summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms/IPO
Commit message (Collapse)AuthorAgeFilesLines
* Use the Attributes::get method which takes an AttrVal value directly to ↵Bill Wendling2012-10-161-4/+2
| | | | | | simplify the code a bit. No functionality change. llvm-svn: 166009
* Fix filename in file header.Craig Topper2012-10-161-1/+1
| | | | llvm-svn: 166004
* Move the Attributes::Builder outside of the Attributes class and into its ↵Bill Wendling2012-10-154-7/+7
| | | | | | own class named AttrBuilder. No functionality change. llvm-svn: 165960
* Add an enum for the return and function indexes into the AttrListPtr object. ↵Bill Wendling2012-10-153-11/+22
| | | | | | This gets rid of some magic numbers. llvm-svn: 165924
* Attributes RewriteBill Wendling2012-10-154-10/+13
| | | | | | | | | | Convert the internal representation of the Attributes class into a pointer to an opaque object that's uniqued by and stored in the LLVMContext object. The Attributes class then becomes a thin wrapper around this opaque object. Eventually, the internal representation will be expanded to include attributes that represent code generation options, etc. llvm-svn: 165917
* Remove operator cast method in favor of querying with the correct method.Bill Wendling2012-10-142-13/+24
| | | | llvm-svn: 165899
* Remove the bitwise AND operators from the Attributes class. Replace it with ↵Bill Wendling2012-10-141-2/+3
| | | | | | the equivalent from the builder class. llvm-svn: 165896
* Remove the bitwise assignment OR operator from the Attributes class. Replace ↵Bill Wendling2012-10-141-1/+2
| | | | | | it with the equivalent from the builder class. llvm-svn: 165895
* Remove the bitwise NOT operator from the Attributes class. Replace it with ↵Bill Wendling2012-10-142-6/+10
| | | | | | the equivalent from the builder class. llvm-svn: 165892
* Remove the final bits of Attributes being declared in the AttributeBill Wendling2012-10-102-8/+19
| | | | | | | namespace. Use the attribute's enum value instead. No functionality change intended. llvm-svn: 165610
* Have 'addFnAttr' take the attribute enum value. Then have it build the ↵Bill Wendling2012-10-101-2/+2
| | | | | | attribute object and add it appropriately. No functionality change. llvm-svn: 165595
* Use the attribute enums to query if a parameter has an attribute.Bill Wendling2012-10-091-1/+1
| | | | llvm-svn: 165550
* Fixup for r165490: Use DenseMap instead of std::map. Simplify the loop in ↵Alexey Samsonov2012-10-091-13/+16
| | | | | | CollectFunctionDIs. llvm-svn: 165498
* Use the enum value of the attributes when adding them to the attributes builder.Bill Wendling2012-10-091-2/+2
| | | | llvm-svn: 165494
* Fix PR14016.Alexey Samsonov2012-10-091-0/+50
| | | | | | | | | | DeadArgumentElimination pass can replace one LLVM function with another, invalidating a pointer stored in debug info metadata entry for this function. To fix this, we collect debug info descriptors for functions before running a DeadArgumentElimination pass and "patch" pointers in metadata nodes if we replace a function. llvm-svn: 165490
* Create enums for the different attributes.Bill Wendling2012-10-094-10/+12
| | | | | | | We use the enums to query whether an Attributes object has that attribute. The opaque layer is responsible for knowing where that specific attribute is stored. llvm-svn: 165488
* Convert to using the Attributes::Builder class to create attributes.Bill Wendling2012-10-091-4/+4
| | | | llvm-svn: 165468
* Give CaptureTracker::shouldExplore a base implementation. Most users want to doNick Lewycky2012-10-081-2/+0
| | | | | | the same thing. No functionality change. llvm-svn: 165435
* Move TargetData to DataLayout.Micah Villmow2012-10-086-40/+40
| | | | llvm-svn: 165402
* Use method to query for attributes.Bill Wendling2012-10-041-1/+1
| | | | llvm-svn: 165209
* Add method to query for 'NoAlias' attribute on call/invoke instructions.Bill Wendling2012-10-041-1/+1
| | | | llvm-svn: 165208
* Use method to query for attributes.Bill Wendling2012-10-041-1/+1
| | | | llvm-svn: 165207
* Query for attributes via the correct method call.Bill Wendling2012-10-041-2/+2
| | | | llvm-svn: 165206
* Turn the new SROA pass back on. Let's see if it sticks this time. =]Chandler Carruth2012-10-021-1/+1
| | | | | | Again, let me know if anything breaks due to this! llvm-svn: 164986
* ArgumentPromotion: Remove ancient workaround for a bug in the C backend.Benjamin Kramer2012-09-301-19/+1
| | | | | | Fun fact: The CBE learned how to deal with this situation before it was removed. llvm-svn: 164918
* GlobalDCE should be run at -O2 / -Os to eliminate unused dtor, etc. ↵Evan Cheng2012-09-281-4/+3
| | | | | | rdar://9142819 llvm-svn: 164850
* GlobalOpt: non-constexpr bitcasts or GEPs can occur even if the global value ↵Benjamin Kramer2012-09-281-1/+3
| | | | | | | | is only stored once. Fixes PR13968. llvm-svn: 164815
* Revert 'Fix a typo 'iff' => 'if''. iff is an abreviation of if and only if. ↵Sylvestre Ledru2012-09-273-3/+3
| | | | | | See: http://en.wikipedia.org/wiki/If_and_only_if Commit 164767 llvm-svn: 164768
* Fix a typo 'iff' => 'if'Sylvestre Ledru2012-09-273-3/+3
| | | | llvm-svn: 164767
* Disable the new SROA pass to get the tree back in working order. We don't yetNick Lewycky2012-09-261-1/+1
| | | | | | have testcases for the current problems. llvm-svn: 164731
* Remove the `hasFnAttr' method from Function.Bill Wendling2012-09-262-8/+8
| | | | | | | The hasFnAttr method has been replaced by querying the Attributes explicitly. No intended functionality change. llvm-svn: 164725
* Move Attribute::typeIncompatible inside of the Attributes class.Bill Wendling2012-09-251-3/+3
| | | | llvm-svn: 164629
* Enable the new SROA pass by default.Chandler Carruth2012-09-241-1/+1
| | | | | | Queue the fallout. ;] llvm-svn: 164480
* LNT builders have picked up new SROA, disable it to get the remaining ↵Benjamin Kramer2012-09-181-1/+1
| | | | | | builders green again. llvm-svn: 164124
* Add a major missing piece to the new SROA pass: aggressive splitting ofChandler Carruth2012-09-181-1/+1
| | | | | | | | | | | | | | | | | | | | | | | FCAs. This is essential in order to promote allocas that are used in struct returns by frontends like Clang. The FCA load would block the rest of the pass from firing, resulting is significant regressions with the bullet benchmark in the nightly test suite. Thanks to Duncan for repeated discussions about how best to do this, and to both him and Benjamin for review. This appears to have blocked many places where the pass tries to fire, and so I'm expect somewhat different results with this fix added. As with the last big patch, I'm including a change to enable the SROA by default *temporarily*. Ben is going to remove this as soon as the LNT bots pick up the patch. I'm just trying to get a round of LNT numbers from the stable machines in the lab. NOTE: Four clang tests are expected to fail in the brief window where this is enabled. Sorry for the noise! llvm-svn: 164119
* Disable new sroa now that all buildbots have tested it.Benjamin Kramer2012-09-151-1/+1
| | | | | | | | | | | | | | | | | What we have so far: - Some clang test failures (these were known already) - Perf results are mixed, some big regressions http://llvm.org/perf/db_default/v4/nts/3844 http://llvm.org/perf/db_default/v4/nts/3845 bullet suffers a lot. matmul is interesting: slower scalar code, faster with -vectorize. - Some dragonegg selfhost bots crash in SROA during selfhost now http://lab.llvm.org:8011/builders/dragonegg-x86_64-linux-gcc-4.6-self-host-checks/builds/1632 http://lab.llvm.org:8011/builders/dragonegg-x86_64-linux-gcc-4.5-self-host/builds/1891 llvm-svn: 163968
* Port the SSAUpdater-based promotion logic from the old SROA pass to theChandler Carruth2012-09-151-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | new one, and add support for running the new pass in that mode and in that slot of the pass manager. With this the new pass can completely replace the old one within the pipeline. The strategy for enabling or disabling the SSAUpdater logic is to do it by making the requirement of the domtree analysis optional. By default, it is required and we get the standard mem2reg approach. This is usually the desired strategy when run in stand-alone situations. Within the CGSCC pass manager, we disable requiring of the domtree analysis and consequentially trigger fallback to the SSAUpdater promotion. In theory this would allow the pass to re-use a domtree if one happened to be available even when run in a mode that doesn't require it. In practice, it lets us have a single pass rather than two which was simpler for me to wrap my head around. There is a hidden flag to force the use of the SSAUpdater code path for the purpose of testing. The primary testing strategy is just to run the existing tests through that path. One notable difference is that it has custom code to handle lifetime markers, and one of the tests has been enhanced to exercise that code. This has survived a bootstrap and the test suite without serious correctness issues, however my run of the test suite produced *very* alarming performance numbers. I don't entirely understand or trust them though, so more investigation is on-going. To aid my understanding of the performance impact of the new SROA now that it runs throughout the optimization pipeline, I'm enabling it by default in this commit, and will disable it again once the LNT bots have picked up one iteration with it. I want to get those bots (which are much more stable) to evaluate the impact of the change before I jump to any conclusions. NOTE: Several Clang tests will fail because they run -O3 and check the result's order of output. They'll go back to passing once I disable it again. llvm-svn: 163965
* Actually keep the flag default-off for now. =/ That's what I get forChandler Carruth2012-09-141-1/+1
| | | | | | being busy testing this... llvm-svn: 163890
* Introduce a new SROA implementation.Chandler Carruth2012-09-141-2/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is essentially a ground up re-think of the SROA pass in LLVM. It was initially inspired by a few problems with the existing pass: - It is subject to the bane of my existence in optimizations: arbitrary thresholds. - It is overly conservative about which constructs can be split and promoted. - The vector value replacement aspect is separated from the splitting logic, missing many opportunities where splitting and vector value formation can work together. - The splitting is entirely based around the underlying type of the alloca, despite this type often having little to do with the reality of how that memory is used. This is especially prevelant with unions and base classes where we tail-pack derived members. - When splitting fails (often due to the thresholds), the vector value replacement (again because it is separate) can kick in for preposterous cases where we simply should have split the value. This results in forming i1024 and i2048 integer "bit vectors" that tremendously slow down subsequnet IR optimizations (due to large APInts) and impede the backend's lowering. The new design takes an approach that fundamentally is not susceptible to many of these problems. It is the result of a discusison between myself and Duncan Sands over IRC about how to premptively avoid these types of problems and how to do SROA in a more principled way. Since then, it has evolved and grown, but this remains an important aspect: it fixes real world problems with the SROA process today. First, the transform of SROA actually has little to do with replacement. It has more to do with splitting. The goal is to take an aggregate alloca and form a composition of scalar allocas which can replace it and will be most suitable to the eventual replacement by scalar SSA values. The actual replacement is performed by mem2reg (and in the future SSAUpdater). The splitting is divided into four phases. The first phase is an analysis of the uses of the alloca. This phase recursively walks uses, building up a dense datastructure representing the ranges of the alloca's memory actually used and checking for uses which inhibit any aspects of the transform such as the escape of a pointer. Once we have a mapping of the ranges of the alloca used by individual operations, we compute a partitioning of the used ranges. Some uses are inherently splittable (such as memcpy and memset), while scalar uses are not splittable. The goal is to build a partitioning that has the minimum number of splits while placing each unsplittable use in its own partition. Overlapping unsplittable uses belong to the same partition. This is the target split of the aggregate alloca, and it maximizes the number of scalar accesses which become accesses to their own alloca and candidates for promotion. Third, we re-walk the uses of the alloca and assign each specific memory access to all the partitions touched so that we have dense use-lists for each partition. Finally, we build a new, smaller alloca for each partition and rewrite each use of that partition to use the new alloca. During this phase the pass will also work very hard to transform uses of an alloca into a form suitable for promotion, including forming vector operations, speculating loads throguh PHI nodes and selects, etc. After splitting is complete, each newly refined alloca that is a candidate for promotion to a scalar SSA value is run through mem2reg. There are lots of reasonably detailed comments in the source code about the design and algorithms, and I'm going to be trying to improve them in subsequent commits to ensure this is well documented, as the new pass is in many ways more complex than the old one. Some of this is still a WIP, but the current state is reasonbly stable. It has passed bootstrap, the nightly test suite, and Duncan has run it successfully through the ACATS and DragonEgg test suites. That said, it remains behind a default-off flag until the last few pieces are in place, and full testing can be done. Specific areas I'm looking at next: - Improved comments and some code cleanup from reviews. - SSAUpdater and enabling this pass inside the CGSCC pass manager. - Some datastructure tuning and compile-time measurements. - More aggressive FCA splitting and vector formation. Many thanks to Duncan Sands for the thorough final review, as well as Benjamin Kramer for lots of review during the process of writing this pass, and Daniel Berlin for reviewing the data structures and algorithms and general theory of the pass. Also, several other people on IRC, over lunch tables, etc for lots of feedback and advice. llvm-svn: 163883
* Fix an 80 char line limit.Nadav Rotem2012-09-131-1/+2
| | | | llvm-svn: 163808
* Make MemoryBuiltins aware of TargetLibraryInfo.Benjamin Kramer2012-08-292-16/+21
| | | | | | | | | | | | | | | | This disables malloc-specific optimization when -fno-builtin (or -ffreestanding) is specified. This has been a problem for a long time but became more severe with the recent memory builtin improvements. Since the memory builtin functions are used everywhere, this required passing TLI in many places. This means that functions that now have an optional TLI argument, like RecursivelyDeleteTriviallyDeadFunctions, won't remove dead mallocs anymore if the TLI argument is missing. I've updated most passes to do the right thing. Fixes PR13694 and probably others. llvm-svn: 162841
* Move the "findUsedStructTypes" functionality outside of the Module class.Bill Wendling2012-08-031-2/+3
| | | | | | | | | The "findUsedStructTypes" method is very expensive to run. It needs to be optimized so that LTO can run faster. Splitting this method out of the Module class will help this occur. For instance, it can keep a list of seen objects so that it doesn't process them over and over again. llvm-svn: 161228
* It's not safe to blindly remove invoke instructions. This happens when weNick Lewycky2012-07-251-1/+2
| | | | | | | encounter an invoke of an allocation function. This should fix the dragonegg bootstrap. Testcase to follow, later. llvm-svn: 160757
* Don't delete one more instruction than we're allowed to. This should fix theNick Lewycky2012-07-241-1/+3
| | | | | | | Darwin bootstrap. Testcase exists but isn't fully reduced, I expect to commit the testcase this evening. llvm-svn: 160693
* Teach globalopt to not nuke all stores to globals. Keep them around of theyNick Lewycky2012-07-241-8/+177
| | | | | | | | | might be deliberate "one time" leaks, so that leak checkers can find them. This is a reapply of r160602 with the fix that this time I'm committing the code I thought I was committing last time; the I->eraseFromParent() goes *after* the break out of the loop. llvm-svn: 160664
* Revert r160602.Nick Lewycky2012-07-211-177/+8
| | | | llvm-svn: 160603
* Teach globalopt to play nice with leak checkers. This is a reapplication ofNick Lewycky2012-07-211-8/+177
| | | | | | | | r160529 that was subsequently reverted. The fix was to not call GV->eraseFromParent() right before the caller does the same. The existing testcases already caught this bug if run under valgrind. llvm-svn: 160602
* Revert r160529 due to crashes.Nick Lewycky2012-07-191-171/+8
| | | | llvm-svn: 160532
* Don't wipe out global variables that are probably storing pointers to heapNick Lewycky2012-07-191-8/+171
| | | | | | memory. This makes clang play nice with leak checkers. llvm-svn: 160529
* Replace some explicit compare loops with std::equal.Benjamin Kramer2012-07-191-4/+1
| | | | | | No functionality change. llvm-svn: 160501
OpenPOWER on IntegriCloud