summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms/IPO
Commit message (Collapse)AuthorAgeFilesLines
* Add some optional passes around the vectorizer to both better prepareChandler Carruth2014-10-141-1/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the IR going into it and to clean up the IR produced by the vectorizers. Note that these are *off by default* right now while folks collect data on whether the performance tradeoff is reasonable. In a build of the 'opt' binary, I see about 2% compile time regression due to this change on average. This is in my mind essentially the worst expected case: very little of the opt binary is going to *benefit* from these extra passes. I've seen several benchmarks improve in performance my small amounts due to running these passes, and there are certain (rare) cases where these passes make a huge difference by either enabling the vectorizer at all or by hoisting runtime checks out of the outer loop. My primary motivation is to prevent people from seeing runtime check overhead in benchmarks where the existing passes and optimizers would be able to eliminate that. I've chosen the sequence of passes based on the kinds of things that seem likely to be relevant for the code at each stage: rotaing loops for the vectorizer, finding correlated values, loop invariants, and unswitching opportunities from any runtime checks, and cleaning up commonalities exposed by the SLP vectorizer. I'll be pinging existing threads where some of these issues have come up and will start new threads to get folks to benchmark and collect data on whether this is the right tradeoff or we should do something else. llvm-svn: 219644
* Inliner: Non-local functions in COMDATs shouldn't be droppedDavid Majnemer2014-10-081-0/+7
| | | | | | | | | | A function with discardable linkage cannot be discarded if its a member of a COMDAT group without considering all the other COMDAT members as well. This sort of thing is already handled by GlobalOpt/GlobalDCE. This fixes PR21206. llvm-svn: 219335
* GlobalOpt: Don't drop unused memberes of a ComdatDavid Majnemer2014-10-081-8/+20
| | | | | | | | | A linkonce_odr member of a COMDAT shouldn't be dropped if we need to keep the entire COMDAT group. This fixes PR21191. llvm-svn: 219283
* DebugInfo+DeadArgElimination: Ensure llvm::Function*s from debug info are ↵David Blaikie2014-10-072-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | updated even when DAE removes both varargs and non-varargs arguments on the same function. After some stellar (& inspired) help from Reid Kleckner providing a test case for some rather unstable undefined behavior showing up as assertions produced by r214761, I was able to fix this issue in DAE involving the application of both varargs removal, followed by normal argument removal. Indeed I introduced this same bug into ArgumentPromotion (r212128) by copying the code from DAE, and when I fixed the bug in ArgPromo (r213805) and commented in that patch that I didn't need to address the same issue in DAE because it was a single pass. Turns out it's two pass, one for the varargs and one for the normal arguments, so the same fix is needed (at least during varargs removal). So here it is. (the observable/net effect of this bug, even when it didn't result in assertion failure, is that debug info would describe the DAE'd function in the abstract, but wouldn't provide high/low_pc, variable locations, line table, etc (it would appear as though the function had been entirely optimized away), see the original PR14016 for details of the general problem) I'm not recommitting the assertion just yet, as there's been another regression of it since I last tried. It might just be a few test cases weren't adequately updated after Adrian or Duncan's recent schema changes. llvm-svn: 219210
* GlobalDCE: Don't drop any COMDAT membersDavid Majnemer2014-10-071-25/+14
| | | | | | | | | If we require a single member of a comdat, require all of the other members as well. This fixes PR20981. llvm-svn: 219191
* range-for some loops in DAEDavid Blaikie2014-10-061-6/+3
| | | | llvm-svn: 219167
* Revert "Revert "DI: Fold constant arguments into a single MDString""Duncan P. N. Exon Smith2014-10-031-18/+2
| | | | | | | | | | | | | | | | | | | | | | This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
* Eliminate some deep std::vector copies. NFC.Benjamin Kramer2014-10-031-1/+1
| | | | llvm-svn: 218999
* Revert "DI: Fold constant arguments into a single MDString"Duncan P. N. Exon Smith2014-10-021-7/+10
| | | | | | This reverts commit r218914 while I investigate some bots. llvm-svn: 218918
* DI: Fold constant arguments into a single MDStringDuncan P. N. Exon Smith2014-10-021-10/+7
| | | | | | | | | | | | | This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 218914
* Add control of function merging to the PMBuilder.Nick Lewycky2014-09-131-0/+10
| | | | llvm-svn: 217731
* Add doInitialization/doFinalization to DataLayoutPass.Rafael Espindola2014-09-101-2/+1
| | | | | | | | | | | | | With this a DataLayoutPass can be reused for multiple modules. Once we have doInitialization/doFinalization, it doesn't seem necessary to pass a Module to the constructor. Overall this change seems in line with the idea of making DataLayout a required part of Module. With it the only way of having a DataLayout used is to add it to the Module. llvm-svn: 217548
* [PassManager] Adding Hidden attribute to EnableMLSM optionGerolf Hoflehner2014-09-101-2/+2
| | | | llvm-svn: 217539
* [MergedLoadStoreMotion] Move pass enabling option to PassManagerBuilderGerolf Hoflehner2014-09-101-2/+8
| | | | llvm-svn: 217538
* MergeFunctions: FunctionPtr has been renamed to FunctionNode.Stepan Dyatkovskiy2014-09-101-7/+7
| | | | | | | It's supposed to store additional pass information for current function here. That was the reason for name change. llvm-svn: 217483
* Add an AlignmentFromAssumptions PassHal Finkel2014-09-071-0/+8
| | | | | | | | | | | | | | | | | | | | | This adds a ScalarEvolution-powered transformation that updates load, store and memory intrinsic pointer alignments based on invariant((a+q) & b == 0) expressions. Many of the simple cases we can get with ValueTracking, but we still need something like this for the more complicated cases (such as those with an offset) that require some algebra. Note that gcc's __builtin_assume_aligned's optional third argument provides exactly for this kind of 'misalignment' offset for which this kind of logic is necessary. The primary motivation is to fixup alignments for vector loads/stores after vectorization (and unrolling). This pass is added to the optimization pipeline just after the SLP vectorizer runs (which, admittedly, does not preserve SE, although I imagine it could). Regardless, I actually don't think that the preservation matters too much in this case: SE computes lazily, and this pass won't issue any SE queries unless there are any assume intrinsics, so there should be no real additional cost in the common case (SLP does preserve DT and LoopInfo). llvm-svn: 217344
* Add an Assumption-Tracking PassHal Finkel2014-09-073-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | This adds an immutable pass, AssumptionTracker, which keeps a cache of @llvm.assume call instructions within a module. It uses callback value handles to keep stale functions and intrinsics out of the map, and it relies on any code that creates new @llvm.assume calls to notify it of the new instructions. The benefit is that code needing to find @llvm.assume intrinsics can do so directly, without scanning the function, thus allowing the cost of @llvm.assume handling to be negligible when none are present. The current design is intended to be lightweight. We don't keep track of anything until we need a list of assumptions in some function. The first time this happens, we scan the function. After that, we add/remove @llvm.assume calls from the cache in response to registration calls and ValueHandle callbacks. There are no new direct test cases for this pass, but because it calls it validation function upon module finalization, we'll pick up detectable inconsistencies from the other tests that touch @llvm.assume calls. This pass will be used by follow-up commits that make use of @llvm.assume. llvm-svn: 217334
* Enable noalias metadata by default and swap the order of the SLP and Loop ↵James Molloy2014-09-041-1/+1
| | | | | | | | vectorizers by default. After some time maturing, hopefully the flags themselves will be removed. llvm-svn: 217144
* Add pass-manager flags to use CFL AAHal Finkel2014-09-021-0/+5
| | | | | | | Add -use-cfl-aa (and -use-cfl-aa-in-codegen) to add CFL AA in the default pass managers (for easy testing). llvm-svn: 216978
* Feed AA to the inliner and use AA->getModRefBehavior in AddAliasScopeMetadataHal Finkel2014-09-013-1/+8
| | | | | | | | | | | | This feeds AA through the IFI structure into the inliner so that AddAliasScopeMetadata can use AA->getModRefBehavior to figure out which functions only access their arguments (instead of just hard-coding some knowledge of memory intrinsics). Most of the information is only available from BasicAA; this is important for preserving alias scoping information for target-specific intrinsics when doing the noalias parameter attribute to metadata conversion. llvm-svn: 216866
* Don't promote byval pointer arguments when padding mattersReid Kleckner2014-08-281-3/+81
| | | | | | | | | | | | | | | | Don't promote byval pointer arguments when when their size in bits is not equal to their alloc size in bits. This can happen for x86_fp80, where the size in bits is 80 but the alloca size in bits in 128. Promoting these types can break passing unions of x86_fp80s and other types. Patch by Thomas Jablin! Reviewed By: rnk Differential Revision: http://reviews.llvm.org/D5057 llvm-svn: 216693
* Simplify creation of a bunch of ArrayRefs by using None, makeArrayRef or ↵Craig Topper2014-08-271-2/+2
| | | | | | just letting them be implicitly created. llvm-svn: 216525
* musttail: Don't eliminate varargs packs if there is a forwarding callReid Kleckner2014-08-261-2/+7
| | | | | | Also clean up and beef up this grep test for the feature. llvm-svn: 216425
* ArgPromotion: Don't touch variadic functionsReid Kleckner2014-08-251-0/+7
| | | | | | | | | | | | | | | | Adding, removing, or changing non-pack parameters can change the ABI classification of pack parameters. Clang and other frontends encode the classification in the IR of the call site, but the callee side determines it dynamically based on the number of registers consumed so far. Changing the prototype affects the number of registers consumed would break such code. Dead argument elimination performs a similar task and already has a similar check to avoid this problem. Patch by Thomas Jablin! llvm-svn: 216421
* Remove dangling initializers in GlobalDCEBruno Cardoso Lopes2014-08-251-1/+7
| | | | | | | | | | | | | | GlobalDCE deletes global vars and updates their initializers to nullptr while leaving underlying constants to be cleaned up later by its uses. The clean up may never happen, fix this by forcing it every time it's safe to destroy constants. Final patch by Rafael Espindola http://reviews.llvm.org/D4931 <rdar://problem/17523868> llvm-svn: 216390
* MergeFunctions, tiny refactoring:Stepan Dyatkovskiy2014-08-251-3/+3
| | | | | | cmpAPFloat has been renamed to cmpAPFloats (multiple form). llvm-svn: 216376
* MergeFunctions, tiny refactoring:Stepan Dyatkovskiy2014-08-251-5/+5
| | | | | | cmpAPInt has been renamed to cmpAPInts (multiple form). llvm-svn: 216375
* MergeFunctions, tiny refactoring:Stepan Dyatkovskiy2014-08-251-12/+11
| | | | | | cmpType has been renamed to cmpTypes (multiple form). llvm-svn: 216374
* MergeFunctions, tiny refactoring:Stepan Dyatkovskiy2014-08-251-5/+5
| | | | | | cmpGEP has been renamed to cmpGEPs (multiple form). llvm-svn: 216373
* Use range based for loops to avoid needing to re-mention SmallPtrSet size.Craig Topper2014-08-244-19/+17
| | | | llvm-svn: 216351
* Move some logic to populateLTOPassManager.Rafael Espindola2014-08-211-5/+36
| | | | | | | This will avoid code duplication in the next commit which calls it directly from the gold plugin. llvm-svn: 216211
* Respect LibraryInfo in populateLTOPassManager and use it. NFC.Rafael Espindola2014-08-211-0/+4
| | | | llvm-svn: 216203
* Handle inlining in populateLTOPassManager like in populateModulePassManager.Rafael Espindola2014-08-211-5/+13
| | | | | | No functionality change. llvm-svn: 216178
* Move DisableGVNLoadPRE from populateLTOPassManager to PassManagerBuilder.Rafael Espindola2014-08-211-6/+6
| | | | llvm-svn: 216174
* Repace SmallPtrSet with SmallPtrSetImpl in function arguments to avoid ↵Craig Topper2014-08-214-21/+19
| | | | | | needing to mention the size. llvm-svn: 216158
* Const-correct and prevent a copy of a SmallPtrSet.Craig Topper2014-08-191-2/+2
| | | | llvm-svn: 215973
* Revert "Repace SmallPtrSet with SmallPtrSetImpl in function arguments to ↵Craig Topper2014-08-184-15/+15
| | | | | | | | avoid needing to mention the size." Getting a weird buildbot failure that I need to investigate. llvm-svn: 215870
* Repace SmallPtrSet with SmallPtrSetImpl in function arguments to avoid ↵Craig Topper2014-08-174-15/+15
| | | | | | needing to mention the size. llvm-svn: 215868
* [optnone] Make the optnone attribute effective at suppressing functionChandler Carruth2014-08-131-7/+13
| | | | | | | | | | | | | attribute and function argument attribute synthesizing and propagating. As with the other uses of this attribute, the goal remains a best-effort (no guarantees) attempt to not optimize the function or assume things about the function when optimizing. This is particularly useful for compiler testing, bisecting miscompiles, triaging things, etc. I was hitting specific issues using optnone to isolate test code from a test driver for my fuzz testing, and this is one step of fixing that. llvm-svn: 215538
* GlobalOpt: Optimize in the face of insertvalue/extractvalueDavid Majnemer2014-08-081-0/+11
| | | | | | | | | | GlobalOpt didn't know how to simulate InsertValueInst or ExtractValueInst. Optimizing these is pretty straightforward. N.B. This came up when looking at clang's IRGen for MS ABI member pointers; they are represented as aggregates. llvm-svn: 215184
* Add a new option -run-slp-after-loop-vectorization.James Molloy2014-08-061-15/+44
| | | | | | This swaps the order of the loop vectorizer and the SLP/BB vectorizers. It is disabled by default so we can do performance testing - ideally we want to change to having the loop vectorizer running first, and the SLP vectorizer using its leftovers instead of the other way around. llvm-svn: 214963
* Don't internalize all but main by default.Rafael Espindola2014-08-051-8/+1
| | | | | | | | | | | | | | | This is mostly a cleanup, but it changes a fairly old behavior. Every "real" LTO user was already disabling the silly internalize pass and creating the internalize pass itself. The difference with this patch is for "opt -std-link-opts" and the C api. Now to get a usable behavior out of opt one doesn't need the funny looking command line: opt -internalize -disable-internalize -internalize-public-api-list=foo,bar -std-link-opts llvm-svn: 214919
* MergeFunctions, tiny refactoring:Stepan Dyatkovskiy2014-07-311-4/+4
| | | | | | cmpOperation has been renamed to cmpOperations (multiple form). llvm-svn: 214392
* Fixing a few -Woverloaded-virtual warnings by exposing the hidden virtual ↵Aaron Ballman2014-07-301-0/+2
| | | | | | function as well. No functional changes intended. llvm-svn: 214325
* Add the missing hasLinkOnceODRLinkage predicate.Rafael Espindola2014-07-301-2/+1
| | | | llvm-svn: 214312
* Move -verify-use-list-order into llvm-uselistorderDuncan P. N. Exon Smith2014-07-253-372/+0
| | | | | | | | | | | | | | | | | Ugh. Turns out not even transformation passes link in how to read IR. I sincerely believe the buildbots will finally agree with my system after this though. (I don't really understand why all of this has been working on my system, but not on all the buildbots.) Create a new tool called llvm-uselistorder to use for verifying use-list order. For now, just dump everything from the (now defunct) -verify-use-list-order pass into the tool. This might be a better way to test use-list order anyway. Part of PR5680. llvm-svn: 213957
* Try to fix a layering violation introduced by r213945Duncan P. N. Exon Smith2014-07-251-0/+309
| | | | | | | | | | | The dragonegg buildbot (and others?) started failing after r213945/r213946 because `llvm-as` wasn't linking in the bitcode reader. I think moving the verify functions to the same file as the verify pass should fix the build. Adding a command-line option for maintaining use-list order in assembly as a drive-by to prevent warnings about unused static functions. llvm-svn: 213947
* IPO: Add use-list-order verifierDuncan P. N. Exon Smith2014-07-253-0/+63
| | | | | | | | | | | | | | | | | | | | Add a -verify-use-list-order pass, which shuffles use-list order, writes to bitcode, reads back, and verifies that the (shuffled) order matches. - The utility functions live in lib/IR/UseListOrder.cpp. - Moved (and renamed) the command-line option to enable writing use-lists, so that this pass can return early if the use-list orders aren't being serialized. It's not clear that this pass is the right direction long-term (perhaps a separate tool instead?), but short-term it's a great way to test the use-list order prototype. I've added an XFAIL-ed testcase that I'm hoping to get working pretty quickly. This is part of PR5680. llvm-svn: 213945
* Add scoped-noalias metadataHal Finkel2014-07-241-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit adds scoped noalias metadata. The primary motivations for this feature are: 1. To preserve noalias function attribute information when inlining 2. To provide the ability to model block-scope C99 restrict pointers Neither of these two abilities are added here, only the necessary infrastructure. In fact, there should be no change to existing functionality, only the addition of new features. The logic that converts noalias function parameters into this metadata during inlining will come in a follow-up commit. What is added here is the ability to generally specify noalias memory-access sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA nodes: !scope0 = metadata !{ metadata !"scope of foo()" } !scope1 = metadata !{ metadata !"scope 1", metadata !scope0 } !scope2 = metadata !{ metadata !"scope 2", metadata !scope0 } !scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 } !scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 } Loads and stores can be tagged with an alias-analysis scope, and also, with a noalias tag for a specific scope: ... = load %ptr1, !alias.scope !{ !scope1 } ... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 } When evaluating an aliasing query, if one of the instructions is associated with an alias.scope id that is identical to the noalias scope associated with the other instruction, or is a descendant (in the scope hierarchy) of the noalias scope associated with the other instruction, then the two memory accesses are assumed not to alias. Note that is the first element of the scope metadata is a string, then it can be combined accross functions and translation units. The string can be replaced by a self-reference to create globally unqiue scope identifiers. [Note: This overview is slightly stylized, since the metadata nodes really need to just be numbers (!0 instead of !scope0), and the scope lists are also global unnamed metadata.] Existing noalias metadata in a callee is "cloned" for use by the inlined code. This is necessary because the aliasing scopes are unique to each call site (because of possible control dependencies on the aliasing properties). For example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } -- now just because we know that a1 does not alias with b1 at the first call site, and a2 does not alias with b2 at the second call site, we cannot let inlining these functons have the metadata imply that a1 does not alias with b2. llvm-svn: 213864
* AA metadata refactoring (introduce AAMDNodes)Hal Finkel2014-07-242-5/+9
| | | | | | | | | | | | | | | | | | | | In order to enable the preservation of noalias function parameter information after inlining, and the representation of block-level __restrict__ pointer information (etc.), additional kinds of aliasing metadata will be introduced. This metadata needs to be carried around in AliasAnalysis::Location objects (and MMOs at the SDAG level), and so we need to generalize the current scheme (which is hard-coded to just one TBAA MDNode*). This commit introduces only the necessary refactoring to allow for the introduction of other aliasing metadata types, but does not actually introduce any (that will come in a follow-up commit). What it does introduce is a new AAMDNodes structure to hold all of the aliasing metadata nodes associated with a particular memory-accessing instruction, and uses that structure instead of the raw MDNode* in AliasAnalysis::Location, etc. No functionality change intended. llvm-svn: 213859
OpenPOWER on IntegriCloud