RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-28 22:42:52 +01:00
Code Issues Projects Releases Wiki Activity
43,725 Commits 4 Branches 5 Tags 704 MiB
0030a3f5d4
Commit Graph

228 Commits

Author SHA1 Message Date
Chris Lattner
87c89cafb2 switch LSR to use isLegalAddressingMode instead of other simpler hooks 
llvm-svn: 35837
 2007-04-09 22:20:14 +00:00
Owen Anderson
85b0e20f2a Completely purge DomSet. This is the (hopefully) final patch for PR1171. 
llvm-svn: 35731
 2007-04-07 07:17:27 +00:00
Chris Lattner
4302c8bcdb split some code out into a helper function 
llvm-svn: 35615
 2007-04-03 05:11:24 +00:00
Chris Lattner
3df2382275 allow -1 strides to reuse "1" strides. 
llvm-svn: 35607
 2007-04-02 22:51:58 +00:00
Chris Lattner
d7c96e25e4 Pass the type of the store access, not the type of the store, into the 
target hook.  This allows us to codegen a loop as:

LBB1_1: @cond_next
        mov r2, #0
        str r2, [r0, +r3, lsl #2]
        add r3, r3, #1
        cmn r3, #1
        bne LBB1_1      @cond_next

instead of:

LBB1_1: @cond_next
        mov r2, #0
        str r2, [r0], #+4
        add r3, r3, #1
        cmn r3, #1
        bne LBB1_1      @cond_next

This looks the same, but has one fewer induction variable (and therefore,
one fewer register) live in the loop.

llvm-svn: 35592
 2007-04-02 06:34:44 +00:00
Chris Lattner
ac77daac29 print the type of an inserted IV in -debug mode. 
llvm-svn: 35563
 2007-04-01 22:21:39 +00:00
Dale Johannesen
43fc14d6e4 Look through bitcast when finding IVs. (Chris' patch really.) 
llvm-svn: 35347
 2007-03-26 03:01:27 +00:00
Dale Johannesen
3e422e3b49 do not share old induction variables when this would result in invalid 
instructions (that would have to be split later)

llvm-svn: 35227
 2007-03-20 21:54:54 +00:00
Jeff Cohen
8dea6ada6f Fix some VC++ warnings. 
llvm-svn: 35224
 2007-03-20 20:43:18 +00:00
Dale Johannesen
516d98c710 use types of loads and stores, not address, in CheckForIVReuse 
llvm-svn: 35197
 2007-03-20 00:47:50 +00:00
Evan Cheng
bd964bd8eb Correct type info for isLegalAddressImmediate() check. 
llvm-svn: 35086
 2007-03-13 20:34:37 +00:00
Evan Cheng
1fcdd7e1e5 Use new TargetLowering addressing modes hooks. 
llvm-svn: 35072
 2007-03-12 23:27:37 +00:00
Devang Patel
ca60446cb4 Increment iterator now because IVUseShouldUsePostIncValue may remove 
User from the list of I users.

llvm-svn: 35051
 2007-03-09 21:19:53 +00:00
Devang Patel
0b82378a28 Now LoopStrengthReduce is a LoopPass. 
llvm-svn: 34984
 2007-03-06 21:14:09 +00:00
Reid Spencer
6f7b9afc1c Finally get this patch right :) 
Replace expensive getZExtValue() == 0 calls with isZero() calls.

llvm-svn: 34861
 2007-03-02 23:51:25 +00:00
Reid Spencer
438d17b5ad Dang, I've done that twice now! Undo previous commit. 
llvm-svn: 34860
 2007-03-02 23:37:53 +00:00
Reid Spencer
70fbfc75e8 Use more efficient test for one value in a ConstantInt. 
llvm-svn: 34859
 2007-03-02 23:35:28 +00:00
Reid Spencer
2814c17704 Reverse a premature commital. 
llvm-svn: 34822
 2007-03-02 00:31:39 +00:00
Reid Spencer
1f970769e8 Prefer non-virtual calls to ConstantInt::isZero over virtual calls to 
Constant::isNullValue() in situations where it is possible.

llvm-svn: 34821
 2007-03-02 00:28:52 +00:00
Chris Lattner
9dafd9fe90 Privatize StructLayout::MemberOffsets, adding an accessor 
llvm-svn: 34156
 2007-02-10 19:55:17 +00:00
Reid Spencer
8749d297f5 Apply the VISIBILITY_HIDDEN field to the remaining anonymous classes in 
the Transforms library. This reduces debug library size by 132 KB, debug
binary size by 376 KB, and reduces link time for llvm tools slightly.

llvm-svn: 33939
 2007-02-05 23:32:05 +00:00
Chris Lattner
c5e1611848 rename Type::isIntegral to Type::isInteger, eliminating the old Type::isInteger. 
rename Type::getIntegralTypeMask to Type::getIntegerTypeMask.

This makes naming much more consistent.  For example, there are now no longer any
instances of IntegerType that are not considered isInteger! :)

llvm-svn: 33225
 2007-01-15 02:27:26 +00:00
Chris Lattner
8fbcb083aa Eliminate calls to isInteger, generalizing code and tightening checks as needed. 
llvm-svn: 33218
 2007-01-15 01:55:30 +00:00
Reid Spencer
405e204fb3 For PR1097: 
Enable complex addressing modes on 64-bit platforms involving two induction
variables by keeping a size and scale in 64-bits not 32.
Patch by Dan Gohman.

llvm-svn: 33011
 2007-01-08 16:17:51 +00:00
Chris Lattner
176b690f07 no need to worry about int vs uint any more. 
llvm-svn: 32946
 2007-01-06 01:37:35 +00:00
Reid Spencer
0fb7fa82d4 For PR950: 
This patch replaces signed integer types with signless ones:
1. [US]Byte -> Int8
2. [U]Short -> Int16
3. [U]Int   -> Int32
4. [U]Long  -> Int64.
5. Removal of isSigned, isUnsigned, getSignedVersion, getUnsignedVersion
   and other methods related to signedness. In a few places this warranted
   identifying the signedness information from other sources.

llvm-svn: 32785
 2006-12-31 05:48:39 +00:00
Reid Spencer
4428c3483b For PR950: 
This patch removes the SetCC instructions and replaces them with the ICmp
and FCmp instructions. The SetCondInst instruction has been removed and
been replaced with ICmpInst and FCmpInst.

llvm-svn: 32751
 2006-12-23 06:05:41 +00:00
Chris Lattner
9e7640cf02 Switch over Transforms/Scalar to use the STATISTIC macro. For each statistic 
converted, we lose a static initializer.  This also allows GCC to emit warnings
about unused statistics.

llvm-svn: 32690
 2006-12-19 21:40:18 +00:00
Reid Spencer
ad772bfec3 Change the interface to SCEVExpander::InsertCastOfTo to take a cast opcode 
so the decision of which opcode to use is pushed upward to the caller.
Adjust the callers to pass the expected opcode.

llvm-svn: 32535
 2006-12-13 08:06:42 +00:00
Reid Spencer
562b83c7df Change inferred getCast into specific getCast. Passes all tests. 
llvm-svn: 32469
 2006-12-12 05:05:00 +00:00
Bill Wendling
a3246c4272 Changed llvm_ostream et all to OStream. llvm_cerr, llvm_cout, llvm_null, are 
now cerr, cout, and NullStream resp.

llvm-svn: 32298
 2006-12-07 01:30:32 +00:00
Chris Lattner
a531ce882e Detemplatize the Statistic class. The only type it is instantiated with 
is 'unsigned'.

llvm-svn: 32279
 2006-12-06 17:46:33 +00:00
Reid Spencer
992d9788b3 For PR950: 
The long awaited CAST patch. This introduces 12 new instructions into LLVM
to replace the cast instruction. Corresponding changes throughout LLVM are
provided. This passes llvm-test, llvm/test, and SPEC CPUINT2000 with the
exception of 175.vpr which fails only on a slight floating point output
difference.

llvm-svn: 31931
 2006-11-27 01:05:10 +00:00
Bill Wendling
1b3a86000a Removed #include <iostream> and replaced with llvm_* streams. 
llvm-svn: 31923
 2006-11-26 09:46:52 +00:00
Chris Lattner
cc4df7e0ab If an indvar with a variable stride is used by the exit condition, go ahead 
and handle it like constant stride vars.  This fixes some bad codegen in
variable stride cases.  For example, it compiles this:

void foo(int k, int i) {
  for (k=i+i; k <= 8192; k+=i)
    flags2[k] = 0;
}

to:

LBB1_1: #bb.preheader
        movl %eax, %ecx
        addl %ecx, %ecx
        movl L_flags2$non_lazy_ptr, %edx
LBB1_2: #bb
        movb $0, (%edx,%ecx)
        addl %eax, %ecx
        cmpl $8192, %ecx
        jle LBB1_2      #bb
LBB1_5: #return
        ret

or (if the array is local and we are in dynamic-nonpic or static mode):

LBB3_2: #bb
        movb $0, _flags2(%ecx)
        addl %eax, %ecx
        cmpl $8192, %ecx
        jle LBB3_2      #bb

and:

        lis r2, ha16(L_flags2$non_lazy_ptr)
        lwz r2, lo16(L_flags2$non_lazy_ptr)(r2)
        slwi r3, r4, 1
LBB1_2: ;bb
        li r5, 0
        add r6, r4, r3
        stbx r5, r2, r3
        cmpwi cr0, r6, 8192
        bgt cr0, LBB1_5 ;return

instead of:

        leal (%eax,%eax,2), %ecx
        movl %eax, %edx
        addl %edx, %edx
        addl L_flags2$non_lazy_ptr, %edx
        xorl %esi, %esi
LBB1_2: #bb
        movb $0, (%edx,%esi)
        movl %eax, %edi
        addl %esi, %edi
        addl %ecx, %esi
        cmpl $8192, %esi
        jg LBB1_5       #return

and:

        lis r2, ha16(L_flags2$non_lazy_ptr)
        lwz r2, lo16(L_flags2$non_lazy_ptr)(r2)
        mulli r3, r4, 3
        slwi r5, r4, 1
        li r6, 0
        add r2, r2, r5
LBB1_2: ;bb
        li r5, 0
        add r7, r3, r6
        stbx r5, r2, r6
        add r6, r4, r6
        cmpwi cr0, r7, 8192
        ble cr0, LBB1_2 ;bb

This speeds up Benchmarks/Shootout/sieve from 8.533s to 6.464s and
implements LoopStrengthReduce/var_stride_used_by_compare.ll

llvm-svn: 31809
 2006-11-17 06:17:33 +00:00
Reid Spencer
4bafa71dc1 For PR786: 
Turn on -Wunused and -Wno-unused-parameter. Clean up most of the resulting
fall out by removing unused variables. Remaining warnings have to do with
unused functions (I didn't want to delete code without review) and unused
variables in generated code. Maintainers should clean up the remaining
issues when they see them. All changes pass DejaGnu tests and Olden.

llvm-svn: 31380
 2006-11-02 20:25:50 +00:00
Chris Lattner
a1da382ad3 break edges more intelligently 
llvm-svn: 31257
 2006-10-28 06:45:33 +00:00
Chris Lattner
93414d06c4 prepare for a change I'm about to make 
llvm-svn: 31248
 2006-10-28 00:59:20 +00:00
Reid Spencer
d414793dbc For PR950: 
This patch implements the first increment for the Signless Types feature.
All changes pertain to removing the ConstantSInt and ConstantUInt classes
in favor of just using ConstantInt.

llvm-svn: 31063
 2006-10-20 07:07:24 +00:00
Chris Lattner
a39dcb5377 eliminate RegisterOpt. It does the same thing as RegisterPass. 
llvm-svn: 29925
 2006-08-27 22:42:52 +00:00
Chris Lattner
33bd5dcfb7 s|llvm/Support/Visibility.h|llvm/Support/Compiler.h| 
llvm-svn: 29911
 2006-08-27 12:54:02 +00:00
Chris Lattner
52419ac93e Changes: 
1. Update an obsolete comment.
  2. Make the sorting by base an explicit (though still N^2) step, so
     that the code is more clear on what it is doing.
  3. Partition uses so that uses inside the loop are handled before uses
     outside the loop.

Note that none of these changes currently changes the code inserted by LSR,
but they are a stepping stone to getting there.

This code is the result of some crazy pair programming with Nate. :)

llvm-svn: 29493
 2006-08-03 06:34:50 +00:00
Evan Cheng
725fc9e73d Only reuse a previous IV if it would not require a type conversion. 
llvm-svn: 29186
 2006-07-18 19:07:58 +00:00
Chris Lattner
496bd3fbf6 Use hidden visibility to make symbols in an anonymous namespace get 
dropped.  This shrinks libllvmgcc.dylib another 67K

llvm-svn: 28975
 2006-06-28 23:17:24 +00:00
Evan Cheng
6039769dc1 RewriteExpr, either the new PHI node of induction variable or the 
post-increment value, should be first cast to the appropriated type (to the
type of the common expr). Otherwise, the rewrite of a use based on (common +
iv) may end up with an incorrect type.

llvm-svn: 28735
 2006-06-09 00:12:42 +00:00
Reid Spencer
56aa7c79b7 Get rid of a signed/unsigned compare warning. 
llvm-svn: 27625
 2006-04-12 19:28:15 +00:00
Chris Lattner
826dbd7173 Fix spello 
llvm-svn: 27052
 2006-03-24 07:14:34 +00:00
Chris Lattner
a779db7473 silence a bogus gcc warning 
llvm-svn: 26953
 2006-03-22 17:27:24 +00:00
Evan Cheng
7955a187b1 - Fixed a bogus if condition. 
- Added more debugging info.
- Allow reuse of IV of negative stride. e.g. -4 stride == 2 * iv of -2 stride.

llvm-svn: 26841
 2006-03-18 08:03:12 +00:00
Evan Cheng
afe1ee3496 Sort StrideOrder so we can process the smallest strides first. This allows 
for more IV reuses.

llvm-svn: 26837
 2006-03-18 00:44:49 +00:00
Evan Cheng
fa1b885135 Allow users of iv / stride to be rewritten with expression that is a multiply 
of a smaller stride even if they have a common loop invariant expression part.

llvm-svn: 26828
 2006-03-17 19:52:23 +00:00
Evan Cheng
db35180b27 For each loop, keep track of all the IV expressions inserted indexed by 
stride. For a set of uses of the IV of a stride which is a multiple
of another stride, do not insert a new IV expression. Rather, reuse the
previous IV and rewrite the uses as uses of IV expression multiplied by
the factor.

e.g.
x = 0 ...; x ++
y = 0 ...; y += 4
then use of y can be rewritten as use of 4*x for x86.

llvm-svn: 26803
 2006-03-16 21:53:05 +00:00
Evan Cheng
692235499c Added target lowering hooks which LSR consults to make more intelligent 
transformation decisions.

llvm-svn: 26738
 2006-03-13 23:14:23 +00:00
Chris Lattner
e4dc660cbb Use SCEVExpander::InsertCastOfTo instead of our own code. This reduces 
#LLVM LOC, and auto-cse's cast instructions.

llvm-svn: 25974
 2006-02-04 09:52:43 +00:00
Chris Lattner
8741126ea7 Fix two significant bugs in LSR: 
1. When rewriting code in outer loops, sometimes we would insert code into
   inner loops that is invariant in that loop.
2. Notice that 4*(2+x) is 8+4*x and use that to simplify expressions.

This is a performance neutral change.

llvm-svn: 25964
 2006-02-04 07:36:50 +00:00
Chris Lattner
2588f0eb8f Make iostream #inclusion explicit 
llvm-svn: 25514
 2006-01-22 23:32:06 +00:00
Chris Lattner
7b7fee2d92 Switch these to using ETForest instead of DominatorSet to compute itself. 
Patch written by Daniel Berlin!

llvm-svn: 25202
 2006-01-11 05:10:20 +00:00
Chris Lattner
3d993f7e4d getRawValue zero extens for unsigned values, use getsextvalue so that we 
know that small negative values fit into the immediate field of addressing
modes.

llvm-svn: 24608
 2005-12-05 18:23:57 +00:00
Chris Lattner
63aa74e804 My previous patch was too conservative. Reject FP and void types, but do 
allow pointer types.

llvm-svn: 23859
 2005-10-21 05:45:41 +00:00
Chris Lattner
3451ac7691 Do NOT touch FP ops with LSR. This fixes a testcase Nate sent me from an 
inner loop like this:

LBB_RateConvertMono8AltiVec_2:  ; no_exit
        lis r2, ha16(.CPI_RateConvertMono8AltiVec_0)
        lfs f3, lo16(.CPI_RateConvertMono8AltiVec_0)(r2)
        fmr f3, f3
        fadd f0, f2, f0
        fadd f3, f0, f3
        fcmpu cr0, f3, f1
        bge cr0, LBB_RateConvertMono8AltiVec_2  ; no_exit

to an inner loop like this:

LBB_RateConvertMono8AltiVec_1:  ; no_exit
        fsub f2, f2, f1
        fcmpu cr0, f2, f1
        fmr f0, f2
        bge cr0, LBB_RateConvertMono8AltiVec_1  ; no_exit

Doh! good catch!

llvm-svn: 23838
 2005-10-20 04:47:10 +00:00
Chris Lattner
23ecce5c82 Fix (hopefully the last) issue where LSR is nondeterminstic. When pulling 
out CSE's of base expressions it could build a result whose order was
nondet.

llvm-svn: 23698
 2005-10-11 18:41:04 +00:00
Chris Lattner
7effc54496 Fix another problem where LSR was being nondeterminstic. Also remove elements 
from the end of a vector instead of the beginning

llvm-svn: 23697
 2005-10-11 18:30:57 +00:00
Chris Lattner
1ec477da8f Fix another lsr-is-nondeterministic case 
llvm-svn: 23695
 2005-10-11 18:17:57 +00:00
Chris Lattner
f6c2411c40 Hrm, you didn't see this. 
llvm-svn: 23673
 2005-10-09 06:24:02 +00:00
Chris Lattner
6fa98d5979 Fix a source of non-determinism in the backend: the order of processing 
IV strides dependend on the pointer order of the strides in memory.
Non-determinism is bad.

llvm-svn: 23672
 2005-10-09 06:20:55 +00:00
Chris Lattner
be6f88a2e8 Make IVUseShouldUsePostIncValue more aggressive when the use is a PHI. In 
particular, it should realize that phi's use their values in the pred block
not the phi block itself.  This change turns our em3d loop from this:

_test:
        cmpwi cr0, r4, 0
        bgt cr0, LBB_test_2     ; entry.no_exit_crit_edge
LBB_test_1:     ; entry.loopexit_crit_edge
        li r2, 0
        b LBB_test_6    ; loopexit
LBB_test_2:     ; entry.no_exit_crit_edge
        li r6, 0
LBB_test_3:     ; no_exit
        or r2, r6, r6
        lwz r6, 0(r3)
        cmpw cr0, r6, r5
        beq cr0, LBB_test_6     ; loopexit
LBB_test_4:     ; endif
        addi r3, r3, 4
        addi r6, r2, 1
        cmpw cr0, r6, r4
        blt cr0, LBB_test_3     ; no_exit
LBB_test_5:     ; endif.loopexit.loopexit_crit_edge
        addi r3, r2, 1
        blr
LBB_test_6:     ; loopexit
        or r3, r2, r2
        blr

into:

_test:
        cmpwi cr0, r4, 0
        bgt cr0, LBB_test_2     ; entry.no_exit_crit_edge
LBB_test_1:     ; entry.loopexit_crit_edge
        li r2, 0
        b LBB_test_5    ; loopexit
LBB_test_2:     ; entry.no_exit_crit_edge
        li r6, 0
LBB_test_3:     ; no_exit
        lwz r2, 0(r3)
        cmpw cr0, r2, r5
        or r2, r6, r6
        beq cr0, LBB_test_5     ; loopexit
LBB_test_4:     ; endif
        addi r3, r3, 4
        addi r6, r6, 1
        cmpw cr0, r6, r4
        or r2, r6, r6
        blt cr0, LBB_test_3     ; no_exit
LBB_test_5:     ; loopexit
        or r3, r2, r2
        blr


Unfortunately, this is actually worse code, because the register coallescer
is getting confused somehow.  If it were doing its job right, it could turn the
code into this:

_test:
        cmpwi cr0, r4, 0
        bgt cr0, LBB_test_2     ; entry.no_exit_crit_edge
LBB_test_1:     ; entry.loopexit_crit_edge
        li r6, 0
        b LBB_test_5    ; loopexit
LBB_test_2:     ; entry.no_exit_crit_edge
        li r6, 0
LBB_test_3:     ; no_exit
        lwz r2, 0(r3)
        cmpw cr0, r2, r5
        beq cr0, LBB_test_5     ; loopexit
LBB_test_4:     ; endif
        addi r3, r3, 4
        addi r6, r6, 1
        cmpw cr0, r6, r4
        blt cr0, LBB_test_3     ; no_exit
LBB_test_5:     ; loopexit
        or r3, r6, r6
        blr

... which I'll work on next. :)

llvm-svn: 23604
 2005-10-03 02:50:05 +00:00
Chris Lattner
6a5ace34da Refactor some code into a function 
llvm-svn: 23603
 2005-10-03 01:04:44 +00:00
Chris Lattner
5a865c8598 This break is bogus and I have no idea why it was there. Basically it prevents 
memoizing code when IV's are used by phinodes outside of loops.  In a simple
example, we were getting this code before (note that r6 and r7 are isomorphic
IV's):

        li r6, 0
        or r7, r6, r6
LBB_test_3:     ; no_exit
        lwz r2, 0(r3)
        cmpw cr0, r2, r5
        or r2, r7, r7
        beq cr0, LBB_test_5     ; loopexit
LBB_test_4:     ; endif
        addi r2, r7, 1
        addi r7, r7, 1
        addi r3, r3, 4
        addi r6, r6, 1
        cmpw cr0, r6, r4
        blt cr0, LBB_test_3     ; no_exit

Now we get:

        li r6, 0
LBB_test_3:     ; no_exit
        or r2, r6, r6
        lwz r6, 0(r3)
        cmpw cr0, r6, r5
        beq cr0, LBB_test_6     ; loopexit
LBB_test_4:     ; endif
        addi r3, r3, 4
        addi r6, r2, 1
        cmpw cr0, r6, r4
        blt cr0, LBB_test_3     ; no_exit

this was noticed in em3d.

llvm-svn: 23602
 2005-10-03 00:37:33 +00:00
Chris Lattner
af938ddbe5 when checking if we should move a split edge block outside of a loop, 
check the presplit pred, not the post-split pred.  This was causing us
to make the wrong decision in some cases, leaving the critical edge block
in the loop.

llvm-svn: 23601
 2005-10-03 00:31:52 +00:00
Chris Lattner
29f2ba1384 Make the pass name simpler 
llvm-svn: 23476
 2005-09-27 21:10:32 +00:00
Chris Lattner
26aef8992f Fix an issue where LSR would miss rewriting a use of an IV expression by a PHI node that is not the original PHI. 
This fixes up a dot-product loop in galgel, speeding it up from 18.47s to
16.13s.

llvm-svn: 23327
 2005-09-13 02:09:55 +00:00
Chris Lattner
a949a2836d Fix a regression from last night, which caused this pass to create invalid 
code for IV uses outside of loops that are not dominated by the latch block.
We should only convert these uses to use the post-inc value if they ARE
dominated by the latch block.

Also use a new LoopInfo method to simplify some code.

This fixes Transforms/LoopStrengthReduce/2005-09-12-UsesOutOutsideOfLoop.ll

llvm-svn: 23318
 2005-09-12 17:11:27 +00:00
Chris Lattner
780ffd9c1f _test: 
li r2, 0
LBB_test_1:     ; no_exit.2
        li r5, 0
        stw r5, 0(r3)
        addi r2, r2, 1
        addi r3, r3, 4
        cmpwi cr0, r2, 701
        blt cr0, LBB_test_1     ; no_exit.2
LBB_test_2:     ; loopexit.2.loopexit
        addi r2, r2, 1
        stw r2, 0(r4)
        blr
[zion ~/llvm]$ cat > ~/xx
Uses of IV's outside of the loop should use hte post-incremented version
of the IV, not the preincremented version.  This helps many loops (e.g. in sixtrack)
which used to generate code like this (this is the code from the
dont-hoist-simple-loop-constants.ll testcase):

_test:
        li r2, 0                 **** IV starts at 0
LBB_test_1:     ; no_exit.2
        or r5, r2, r2            **** Copy for loop exit
        li r2, 0
        stw r2, 0(r3)
        addi r3, r3, 4
        addi r2, r5, 1
        addi r6, r5, 2           **** IV+2
        cmpwi cr0, r6, 701
        blt cr0, LBB_test_1     ; no_exit.2
LBB_test_2:     ; loopexit.2.loopexit
        addi r2, r5, 2       ****  IV+2
        stw r2, 0(r4)
        blr

And now generated code like this:

_test:
        li r2, 1               *** IV starts at 1
LBB_test_1:     ; no_exit.2
        li r5, 0
        stw r5, 0(r3)
        addi r2, r2, 1
        addi r3, r3, 4
        cmpwi cr0, r2, 701     *** IV.postinc + 0
        blt cr0, LBB_test_1
LBB_test_2:     ; loopexit.2.loopexit
        stw r2, 0(r4)          *** IV.postinc + 0
        blr

llvm-svn: 23313
 2005-09-12 06:04:47 +00:00
Chris Lattner
ddec75fdf6 implement Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll. 
We used to emit this code for it:

_test:
        li r2, 1     ;; Value tying up a register for the whole loop
        li r5, 0
LBB_test_1:     ; no_exit.2
        or r6, r5, r5
        li r5, 0
        stw r5, 0(r3)
        addi r5, r6, 1
        addi r3, r3, 4
        add r7, r2, r5  ;; should be addi r7, r5, 1
        cmpwi cr0, r7, 701
        blt cr0, LBB_test_1     ; no_exit.2
LBB_test_2:     ; loopexit.2.loopexit
        addi r2, r6, 2
        stw r2, 0(r4)
        blr

now we emit this:

_test:
        li r2, 0
LBB_test_1:     ; no_exit.2
        or r5, r2, r2
        li r2, 0
        stw r2, 0(r3)
        addi r3, r3, 4
        addi r2, r5, 1
        addi r6, r5, 2   ;; whoa, fold those adds!
        cmpwi cr0, r6, 701
        blt cr0, LBB_test_1     ; no_exit.2
LBB_test_2:     ; loopexit.2.loopexit
        addi r2, r5, 2
        stw r2, 0(r4)
        blr

more improvement coming.

llvm-svn: 23306
 2005-09-10 01:18:45 +00:00
Chris Lattner
a852093954 Fix Transforms/LoopStrengthReduce/2005-08-17-OutOfLoopVariant.ll, a crash 
on 177.mesa

llvm-svn: 22843
 2005-08-17 21:22:41 +00:00
Chris Lattner
969232d2ec Use a new helper to split critical edges, making the code simpler. 
Do not claim to not change the CFG.  We do change the cfg to split critical
edges.  This isn't causing us a problem now, but could likely do so in the
future.

llvm-svn: 22824
 2005-08-17 06:35:16 +00:00
Chris Lattner
bc70c99aef Fix a bad case in gzip where we put lots of things in registers across the 
loop, because a IV-dependent value was used outside of the loop and didn't
have immediate-folding capability

llvm-svn: 22798
 2005-08-16 00:38:11 +00:00
Chris Lattner
57a2a74e99 Ooops, don't forget to clear this. The real inner loop is now: 
.LBB_foo_3:     ; no_exit.1
        lfd f2, 0(r9)
        lfd f3, 8(r9)
        fmul f4, f1, f2
        fmadd f4, f0, f3, f4
        stfd f4, 8(r9)
        fmul f3, f1, f3
        fmsub f2, f0, f2, f3
        stfd f2, 0(r9)
        addi r9, r9, 16
        addi r8, r8, 1
        cmpw cr0, r8, r4
        ble .LBB_foo_3  ; no_exit.1

llvm-svn: 22782
 2005-08-13 07:42:01 +00:00
Chris Lattner
f59e855dbc Recursively scan scev expressions for common subexpressions. This allows us 
to handle nested loops much better, for example, by being able to tell that
these two expressions:

{( 8 + ( 16 * ( 1 +  %Tmp11 +  %Tmp12)) +  %c_),+,( 16 *  %Tmp 12)}<loopentry.1>

{(( 16 * ( 1 +  %Tmp11 +  %Tmp12)) +  %c_),+,( 16 *  %Tmp12)}<loopentry.1>

Have the following common part that can be shared:
{(( 16 * ( 1 +  %Tmp11 +  %Tmp12)) +  %c_),+,( 16 *  %Tmp12)}<loopentry.1>

This allows us to codegen an important inner loop in 168.wupwise as:

.LBB_foo_4:     ; no_exit.1
        lfd f2, 16(r9)
        fmul f3, f0, f2
        fmul f2, f1, f2
        fadd f4, f3, f2
        stfd f4, 8(r9)
        fsub f2, f3, f2
        stfd f2, 16(r9)
        addi r8, r8, 1
        addi r9, r9, 16
        cmpw cr0, r8, r4
        ble .LBB_foo_4  ; no_exit.1

instead of:

.LBB_foo_3:     ; no_exit.1
        lfdx f2, r6, r9
        add r10, r6, r9
        lfd f3, 8(r10)
        fmul f4, f1, f2
        fmadd f4, f0, f3, f4
        stfd f4, 8(r10)
        fmul f3, f1, f3
        fmsub f2, f0, f2, f3
        stfdx f2, r6, r9
        addi r9, r9, 16
        addi r8, r8, 1
        cmpw cr0, r8, r4
        ble .LBB_foo_3  ; no_exit.1

llvm-svn: 22781
 2005-08-13 07:27:18 +00:00
Chris Lattner
811ef4cce0 When splitting critical edges, make sure not to leave the new block in the 
middle of the loop.  This turns a critical loop in gzip into this:

.LBB_test_1:    ; loopentry
        or r27, r28, r28
        add r28, r3, r27
        lhz r28, 3(r28)
        add r26, r4, r27
        lhz r26, 3(r26)
        cmpw cr0, r28, r26
        bne .LBB_test_8 ; loopentry.loopexit_crit_edge
.LBB_test_2:    ; shortcirc_next.0
        add r28, r3, r27
        lhz r28, 5(r28)
        add r26, r4, r27
        lhz r26, 5(r26)
        cmpw cr0, r28, r26
        bne .LBB_test_7 ; shortcirc_next.0.loopexit_crit_edge
.LBB_test_3:    ; shortcirc_next.1
        add r28, r3, r27
        lhz r28, 7(r28)
        add r26, r4, r27
        lhz r26, 7(r26)
        cmpw cr0, r28, r26
        bne .LBB_test_6 ; shortcirc_next.1.loopexit_crit_edge
.LBB_test_4:    ; shortcirc_next.2
        add r28, r3, r27
        lhz r26, 9(r28)
        add r28, r4, r27
        lhz r25, 9(r28)
        addi r28, r27, 8
        cmpw cr7, r26, r25
        mfcr r26, 1
        rlwinm r26, r26, 31, 31, 31
        add r25, r8, r27
        cmpw cr7, r25, r7
        mfcr r25, 1
        rlwinm r25, r25, 29, 31, 31
        and. r26, r26, r25
        bne .LBB_test_1 ; loopentry

instead of this:

.LBB_test_1:    ; loopentry
        or r27, r28, r28
        add r28, r3, r27
        lhz r28, 3(r28)
        add r26, r4, r27
        lhz r26, 3(r26)
        cmpw cr0, r28, r26
        beq .LBB_test_3 ; shortcirc_next.0
.LBB_test_2:    ; loopentry.loopexit_crit_edge
        add r2, r30, r27
        add r8, r29, r27
        b .LBB_test_9   ; loopexit
.LBB_test_3:    ; shortcirc_next.0
        add r28, r3, r27
        lhz r28, 5(r28)
        add r26, r4, r27
        lhz r26, 5(r26)
        cmpw cr0, r28, r26
        beq .LBB_test_5 ; shortcirc_next.1
.LBB_test_4:    ; shortcirc_next.0.loopexit_crit_edge
        add r2, r11, r27
        add r8, r12, r27
        b .LBB_test_9   ; loopexit
.LBB_test_5:    ; shortcirc_next.1
        add r28, r3, r27
        lhz r28, 7(r28)
        add r26, r4, r27
        lhz r26, 7(r26)
        cmpw cr0, r28, r26
        beq .LBB_test_7 ; shortcirc_next.2
.LBB_test_6:    ; shortcirc_next.1.loopexit_crit_edge
        add r2, r9, r27
        add r8, r10, r27
        b .LBB_test_9   ; loopexit
.LBB_test_7:    ; shortcirc_next.2
        add r28, r3, r27
        lhz r26, 9(r28)
        add r28, r4, r27
        lhz r25, 9(r28)
        addi r28, r27, 8
        cmpw cr7, r26, r25
        mfcr r26, 1
        rlwinm r26, r26, 31, 31, 31
        add r25, r8, r27
        cmpw cr7, r25, r7
        mfcr r25, 1
        rlwinm r25, r25, 29, 31, 31
        and. r26, r26, r25
        bne .LBB_test_1 ; loopentry

Next up, improve the code for the loop.

llvm-svn: 22769
 2005-08-12 22:22:17 +00:00
Chris Lattner
a5c0038c25 Fix a FIXME: if we are inserting code for a PHI argument, split the critical 
edge so that the code is not always executed for both operands.  This
prevents LSR from inserting code into loops whose exit blocks contain
PHI uses of IV expressions (which are outside of loops).  On gzip, for
example, we turn this ugly code:

.LBB_test_1:    ; loopentry
        add r27, r3, r28
        lhz r27, 3(r27)
        add r26, r4, r28
        lhz r26, 3(r26)
        add r25, r30, r28    ;; Only live if exiting the loop
        add r24, r29, r28    ;; Only live if exiting the loop
        cmpw cr0, r27, r26
        bne .LBB_test_5 ; loopexit

into this:

.LBB_test_1:    ; loopentry
        or r27, r28, r28
        add r28, r3, r27
        lhz r28, 3(r28)
        add r26, r4, r27
        lhz r26, 3(r26)
        cmpw cr0, r28, r26
        beq .LBB_test_3 ; shortcirc_next.0
.LBB_test_2:    ; loopentry.loopexit_crit_edge
        add r2, r30, r27
        add r8, r29, r27
        b .LBB_test_9   ; loopexit
.LBB_test_2:    ; shortcirc_next.0
        ...
        blt .LBB_test_1


into this:

.LBB_test_1:    ; loopentry
        or r27, r28, r28
        add r28, r3, r27
        lhz r28, 3(r28)
        add r26, r4, r27
        lhz r26, 3(r26)
        cmpw cr0, r28, r26
        beq .LBB_test_3 ; shortcirc_next.0
.LBB_test_2:    ; loopentry.loopexit_crit_edge
        add r2, r30, r27
        add r8, r29, r27
        b .LBB_t_3:    ; shortcirc_next.0
.LBB_test_3:    ; shortcirc_next.0
        ...
        blt .LBB_test_1


Next step: get the block out of the loop so that the loop is all
fall-throughs again.

llvm-svn: 22766
 2005-08-12 22:06:11 +00:00
Chris Lattner
8c7e769325 Teach LSR to strength reduce IVs that have a loop-invariant but non-constant stride. 
For code like this:

void foo(float *a, float *b, int n, int stride_a, int stride_b) {
  int i;
  for (i=0; i<n; i++)
      a[i*stride_a] = b[i*stride_b];
}

we now emit:

.LBB_foo2_2:    ; no_exit
        lfs f0, 0(r4)
        stfs f0, 0(r3)
        addi r7, r7, 1
        add r4, r2, r4
        add r3, r6, r3
        cmpw cr0, r7, r5
        blt .LBB_foo2_2 ; no_exit

instead of:

.LBB_foo_2:     ; no_exit
        mullw r8, r2, r7     ;; multiply!
        slwi r8, r8, 2
        lfsx f0, r4, r8
        mullw r8, r2, r6     ;; multiply!
        slwi r8, r8, 2
        stfsx f0, r3, r8
        addi r2, r2, 1
        cmpw cr0, r2, r5
        blt .LBB_foo_2  ; no_exit

loops with variable strides occur pretty often.  For example, in SPECFP2K
there are 317 variable strides in 177.mesa, 3 in 179.art, 14 in 188.ammp,
56 in 168.wupwise, 36 in 172.mgrid.

Now we can allow indvars to turn functions written like this:

void foo2(float *a, float *b, int n, int stride_a, int stride_b) {
  int i, ai = 0, bi = 0;
  for (i=0; i<n; i++)
    {
      a[ai] = b[bi];
      ai += stride_a;
      bi += stride_b;
    }
}

into code like the above for better analysis.  With this patch, they generate
identical code.

llvm-svn: 22740
 2005-08-10 00:45:21 +00:00
Chris Lattner
3d251b90f3 Fix Regression/Transforms/LoopStrengthReduce/phi_node_update_multiple_preds.ll 
by being more careful about updating PHI nodes

llvm-svn: 22739
 2005-08-10 00:35:32 +00:00
Chris Lattner
24f927cfe9 Fix some 80 column violations. 
Once we compute the evolution for a GEP, tell SE about it.  This allows users
of the GEP to know it, if the users are not direct.  This allows us to compile
this testcase:

void fbSolidFillmmx(int w, unsigned char *d) {
    while (w >= 64) {
        *(unsigned long long *) (d +  0) = 0;
        *(unsigned long long *) (d +  8) = 0;
        *(unsigned long long *) (d + 16) = 0;
        *(unsigned long long *) (d + 24) = 0;
        *(unsigned long long *) (d + 32) = 0;
        *(unsigned long long *) (d + 40) = 0;
        *(unsigned long long *) (d + 48) = 0;
        *(unsigned long long *) (d + 56) = 0;
        w -= 64;
        d += 64;
    }
}

into:

.LBB_fbSolidFillmmx_2:  ; no_exit
        li r2, 0
        stw r2, 0(r4)
        stw r2, 4(r4)
        stw r2, 8(r4)
        stw r2, 12(r4)
        stw r2, 16(r4)
        stw r2, 20(r4)
        stw r2, 24(r4)
        stw r2, 28(r4)
        stw r2, 32(r4)
        stw r2, 36(r4)
        stw r2, 40(r4)
        stw r2, 44(r4)
        stw r2, 48(r4)
        stw r2, 52(r4)
        stw r2, 56(r4)
        stw r2, 60(r4)
        addi r4, r4, 64
        addi r3, r3, -64
        cmpwi cr0, r3, 63
        bgt .LBB_fbSolidFillmmx_2       ; no_exit

instead of:

.LBB_fbSolidFillmmx_2:  ; no_exit
        li r11, 0
        stw r11, 0(r4)
        stw r11, 4(r4)
        stwx r11, r10, r4
        add r12, r10, r4
        stw r11, 4(r12)
        stwx r11, r9, r4
        add r12, r9, r4
        stw r11, 4(r12)
        stwx r11, r8, r4
        add r12, r8, r4
        stw r11, 4(r12)
        stwx r11, r7, r4
        add r12, r7, r4
        stw r11, 4(r12)
        stwx r11, r6, r4
        add r12, r6, r4
        stw r11, 4(r12)
        stwx r11, r5, r4
        add r12, r5, r4
        stw r11, 4(r12)
        stwx r11, r2, r4
        add r12, r2, r4
        stw r11, 4(r12)
        addi r4, r4, 64
        addi r3, r3, -64
        cmpwi cr0, r3, 63
        bgt .LBB_fbSolidFillmmx_2       ; no_exit

llvm-svn: 22737
 2005-08-09 23:39:36 +00:00
Chris Lattner
2872f369f0 SCEVAddExpr::get() of an empty list is invalid. 
llvm-svn: 22724
 2005-08-09 01:13:47 +00:00
Chris Lattner
11dd32a826 Implement: LoopStrengthReduce/share_ivs.ll 
Two changes:
  * Only insert one PHI node for each stride.  Other values are live in
    values.  This cannot introduce higher register pressure than the
    previous approach, and can take advantage of reg+reg addressing modes.
  * Factor common base values out of uses before moving values from the
    base to the immediate fields.  This improves codegen by starting the
    stride-specific PHI node out at a common place for each IV use.

As an example, we used to generate this for a loop in swim:

.LBB_main_no_exit_2E_6_2E_i_no_exit_2E_7_2E_i_2:        ; no_exit.7.i
        lfd f0, 0(r8)
        stfd f0, 0(r3)
        lfd f0, 0(r6)
        stfd f0, 0(r7)
        lfd f0, 0(r2)
        stfd f0, 0(r5)
        addi r9, r9, 1
        addi r2, r2, 8
        addi r5, r5, 8
        addi r6, r6, 8
        addi r7, r7, 8
        addi r8, r8, 8
        addi r3, r3, 8
        cmpw cr0, r9, r4
        bgt .LBB_main_no_exit_2E_6_2E_i_no_exit_2E_7_2E_i_1

now we emit:

.LBB_main_no_exit_2E_6_2E_i_no_exit_2E_7_2E_i_2:        ; no_exit.7.i
        lfdx f0, r8, r2
        stfdx f0, r9, r2
        lfdx f0, r5, r2
        stfdx f0, r7, r2
        lfdx f0, r3, r2
        stfdx f0, r6, r2
        addi r10, r10, 1
        addi r2, r2, 8
        cmpw cr0, r10, r4
        bgt .LBB_main_no_exit_2E_6_2E_i_no_exit_2E_7_2E_i_1

As another more dramatic example, we used to emit this:

.LBB_main_L_90_no_exit_2E_0_2E_i16_no_exit_2E_1_2E_i19_2:       ; no_exit.1.i19
        lfd f0, 8(r21)
        lfd f4, 8(r3)
        lfd f5, 8(r27)
        lfd f6, 8(r22)
        lfd f7, 8(r5)
        lfd f8, 8(r6)
        lfd f9, 8(r30)
        lfd f10, 8(r11)
        lfd f11, 8(r12)
        fsub f10, f10, f11
        fadd f5, f4, f5
        fmul f5, f5, f1
        fadd f6, f6, f7
        fadd f6, f6, f8
        fadd f6, f6, f9
        fmadd f0, f5, f6, f0
        fnmsub f0, f10, f2, f0
        stfd f0, 8(r4)
        lfd f0, 8(r25)
        lfd f5, 8(r26)
        lfd f6, 8(r23)
        lfd f9, 8(r28)
        lfd f10, 8(r10)
        lfd f12, 8(r9)
        lfd f13, 8(r29)
        fsub f11, f13, f11
        fadd f4, f4, f5
        fmul f4, f4, f1
        fadd f5, f6, f9
        fadd f5, f5, f10
        fadd f5, f5, f12
        fnmsub f0, f4, f5, f0
        fnmsub f0, f11, f3, f0
        stfd f0, 8(r24)
        lfd f0, 8(r8)
        fsub f4, f7, f8
        fsub f5, f12, f10
        fnmsub f0, f5, f2, f0
        fnmsub f0, f4, f3, f0
        stfd f0, 8(r2)
        addi r20, r20, 1
        addi r2, r2, 8
        addi r8, r8, 8
        addi r10, r10, 8
        addi r12, r12, 8
        addi r6, r6, 8
        addi r29, r29, 8
        addi r28, r28, 8
        addi r26, r26, 8
        addi r25, r25, 8
        addi r24, r24, 8
        addi r5, r5, 8
        addi r23, r23, 8
        addi r22, r22, 8
        addi r3, r3, 8
        addi r9, r9, 8
        addi r11, r11, 8
        addi r30, r30, 8
        addi r27, r27, 8
        addi r21, r21, 8
        addi r4, r4, 8
        cmpw cr0, r20, r7
        bgt .LBB_main_L_90_no_exit_2E_0_2E_i16_no_exit_2E_1_2E_i19_1

we now emit:

.LBB_main_L_90_no_exit_2E_0_2E_i16_no_exit_2E_1_2E_i19_2:       ; no_exit.1.i19
        lfdx f0, r21, r20
        lfdx f4, r3, r20
        lfdx f5, r27, r20
        lfdx f6, r22, r20
        lfdx f7, r5, r20
        lfdx f8, r6, r20
        lfdx f9, r30, r20
        lfdx f10, r11, r20
        lfdx f11, r12, r20
        fsub f10, f10, f11
        fadd f5, f4, f5
        fmul f5, f5, f1
        fadd f6, f6, f7
        fadd f6, f6, f8
        fadd f6, f6, f9
        fmadd f0, f5, f6, f0
        fnmsub f0, f10, f2, f0
        stfdx f0, r4, r20
        lfdx f0, r25, r20
        lfdx f5, r26, r20
        lfdx f6, r23, r20
        lfdx f9, r28, r20
        lfdx f10, r10, r20
        lfdx f12, r9, r20
        lfdx f13, r29, r20
        fsub f11, f13, f11
        fadd f4, f4, f5
        fmul f4, f4, f1
        fadd f5, f6, f9
        fadd f5, f5, f10
        fadd f5, f5, f12
        fnmsub f0, f4, f5, f0
        fnmsub f0, f11, f3, f0
        stfdx f0, r24, r20
        lfdx f0, r8, r20
        fsub f4, f7, f8
        fsub f5, f12, f10
        fnmsub f0, f5, f2, f0
        fnmsub f0, f4, f3, f0
        stfdx f0, r2, r20
        addi r19, r19, 1
        addi r20, r20, 8
        cmpw cr0, r19, r7
        bgt .LBB_main_L_90_no_exit_2E_0_2E_i16_no_exit_2E_1_2E_i19_1

llvm-svn: 22722
 2005-08-09 00:18:09 +00:00
Chris Lattner
ea50bf5aca Suck the base value out of the UsersToProcess vector into the BasedUser 
class to simplify the code.  Fuse two loops.

llvm-svn: 22721
 2005-08-08 22:56:21 +00:00
Chris Lattner
b9d13099a7 Split MoveLoopVariantsToImediateField out from MoveImmediateValues. The 
first is a correctness thing, and the later is an optzn thing.  This also
is needed to support a future change.

llvm-svn: 22720
 2005-08-08 22:32:34 +00:00
Chris Lattner
f6e6e25039 Not all constants are legal immediates in load/store instructions. 
llvm-svn: 22704
 2005-08-08 06:25:50 +00:00
Chris Lattner
ab45a77fed Implement LoopStrengthReduce/share_code_in_preheader.ll by having one 
rewriter for all code inserted into the preheader, which is never flushed.

llvm-svn: 22702
 2005-08-08 05:47:49 +00:00
Chris Lattner
dd97325bc0 Implement a simple optimization for the termination condition of the loop. 
The termination condition actually wants to use the post-incremented value
of the loop, not a new indvar with an unusual base.

On PPC, for example, this allows us to compile
LoopStrengthReduce/exit_compare_live_range.ll to:

_foo:
        li r2, 0
.LBB_foo_1:     ; no_exit
        li r5, 0
        stw r5, 0(r3)
        addi r2, r2, 1
        cmpw cr0, r2, r4
        bne .LBB_foo_1  ; no_exit
        blr

instead of:

_foo:
        li r2, 1                ;; IV starts at 1, not 0
.LBB_foo_1:     ; no_exit
        li r5, 0
        stw r5, 0(r3)
        addi r5, r2, 1
        cmpw cr0, r2, r4
        or r2, r5, r5           ;; Reg-reg copy, extra live range
        bne .LBB_foo_1  ; no_exit
        blr

This implements LoopStrengthReduce/exit_compare_live_range.ll

llvm-svn: 22699
 2005-08-08 05:28:22 +00:00
Chris Lattner
547cf56441 Make sure to clean CastedPointers after casts are potentially deleted. 
This fixes LSR crashes on 301.apsi, 191.fma3d, and 189.lucas

llvm-svn: 22673
 2005-08-05 01:30:11 +00:00
Chris Lattner
855cfc4e90 Modify how immediates are removed from base expressions to deal with the fact 
that the symbolic evaluator is not always able to use subtraction to remove
expressions.  This makes the code faster, and fixes the last crash on 178.galgel.
Finally, add a statistic to see how many phi nodes are inserted.

On 178.galgel, we get the follow stats:

2562 loop-reduce  - Number of PHIs inserted
3927 loop-reduce  - Number of GEPs strength reduced

llvm-svn: 22662
 2005-08-04 22:34:05 +00:00
Chris Lattner
07466ea612 * Refactor some code into a new BasedUser::RewriteInstructionToUseNewBase 
method.
* Fix a crash on 178.galgel, where we would insert expressions before PHI
  nodes instead of into the PHI node predecessor blocks.

llvm-svn: 22657
 2005-08-04 20:03:32 +00:00
Chris Lattner
c4beaec288 Fix a case that caused this to crash on 178.galgel 
llvm-svn: 22653
 2005-08-04 19:26:19 +00:00
Chris Lattner
5a0587224a Teach LSR about loop-variant expressions, such as loops like this: 
for (i = 0; i < N; ++i)
    A[i][foo()] = 0;

here we still want to strength reduce the A[i] part, even though foo() is
l-v.

This also simplifies some of the 'CanReduce' logic.

This implements Transforms/LoopStrengthReduce/ops_after_indvar.ll

llvm-svn: 22652
 2005-08-04 19:08:16 +00:00
Nate Begeman
0ab0a0f72d Remove some more dead code. 
llvm-svn: 22650
 2005-08-04 18:13:56 +00:00
Chris Lattner
f63b85d9c2 Refactor this code substantially with the following improvements: 
1. We only analyze instructions once, guaranteed
  2. AnalyzeGetElementPtrUsers has been ripped apart and replaced with
     something much simpler.

The next step is to handle expressions that are not all indvar+loop-invariant
values (e.g. handling indvar+loopvariant).

llvm-svn: 22649
 2005-08-04 17:40:30 +00:00
Chris Lattner
ea82ef5db1 refactor some code 
llvm-svn: 22643
 2005-08-04 01:19:13 +00:00
Chris Lattner
09241be4c7 invert to if's to make the logic simpler 
llvm-svn: 22641
 2005-08-04 00:40:47 +00:00
Chris Lattner
df7961ec73 When processing outer loops and we find uses of an IV in inner loops, make 
sure to handle the use, just don't recurse into it.

This permits us to generate this code for a simple nested loop case:

.LBB_foo_0:     ; entry
        stwu r1, -48(r1)
        stw r29, 44(r1)
        stw r30, 40(r1)
        mflr r11
        stw r11, 56(r1)
        lis r2, ha16(L_A$non_lazy_ptr)
        lwz r30, lo16(L_A$non_lazy_ptr)(r2)
        li r29, 1
.LBB_foo_1:     ; no_exit.0
        bl L_bar$stub
        li r2, 1
        or r3, r30, r30
.LBB_foo_2:     ; no_exit.1
        lfd f0, 8(r3)
        stfd f0, 0(r3)
        addi r4, r2, 1
        addi r3, r3, 8
        cmpwi cr0, r2, 100
        or r2, r4, r4
        bne .LBB_foo_2  ; no_exit.1
.LBB_foo_3:     ; loopexit.1
        addi r30, r30, 800
        addi r2, r29, 1
        cmpwi cr0, r29, 100
        or r29, r2, r2
        bne .LBB_foo_1  ; no_exit.0
.LBB_foo_4:     ; return
        lwz r11, 56(r1)
        mtlr r11
        lwz r30, 40(r1)
        lwz r29, 44(r1)
        lwz r1, 0(r1)
        blr

instead of this:

_foo:
.LBB_foo_0:     ; entry
        stwu r1, -48(r1)
        stw r28, 44(r1)                   ;; uses an extra register.
        stw r29, 40(r1)
        stw r30, 36(r1)
        mflr r11
        stw r11, 56(r1)
        li r30, 1
        li r29, 0
        or r28, r29, r29
.LBB_foo_1:     ; no_exit.0
        bl L_bar$stub
        mulli r2, r28, 800           ;; unstrength-reduced multiply
        lis r3, ha16(L_A$non_lazy_ptr)   ;; loop invariant address computation
        lwz r3, lo16(L_A$non_lazy_ptr)(r3)
        add r2, r2, r3
        mulli r4, r29, 800           ;; unstrength-reduced multiply
        addi r3, r3, 8
        add r3, r4, r3
        li r4, 1
.LBB_foo_2:     ; no_exit.1
        lfd f0, 0(r3)
        stfd f0, 0(r2)
        addi r5, r4, 1
        addi r2, r2, 8                 ;; multiple stride 8 IV's
        addi r3, r3, 8
        cmpwi cr0, r4, 100
        or r4, r5, r5
        bne .LBB_foo_2  ; no_exit.1
.LBB_foo_3:     ; loopexit.1
        addi r28, r28, 1               ;;; Many IV's with stride 1
        addi r29, r29, 1
        addi r2, r30, 1
        cmpwi cr0, r30, 100
        or r30, r2, r2
        bne .LBB_foo_1  ; no_exit.0
.LBB_foo_4:     ; return
        lwz r11, 56(r1)
        mtlr r11
        lwz r30, 36(r1)
        lwz r29, 40(r1)
        lwz r28, 44(r1)
        lwz r1, 0(r1)
        blr

llvm-svn: 22640
 2005-08-04 00:14:11 +00:00
Chris Lattner
8b1b7c9e7d Teach loop-reduce to see into nested loops, to pull out immediate values 
pushed down by SCEV.

In a nested loop case, this allows us to emit this:

        lis r3, ha16(L_A$non_lazy_ptr)
        lwz r3, lo16(L_A$non_lazy_ptr)(r3)
        add r2, r2, r3
        li r3, 1
.LBB_foo_2:     ; no_exit.1
        lfd f0, 8(r2)        ;; Uses offset of 8 instead of 0
        stfd f0, 0(r2)
        addi r4, r3, 1
        addi r2, r2, 8
        cmpwi cr0, r3, 100
        or r3, r4, r4
        bne .LBB_foo_2  ; no_exit.1

instead of this:

        lis r3, ha16(L_A$non_lazy_ptr)
        lwz r3, lo16(L_A$non_lazy_ptr)(r3)
        add r2, r2, r3
        addi r3, r3, 8
        li r4, 1
.LBB_foo_2:     ; no_exit.1
        lfd f0, 0(r3)
        stfd f0, 0(r2)
        addi r5, r4, 1
        addi r2, r2, 8
        addi r3, r3, 8
        cmpwi cr0, r4, 100
        or r4, r5, r5
        bne .LBB_foo_2  ; no_exit.1

llvm-svn: 22639
 2005-08-03 23:44:42 +00:00
Chris Lattner
512f74d445 improve debug output 
llvm-svn: 22638
 2005-08-03 23:30:08 +00:00
Chris Lattner
1dcd811d36 Move from Stage 0 to Stage 1. 
Only emit one PHI node for IV uses with identical bases and strides (after
moving foldable immediates to the load/store instruction).

This implements LoopStrengthReduce/dont_insert_redundant_ops.ll, allowing
us to generate this PPC code for test1:

        or r30, r3, r3
.LBB_test1_1:   ; Loop
        li r2, 0
        stw r2, 0(r30)
        stw r2, 4(r30)
        bl L_pred$stub
        addi r30, r30, 8
        cmplwi cr0, r3, 0
        bne .LBB_test1_1        ; Loop

instead of this code:

        or r30, r3, r3
        or r29, r3, r3
.LBB_test1_1:   ; Loop
        li r2, 0
        stw r2, 0(r29)
        stw r2, 4(r30)
        bl L_pred$stub
        addi r30, r30, 8        ;; Two iv's with step of 8
        addi r29, r29, 8
        cmplwi cr0, r3, 0
        bne .LBB_test1_1        ; Loop

llvm-svn: 22635
 2005-08-03 22:51:21 +00:00
Chris Lattner
96367799a2 Rename IVUse to IVUsersOfOneStride, use a struct instead of a pair to 
unify some parallel vectors and get field names more descriptive than
"first" and "second".  This isn't lisp afterall :)

llvm-svn: 22633
 2005-08-03 22:21:05 +00:00
Chris Lattner
230700ef26 Fix a nasty dangling pointer issue. The ScalarEvolution pass would keep a 
map from instruction* to SCEVHandles.  When we delete instructions, we have
to tell it about it.  We would run into nasty cases where new instructions
were reallocated at old instruction addresses and get the old map values.
Bad bad bad :(

llvm-svn: 22632
 2005-08-03 21:36:09 +00:00
Chris Lattner
05431d5ca5 Like the comment says, do not insert cast instructions before phi nodes 
llvm-svn: 22586
 2005-08-02 03:31:14 +00:00
Chris Lattner
4187b8bebe add a comment, make a check more lenient 
llvm-svn: 22581
 2005-08-02 02:52:02 +00:00
Chris Lattner
38b0e93bd4 Simplify for loop, clear a per-loop map after processing each loop 
llvm-svn: 22580
 2005-08-02 02:44:31 +00:00
Chris Lattner
954842274e Add a comment 
Make LSR ignore GEP's that have loop variant base values, as we currently
cannot codegen them

llvm-svn: 22576
 2005-08-02 01:32:29 +00:00
Chris Lattner
ec624cdcb5 Fix an iterator invalidation problem 
llvm-svn: 22575
 2005-08-02 00:41:11 +00:00
Jeff Cohen
019104459d Keep tabs and trailing spaces out. 
llvm-svn: 22565
 2005-07-30 18:33:25 +00:00
Jeff Cohen
4f69b0d5cd Fix VC++ build problems. 
llvm-svn: 22564
 2005-07-30 18:22:27 +00:00
Nate Begeman
13bd25dc1d Ack, typo 
llvm-svn: 22560
 2005-07-30 00:21:31 +00:00
Nate Begeman
454caae5bd Commit a new LoopStrengthReduce pass that can use scalar evolutions and 
target data to decide which loop induction variables to strength reduce
and how to do so.  This work is mostly by Chris Lattner, with tweaks by
me to get it working on some of MultiSource.

llvm-svn: 22558
 2005-07-30 00:15:07 +00:00
Misha Brukman
53e199440e Remove trailing whitespace 
llvm-svn: 21427
 2005-04-21 23:48:37 +00:00
Chris Lattner
6d6f980ca7 fix a bug where we thought arguments were constants :( 
llvm-svn: 20506
 2005-03-06 22:52:29 +00:00
Chris Lattner
2174b70c68 Fix Regression/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll, 
hopefully not breaking too many other things.

llvm-svn: 20505
 2005-03-06 22:36:12 +00:00
Chris Lattner
a29e319d54 implement Transforms/LoopStrengthReduce/invariant_value_first_arg.ll 
llvm-svn: 20501
 2005-03-06 22:06:22 +00:00
Chris Lattner
18fdd1e9f6 minor simplifications of the code. 
llvm-svn: 20497
 2005-03-06 21:58:22 +00:00
Jeff Cohen
db51926d4a Reformat comments to fix 80 columns. 
llvm-svn: 20467
 2005-03-05 22:45:40 +00:00
Jeff Cohen
023f10be12 Reuse induction variables created for strength-reduced GEPs by other similar GEPs. 
llvm-svn: 20466
 2005-03-05 22:40:34 +00:00
Jeff Cohen
29ecafb2e3 Add support for not strength reducing GEPs where the element size is a small 
power of two.  This emphatically includes the zeroeth power of two.

llvm-svn: 20429
 2005-03-04 04:04:26 +00:00
Jeff Cohen
6d82d5b23e Fixed the following LSR bugs: 
* Loop invariant code does not dominate the loop header, but rather
    the end of the loop preheader.

  * The base for a reduced GEP isn't a constant unless all of its
    operands (preceding the induction variable) are constant.

  * Allow induction variable elimination for the simple case after all.

Also made changes recommended by Chris for properly deleting
instructions.

llvm-svn: 20383
 2005-03-01 03:46:11 +00:00
Jeff Cohen
d5b1827c3f Fix crash in LSR due to attempt to remove original induction variable. However, 
for reasons explained in the comments, I also deactivated this code as it needs
more thought.

llvm-svn: 20367
 2005-02-28 00:08:56 +00:00
Jeff Cohen
fd9504c7d9 PHI nodes were incorrectly placed when more than one GEP is reduced in a loop. 
llvm-svn: 20360
 2005-02-27 21:08:04 +00:00
Jeff Cohen
6258d4a431 First pass at improved Loop Strength Reduction. Still not yet ready for prime time. 
llvm-svn: 20358
 2005-02-27 19:37:07 +00:00
Nate Begeman
49e38c2185 Initial implementation of the strength reduction for GEP instructions in 
loops.  This optimization is not turned on by default yet, but may be run
with the opt tool's -loop-reduce flag.  There are many FIXMEs listed in the
code that will make it far more applicable to a wide range of code, but you
have to start somewhere :)

This limited version currently triggers on the following tests in the
MultiSource directory:
pcompress2: 7 times
cfrac: 5 times
anagram: 2 times
ks: 6 times
yacr2: 2 times

llvm-svn: 17134
 2004-10-18 21:08:22 +00:00