned TrailingBytes = NumBytes % sizeof(uint64_t);
const uint64_t *p = API.getRawData();
Allow the asm printer to print fp128 values properly. llvm-svn: 171866
2013-01-09 00:56:23 +08:00
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
// PPC's long double has odd notions of endianness compared to how LLVM
// handles it: p[0] goes first for *big* endian on PPC.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
if (AP.getDataLayout().isBigEndian() && !CFP->getType()->isPPC_FP128Ty()) {
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
int Chunk = API.getNumWords() - 1;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
if (TrailingBytes)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(p[Chunk--], TrailingBytes);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
for (; Chunk >= 0; --Chunk)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(p[Chunk], sizeof(uint64_t));
switch ConstantFP emission to use MCStreamer, significantly simplifying the code. llvm-svn: 93988
2010-01-20 14:53:37 +08:00
} else {
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
unsigned Chunk;
for (Chunk = 0; Chunk < NumBytes / sizeof(uint64_t); ++Chunk)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(p[Chunk], sizeof(uint64_t));
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
if (TrailingBytes)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(p[Chunk], TrailingBytes);
simplify the rest of fp constant printing. llvm-svn: 93929
2010-01-20 06:16:33 +08:00
}
Simplify writing floating types to assembly. This removes previous special cases for each floating-point type in favour of a shared codepath. llvm-svn: 172189
2013-01-11 18:36:13 +08:00
// Emit the tail padding for the long double.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = AP.getDataLayout();
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitZeros(DL.getTypeAllocSize(CFP->getType()) -
DL.getTypeStoreSize(CFP->getType()));
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
Remove address spaces from MC. This is dead code since PIC16 was removed in 2010. The result was an odd mix, where some parts would carefully pass it along and others would assert it was zero (most of the object streamer for example). llvm-svn: 185436
2013-07-02 23:49:13 +08:00
static void emitGlobalConstantLargeInt(const ConstantInt *CI, AsmPrinter &AP) {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = AP.getDataLayout();
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
unsigned BitWidth = CI->getBitWidth();
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
// Copy the value as we may massage the layout for constants whose bit width
// is not a multiple of 64-bits.
APInt Realigned(CI->getValue());
uint64_t ExtraBits = 0;
unsigned ExtraBitsSize = BitWidth & 63;
if (ExtraBitsSize) {
// The bit width of the data is not a multiple of 64-bits.
// The extra bits are expected to be at the end of the chunk of the memory.
// Little endian:
// * Nothing to be done, just record the extra bits to emit.
// Big endian:
// * Record the extra bits to emit.
// * Realign the raw data to emit the chunks of 64-bits.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
if (DL.isBigEndian()) {
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
// Basically the structure of the raw data is a chunk of 64-bits cells:
// 0 1 BitWidth / 64
// [chunk1][chunk2] ... [chunkN].
// The most significant chunk is chunkN and it should be emitted first.
// However, due to the alignment issue chunkN contains useless bits.
// Realign the chunks so that they contain only useless information:
// ExtraBits 0 1 (BitWidth / 64) - 1
// chu[nk1 chu][nk2 chu] ... [nkN-1 chunkN]
ExtraBits = Realigned.getRawData()[0] &
(((uint64_t)-1) >> (64 - ExtraBitsSize));
[APInt] Use lshrInPlace to replace lshr where possible This patch uses lshrInPlace to replace code where the object that lshr is called on is being overwritten with the result. This adds an lshrInPlace(const APInt &) version as well. Differential Revision: https://reviews.llvm.org/D32155 llvm-svn: 300566
2017-04-19 01:14:21 +08:00
Realigned.lshrInPlace(ExtraBitsSize);
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
} else
ExtraBits = Realigned.getRawData()[BitWidth / 64];
}
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
// We don't expect assemblers to support integer data directives
// for more than 64 bits, so we emit the data in at most 64-bit
// quantities at a time.
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
const uint64_t *RawData = Realigned.getRawData();
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
for (unsigned i = 0, e = BitWidth / 64; i != e; ++i) {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
uint64_t Val = DL.isBigEndian() ? RawData[e - i - 1] : RawData[i];
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(Val, 8);
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
if (ExtraBitsSize) {
// Emit the extra bits after the 64-bits chunks.
// Emit a directive that fills the expected size.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
uint64_t Size = AP.getDataLayout().getTypeAllocSize(CI->getType());
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
Size -= (BitWidth / 64) * 8;
assert(Size && Size * 8 >= ExtraBitsSize &&
(ExtraBits & (((uint64_t)-1) >> (64 - ExtraBitsSize)))
== ExtraBits && "Directive too small for extra bits.");
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(ExtraBits, Size);
Teach AsmPrinter how to print odd constants. Fix an assertion when the compiler encounters big constants whose bit width is not a multiple of 64-bits. Although clang would never generate something like this, the backend should be able to handle any legal IR. <rdar://problem/13363576> llvm-svn: 183544
2013-06-08 02:36:03 +08:00
}
Refactor a bunch of code out of AsmPrinter::EmitGlobalConstant into separate functions. llvm-svn: 61345
2008-12-23 05:14:27 +08:00
}
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
/// \brief Transform a not absolute MCExpr containing a reference to a GOT
/// equivalent global, by a target specific GOT pc relative access to the
/// final symbol.
static void handleIndirectSymViaGOTPCRel(AsmPrinter &AP, const MCExpr **ME,
const Constant *BaseCst,
uint64_t Offset) {
// The global @foo below illustrates a global that uses a got equivalent.
//
// @bar = global i32 42
// @gotequiv = private unnamed_addr constant i32* @bar
// @foo = i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequiv to i64),
// i64 ptrtoint (i32* @foo to i64))
// to i32)
//
// The cstexpr in @foo is converted into the MCExpr `ME`, where we actually
// check whether @foo is suitable to use a GOTPCREL. `ME` is usually in the
// form:
//
// foo = cstexpr, where
// cstexpr := <gotequiv> - "." + <cst>
// cstexpr := <gotequiv> - (<foo> - <offset from @foo base>) + <cst>
//
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
// After canonicalization by evaluateAsRelocatable `ME` turns into:
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
//
// cstexpr := <gotequiv> - <foo> + gotpcrelcst, where
// gotpcrelcst := <offset from @foo base> + <cst>
//
MCValue MV;
MC: Clean up MCExpr naming. NFC. llvm-svn: 238634
2015-05-30 09:25:56 +08:00
if (!(*ME)->evaluateAsRelocatable(MV, nullptr, nullptr) || MV.isAbsolute())
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
return;
[AsmPrinter] Fix crash in handleIndirectSymViaGOTPCRel Check for symbols in MCValue before using them. Bail out early in case they are null. This fixes PR23779. Differential Revision: http://reviews.llvm.org/D10712 rdar://problem/21532830 llvm-svn: 240649
2015-06-25 23:17:23 +08:00
const MCSymbolRefExpr *SymA = MV.getSymA();
if (!SymA)
return;
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
[AsmPrinter] Fix crash in handleIndirectSymViaGOTPCRel Check for symbols in MCValue before using them. Bail out early in case they are null. This fixes PR23779. Differential Revision: http://reviews.llvm.org/D10712 rdar://problem/21532830 llvm-svn: 240649
2015-06-25 23:17:23 +08:00
// Check that GOT equivalent symbol is cached.
const MCSymbol *GOTEquivSym = &SymA->getSymbol();
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
if (!AP.GlobalGOTEquivs.count(GOTEquivSym))
return;
[AsmPrinter] Check for valid constants in handleIndirectSymViaGOTPCRel Check whether BaseCst is valid before extracting a GlobalValue. This fixes PR24163. Patch by David Majnemer. llvm-svn: 242840
2015-07-22 05:45:42 +08:00
const GlobalValue *BaseGV = dyn_cast_or_null<GlobalValue>(BaseCst);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
if (!BaseGV)
return;
[AsmPrinter] Fix crash in handleIndirectSymViaGOTPCRel Check for symbols in MCValue before using them. Bail out early in case they are null. This fixes PR23779. Differential Revision: http://reviews.llvm.org/D10712 rdar://problem/21532830 llvm-svn: 240649
2015-06-25 23:17:23 +08:00
// Check for a valid base symbol
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const MCSymbol *BaseSym = AP.getSymbol(BaseGV);
[AsmPrinter] Fix crash in handleIndirectSymViaGOTPCRel Check for symbols in MCValue before using them. Bail out early in case they are null. This fixes PR23779. Differential Revision: http://reviews.llvm.org/D10712 rdar://problem/21532830 llvm-svn: 240649
2015-06-25 23:17:23 +08:00
const MCSymbolRefExpr *SymB = MV.getSymB();
if (!SymB || BaseSym != &SymB->getSymbol())
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
return;
// Make sure to match:
//
// gotpcrelcst := <offset from @foo base> + <cst>
//
[AsmPrinter][TLOF] 32-bit MachO support for replacing GOT equivalents Add MachO 32-bit (i.e. arm and x86) support for replacing global GOT equivalent symbol accesses. Unlike 64-bit targets, there's no GOTPCREL relocation, and access through a non_lazy_symbol_pointers section is used instead. -- before _extgotequiv: .long _extfoo _delta: .long _extgotequiv-_delta -- after _delta: .long L_extfoo$non_lazy_ptr-_delta .section __IMPORT,__pointers,non_lazy_symbol_pointers L_extfoo$non_lazy_ptr: .indirect_symbol _extfoo .long 0 llvm-svn: 231475
2015-03-06 21:49:05 +08:00
// If gotpcrelcst is positive it means that we can safely fold the pc rel
// displacement into the GOTPCREL. We can also can have an extra offset <cst>
// if the target knows how to encode it.
[AsmPrinter][TLOF] ARM64 MachO support for replacing GOT equivalents Follow up r230264 and add ARM64 support for replacing global GOT equivalent symbol accesses by references to the GOT entry for the final symbol instead, example: -- before .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta -- after .globl _foo _foo: .long 42 .globl _delta Ltmp3: .long _foo@GOT-Ltmp3 llvm-svn: 231474
2015-03-06 21:48:45 +08:00
//
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
int64_t GOTPCRelCst = Offset + MV.getConstant();
if (GOTPCRelCst < 0)
return;
[AsmPrinter][TLOF] ARM64 MachO support for replacing GOT equivalents Follow up r230264 and add ARM64 support for replacing global GOT equivalent symbol accesses by references to the GOT entry for the final symbol instead, example: -- before .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta -- after .globl _foo _foo: .long 42 .globl _delta Ltmp3: .long _foo@GOT-Ltmp3 llvm-svn: 231474
2015-03-06 21:48:45 +08:00
if (!AP.getObjFileLowering().supportGOTPCRelWithOffset() && GOTPCRelCst != 0)
return;
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Emit the GOT PC relative to replace the got equivalent global, i.e.:
//
// bar:
// .long 42
// gotequiv:
// .quad bar
// foo:
// .long gotequiv - "." + <cst>
//
// is replaced by the target specific equivalent to:
//
// bar:
// .long 42
// foo:
// .long bar@GOTPCREL+<gotpcrelcst>
//
AsmPrinter::GOTEquivUsePair Result = AP.GlobalGOTEquivs[GOTEquivSym];
const GlobalVariable *GV = Result.first;
[AsmPrinter][TLOF] Reintroduce AArch64 test Follow up from r231505. Fix the non-determinism by using a MapVector and reintroduce the AArch64 testcase. Defer deleting the got candidates up to the end and remove them in a bulk, avoiding linear time removal of each element. Thanks to Renato Golin for trying it out on other platforms. llvm-svn: 231830
2015-03-11 04:05:23 +08:00
int NumUses = (int)Result.second;
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const GlobalValue *FinalGV = dyn_cast<GlobalValue>(GV->getOperand(0));
const MCSymbol *FinalSym = AP.getSymbol(FinalGV);
[AsmPrinter][TLOF] 32-bit MachO support for replacing GOT equivalents Add MachO 32-bit (i.e. arm and x86) support for replacing global GOT equivalent symbol accesses. Unlike 64-bit targets, there's no GOTPCREL relocation, and access through a non_lazy_symbol_pointers section is used instead. -- before _extgotequiv: .long _extfoo _delta: .long _extgotequiv-_delta -- after _delta: .long L_extfoo$non_lazy_ptr-_delta .section __IMPORT,__pointers,non_lazy_symbol_pointers L_extfoo$non_lazy_ptr: .indirect_symbol _extfoo .long 0 llvm-svn: 231475
2015-03-06 21:49:05 +08:00
*ME = AP.getObjFileLowering().getIndirectSymViaGOTPCRel(
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
FinalSym, MV, Offset, AP.MMI, *AP.OutStreamer);
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Update GOT equivalent usage information
--NumUses;
[AsmPrinter][TLOF] Reintroduce AArch64 test Follow up from r231505. Fix the non-determinism by using a MapVector and reintroduce the AArch64 testcase. Defer deleting the got candidates up to the end and remove them in a bulk, avoiding linear time removal of each element. Thanks to Renato Golin for trying it out on other platforms. llvm-svn: 231830
2015-03-11 04:05:23 +08:00
if (NumUses >= 0)
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
AP.GlobalGOTEquivs[GOTEquivSym] = std::make_pair(GV, NumUses);
}
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
static void emitGlobalConstantImpl(const DataLayout &DL, const Constant *CV,
AsmPrinter &AP, const Constant *BaseCV,
uint64_t Offset) {
uint64_t Size = DL.getTypeAllocSize(CV->getType());
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
// Globals with sub-elements such as combinations of arrays and structs
// are handled recursively by emitGlobalConstantImpl. Keep track of the
// constant symbol base and the current position with BaseCV and Offset.
if (!BaseCV && CV->hasOneUse())
BaseCV = dyn_cast<Constant>(CV->user_back());
It's possible to have a constant expression who's size is quite big (e.g., i128). In that case, we may not be able to print out the MCExpr as an expression. For instance, we could have an MCExpr like this: 0xBEEF0000BEEF0000 | (0xBEEF0000BEEF0000 << 64) The MCExpr printer handles sizes up to 64-bits, but this expression would require 128-bits. In this situation, try to evaluate the constant expression and emit that as the value into 64-bit chunks. <rdar://problem/11070338> llvm-svn: 153081
2012-03-20 16:56:43 +08:00
if (isa<ConstantAggregateZero>(CV) || isa<UndefValue>(CV))
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
return AP.OutStreamer->EmitZeros(Size);
signficant cleanups to EmitGlobalConstant (including streamerization of int initializers), change some methods to be static functions, use raw_ostream::write_hex instead of a smallstring dance with APValue::toStringUnsigned(S, 16). llvm-svn: 93991
2010-01-20 15:11:32 +08:00
if (const ConstantInt *CI = dyn_cast<ConstantInt>(CV)) {
switch (Size) {
case 1:
case 2:
case 4:
case 8:
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
if (AP.isVerbose())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->GetCommentOS() << format("0x%" PRIx64 "\n",
CI->getZExtValue());
AP.OutStreamer->EmitIntValue(CI->getZExtValue(), Size);
signficant cleanups to EmitGlobalConstant (including streamerization of int initializers), change some methods to be static functions, use raw_ostream::write_hex instead of a smallstring dance with APValue::toStringUnsigned(S, 16). llvm-svn: 93991
2010-01-20 15:11:32 +08:00
return;
default:
Remove address spaces from MC. This is dead code since PIC16 was removed in 2010. The result was an odd mix, where some parts would carefully pass it along and others would assert it was zero (most of the object streamer for example). llvm-svn: 185436
2013-07-02 23:49:13 +08:00
emitGlobalConstantLargeInt(CI, AP);
signficant cleanups to EmitGlobalConstant (including streamerization of int initializers), change some methods to be static functions, use raw_ostream::write_hex instead of a smallstring dance with APValue::toStringUnsigned(S, 16). llvm-svn: 93991
2010-01-20 15:11:32 +08:00
return;
}
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
refactor code to be static functions instead of methods on AsmPrinter. This fixes some bugs handling address spaces. llvm-svn: 93891
2010-01-20 03:10:44 +08:00
if (const ConstantFP *CFP = dyn_cast<ConstantFP>(CV))
Remove address spaces from MC. This is dead code since PIC16 was removed in 2010. The result was an odd mix, where some parts would carefully pass it along and others would assert it was zero (most of the object streamer for example). llvm-svn: 185436
2013-07-02 23:49:13 +08:00
return emitGlobalConstantFP(CFP, AP);
Implement emitGlobalConstant llvm-svn: 15868
2004-08-17 14:36:49 +08:00
don't send null pointers through the constantexpr codepath unneededly. llvm-svn: 94010
2010-01-21 01:57:50 +08:00
if (isa<ConstantPointerNull>(CV)) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitIntValue(0, Size);
don't send null pointers through the constantexpr codepath unneededly. llvm-svn: 94010
2010-01-21 01:57:50 +08:00
return;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (const ConstantDataSequential *CDS = dyn_cast<ConstantDataSequential>(CV))
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantDataSequential(DL, CDS, AP);
Whitespace. llvm-svn: 168383
2012-11-21 04:34:44 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (const ConstantArray *CVA = dyn_cast<ConstantArray>(CV))
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantArray(DL, CVA, AP, BaseCV, Offset);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (const ConstantStruct *CVS = dyn_cast<ConstantStruct>(CV))
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantStruct(DL, CVS, AP, BaseCV, Offset);
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
It's possible to have a constant expression who's size is quite big (e.g., i128). In that case, we may not be able to print out the MCExpr as an expression. For instance, we could have an MCExpr like this: 0xBEEF0000BEEF0000 | (0xBEEF0000BEEF0000 << 64) The MCExpr printer handles sizes up to 64-bits, but this expression would require 128-bits. In this situation, try to evaluate the constant expression and emit that as the value into 64-bit chunks. <rdar://problem/11070338> llvm-svn: 153081
2012-03-20 16:56:43 +08:00
if (const ConstantExpr *CE = dyn_cast<ConstantExpr>(CV)) {
// Look through bitcasts, which might not be able to be MCExpr'ized (e.g. of
// vectors).
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (CE->getOpcode() == Instruction::BitCast)
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantImpl(DL, CE->getOperand(0), AP);
It's possible to have a constant expression who's size is quite big (e.g., i128). In that case, we may not be able to print out the MCExpr as an expression. For instance, we could have an MCExpr like this: 0xBEEF0000BEEF0000 | (0xBEEF0000BEEF0000 << 64) The MCExpr printer handles sizes up to 64-bits, but this expression would require 128-bits. In this situation, try to evaluate the constant expression and emit that as the value into 64-bit chunks. <rdar://problem/11070338> llvm-svn: 153081
2012-03-20 16:56:43 +08:00
if (Size > 8) {
// If the constant expression's size is greater than 64-bits, then we have
// to emit the value in chunks. Try to constant fold the value and emit it
// that way.
[ConstnatFolding] Teach the folder how to fold ConstantVector A ConstantVector can have ConstantExpr operands and vice versa. However, the folder had no ability to fold ConstantVectors which, in some cases, was an optimization barrier. Instead, rephrase the folder in terms of Constants instead of ConstantExprs and teach callers how to deal with failure. llvm-svn: 277099
2016-07-29 11:27:26 +08:00
Constant *New = ConstantFoldConstant(CE, DL);
It's possible to have a constant expression who's size is quite big (e.g., i128). In that case, we may not be able to print out the MCExpr as an expression. For instance, we could have an MCExpr like this: 0xBEEF0000BEEF0000 | (0xBEEF0000BEEF0000 << 64) The MCExpr printer handles sizes up to 64-bits, but this expression would require 128-bits. In this situation, try to evaluate the constant expression and emit that as the value into 64-bit chunks. <rdar://problem/11070338> llvm-svn: 153081
2012-03-20 16:56:43 +08:00
if (New && New != CE)
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantImpl(DL, New, AP);
It's possible to have a constant expression who's size is quite big (e.g., i128). In that case, we may not be able to print out the MCExpr as an expression. For instance, we could have an MCExpr like this: 0xBEEF0000BEEF0000 | (0xBEEF0000BEEF0000 << 64) The MCExpr printer handles sizes up to 64-bits, but this expression would require 128-bits. In this situation, try to evaluate the constant expression and emit that as the value into 64-bit chunks. <rdar://problem/11070338> llvm-svn: 153081
2012-03-20 16:56:43 +08:00
}
}
Whitespace. llvm-svn: 168383
2012-11-21 04:34:44 +08:00
Add AsmPrinter (aka MCLowering) support for ConstantDataSequential, and clean up some other misc stuff. Unlike ConstantArray, we will prefer to emit .fill directives for "String" arrays that all have the same value, since they are denser than emitting a .ascii llvm-svn: 148793
2012-01-24 17:31:43 +08:00
if (const ConstantVector *V = dyn_cast<ConstantVector>(CV))
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
return emitGlobalConstantVector(DL, V, AP);
Whitespace. llvm-svn: 168383
2012-11-21 04:34:44 +08:00
Change constantexpr global variable initializers to convert the constants to MCExpr then emit them through MCStreamer with EmitValue. I think all global variable initializers are now going through mcstreamer. llvm-svn: 94293
2010-01-23 14:17:14 +08:00
// Otherwise, it must be a ConstantExpr. Lower it to an MCExpr, then emit it
// thread the streamer with EmitValue.
[AsmPrinter] Access pointers to globals via pcrel GOT entries Front-ends could use global unnamed_addr to hold pointers to other symbols, like @gotequivalent below: @foo = global i32 42 @gotequivalent = private unnamed_addr constant i32* @foo @delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64), i64 ptrtoint (i32* @delta to i64)) to i32) The global @delta holds a data "PC"-relative offset to @gotequivalent, an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows: .globl _foo _foo: .long 42 .globl _gotequivalent _gotequivalent: .quad _foo .globl _delta _delta: .long _gotequivalent-_delta Since unnamed_addr indicates that the address is not significant, only the content, we can optimize the case above by replacing pc-relative accesses to "GOT equivalent" globals, by a PC relative access to the GOT entry of the final symbol instead. Therefore, "delta" can contain a pc relative relocation to foo's GOT entry and we avoid the emission of "gotequivalent", yielding the assembly code below: .globl _foo _foo: .long 42 .globl _delta _delta: .long _foo@GOTPCREL+4 There are a couple of advantages of doing this: (1) Front-ends that need to emit a great deal of data to store pointers to external symbols could save space by not emitting such "got equivalent" globals and (2) IR constructs combined with this opt opens a way to represent GOT pcrel relocations by using the LLVM IR, which is something we previously had no way to express. Differential Revision: http://reviews.llvm.org/D6922 rdar://problem/18534217 llvm-svn: 230264
2015-02-24 05:26:18 +08:00
const MCExpr *ME = AP.lowerConstant(CV);
// Since lowerConstant already folded and got rid of all IR pointer and
// integer casts, detect GOT equivalent accesses by looking into the MCExpr
// directly.
if (AP.getObjFileLowering().supportIndirectSymViaGOTPCRel())
handleIndirectSymViaGOTPCRel(AP, &ME, BaseCV, Offset);
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->EmitValue(ME, Size);
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
}
/// EmitGlobalConstant - Print a general LLVM constant to the .s file.
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
void AsmPrinter::EmitGlobalConstant(const DataLayout &DL, const Constant *CV) {
uint64_t Size = DL.getTypeAllocSize(CV->getType());
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
if (Size)
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
emitGlobalConstantImpl(DL, CV, *this);
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
else if (MAI->hasSubsectionsViaSymbols()) {
// If the global has zero size, emit a single byte so that two labels don't
// look like they are at the same location.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitIntValue(0, 1);
Bill's change in r95336 broke empty aggregates embedded in other types. fix this by only bumping zero-byte globals up to a single byte if the *entire global* is zero size, fixing PR6340. This also fixes empty arrays etc to be handled correctly, and only does this on subsection-via-symbols targets (aka darwin) which is the only place where this matters. llvm-svn: 101879
2010-04-20 14:20:21 +08:00
}
Implement emitGlobalConstant llvm-svn: 15868
2004-08-17 14:36:49 +08:00
}
Stub out a method llvm-svn: 25676
2006-01-27 10:10:10 +08:00
Rework the routines that convert AP[S]Int into a string. Now, instead of returning an std::string by value, it fills in a SmallString/SmallVector passed in. This significantly reduces string thrashing in some cases. More specifically, this: - Adds an operator<< and a print method for APInt that allows you to directly send them to an ostream. - Reimplements APInt::toString to be much simpler and more efficient algorithmically in addition to not thrashing strings quite as much. This speeds up llvm-dis on kc++ by 7%, and may also slightly speed up the asmprinter. This also fixes a bug I introduced into the asmwriter in a previous patch w.r.t. alias printing. llvm-svn: 54873
2008-08-17 15:19:36 +08:00
void AsmPrinter::EmitMachineConstantPoolValue(MachineConstantPoolValue *MCPV) {
Added support for machine specific constantpool values. These are useful for representing expressions that can only be resolved at link time, etc. llvm-svn: 30278
2006-09-13 05:00:35 +08:00
// Target doesn't support this yet!
llvm_unreachable->llvm_unreachable(0), LLVM_UNREACHABLE->llvm_unreachable. This adds location info for all llvm_unreachable calls (which is a macro now) in !NDEBUG builds. In NDEBUG builds location info and the message is off (it only prints "UREACHABLE executed"). llvm-svn: 75640
2009-07-15 00:55:14 +08:00
llvm_unreachable("Target does not support EmitMachineConstantPoolValue");
Added support for machine specific constantpool values. These are useful for representing expressions that can only be resolved at link time, etc. llvm-svn: 30278
2006-09-13 05:00:35 +08:00
}
minor tidying. llvm-svn: 100338
2010-04-05 02:16:38 +08:00
void AsmPrinter::printOffset(int64_t Offset, raw_ostream &OS) const {
if (Offset > 0)
OS << '+' << Offset;
else if (Offset < 0)
OS << Offset;
}
move gettemplabel and getdwlabel to AsmPrinter and rename them for consistency. llvm-svn: 100345
2010-04-05 03:25:43 +08:00
//===----------------------------------------------------------------------===//
// Symbol Lowering Routines.
//===----------------------------------------------------------------------===//
Centralize the handling of unique ids for temporary labels. Before this patch code wanting to create temporary labels for a given entity (function, cu, exception range, etc) had to keep its own counter to have stable symbol names. createTempSymbol would still add a suffix to make sure a new symbol was always returned, but it kept a single counter. Because of that, if we were to use just createTempSymbol("cu_begin"), the label could change from cu_begin42 to cu_begin43 because some other code started using temporary labels. Simplify this by just keeping one counter per prefix and removing the various specialized counters. llvm-svn: 232535
2015-03-18 04:07:06 +08:00
MCSymbol *AsmPrinter::createTempSymbol(const Twine &Name) const {
return OutContext.createTempSymbol(Name, true);
Centralize handling of the eh_begin and eh_end labels. This removes a bit of duplicated code and more importantly, remembers the labels so that they don't need to be looked up by name. This in turn allows for any name to be used and avoids a crash if the name we wanted was already taken. llvm-svn: 230772
2015-02-28 02:18:39 +08:00
}
Implement smart printing of inline asm strings, handling variants and substituted operands. For this testcase: int %test(int %A, int %B) { %C = call int asm "xyz $0, $1, $2", "=r,r,r"(int %A, int %B) ret int %C } we now emit: _test: or r2, r3, r3 or r3, r4, r4 xyz r2, r2, r3 ;; look here or r3, r2, r2 blr ... note the substituted operands. :) llvm-svn: 25886
2006-02-02 06:41:11 +08:00
now that @GOTOFF is no longer represented as a suffix on a MCSymbol, we can remove the 'suffix' argument of GetBlockAddressSymbol. Do so. llvm-svn: 95601
2010-02-09 07:10:08 +08:00
MCSymbol *AsmPrinter::GetBlockAddressSymbol(const BlockAddress *BA) const {
fix AsmPrinter::GetBlockAddressSymbol to always return a unique label instead of trying to form one based on the BB name (which causes collisions if the name is empty). This fixes PR6608 llvm-svn: 98495
2010-03-15 01:53:23 +08:00
return MMI->getAddrLabelSymbol(BA->getBasicBlock());
Initial target-independent CodeGen support for BlockAddresses. llvm-svn: 85556
2009-10-30 09:27:03 +08:00
}
fix AsmPrinter::GetBlockAddressSymbol to always return a unique label instead of trying to form one based on the BB name (which causes collisions if the name is empty). This fixes PR6608 llvm-svn: 98495
2010-03-15 01:53:23 +08:00
MCSymbol *AsmPrinter::GetBlockAddressSymbol(const BasicBlock *BB) const {
return MMI->getAddrLabelSymbol(BB);
Initial target-independent CodeGen support for BlockAddresses. llvm-svn: 85556
2009-10-30 09:27:03 +08:00
}
mcize jump table and constant pool entry labels, .local on elf, and some .weak directives. llvm-svn: 94284
2010-01-23 13:19:23 +08:00
/// GetCPISymbol - Return the symbol for the specified constant pool entry.
MCSymbol *AsmPrinter::GetCPISymbol(unsigned CPID) const {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = getDataLayout();
return OutContext.getOrCreateSymbol(Twine(DL.getPrivateGlobalPrefix()) +
"CPI" + Twine(getFunctionNumber()) + "_" +
Twine(CPID));
mcize jump table and constant pool entry labels, .local on elf, and some .weak directives. llvm-svn: 94284
2010-01-23 13:19:23 +08:00
}
/// GetJTISymbol - Return the symbol for the specified jump table entry.
Revert r107205 and r107207. llvm-svn: 107215
2010-06-30 06:34:52 +08:00
MCSymbol *AsmPrinter::GetJTISymbol(unsigned JTID, bool isLinkerPrivate) const {
return MF->getJTISymbol(JTID, OutContext, isLinkerPrivate);
devirtualize AsmPrinter::printBasicBlockLabel since it is never overridden. Move GetMBBSymbol up to AsmPrinter and make printBasicBlockLabel use it so that we only have one place that decides what to name bb labels. Hopefully various clients of printBasicBlockLabel can start using GetMBBSymbol instead. llvm-svn: 81652
2009-09-13 07:02:08 +08:00
}
handle the _set_ symbol with an MCSymbol. llvm-svn: 94449
2010-01-26 05:17:10 +08:00
/// GetJTSetSymbol - Return the symbol for the specified jump table .set
/// FIXME: privatize to AsmPrinter.
MCSymbol *AsmPrinter::GetJTSetSymbol(unsigned UID, unsigned MBBID) const {
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
const DataLayout &DL = getDataLayout();
return OutContext.getOrCreateSymbol(Twine(DL.getPrivateGlobalPrefix()) +
Twine(getFunctionNumber()) + "_" +
Twine(UID) + "_set_" + Twine(MBBID));
handle the _set_ symbol with an MCSymbol. llvm-svn: 94449
2010-01-26 05:17:10 +08:00
}
Move getSymbolWithGlobalValueBase to TargetLoweringObjectFile. This allows it to be used in TargetLoweringObjectFileImpl.cpp. llvm-svn: 196117
2013-12-03 00:25:47 +08:00
MCSymbol *AsmPrinter::getSymbolWithGlobalValueBase(const GlobalValue *GV,
Remove an always true parameter. llvm-svn: 195931
2013-11-29 03:35:07 +08:00
StringRef Suffix) const {
Move the Mangler from the AsmPrinter down to TLOF and clean up the TLOF API accordingly. llvm-svn: 281708
2016-09-16 15:33:15 +08:00
return getObjFileLowering().getSymbolWithGlobalValueBase(GV, Suffix, TM);
add another helper llvm-svn: 93577
2010-01-16 07:25:11 +08:00
}
Simplify the Mangler interface now that DataLayout is mandatory. We only need to pass in a DataLayout when mangling a raw string, not when constructing the mangler. llvm-svn: 240405
2015-06-23 21:59:29 +08:00
/// Return the MCSymbol for the specified ExternalSymbol.
add a AsmPrinter::GetGlobalValueSymbol and GetExternalSymbolSymbol helper method, use it to simplify some code. llvm-svn: 93575
2010-01-16 07:18:17 +08:00
MCSymbol *AsmPrinter::GetExternalSymbolSymbol(StringRef Sym) const {
SmallString<60> NameStr;
Move most user of TargetMachine::getDataLayout to the Module one Summary: This change is part of a series of commits dedicated to have a single DataLayout during compilation by using always the one owned by the module. This patch is quite boring overall, except for some uglyness in ASMPrinter which has a getDataLayout function but has some clients that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so some methods are taking a DataLayout as parameter. Reviewers: echristo Subscribers: yaron.keren, rafael, llvm-commits, jholewinski Differential Revision: http://reviews.llvm.org/D11090 From: Mehdi Amini <mehdi.amini@apple.com> llvm-svn: 242386
2015-07-16 14:11:10 +08:00
Mangler::getNameWithPrefix(NameStr, Sym, getDataLayout());
MC: Clean up method names in MCContext. The naming was a mish-mash of old and new style. Update to be consistent with the new. NFC. llvm-svn: 237594
2015-05-19 02:43:14 +08:00
return OutContext.getOrCreateSymbol(NameStr);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
}
add a AsmPrinter::GetGlobalValueSymbol and GetExternalSymbolSymbol helper method, use it to simplify some code. llvm-svn: 93575
2010-01-16 07:18:17 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
/// PrintParentLoopComment - Print comments about parent loops of this one.
static void PrintParentLoopComment(raw_ostream &OS, const MachineLoop *Loop,
unsigned FunctionNumber) {
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (!Loop) return;
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
PrintParentLoopComment(OS, Loop->getParentLoop(), FunctionNumber);
OS.indent(Loop->getLoopDepth()*2)
<< "Parent Loop BB" << FunctionNumber << "_"
<< Loop->getHeader()->getNumber()
<< " Depth=" << Loop->getLoopDepth() << '\n';
}
/// PrintChildLoopComment - Print comments about child loops within
/// the loop for this basic block, with nesting.
static void PrintChildLoopComment(raw_ostream &OS, const MachineLoop *Loop,
unsigned FunctionNumber) {
// Add child loop information
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const MachineLoop *CL : *Loop) {
OS.indent(CL->getLoopDepth()*2)
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
<< "Child Loop BB" << FunctionNumber << "_"
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
<< CL->getHeader()->getNumber() << " Depth " << CL->getLoopDepth()
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
<< '\n';
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
PrintChildLoopComment(OS, CL, FunctionNumber);
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
}
}
Update function names to conform to guidelines. No functional change intended. llvm-svn: 163401
2012-09-08 02:16:38 +08:00
/// emitBasicBlockLoopComments - Pretty-print comments for basic blocks.
static void emitBasicBlockLoopComments(const MachineBasicBlock &MBB,
minor tidying. llvm-svn: 100338
2010-04-05 02:16:38 +08:00
const MachineLoopInfo *LI,
const AsmPrinter &AP) {
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
// Add loop depth information
const MachineLoop *Loop = LI->getLoopFor(&MBB);
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
if (!Loop) return;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
MachineBasicBlock *Header = Loop->getHeader();
assert(Header && "No header for loop");
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
// If this block is not a loop header, just print out what is the loop header
// and return.
if (Header != &MBB) {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
AP.OutStreamer->AddComment(" in Loop: Header=BB" +
Twine(AP.getFunctionNumber())+"_" +
Twine(Loop->getHeader()->getNumber())+
" Depth="+Twine(Loop->getLoopDepth()));
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
return;
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
// Otherwise, it is a loop header. Print out information about child and
// parent loops.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
raw_ostream &OS = AP.OutStreamer->GetCommentOS();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
PrintParentLoopComment(OS, Loop->getParentLoop(), AP.getFunctionNumber());
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
OS << "=>";
OS.indent(Loop->getLoopDepth()*2-2);
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
OS << "This ";
if (Loop->empty())
OS << "Inner ";
OS << "Loop Header: Depth=" + Twine(Loop->getLoopDepth()) << '\n';
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
change this to be a static function instead of a method on asmprinter. llvm-svn: 94227
2010-01-23 05:50:41 +08:00
PrintChildLoopComment(OS, Loop, AP.getFunctionNumber());
}
replace printBasicBlockLabel with EmitBasicBlockStart, now that printBasicBlockLabel is only used for starting a MBB. This allows elimination of a bunch of arguments. llvm-svn: 81684
2009-09-14 02:25:37 +08:00
/// EmitBasicBlockStart - This method prints the label for the specified
/// MachineBasicBlock, an alignment (if present) and a comment describing
/// it if appropriate.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
void AsmPrinter::EmitBasicBlockStart(const MachineBasicBlock &MBB) const {
[WinEH] Teach AsmPrinter about funclets Summary: Funclets have been turned into functions by the time they hit the object file. Make sure that they have decent names for the symbol table and CFI directives explaining how to reason about their prologues. Differential Revision: http://reviews.llvm.org/D13261 llvm-svn: 248824
2015-09-30 04:12:33 +08:00
// End the previous funclet and start a new one.
if (MBB.isEHFuncletEntry()) {
for (const HandlerInfo &HI : Handlers) {
HI.Handler->endFunclet();
HI.Handler->beginFunclet(MBB);
}
}
Add some comments. llvm-svn: 85558
2009-10-30 09:34:35 +08:00
// Emit an alignment directive for this block, if needed.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (unsigned Align = MBB.getAlignment())
Use logarithmic units for basic block alignment. This was actually a bit of a mess. TLI.setPrefLoopAlignment was clearly documented as taking log2(bytes) units, but the x86 target would still set a preferred loop alignment of '16'. CodePlacementOpt passed this number on to the basic block, and AsmPrinter interpreted it as bytes. Now both MachineFunction and MachineBasicBlock use logarithmic alignments. Obviously, MachineConstantPool still measures alignments in bytes, so we can emulate the thrill of using as. llvm-svn: 145889
2011-12-06 09:26:19 +08:00
EmitAlignment(Align);
Add a quick and dirty "loop aligner pass". x86 uses it to align its loops to 16-byte boundaries. llvm-svn: 47703
2008-02-28 08:43:03 +08:00
Fix the third (and last known) case of code update problems due to LLVM IR changes with addr label weirdness. In the testcase, we generate references to the two bb's when codegen'ing the first function: _test1: ## @test1 leaq Ltmp0(%rip), %rax .. leaq Ltmp1(%rip), %rax Then continue to codegen the second function where the blocks get merged. We're now smart enough to emit both labels, producing this code: _test_fun: ## @test_fun ## BB#0: ## %entry Ltmp1: ## Block address taken Ltmp0: ## BB#1: ## %ret movl $-1, %eax ret Rejoice. llvm-svn: 98595
2010-03-16 08:29:39 +08:00
// If the block has its address taken, emit any labels that were used to
// reference the block. It is possible that there is more than one label
// here, because multiple LLVM BB's may have been RAUW'd to this block after
// the references were generated.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (MBB.hasAddressTaken()) {
const BasicBlock *BB = MBB.getBasicBlock();
clean up the asmprinter header and privatize some stuff. llvm-svn: 100342
2010-04-05 02:52:31 +08:00
if (isVerbose())
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->AddComment("Block address taken");
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
[CodeGen] Mark setjmp/catchret MBBs address-taken Summary: This ensures that BranchFolding (and similar) won't remove these blocks. Also allow AsmPrinter::EmitBasicBlockStart to process MBBs which are address-taken but do not have BBs that are address-taken, since otherwise its call to getAddrLabelSymbolTableToEmit would fail an assertion on such blocks. I audited the other callers of getAddrLabelSymbolTableToEmit (and getAddrLabelSymbol); they all have BBs known to be address-taken except for the call through getAddrLabelSymbol from WinException::create32bitRef; that call is actually now unreachable, so I've removed it and updated the signature of create32bitRef. This fixes PR25168. Reviewers: majnemer, andrew.w.kaylor, rnk Subscribers: pgavlin, llvm-commits Differential Revision: http://reviews.llvm.org/D13774 llvm-svn: 251113
2015-10-23 23:06:05 +08:00
// MBBs can have their address taken as part of CodeGen without having
// their corresponding BB's address taken in IR
if (BB->hasAddressTaken())
for (MCSymbol *Sym : MMI->getAddrLabelSymbolToEmit(BB))
OutStreamer->EmitLabel(Sym);
Initial target-independent CodeGen support for BlockAddresses. llvm-svn: 85556
2009-10-30 09:27:03 +08:00
}
Hoist common code out of if statement. llvm-svn: 152153
2012-03-07 06:27:13 +08:00
// Print some verbose block comments.
if (isVerbose()) {
[CodeGen] Use generic printAsOperand machinery instead of hand rolling it We already know how to properly print out basic blocks in printAsOperand, we should not roll it ourselves in AsmPrinter::EmitBasicBlockStart. No functionality change is intended. llvm-svn: 256413
2015-12-25 17:37:26 +08:00
if (const BasicBlock *BB = MBB.getBasicBlock()) {
if (BB->hasName()) {
BB->printAsOperand(OutStreamer->GetCommentOS(),
/*PrintType=*/false, BB->getModule());
OutStreamer->GetCommentOS() << '\n';
}
}
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
emitBasicBlockLoopComments(MBB, LI, *this);
Hoist common code out of if statement. llvm-svn: 152153
2012-03-07 06:27:13 +08:00
}
Add some comments. llvm-svn: 85558
2009-10-30 09:34:35 +08:00
// Print the main label for the block.
[WinEH] Add some support for code generating catchpad We can now run 32-bit programs with empty catch bodies. The next step is to change PEI so that we get funclet prologues and epilogues. llvm-svn: 246235
2015-08-28 07:27:47 +08:00
if (MBB.pred_empty() ||
(isBlockOnlyReachableByFallthrough(&MBB) && !MBB.isEHFuncletEntry())) {
Remove a redundant call to hasRawTextSupport. The code path it was guarding was already using emitRawComment. llvm-svn: 200591
2014-02-01 07:14:01 +08:00
if (isVerbose()) {
add a twine form of MCStreamer::EmitRawText, and mc'ize a few more things in AsmPrinter.cpp. llvm-svn: 100294
2010-04-04 06:12:35 +08:00
// NOTE: Want this comment at start of line, don't emit with AddComment.
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->emitRawComment(" BB#" + Twine(MBB.getNumber()) + ":", false);
Convert some more random-comment-printing stuff to use AddComment and GetCommentOS. Add a blank line between globals (even in non-verbose mode) to make the assembly more readable. llvm-svn: 94202
2010-01-23 03:52:01 +08:00
}
Instead of printing unnecessary basic block labels as labels in verbose-asm mode, print comments instead. This eliminates a non-comment difference between verbose-asm mode and non-verbose-asm mode. Also, factor out the relevant code out of all the targets and into target-independent code. llvm-svn: 83392
2009-10-07 01:38:38 +08:00
} else {
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitLabel(MBB.getSymbol());
Instead of printing unnecessary basic block labels as labels in verbose-asm mode, print comments instead. This eliminates a non-comment difference between verbose-asm mode and non-verbose-asm mode. Also, factor out the relevant code out of all the targets and into target-independent code. llvm-svn: 83392
2009-10-07 01:38:38 +08:00
}
JumpTable support! What this represents is working asm and jit support for x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947
2006-04-23 02:53:45 +08:00
}
Emit .set directives for jump table entries when possible, which reduces the number of relocations in object files, shrinkifying them. llvm-svn: 29650
2006-08-13 05:29:52 +08:00
Omit private_extern declarations of extern symbols; followup to r124468. Patch by Rafael Avila de Espindola! llvm-svn: 126297
2011-02-23 10:27:05 +08:00
void AsmPrinter::EmitVisibility(MCSymbol *Sym, unsigned Visibility,
bool IsDefinition) const {
mcize visibility directives. llvm-svn: 94295
2010-01-23 14:53:23 +08:00
MCSymbolAttr Attr = MCSA_Invalid;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
mcize visibility directives. llvm-svn: 94295
2010-01-23 14:53:23 +08:00
switch (Visibility) {
default: break;
case GlobalValue::HiddenVisibility:
Omit private_extern declarations of extern symbols; followup to r124468. Patch by Rafael Avila de Espindola! llvm-svn: 126297
2011-02-23 10:27:05 +08:00
if (IsDefinition)
Attr = MAI->getHiddenVisibilityAttr();
else
Attr = MAI->getHiddenDeclarationVisibilityAttr();
mcize visibility directives. llvm-svn: 94295
2010-01-23 14:53:23 +08:00
break;
case GlobalValue::ProtectedVisibility:
Attr = MAI->getProtectedVisibilityAttr();
break;
add a version of AsmPrinter::printVisibility that takes an MCSymbol. llvm-svn: 93587
2010-01-16 07:38:51 +08:00
}
mcize visibility directives. llvm-svn: 94295
2010-01-23 14:53:23 +08:00
if (Attr != MCSA_Invalid)
[AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr. AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a reference for this is crufty. llvm-svn: 235752
2015-04-25 03:11:51 +08:00
OutStreamer->EmitSymbolAttribute(Sym, Attr);
add a version of AsmPrinter::printVisibility that takes an MCSymbol. llvm-svn: 93587
2010-01-16 07:38:51 +08:00
}
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
/// isBlockOnlyReachableByFallthough - Return true if the basic block has
/// exactly one predecessor and the control transfer mechanism between
/// the predecessor and this block is a fall-through.
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
bool AsmPrinter::
isBlockOnlyReachableByFallthrough(const MachineBasicBlock *MBB) const {
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
// If this is a landing pad, it isn't a fall through. If it has no preds,
// then nothing falls through to it.
[WinEH] Add some support for code generating catchpad We can now run 32-bit programs with empty catch bodies. The next step is to change PEI so that we get funclet prologues and epilogues. llvm-svn: 246235
2015-08-28 07:27:47 +08:00
if (MBB->isEHPad() || MBB->pred_empty())
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
return false;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
// If there isn't exactly one predecessor, it can't be a fall through.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
if (MBB->pred_size() > 1)
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
return false;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
// The predecessor has to be immediately before this block.
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
MachineBasicBlock *Pred = *MBB->pred_begin();
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
if (!Pred->isLayoutSuccessor(MBB))
return false;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
// If the block is completely empty, then it definitely does fall through.
if (Pred->empty())
return true;
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Revert r133004 ; it's breaking nightly tests. llvm-svn: 133007
2011-06-15 03:30:33 +08:00
// Check the terminators in the previous blocks
[c++11] Tidy up AsmPrinter.cpp. Range'ify loops and tidy up some by-reference handling. No functional change. llvm-svn: 206422
2014-04-17 06:38:02 +08:00
for (const auto &MI : Pred->terminators()) {
Revert r133004 ; it's breaking nightly tests. llvm-svn: 133007
2011-06-15 03:30:33 +08:00
// If it is not a simple branch, we are in a table somewhere.
Add bundle aware API for querying instruction properties and switch the code generator to it. For non-bundle instructions, these behave exactly the same as the MC layer API. For properties like mayLoad / mayStore, look into the bundle and if any of the bundled instructions has the property it would return true. For properties like isPredicable, only return true if *all* of the bundled instructions have the property. For properties like canFoldAsLoad, isCompare, conservatively return false for bundles. llvm-svn: 146026
2011-12-07 15:15:52 +08:00
if (!MI.isBranch() || MI.isIndirectBranch())
Revert r133004 ; it's breaking nightly tests. llvm-svn: 133007
2011-06-15 03:30:33 +08:00
return false;
Handle bundled terminators in isBlockOnlyReachableByFallthrough. Targets like SPARC and MIPS have delay slots and normally bundle the delay slot instruction with the corresponding terminator. Teach isBlockOnlyReachableByFallthrough to find any MBB operands on bundled terminators so SPARC doesn't need to specialize this function. llvm-svn: 199061
2014-01-13 03:24:08 +08:00
// If we are the operands of one of the branches, this is not a fall
// through. Note that targets with delay slots will usually bundle
// terminators with the delay slot instruction.
WIP: CodeGen: Use MachineInstr& in MachineInstrBundle.h, NFC Update APIs in MachineInstrBundle.h to take and return MachineInstr& instead of MachineInstr* when the instruction cannot be null. Besides being a nice cleanup, this is tacking toward a fix for PR26753. llvm-svn: 262141
2016-02-28 01:05:33 +08:00
for (ConstMIBundleOperands OP(MI); OP.isValid(); ++OP) {
Handle bundled terminators in isBlockOnlyReachableByFallthrough. Targets like SPARC and MIPS have delay slots and normally bundle the delay slot instruction with the corresponding terminator. Teach isBlockOnlyReachableByFallthrough to find any MBB operands on bundled terminators so SPARC doesn't need to specialize this function. llvm-svn: 199061
2014-01-13 03:24:08 +08:00
if (OP->isJTI())
Handle jump tables. Test to follow soon. llvm-svn: 133083
2011-06-16 05:00:28 +08:00
return false;
Handle bundled terminators in isBlockOnlyReachableByFallthrough. Targets like SPARC and MIPS have delay slots and normally bundle the delay slot instruction with the corresponding terminator. Teach isBlockOnlyReachableByFallthrough to find any MBB operands on bundled terminators so SPARC doesn't need to specialize this function. llvm-svn: 199061
2014-01-13 03:24:08 +08:00
if (OP->isMBB() && OP->getMBB() == MBB)
Revert r133004 ; it's breaking nightly tests. llvm-svn: 133007
2011-06-15 03:30:33 +08:00
return false;
}
}
return true;
move isOnlyReachableByFallthrough out of MachineBasicBlock into AsmPrinter, and add a sparc implementation that knows about delay slots. Patch by Nathan Keynes! llvm-svn: 96492
2010-02-18 02:52:56 +08:00
}
Use unique_ptr to manage ownership of GCStrategy objects in GCMetadata llvm-svn: 206246
2014-04-15 13:34:49 +08:00
GCMetadataPrinter *AsmPrinter::GetOrCreateGCPrinter(GCStrategy &S) {
if (!S.usesMetadata())
[C++] Use 'nullptr'. llvm-svn: 207083
2014-04-24 14:44:33 +08:00
return nullptr;
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
Introduce an example statepoint GC strategy This change includes the most basic possible GCStrategy for a GC which is using the statepoint lowering code. At the moment, this GCStrategy doesn't really do much - aside from actually generate correct stackmaps that is - but I went ahead and added a few extra correctness checks as proof of concept. It's mostly here to provide documentation on how to do one, and to provide a point for various optimization legality hooks I'd like to add going forward. (For context, see the TODOs in InstCombine around gc.relocate.) Most of the validation logic added here as proof of concept will soon move in to the Verifier. That move is dependent on http://reviews.llvm.org/D6811 There was discussion in the review thread about addrspace(1) being reserved for something. I'm going to follow up on a seperate llvmdev thread. If needed, I'll update all the code at once. Note that I am deliberately not making a GCStrategy required to use gc.statepoints with this change. I want to give folks out of tree - including myself - a chance to migrate. In a week or two, I'll make having a GCStrategy be required for gc.statepoints. To this end, I added the gc tag to one of the test cases but not others. Differential Revision: http://reviews.llvm.org/D6808 llvm-svn: 225365
2015-01-08 03:07:50 +08:00
assert(!S.useStatepoints() && "statepoints do not currently support custom"
" stackmap formats, please see the documentation for a description of"
" the default format. If you really need a custom serialized format,"
" please file a bug");
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
gcp_map_type &GCMap = getGCMap(GCMetadataPrinters);
Use unique_ptr to manage ownership of GCStrategy objects in GCMetadata llvm-svn: 206246
2014-04-15 13:34:49 +08:00
gcp_map_type::iterator GCPI = GCMap.find(&S);
lazily allocate the GCMetadataPrinters map and remove DenseMap from the AsmPrinter interface. llvm-svn: 100331
2010-04-05 01:57:56 +08:00
if (GCPI != GCMap.end())
Use unique_ptr for the result of Registry entries. llvm-svn: 206248
2014-04-15 13:53:26 +08:00
return GCPI->second.get();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Use StringRef in Registry API (NFC) llvm-svn: 283039
2016-10-01 23:44:54 +08:00
auto Name = S.getName();
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
Factor GC metadata table assembly generation out of Collector in preparation for splitting AsmPrinter into its own library. llvm-svn: 54880
2008-08-17 20:08:44 +08:00
for (GCMetadataPrinterRegistry::iterator
I = GCMetadataPrinterRegistry::begin(),
E = GCMetadataPrinterRegistry::end(); I != E; ++I)
Use StringRef in Registry API (NFC) llvm-svn: 283039
2016-10-01 23:44:54 +08:00
if (Name == I->getName()) {
Use unique_ptr for the result of Registry entries. llvm-svn: 206248
2014-04-15 13:53:26 +08:00
std::unique_ptr<GCMetadataPrinter> GMP = I->instantiate();
Use unique_ptr to manage ownership of GCStrategy objects in GCMetadata llvm-svn: 206246
2014-04-15 13:34:49 +08:00
GMP->S = &S;
Use unique_ptr for the result of Registry entries. llvm-svn: 206248
2014-04-15 13:53:26 +08:00
auto IterBool = GCMap.insert(std::make_pair(&S, std::move(GMP)));
return IterBool.first->second.get();
Factor GC metadata table assembly generation out of Collector in preparation for splitting AsmPrinter into its own library. llvm-svn: 54880
2008-08-17 20:08:44 +08:00
}
Tidy up. 80 columns and trailing whitespace. llvm-svn: 128504
2011-03-30 07:20:22 +08:00
rename llvm::llvm_report_error -> llvm::report_fatal_error llvm-svn: 100709
2010-04-08 06:58:41 +08:00
report_fatal_error("no GCMetadataPrinter registered for GC: " + Twine(Name));
Factor GC metadata table assembly generation out of Collector in preparation for splitting AsmPrinter into its own library. llvm-svn: 54880
2008-08-17 20:08:44 +08:00
}
Remove weak vtables. No functional change. llvm-svn: 197323
2013-12-14 20:23:14 +08:00
/// Pin vtable to this file.
[MC] Fix some Clang-tidy modernize and Include What You Use warnings; other minor fixes (NFC). Same changes in files affected by reduced MC headers dependencies. llvm-svn: 295009
2017-02-14 08:33:36 +08:00
AsmPrinterHandler::~AsmPrinterHandler() = default;
Print jump tables before exception tables. In the case where just tables are part of the function section, this produces more readable assembly by avoiding switching to the eh section and back to .text. This would also break with non unique section names, as trying to switch to a unique section actually creates a new one. llvm-svn: 231677
2015-03-10 02:29:12 +08:00
void AsmPrinterHandler::markFunctionEnd() {}
[XRay] ARM 32-bit no-Thumb support in LLVM This is a port of XRay to ARM 32-bit, without Thumb support yet. The XRay instrumentation support is moving up to AsmPrinter. This is one of 3 commits to different repositories of XRay ARM port. The other 2 are: https://reviews.llvm.org/D23932 (Clang test) https://reviews.llvm.org/D23933 (compiler-rt) Differential Revision: https://reviews.llvm.org/D23931 llvm-svn: 281878
2016-09-19 08:54:35 +08:00
[XRay] Merge instrumentation point table emission code into AsmPrinter. Summary: No need to have this per-architecture. While there, unify 32-bit ARM's behaviour with what changed elsewhere and start function names lowercase as per the coding standards. Individual entry emission code goes to the entry's own class. Fully tested on amd64, cross-builds on both ARMs and PowerPC. Reviewers: dberris Subscribers: aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D28209 llvm-svn: 290858
2017-01-03 12:30:21 +08:00
// In the binary's "xray_instr_map" section, an array of these function entries
// describes each instrumentation point. When XRay patches your code, the index
// into this table will be given to your handler as a patch point identifier.
void AsmPrinter::XRayFunctionEntry::emit(int Bytes, MCStreamer *Out,
const MCSymbol *CurrentFnSym) const {
Out->EmitSymbolValue(Sled, Bytes);
Out->EmitSymbolValue(CurrentFnSym, Bytes);
auto Kind8 = static_cast<uint8_t>(Kind);
Out->EmitBytes(StringRef(reinterpret_cast<const char *>(&Kind8), 1));
Out->EmitBytes(
StringRef(reinterpret_cast<const char *>(&AlwaysInstrument), 1));
Out->EmitZeros(2 * Bytes - 2); // Pad the previous two entries
}
void AsmPrinter::emitXRayTable() {
if (Sleds.empty())
return;
auto PrevSection = OutStreamer->getCurrentSectionOnly();
auto Fn = MF->getFunction();
MCSection *Section = nullptr;
if (MF->getSubtarget().getTargetTriple().isOSBinFormatELF()) {
if (Fn->hasComdat()) {
Section = OutContext.getELFSection("xray_instr_map", ELF::SHT_PROGBITS,
ELF::SHF_ALLOC | ELF::SHF_GROUP, 0,
Fn->getComdat()->getName());
} else {
Section = OutContext.getELFSection("xray_instr_map", ELF::SHT_PROGBITS,
ELF::SHF_ALLOC);
}
} else if (MF->getSubtarget().getTargetTriple().isOSBinFormatMachO()) {
Section = OutContext.getMachOSection("__DATA", "xray_instr_map", 0,
SectionKind::getReadOnlyWithRel());
} else {
llvm_unreachable("Unsupported target");
}
// Before we switch over, we force a reference to a label inside the
// xray_instr_map section. Since this function is always called just
// before the function's end, we assume that this is happening after
// the last return instruction.
Distinguish between code pointer size and DataLayout::getPointerSize() in DWARF info generation llvm-svn: 300463
2017-04-18 01:41:25 +08:00
auto WordSizeBytes = MAI->getCodePointerSize();
[XRay] Merge instrumentation point table emission code into AsmPrinter. Summary: No need to have this per-architecture. While there, unify 32-bit ARM's behaviour with what changed elsewhere and start function names lowercase as per the coding standards. Individual entry emission code goes to the entry's own class. Fully tested on amd64, cross-builds on both ARMs and PowerPC. Reviewers: dberris Subscribers: aemerson, llvm-commits Differential Revision: https://reviews.llvm.org/D28209 llvm-svn: 290858
2017-01-03 12:30:21 +08:00
MCSymbol *Tmp = OutContext.createTempSymbol("xray_synthetic_", true);
OutStreamer->EmitCodeAlignment(16);
OutStreamer->EmitSymbolValue(Tmp, WordSizeBytes, false);
OutStreamer->SwitchSection(Section);
OutStreamer->EmitLabel(Tmp);
for (const auto &Sled : Sleds)
Sled.emit(WordSizeBytes, OutStreamer.get(), CurrentFnSym);
OutStreamer->SwitchSection(PrevSection);
Sleds.clear();
}
[XRay] ARM 32-bit no-Thumb support in LLVM This is a port of XRay to ARM 32-bit, without Thumb support yet. The XRay instrumentation support is moving up to AsmPrinter. This is one of 3 commits to different repositories of XRay ARM port. The other 2 are: https://reviews.llvm.org/D23932 (Clang test) https://reviews.llvm.org/D23933 (compiler-rt) Differential Revision: https://reviews.llvm.org/D23931 llvm-svn: 281878
2016-09-19 08:54:35 +08:00
void AsmPrinter::recordSled(MCSymbol *Sled, const MachineInstr &MI,
SledKind Kind) {
auto Fn = MI.getParent()->getParent()->getFunction();
auto Attr = Fn->getFnAttribute("function-instrument");
[XRay] Allow logging the first argument of a function call. Summary: Functions with the "xray-log-args" attribute will have a special XRay sled kind emitted, for compiler-rt to copy any call arguments to your logging handler. For practical and performance reasons, only the first argument is supported, and only up to 64 bits. Reviewers: dberris Reviewed By: dberris Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D29702 llvm-svn: 296998
2017-03-06 14:48:56 +08:00
bool LogArgs = Fn->hasFnAttribute("xray-log-args");
[XRay] ARM 32-bit no-Thumb support in LLVM This is a port of XRay to ARM 32-bit, without Thumb support yet. The XRay instrumentation support is moving up to AsmPrinter. This is one of 3 commits to different repositories of XRay ARM port. The other 2 are: https://reviews.llvm.org/D23932 (Clang test) https://reviews.llvm.org/D23933 (compiler-rt) Differential Revision: https://reviews.llvm.org/D23931 llvm-svn: 281878
2016-09-19 08:54:35 +08:00
bool AlwaysInstrument =
Attr.isStringAttribute() && Attr.getValueAsString() == "xray-always";
[XRay] Allow logging the first argument of a function call. Summary: Functions with the "xray-log-args" attribute will have a special XRay sled kind emitted, for compiler-rt to copy any call arguments to your logging handler. For practical and performance reasons, only the first argument is supported, and only up to 64 bits. Reviewers: dberris Reviewed By: dberris Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D29702 llvm-svn: 296998
2017-03-06 14:48:56 +08:00
if (Kind == SledKind::FUNCTION_ENTER && LogArgs)
Kind = SledKind::LOG_ARGS_ENTER;
[XRay] ARM 32-bit no-Thumb support in LLVM This is a port of XRay to ARM 32-bit, without Thumb support yet. The XRay instrumentation support is moving up to AsmPrinter. This is one of 3 commits to different repositories of XRay ARM port. The other 2 are: https://reviews.llvm.org/D23932 (Clang test) https://reviews.llvm.org/D23933 (compiler-rt) Differential Revision: https://reviews.llvm.org/D23931 llvm-svn: 281878
2016-09-19 08:54:35 +08:00
Sleds.emplace_back(
XRayFunctionEntry{ Sled, CurrentFnSym, Kind, AlwaysInstrument, Fn });
}
Rely on a single DWARF version instead of having two copies This patch makes AsmPrinter less reliant on DwarfDebug by relying on the DWARF version in the AsmPrinter's MCStreamer's MCContext. This allows us to remove the redundant DWARF version from DwarfDebug. It also lets us change code that used to access the AsmPrinter's DwarfDebug just to get to the DWARF version by changing the DWARF version accessor on AsmPrinter so that it grabs the version from its MCStreamer's MCContext. Differential Revision: https://reviews.llvm.org/D27032 llvm-svn: 287839
2016-11-24 07:30:37 +08:00
uint16_t AsmPrinter::getDwarfVersion() const {
return OutStreamer->getContext().getDwarfVersion();
}
void AsmPrinter::setDwarfVersion(uint16_t Version) {
OutStreamer->getContext().setDwarfVersion(Version);
}