aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms
AgeCommit message (Collapse)Author
2010-01-02Teach the table lookup optimization to generate range comparesChris Lattner
when a consequtive sequence of elements all satisfies the predicate. Like the double compare case, this generates better code than the magic constant case and generalizes to more than 32/64 element array lookups. Here are some examples where it triggers. From 403.gcc, most accesses to the rtx_class array are handled, e.g.: @rtx_class = constant [153 x i8] c"xxxxxmmmmmmmmxxxxxxxxxxxxmxxxxxxiiixxxxxxxxxxxxxxxxxxxooxooooooxxoooooox3x2c21c2222ccc122222ccccaaaaaa<<<<<<<<<<<<<<<<<<111111111111bbooxxxxxxxxxxcc2211x", align 32 ; <[153 x i8]*> [#uses=547] %142 = icmp eq i8 %141, 105 @rtx_class = constant [153 x i8] c"xxxxxmmmmmmmmxxxxxxxxxxxxmxxxxxxiiixxxxxxxxxxxxxxxxxxxooxooooooxxoooooox3x2c21c2222ccc122222ccccaaaaaa<<<<<<<<<<<<<<<<<<111111111111bbooxxxxxxxxxxcc2211x", align 32 ; <[153 x i8]*> [#uses=543] %165 = icmp eq i8 %164, 60 Also, most of the 59-element arrays (mode_class/rid_to_yy, etc) optimized before are actually range compares. This lets 32-bit machines optimize them. 400.perlbmk has stuff like this: 400.perlbmk: PL_regkind, even for 32-bit: @PL_regkind = constant [62 x i8] c"\00\00\02\02\02\06\06\06\06\09\09\0B\0B\0D\0E\0E\0E\11\12\12\14\14\16\16\18\18\1A\1A\1C\1C\1E\1F !!!$$&'((((,-.///88886789:;8$", align 32 ; <[62 x i8]*> [#uses=4] %811 = icmp ne i8 %810, 33 @PL_utf8skip = constant [256 x i8] c"\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\01\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\02\03\03\03\03\03\03\03\03\03\03\03\03\03\03\03\03\04\04\04\04\04\04\04\04\05\05\05\05\06\06\07\0D", align 32 ; <[256 x i8]*> [#uses=94] %12 = icmp ult i8 %10, 2 etc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92426 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02theoretically the negate we find could be in a different function, checkChris Lattner
for this case. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92425 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02use enums for the over/underdefined markers for clarity. SwitchChris Lattner
to using -2/-3 instead of -1/-2 for a future xform. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92423 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02remove the random sampling framework, which is not maintained anymore.Chris Lattner
If there is interest, it can be resurrected from SVN. PR4912. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92422 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02Fix logic error in previous commit. The != case needs to become an or, not anNick Lewycky
and. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92419 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02Optimize pointer comparison into the typesafe form, now that the backends willNick Lewycky
handle them efficiently. This is the opposite direction of the transformation we used to have here. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92418 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02Generalize the previous xform to handle cases where exactlyChris Lattner
two elements match or don't match with two comparisons. For example, the testcase compiles into: define i1 @test5(i32 %X) { %1 = icmp eq i32 %X, 2 ; <i1> [#uses=1] %2 = icmp eq i32 %X, 7 ; <i1> [#uses=1] %R = or i1 %1, %2 ; <i1> [#uses=1] ret i1 %R } This generalizes the previous xforms when the array is larger than 64 elements (and this case matches) and generates better code for cases where it overlaps with the magic bitshift case. This generalizes more cases than you might expect. For example, 400.perlbmk has: @PL_utf8skip = constant [256 x i8] c"\01\01\01\... %15 = icmp ult i8 %7, 7 403.gcc has: @rid_to_yy = internal constant [114 x i16] [i16 259, i16 260, ... %18 = icmp eq i16 %16, 295 and xalancbmk has a bunch of examples, such as _ZN11xercesc_2_5L15gCombiningCharsE and _ZN11xercesc_2_5L10gBaseCharsE. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92417 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02fix a miscompilation I introduced of cdecl with a late change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92416 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02enhance the compare/load/index optimization to work on *any* loadChris Lattner
from a global with 32/64 elements or less (depending on whether i64 is native on the target), generating a bitshift idiom to determine the result. For example, on test4 we produce: define i1 @test4(i32 %X) { %1 = lshr i32 933, %X ; <i32> [#uses=1] %2 = and i32 %1, 1 ; <i32> [#uses=1] %R = icmp ne i32 %2, 0 ; <i1> [#uses=1] ret i1 %R } This triggers in a number of interesting cases, for example, here's an fp case: @A.3255 = internal constant [4 x double] [double 4.100000e+00, double -3.900000e+00, double -1.000000e+00, double 1.000000e+00], align 32 ; <[4 x double]*> [#uses=7] ... %7 = fcmp olt double %3, 0.000000e+00 In this case we make the slen2_tab global dead, which is nice: @slen2_tab = internal constant [16 x i32] [i32 0, i32 1, i32 2, i32 3, i32 0, i32 1, i32 2, i32 3, i32 1, i32 2, i32 3, i32 1, i32 2, i32 3, i32 2, i32 3], align 32 ; <[16 x i32]*> [#uses=1] ... %204 = icmp eq i32 %46, 0 Perl has a bunch of these, also on the 'Perl_regkind' array: @Perl_yygindex = internal constant [51 x i16] [i16 0, i16 0, i16 0, i16 0, i16 374, i16 351, i16 0, i16 -12, i16 0, i16 946, i16 413, i16 -83, i16 0, i16 0, i16 0, i16 -311, i16 -13, i16 4007, i16 2893, i16 0, i16 0, i16 0, i16 0, i16 0, i16 372, i16 -8, i16 0, i16 0, i16 246, i16 -131, i16 43, i16 86, i16 208, i16 -45, i16 -169, i16 987, i16 0, i16 0, i16 0, i16 0, i16 308, i16 0, i16 -271, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0, i16 0], align 32 ; <[51 x i16]*> [#uses=1] ... %1364 = icmp eq i16 %1361, 0 186.crafty really likes this on 64-bit machines, because it triggers on a bunch of globals like this: @white_outpost = internal constant [64 x i8] c"\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\02\02\00\00\00\00\00\04\05\05\04\00\00\00\00\03\06\06\03\00\00\00\00\00\01\01\00\00\00\00\00\00\00\00\00\00\00", align 32 ; <[64 x i8]*> [#uses=2] However the big winner is 403.gcc, which triggers hundreds of times, eliminating all the accesses to the 57-element arrays 'mode_class', mode_unit_size, mode_bitsize, regclass_map, etc. go 64-bit machines :) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92415 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02enhance the previous optimization to work with fcmp in additionChris Lattner
to icmp. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92412 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02Teach instcombine to fold compares of loads from constantChris Lattner
arrays with variable indices into a comparison of the index with a constant. The most common occurrence of this that I see by far is stuff like: if ("foobar"[i] == '\0') ... which we compile into: if (i == 6), saving a load and materialization of the global address. This also exposes loop trip count information to later passes in many cases. This triggers hundreds of times in xalancbmk, which is where I first noticed it, but it also triggers in many other apps. Here are a few interesting ones from various apps: @must_be_connected_without = internal constant [8 x i8*] [i8* getelementptr inbounds ([3 x i8]* @.str64320, i64 0, i64 0), i8* getelementptr inbounds ([3 x i8]* @.str27283, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str71327, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str72328, i64 0, i64 0), i8* getelementptr inbounds ([3 x i8]* @.str18274, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8]* @.str11267, i64 0, i64 0), i8* getelementptr inbounds ([3 x i8]* @.str32288, i64 0, i64 0), i8* null], align 32 ; <[8 x i8*]*> [#uses=2] %scevgep.i = getelementptr [8 x i8*]* @must_be_connected_without, i64 0, i64 %indvar.i ; <i8**> [#uses=1] %17 = load ... %18 = icmp eq i8* %17, null ; <i1> [#uses=1] -> icmp eq i64 %indvar.i, 7 @yytable1095 = internal constant [84 x i8] c"\12\01(\05\06\07\08\09\0A\0B\0C\0D\0E1\0F\10\11266\1D: \10\11,-,0\03'\10\11B6\04\17&\18\1945\05\06\07\08\09\0A\0B\0C\0D\0E\1E\0F\10\11*\1A\1B\1C$3+>#%;<IJ=ADFEGH9KL\00\00\00C", align 32 ; <[84 x i8]*> [#uses=2] %57 = getelementptr inbounds [84 x i8]* @yytable1095, i64 0, i64 %56 ; <i8*> [#uses=1] %mode.0.in = getelementptr inbounds [9 x i32]* @mb_mode_table, i64 0, i64 %.pn ; <i32*> [#uses=1] load ... %64 = icmp eq i8 %58, 4 ; <i1> [#uses=1] -> icmp eq i64 %.pn, 35 ; <i1> [#uses=0] @gsm_DLB = internal constant [4 x i16] [i16 6554, i16 16384, i16 26214, i16 32767] %scevgep.i = getelementptr [4 x i16]* @gsm_DLB, i64 0, i64 %indvar.i ; <i16*> [#uses=1] %425 = load %scevgep.i %426 = icmp eq i16 %425, -32768 ; <i1> [#uses=0] -> false git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92411 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-02remove the instcombine transformations that are inserting nastyChris Lattner
pointer to int casts that confuse later optimizations. See PR3351 for details. This improves but doesn't complete fix 483.xalancbmk because llvm-gcc does this xform in GCC's "fold" routine as well. Clang++ will do better I guess. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92408 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01add a simple instcombine xform, simplify another one to use hasAllZeroIndices()Chris Lattner
instead of hand rolling a loop. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92403 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01generalize the pointer difference optimization to handleChris Lattner
a constantexpr gep on the 'base' side of the expression. This completes comment #4 in PR3351, which comes from 483.xalancbmk. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92402 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01teach instcombine to optimize pointer difference idioms involving constantChris Lattner
expressions. This is a step towards comment #4 in PR3351. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92401 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01use 'match' to simplify some code.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92400 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01implement the transform requested in PR5284Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92398 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01add missing line.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92384 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01add a few trivial instcombines for llvm.powi.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92383 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01When factoring multiply expressions across adds, factor both Chris Lattner
positive and negative forms of constants together. This allows us to compile: int foo(int x, int y) { return (x-y) + (x-y) + (x-y); } into: _foo: ## @foo subl %esi, %edi leal (%rdi,%rdi,2), %eax ret instead of (where the 3 and -3 were not factored): _foo: imull $-3, 8(%esp), %ecx imull $3, 4(%esp), %eax addl %ecx, %eax ret this started out as: movl 12(%ebp), %ecx imull $3, 8(%ebp), %eax subl %ecx, %eax subl %ecx, %eax subl %ecx, %eax ret This comes from PR5359. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92381 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01clean up some comments.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92377 91177308-0d34-0410-b5e6-96231b3b80d8
2010-01-01switch from std::map to DenseMap for rank data structures.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92375 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31reuse negates where possible instead of always creating them from scratch.Chris Lattner
This allows us to optimize test12 into: define i32 @test12(i32 %X) { %factor = mul i32 %X, -3 ; <i32> [#uses=1] %Z = add i32 %factor, 6 ; <i32> [#uses=1] ret i32 %Z } instead of: define i32 @test12(i32 %X) { %Y = sub i32 6, %X ; <i32> [#uses=1] %C = sub i32 %Y, %X ; <i32> [#uses=1] %Z = sub i32 %C, %X ; <i32> [#uses=1] ret i32 %Z } git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92373 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31we don't need a smallptrset to detect duplicates, the values areChris Lattner
sorted, so we can just do a linear scan. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92372 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31make reassociate more careful about not leaving around dead mul'sChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92370 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31remove debugChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92369 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31teach reassociate to factor x+x+x -> x*3. While I'm at it,Chris Lattner
fix RemoveDeadBinaryOp to actually do something. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92368 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31change reassociate to use SmallVector for its key datastructuresChris Lattner
instead of std::vector. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92366 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31change an if to an assert, fix comment.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92364 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31move the rest of the add optimization code out to OptimizeAdd,Chris Lattner
improve some comments, simplify a bit of code. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92363 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31factor statistic updating better.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92362 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31simple fix for an incorrect factoring which causes aChris Lattner
miscompilation, PR5458. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92354 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31factor code out into helper functions.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92347 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31switch some std::vector's to smallvector. Reduce nesting.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92346 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31use more modern datastructures.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92344 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31clean up -debug output.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92343 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31fix Analysis/DebugInfo.h to not include Metadata.h. Do thisChris Lattner
by moving one method out of line and eliminating redundant checks from other methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92337 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-31rename "elements" of metadata to "operands". "Elements" areChris Lattner
things that occur in types. "operands" are things that occur in values. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92322 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-29Use an array instead of a SmallVector.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92264 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-29prune #includes.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92260 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-29Final step in the metadata API restructuring: move the Chris Lattner
getMDKindID/getMDKindNames methods to LLVMContext (and add convenience methods to Module), eliminating MetadataContext. Move the state that it maintains out to LLVMContext. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92259 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-29remove useless argument.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92256 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-28This is a major cleanup of the instruction metadata interfaces thatChris Lattner
I asked Devang to do back on Sep 27. Instead of going through the MetadataContext class with methods like getMD() and getMDs(), just ask the instruction directly for its metadata with getMetadata() and getAllMetadata(). This includes a variety of other fixes and improvements: previously all Value*'s were bloated because the HasMetadata bit was thrown into value, adding a 9th bit to a byte. Now this is properly sunk down to the Instruction class (the only place where it makes sense) and it will be folded away somewhere soon. This also fixes some confusion in getMDs and its clients about whether the returned list is indexed by the MDID or densely packed. This is now returned sorted and densely packed and the comments make this clear. This introduces a number of fixme's which I'll follow up on. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92235 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-28split code that doesn't need to be templated out of IRBuilder into a newChris Lattner
non-templated IRBuilderBase class. Move that large CreateGlobalString out of line, eliminating the need to #include GlobalVariable.h in IRBuilder.h git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92227 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-28rename getMDKind -> getMDKindID, make it autoinsert if an MD KindChris Lattner
doesn't exist already, eliminate registerMDKind. Tidy up a bunch of random stuff. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92225 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-28Metadata.h doesn't need to include ValueHandle.h anymore.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92211 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-24move an optimization for memcmp out of simplifylibcalls and into Chris Lattner
SDISel. This optimization was causing simplifylibcalls to introduce type-unsafe nastiness. This is the first step, I'll be expanding the memcmp optimizations shortly, covering things that we really really wouldn't want simplifylibcalls to do. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92098 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-23reorder to follow a normal fall-through style, no functionality change.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92084 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-23Remove dump routine and the associated Debug.h from a header. Patch upDavid Greene
other files to compensate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@92075 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-23Update objectsize intrinsic and associated dependencies. FixEric Christopher
lowering code and update testcases. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@91979 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-09-27 07:58:38 +0000