3b48431333
Add initial support for immediates. This allows us to compile this:
...
int %rlwnm(int %A, int %B) {
%C = call int asm "rlwnm $0, $1, $2, $3, $4", "=r,r,r,n,n"(int %A, int %B, int 4, int 17)
ret int %C
}
into:
_rlwnm:
or r2, r3, r3
or r3, r4, r4
rlwnm r2, r2, r3, 4, 17 ;; note the immediates :)
or r3, r2, r2
blr
llvm-svn: 25955
2006-02-04 02:26:14 +00:00
65ad53feb3
Initial early support for non-register operands, like immediates
...
llvm-svn: 25952
2006-02-04 02:16:44 +00:00
c93403a7fb
Handle another case exposed on X86.
...
llvm-svn: 25949
2006-02-03 23:50:46 +00:00
71d20c4e18
Fix a nasty problem on two-address machines in the following situation:
...
store EAX -> [ss#0]
[ss#0] += 1
...
use(EAX)
In this case, it is not valid to rewrite this as:
store EAX -> [ss#0]
EAX += 1
store EAX -> [ss#0] ;;; this would also delete the store above
...
use(EAX)
... because EAX is not a dead at that point. Keep track of which registers
we are allowed to clobber, and which ones we aren't, and don't clobber the
ones we're not supposed to. :)
This should resolve the issues on X86 last night.
llvm-svn: 25948
2006-02-03 23:28:46 +00:00
507a3a7bd1
significantly simplify the VirtRegMap code by pulling the SpillSlotsAvailable
...
and PhysRegsAvailable maps out into a new AvailableSpills struct. No
functionality change.
This paves the way for a bugfix, coming up next.
llvm-svn: 25947
2006-02-03 23:13:58 +00:00
dc7bba9ffe
Add a framework for eliminating instructions that produces undemanded bits.
...
llvm-svn: 25945
2006-02-03 22:24:05 +00:00
f68fd20286
remove some #ifdef'd out code, which should properly be in the dag combiner anyway.
...
llvm-svn: 25941
2006-02-03 20:13:59 +00:00
6091407783
remove dead fn
...
llvm-svn: 25935
2006-02-03 06:51:34 +00:00
22e251abf1
Add common code for reassociating ops in the dag combiner
...
llvm-svn: 25934
2006-02-03 06:46:56 +00:00
3276ff7ac6
Fix VC++ compilation error caused by using a std::map iterator variable to receive
...
a std::multimap iterator value. For some reason, GCC doesn't have a problem with this.
llvm-svn: 25927
2006-02-03 03:48:54 +00:00
e18ef0d4a6
Remove move copies and dead stuff by not clobbering the result reg of a noop copy.
...
llvm-svn: 25926
2006-02-03 03:16:14 +00:00
774d4a190b
Simplify some code
...
llvm-svn: 25924
2006-02-03 03:06:49 +00:00
1ef239afb4
Add code that checks for noop copies, which triggers when either:
...
1. a target doesn't know how to fold load/stores into copies, or
2. the spiller rewrites the input to a copy to the same register as the dest
instead of to the reloaded reg.
This will be moved/improved in the near future, but allows elimination of
some ancient x86 hacks. This eliminates 92 copies from SMG2000 on X86 and
163 copies from 252.eon.
llvm-svn: 25922
2006-02-03 02:02:59 +00:00
02b5b9cdd6
Added case HANDLENODE to getOperationName().
...
llvm-svn: 25920
2006-02-03 01:33:01 +00:00
b7f24de4c8
Physregs may hold multiple stack slot values at the same time. Keep track
...
of this, and use it to our advantage (bwahahah). This allows us to eliminate another
60 instructions from smg2000 on PPC (probably significantly more on X86). A common
old-new diff looks like this:
stw r2, 3304(r1)
- lwz r2, 3192(r1)
stw r2, 3300(r1)
- lwz r2, 3192(r1)
stw r2, 3296(r1)
- lwz r2, 3192(r1)
stw r2, 3200(r1)
- lwz r2, 3192(r1)
stw r2, 3196(r1)
- lwz r2, 3192(r1)
+ or r2, r2, r2
stw r2, 3188(r1)
and
- lwz r31, 604(r1)
- lwz r13, 604(r1)
- lwz r14, 604(r1)
- lwz r15, 604(r1)
- lwz r16, 604(r1)
- lwz r30, 604(r1)
+ or r31, r30, r30
+ or r13, r30, r30
+ or r14, r30, r30
+ or r15, r30, r30
+ or r16, r30, r30
+ or r30, r30, r30
Removal of the R = R copies is coming next...
llvm-svn: 25919
2006-02-03 00:36:31 +00:00
f3aef1b004
Fix a deficiency in the spiller that Evan noticed. In particular, consider
...
this code:
store [stack slot #0 ], R10
= add R14, [stack slot #0 ]
The spiller didn't know that the store made the value of [stackslot#0] available
in R10 *IF* the store came from a copy instruction with the store folded into it.
This patch teaches VirtRegMap to look at these stores and recognize the values
they make available. In one case Evan provided, this code:
divsd %XMM0, %XMM1
movsd %XMM1, QWORD PTR [%ESP + 40]
1) movsd QWORD PTR [%ESP + 48], %XMM1
2) movsd %XMM1, QWORD PTR [%ESP + 48]
addsd %XMM1, %XMM0
3) movsd QWORD PTR [%ESP + 48], %XMM1
movsd QWORD PTR [%ESP + 4], %XMM0
turns into:
divsd %XMM0, %XMM1
movsd %XMM1, QWORD PTR [%ESP + 40]
addsd %XMM1, %XMM0
3) movsd QWORD PTR [%ESP + 48], %XMM1
movsd QWORD PTR [%ESP + 4], %XMM0
In this case, instruction #2 was removed because of the value made
available by #1 , and inst #1 was later deleted because it is now
never used before the stack slot is redefined by #3 .
This occurs here and there in a lot of code with high spilling, on PPC
most of the removed loads/stores are LSU-reject-causing loads, which is
nice.
On X86, things are much better (because it spills more), where we nuke
about 1% of the instructions from SMG2000 and several hundred from eon.
More improvements to come...
llvm-svn: 25917
2006-02-02 23:29:36 +00:00
bb53acd03c
Move isLoadFrom/StoreToStackSlot from MRegisterInfo to TargetInstrInfo,a far more logical place. Other methods should also be moved if anyoneis interested. :)
...
llvm-svn: 25913
2006-02-02 20:12:32 +00:00
49beaf40fc
Turn any_extend nodes into zero_extend nodes when it allows us to remove an
...
and instruction. This allows us to compile stuff like this:
bool %X(int %X) {
%Y = add int %X, 14
%Z = setne int %Y, 12345
ret bool %Z
}
to this:
_X:
cmpl $12331, 4(%esp)
setne %al
movzbl %al, %eax
ret
instead of this:
_X:
cmpl $12331, 4(%esp)
setne %al
movzbl %al, %eax
andl $1, %eax
ret
This occurs quite a bit with the X86 backend. For example, 25 times in
lambda, 30 times in 177.mesa, 14 times in galgel, 70 times in fma3d,
25 times in vpr, several hundred times in gcc, ~45 times in crafty,
~60 times in parser, ~140 times in eon, 110 times in perlbmk, 55 on gap,
16 times on bzip2, 14 times on twolf, and 1-2 times in many other SPEC2K
programs.
llvm-svn: 25901
2006-02-02 07:17:31 +00:00
49ce35542f
add two dag combines:
...
(C1-X) == C2 --> X == C1-C2
(X+C1) == C2 --> X == C2-C1
This allows us to compile this:
bool %X(int %X) {
%Y = add int %X, 14
%Z = setne int %Y, 12345
ret bool %Z
}
into this:
_X:
cmpl $12331, 4(%esp)
setne %al
movzbl %al, %eax
andl $1, %eax
ret
not this:
_X:
movl $14, %eax
addl 4(%esp), %eax
cmpl $12345, %eax
setne %al
movzbl %al, %eax
andl $1, %eax
ret
Testcase here: Regression/CodeGen/X86/compare-add.ll
nukage of the and coming up next.
llvm-svn: 25898
2006-02-02 06:36:13 +00:00
0bd74558ae
make -debug output less newliney
...
llvm-svn: 25895
2006-02-02 00:38:08 +00:00
7f5880b1c7
Implement matching constraints. We can now say things like this:
...
%C = call int asm "xyz $0, $1, $2, $3", "=r,r,r,0"(int %A, int %B, int 4)
and get:
xyz r2, r3, r4, r2
note that the r2's are pinned together. Yaay for 2-address instructions.
2342 ----------------------------------------------------------------------
llvm-svn: 25893
2006-02-02 00:25:23 +00:00
aa23fa9f43
Implement smart printing of inline asm strings, handling variants and
...
substituted operands. For this testcase:
int %test(int %A, int %B) {
%C = call int asm "xyz $0, $1, $2", "=r,r,r"(int %A, int %B)
ret int %C
}
we now emit:
_test:
or r2, r3, r3
or r3, r4, r4
xyz r2, r2, r3 ;; look here
or r3, r2, r2
blr
... note the substituted operands. :)
llvm-svn: 25886
2006-02-01 22:41:11 +00:00
01bd9d9911
*** empty log message ***
...
llvm-svn: 25879
2006-02-01 19:05:15 +00:00
1558fc64f9
Implement simple register assignment for inline asms. This allows us to compile:
...
int %test(int %A, int %B) {
%C = call int asm "xyz $0, $1, $2", "=r,r,r"(int %A, int %B)
ret int %C
}
into:
(0x8906130, LLVM BB @0x8902220):
%r2 = OR4 %r3, %r3
%r3 = OR4 %r4, %r4
INLINEASM <es:xyz $0, $1, $2>, %r2<def>, %r2, %r3
%r3 = OR4 %r2, %r2
BLR
which asmprints as:
_test:
or r2, r3, r3
or r3, r4, r4
xyz $0, $1, $2 ;; need to print the operands now :)
or r3, r2, r2
blr
llvm-svn: 25878
2006-02-01 18:59:47 +00:00
7e7f439f85
Fix some of the stuff in the PPC README file, and clean up legalization
...
of the SELECT_CC, BR_CC, and BRTWOWAY_CC nodes.
llvm-svn: 25875
2006-02-01 07:19:44 +00:00
3a5ed55187
adjust to changes in InlineAsm interface. Fix a few minor bugs.
...
llvm-svn: 25865
2006-02-01 01:28:23 +00:00
32be2dc0af
Allow the specification of explicit alignments for constant pool entries.
...
llvm-svn: 25855
2006-01-31 22:23:14 +00:00
2443ab932d
Allow custom lowering of fabs. I forgot to check in this change which
...
caused several test failures.
llvm-svn: 25852
2006-01-31 18:14:25 +00:00
e9721b2984
Only insert an AND when converting from BR_COND to BRCC if needed.
...
llvm-svn: 25832
2006-01-31 05:04:52 +00:00
2e56e89452
Handle physreg input/outputs. We now compile this:
...
int %test_cpuid(int %op) {
%B = alloca int
%C = alloca int
%D = alloca int
%A = call int asm "cpuid", "=eax,==ebx,==ecx,==edx,eax"(int* %B, int* %C, int* %D, int %op)
%Bv = load int* %B
%Cv = load int* %C
%Dv = load int* %D
%x = add int %A, %Bv
%y = add int %x, %Cv
%z = add int %y, %Dv
ret int %z
}
to this:
_test_cpuid:
sub %ESP, 16
mov DWORD PTR [%ESP], %EBX
mov %EAX, DWORD PTR [%ESP + 20]
cpuid
mov DWORD PTR [%ESP + 8], %ECX
mov DWORD PTR [%ESP + 12], %EBX
mov DWORD PTR [%ESP + 4], %EDX
mov %ECX, DWORD PTR [%ESP + 12]
add %EAX, %ECX
mov %ECX, DWORD PTR [%ESP + 8]
add %EAX, %ECX
mov %ECX, DWORD PTR [%ESP + 4]
add %EAX, %ECX
mov %EBX, DWORD PTR [%ESP]
add %ESP, 16
ret
... note the proper register allocation. :)
it is unclear to me why the loads aren't folded into the adds.
llvm-svn: 25827
2006-01-31 02:03:41 +00:00
57ecb561c6
Print the most trivial inline asms.
...
llvm-svn: 25822
2006-01-30 23:00:08 +00:00
f263a23735
Fix a bug in my legalizer reworking that caused the X86 backend to not get
...
a chance to custom legalize setcc, which broke a bunch of C++ Codes.
Testcase here: CodeGen/X86/2006-01-30-LongSetcc.ll
llvm-svn: 25821
2006-01-30 22:43:50 +00:00
d6f5ae4455
don't insert an and node if it isn't needed here, this can prevent folding
...
of lowered target nodes.
llvm-svn: 25804
2006-01-30 04:22:28 +00:00
f0b24d2dc0
Move MaskedValueIsZero from the DAGCombiner to the TargetLowering interface,making isMaskedValueZeroForTargetNode simpler, and useable from other partsof the compiler.
...
llvm-svn: 25803
2006-01-30 04:09:27 +00:00
3b40e64aa3
pass the address of MaskedValueIsZero into isMaskedValueZeroForTargetNode,
...
to permit recursion
llvm-svn: 25799
2006-01-30 03:49:37 +00:00
4d1ea71a31
Fix RET of promoted values on targets that custom expand RET to a target node.
...
llvm-svn: 25794
2006-01-29 21:02:23 +00:00
2c748afd6c
cleanups to the ValueTypeActions interface
...
llvm-svn: 25785
2006-01-29 08:42:06 +00:00
ccb4476c87
Remove some special case hacks for CALLSEQ_*, using UpdateNodeOperands
...
instead.
llvm-svn: 25780
2006-01-29 07:58:15 +00:00
2f292789dc
Allow custom expansion of ConstantVec nodes. PPC will use this in the future.
...
llvm-svn: 25774
2006-01-29 06:34:16 +00:00
758b0ac54b
Legalize ConstantFP into TargetConstantFP when the target allows. Implement
...
custom expansion of ConstantFP nodes.
llvm-svn: 25772
2006-01-29 06:26:56 +00:00
678da98835
eliminate uses of SelectionDAG::getBR2Way_CC
...
llvm-svn: 25767
2006-01-29 06:00:45 +00:00
d02b05473c
Use the new "UpdateNodeOperands" method to simplify LegalizeDAG and make it
...
faster. This cuts about 120 lines of code out of the legalizer (mostly code
checking to see if operands have changed).
It also fixes an ugly performance issue, where the legalizer cloned the entire
graph after any change. Now the "UpdateNodeOperands" method gives it a chance
to reuse nodes if the operands of a node change but not its opcode or valuetypes.
This speeds up instruction selection time on kimwitu++ by about 8.2% with a
release build.
llvm-svn: 25746
2006-01-28 10:58:55 +00:00
580b12ad34
add another method variant
...
llvm-svn: 25744
2006-01-28 10:09:25 +00:00
f34156e8cb
add some methods for updating nodes
...
llvm-svn: 25742
2006-01-28 09:32:45 +00:00
eb63751499
minor tweaks
...
llvm-svn: 25740
2006-01-28 08:31:04 +00:00
689bdcc9cf
move a bunch of code, no other change.
...
llvm-svn: 25739
2006-01-28 08:25:58 +00:00
fcfda5a174
remove a couple more now-extraneous legalizeop's
...
llvm-svn: 25738
2006-01-28 08:22:56 +00:00
364b89a784
fix a bug
...
llvm-svn: 25737
2006-01-28 07:42:08 +00:00
9dcce6da8e
Several major changes:
...
1. Pull out the expand cases for BSWAP and CT* into a separate function,
reducing the size of LegalizeOp.
2. Fix a bug where expand(bswap i64) was wrong when i64 is legal.
3. Changed LegalizeOp/PromoteOp so that the legalizer never needs to be
iterative. It now operates in a single pass over the nodes.
4. Simplify a LOT of code, with a net reduction of ~280 lines.
llvm-svn: 25736
2006-01-28 07:39:30 +00:00
fd4a7f76a9
Eliminate the need for ExpandOp to set 'needsanotheriteration', as it already
...
relegalizes the stuff it returns.
Add the ability to custom expand ADD/SUB, so that targets don't need to deal
with ADD_PARTS/SUB_PARTS if they don't want.
Fix some obscure potential bugs and simplify code.
llvm-svn: 25732
2006-01-28 05:07:51 +00:00
10f677508f
Instead of making callers of ExpandLibCall legalize the result, make
...
ExpandLibCall do it itself.
llvm-svn: 25731
2006-01-28 04:28:26 +00:00
a593acfe66
Eliminate the need to do another iteration of the legalizer after inserting
...
a libcall.
llvm-svn: 25730
2006-01-28 04:23:12 +00:00
98ed05c81d
remove method I just added
...
llvm-svn: 25728
2006-01-28 03:43:09 +00:00
43b867dd3b
add a new callback
...
llvm-svn: 25727
2006-01-28 03:37:03 +00:00
595ec734fc
Implement Promote for VAARG, and allow it to be custom promoted for people
...
who don't want the default behavior (Alpha).
llvm-svn: 25726
2006-01-28 03:14:31 +00:00
af397cec0b
Add a missing case to the dag combiner.
...
llvm-svn: 25723
2006-01-28 01:06:30 +00:00
fb16a62fba
Remove the ISD::CALL and ISD::TAILCALL nodes
...
llvm-svn: 25721
2006-01-28 00:18:58 +00:00
8c47c3a3b1
Remove TLI.LowerReturnTo, and just let targets custom lower ISD::RET for
...
the same functionality. This addresses another piece of bug 680. Next,
on to fixing Alpha VAARG, which I broke last time.
llvm-svn: 25696
2006-01-27 21:09:22 +00:00
0cda006a69
Using bit size of integers instead of ambiguous "long" et all.
...
llvm-svn: 25694
2006-01-27 20:31:25 +00:00
4a84e97421
Sorry - really folowing convention.
...
llvm-svn: 25691
2006-01-27 18:32:41 +00:00
116bb15473
Following convention.
...
llvm-svn: 25689
2006-01-27 18:28:31 +00:00
fc3eca9023
fix build
...
llvm-svn: 25687
2006-01-27 18:16:17 +00:00
ecd7e61a1f
Fix build error that is apparently only a warning with some compilers.
...
llvm-svn: 25686
2006-01-27 17:31:30 +00:00
2b6efa9d41
Forgot the version number.
...
llvm-svn: 25685
2006-01-27 15:46:54 +00:00
f98fc8441c
Improve visibility/correctness of operand indices in "llvm.db" objects.
...
Handle 64 in DIEs.
llvm-svn: 25684
2006-01-27 15:20:54 +00:00
061d9e2cf0
Stub out a method
...
llvm-svn: 25676
2006-01-27 02:10:10 +00:00
4df279cfda
Teach the scheduler to emit the appropriate INLINEASM MachineInstr for an
...
ISD::INLINEASM node.
llvm-svn: 25668
2006-01-26 23:28:04 +00:00
476e67be14
initial selectiondag support for new INLINEASM node. Note that inline asms
...
with outputs or inputs are not supported yet. :)
llvm-svn: 25664
2006-01-26 22:24:51 +00:00
0689dfad23
Use global information to fill out Dwarf compile units.
...
llvm-svn: 25662
2006-01-26 21:22:49 +00:00
0bbdc55333
Set up MachineDebugInfo to scan for debug information form "llvm.db"g globals.
...
Global Variable information is now pulled from "llvm.dbg.globals"
llvm-svn: 25655
2006-01-26 20:21:46 +00:00
c4c339c3d0
Clean up some code; improve efficiency; and fixed a potential bug involving
...
chain successors.
llvm-svn: 25630
2006-01-26 00:30:29 +00:00
5edde66863
Don't break the optimized build (by incorrect placement of #endif)
...
llvm-svn: 25613
2006-01-25 21:49:13 +00:00
1880f8db02
No need to keep track of top and bottom nodes in a group since the vector is
...
already in order. Thanks Jim for pointing it out.
llvm-svn: 25608
2006-01-25 18:54:24 +00:00
e74795cd70
First part of bug 680:
...
Remove TLI.LowerVA* and replace it with SDNodes that are lowered the same
way as everything else.
llvm-svn: 25606
2006-01-25 18:21:52 +00:00
c3cafb8a67
Make it even more portable.
...
llvm-svn: 25605
2006-01-25 17:18:50 +00:00
fb20616aa6
Fix VC++ compilation error.
...
llvm-svn: 25604
2006-01-25 17:17:49 +00:00
ab49556cf4
Bottom up register usage reducing list scheduler.
...
llvm-svn: 25601
2006-01-25 09:14:32 +00:00
fbc88a624a
Keep track of bottom / top element of a set of flagged nodes.
...
llvm-svn: 25600
2006-01-25 09:13:41 +00:00
a6eff8a432
If scheduler choice is the default (-sched=default), use target scheduling
...
preference to determine which scheduler to use. SchedulingForLatency ==
Breadth first; SchedulingForRegPressure == bottom up register reduction list
scheduler.
llvm-svn: 25599
2006-01-25 09:12:57 +00:00
a292744ecc
Portably cast a pointer to an integer.
...
llvm-svn: 25594
2006-01-25 02:40:10 +00:00
1dbc389ad2
fix build on 64 bit hosts
...
llvm-svn: 25591
2006-01-24 21:26:43 +00:00
f9a1e3aadc
Fix an infinite loop I caused by making sure to legalize the flag operand
...
of CALLSEQ_* nodes
llvm-svn: 25582
2006-01-24 05:48:21 +00:00
12f8441c03
Fix VC++ compilation error.
...
llvm-svn: 25577
2006-01-24 04:43:17 +00:00
b56a80061a
Remove unused variables.
...
llvm-svn: 25576
2006-01-24 04:42:53 +00:00
00fcdfef0d
rename method
...
llvm-svn: 25572
2006-01-24 04:16:34 +00:00
3e65f28ffe
Crude Dwarf global variable debugging.
...
llvm-svn: 25569
2006-01-24 00:49:18 +00:00
e3a79268a2
Print file-scope inline asm blocks at the start of the output file.
...
llvm-svn: 25565
2006-01-23 23:47:53 +00:00
683352382e
another couple selects
...
llvm-svn: 25551
2006-01-23 21:51:14 +00:00
c28563874c
another selectto
...
llvm-svn: 25548
2006-01-23 20:59:12 +00:00
b8566fa10a
Typo.
...
llvm-svn: 25545
2006-01-23 13:34:04 +00:00
31272347d4
Skeleton of the list schedule.
...
llvm-svn: 25544
2006-01-23 08:26:10 +00:00
421cfe8006
Minor clean up.
...
llvm-svn: 25543
2006-01-23 08:25:34 +00:00
763dfd7723
Fix Regression/CodeGen/SparcV8/2006-01-22-BitConvertLegalize.ll by making
...
sure that the result of expanding a BIT_CONVERT node is itself legalized.
llvm-svn: 25538
2006-01-23 07:30:46 +00:00
87063b9986
Remove a couple of unnecessary #include's
...
llvm-svn: 25535
2006-01-23 07:21:01 +00:00
c1e1d9724d
Factor out more instruction scheduler code to the base class.
...
llvm-svn: 25532
2006-01-23 07:01:07 +00:00
deda32a786
Fix bugs lowering stackrestore, fixing 2004-08-12-InlinerAndAllocas.c on
...
PPC.
llvm-svn: 25522
2006-01-23 05:22:07 +00:00
de02d7727f
Add explicit #includes of <iostream>
...
llvm-svn: 25515
2006-01-22 23:41:00 +00:00
e23928c67f
Fix a bug in a recent refactor that caused a bunch of programs to miscompile
...
or the compiler to crash.
llvm-svn: 25503
2006-01-21 19:12:11 +00:00
44cab00045
Fix CodeGen/PowerPC/2006-01-20-ShiftPartsCrash.ll
...
llvm-svn: 25496
2006-01-21 04:27:00 +00:00
739a6a456e
Do some code refactoring on Jim's scheduler in preparation of the new list
...
scheduler.
llvm-svn: 25493
2006-01-21 02:32:06 +00:00
3d8f3a55b7
Simplify search for abbreviations.
...
llvm-svn: 25491
2006-01-21 01:13:18 +00:00
45900baadb
Correct some simple errors.
...
llvm-svn: 25490
2006-01-21 00:59:54 +00:00
b17434dc1d
Right size integer values before emitting.
...
llvm-svn: 25489
2006-01-20 21:02:36 +00:00
7a3e9aa282
Reworked how Dwarf debug info entries and abbreviations are handled. Added
...
pubnames and debuy str sections.
llvm-svn: 25487
2006-01-20 20:34:06 +00:00
15afe462a8
remove some unintentionally committed code
...
llvm-svn: 25483
2006-01-20 18:40:10 +00:00
222ceabbee
If the target doesn't support f32 natively, insert the FP_EXTEND in target-indep
...
code, so that the LowerReturn code doesn't have to handle it.
llvm-svn: 25482
2006-01-20 18:38:32 +00:00
13e8c9d6de
Another typo
...
llvm-svn: 25440
2006-01-19 04:54:52 +00:00
7599b6e4af
was ignoring the legalized chain in this case, fixed SPASS on alpha
...
llvm-svn: 25428
2006-01-18 23:19:08 +00:00
569c439567
Get rid of code in the DAGCombiner that is duplicated in SelectionDAG.cpp
...
Now all constant folding in the code generator is in one place.
llvm-svn: 25426
2006-01-18 22:35:16 +00:00
e2ee190821
Temporary work around for a libcall insertion bug: If a target doesn't
...
support FSIN/FCOS nodes, do not lower sin/cos to them.
llvm-svn: 25425
2006-01-18 21:50:14 +00:00
5fee908be5
Fix a backwards conditional that caused an inf loop in some cases. This
...
fixes: test/Regression/CodeGen/Generic/2005-01-18-SetUO-InfLoop.ll
llvm-svn: 25419
2006-01-18 19:13:41 +00:00
194a5268cb
Added minimum Dwarf aranges. Cleaned up some section headers. Line number
...
support now works in gdb.
llvm-svn: 25417
2006-01-18 16:54:26 +00:00
cc9dfecf81
Add frame work for additional dwarf sections. Comments will improve as code
...
is added.
llvm-svn: 25410
2006-01-17 20:41:40 +00:00
03e95af9f7
Support for the insertelement operation.
...
llvm-svn: 25405
2006-01-17 20:06:42 +00:00
6f86a7db07
Bug fix: missing LegalizeOp() on newly created nodes.
...
llvm-svn: 25401
2006-01-17 19:47:13 +00:00
b9966029fe
Adding basic support for Dwarf line number debug information.
...
I promise to keep future commits smaller.
llvm-svn: 25396
2006-01-17 17:31:53 +00:00
b4f9a6f110
For PR411:
...
This patch is an incremental step towards supporting a flat symbol table.
It de-overloads the intrinsic functions by providing type-specific intrinsics
and arranging for automatically upgrading from the old overloaded name to
the new non-overloaded name. Specifically:
llvm.isunordered -> llvm.isunordered.f32, llvm.isunordered.f64
llvm.sqrt -> llvm.sqrt.f32, llvm.sqrt.f64
llvm.ctpop -> llvm.ctpop.i8, llvm.ctpop.i16, llvm.ctpop.i32, llvm.ctpop.i64
llvm.ctlz -> llvm.ctlz.i8, llvm.ctlz.i16, llvm.ctlz.i32, llvm.ctlz.i64
llvm.cttz -> llvm.cttz.i8, llvm.cttz.i16, llvm.cttz.i32, llvm.cttz.i64
New code should not use the overloaded intrinsic names. Warnings will be
emitted if they are used.
llvm-svn: 25366
2006-01-16 21:12:35 +00:00
1e1eb5ee6c
Constant fold ctpop/ctlz/cttz, and a couple other small cleanups
...
llvm-svn: 25357
2006-01-16 08:07:10 +00:00
2642a35f4c
Expand case for 64b Legalize, even though no one should end up using this
...
(itanium supports bswap natively, alpha should custom lower it using the
VAX floating point swapload, ha ha).
llvm-svn: 25356
2006-01-16 07:59:13 +00:00
7d831fa5b9
Add BSWAP stuff to intrinsic lowering for CBE & friends.
...
llvm-svn: 25355
2006-01-16 07:57:00 +00:00
fcdb420baf
Disable two transformations that contribute to bus errors on SparcV8.
...
llvm-svn: 25339
2006-01-15 18:58:59 +00:00
59b82f9848
Allow the target to specify 'expand' if they just require the amount to
...
be subtracted from the stack pointer.
llvm-svn: 25331
2006-01-15 08:54:32 +00:00
2d59142613
Fix custom lowering of dynamic_stackalloc
...
llvm-svn: 25329
2006-01-15 08:43:08 +00:00
9597b33d58
add a missing node name
...
llvm-svn: 25327
2006-01-15 08:39:35 +00:00
02011c9a4f
Token chain results are not always the first or last result. Consider copyfromreg nodes, where they are the middle result (the flag result is last)
...
llvm-svn: 25325
2006-01-14 22:41:46 +00:00
542c3c17a9
Remove some duplicated code
...
llvm-svn: 25313
2006-01-14 03:18:27 +00:00
2fba8a3aaa
bswap implementation
...
llvm-svn: 25312
2006-01-14 03:14:10 +00:00
ed9b3e1c0a
If a target specified a stack pointer with setStackPointerRegisterToSaveRestore,
...
lower STACKSAVE/STACKRESTORE into a copy from/to that register.
llvm-svn: 25276
2006-01-13 17:48:44 +00:00
b32664583b
Compile llvm.stacksave/restore into STACKSAVE/STACKRESTORE nodes, and allow
...
targets to custom expand them as they desire.
llvm-svn: 25273
2006-01-13 02:50:02 +00:00
a5110e854d
add stacksave/stackrestore nodes
...
llvm-svn: 25270
2006-01-13 02:39:42 +00:00
6c9c250dcd
Add "support" for stacksave/stackrestore to the dag isel
...
llvm-svn: 25268
2006-01-13 02:24:42 +00:00
3b2b0aff0c
Add "support" for the llvm.stacksave/stackrestore intrinsics, this is
...
used by the C backend.
llvm-svn: 25267
2006-01-13 02:22:08 +00:00
3470b5dee6
Add a simple missing fold to produce this:
...
subfic r3, r2, 33
instead of this:
subfic r2, r2, 32
addi r3, r2, 1
llvm-svn: 25255
2006-01-12 20:22:43 +00:00
3760e901cf
If using __main, emit global ctor/dtor list like any other global
...
llvm-svn: 25251
2006-01-12 19:17:23 +00:00
b1ee616de9
Don't create rotate instructions in unsupported types, because we don't have
...
promote/expand code yet. This fixes the 177.mesa failure on PPC.
llvm-svn: 25250
2006-01-12 18:57:33 +00:00
7f4ec8274f
Allow custom lowering of DYNAMIC_STACKALLOC.
...
llvm-svn: 25224
2006-01-11 22:14:47 +00:00
982493300e
ignore register #0
...
llvm-svn: 25223
2006-01-11 22:13:48 +00:00
1b8121b227
Add bswap, rotl, and rotr nodes
...
Add dag combiner code to recognize rotl, rotr
Add ppc code to match rotl
Targets should add rotl/rotr patterns if they have them
llvm-svn: 25222
2006-01-11 21:21:00 +00:00
fb5f46541c
silence a warning
...
llvm-svn: 25184
2006-01-10 19:43:26 +00:00
2c966e7617
Added selection DAG support for the extractelement operation.
...
llvm-svn: 25179
2006-01-10 19:04:57 +00:00
b05fce676f
Minor cleanup, no functionality change for current targets
...
llvm-svn: 25173
2006-01-10 05:41:59 +00:00
90ba544826
Fix an exponential function in libcall insertion to not be exponential. :)
...
llvm-svn: 25165
2006-01-09 23:21:49 +00:00
870e4f8e38
* Allow custom lowering of ADD_PARTS, SUB_PARTS, SHL_PARTS, SRA_PARTS,
...
and SRL_PARTS.
* Fix a bug that caused *_PARTS to be custom lowered twice.
llvm-svn: 25157
2006-01-09 18:31:59 +00:00
53a1f57fc5
New getNode() variants.
...
llvm-svn: 25156
2006-01-09 18:29:18 +00:00
fae8afb77f
Unbreak the build :(
...
llvm-svn: 25124
2006-01-06 05:47:48 +00:00
85c973cda9
Revert the previous check-in. Leave shl x, 1 along for target to deal with.
...
llvm-svn: 25121
2006-01-06 01:56:02 +00:00
b03f9b32d2
fold (shl x, 1) -> (add x, x)
...
llvm-svn: 25120
2006-01-06 01:06:31 +00:00
f35b1c837f
Support for custom lowering of ISD::RET.
...
llvm-svn: 25116
2006-01-06 00:41:43 +00:00
762e9ec06c
Added initial support for DEBUG_LABEL allowing debug specific labels to be
...
inserted in the code.
llvm-svn: 25104
2006-01-05 01:25:28 +00:00
219d559824
Applied some recommend changes from sabre. The dominate one beginning "let the
...
pass manager do it's thing." Fixes crash when compiling -g files and suppresses
dwarf statements if no debug info is present.
llvm-svn: 25100
2006-01-04 22:28:25 +00:00
0da76a676a
Add unique id to debug location for debug label use (work in progress.)
...
llvm-svn: 25096
2006-01-04 15:04:11 +00:00
2741e8304c
Add check for debug presence.
...
llvm-svn: 25095
2006-01-04 14:30:12 +00:00
b0609d91c3
Tie dwarf generation to darwin assembler.
...
llvm-svn: 25093
2006-01-04 13:52:30 +00:00
57a5e0b45a
Moving MachineDebugInfo to module level location.
...
llvm-svn: 25090
2006-01-04 13:43:56 +00:00
6f9ff633a6
Change how MachineDebugInfo is fetched.
...
llvm-svn: 25089
2006-01-04 13:42:59 +00:00
44317393f8
Extending MachineDebugInfo.
...
llvm-svn: 25086
2006-01-04 13:36:38 +00:00
227e936650
Add support for targets (like Alpha) that have terminator instructions which
...
use virtual registers. We now allow the first instruction in a block of
terminators to use virtual registers, and update phi elimination to correctly
update livevar when eliminating phi's. This fixes a problem on a testcase
Andrew sent me.
llvm-svn: 25083
2006-01-04 07:12:21 +00:00
0511055276
Add an assertion, update DefInst even though no one uses it (dangling pointers
...
don't help anyone)
llvm-svn: 25081
2006-01-04 06:47:48 +00:00
be45b5e948
Add a LiveVariables::VarInfo::dump method
...
llvm-svn: 25080
2006-01-04 05:40:30 +00:00
b723c33614
Change a variable from being an iterator to a raw MachineInstr*, to make
...
GDB use tolerable
llvm-svn: 25064
2006-01-03 07:41:37 +00:00
164db3a7eb
Make sure to pass the offset into the new node, so that we don't silently
...
drop it on the floor.
llvm-svn: 25044
2005-12-30 00:10:38 +00:00
fb6a914ca7
purity++
...
llvm-svn: 25041
2005-12-29 05:59:19 +00:00
26b037e762
add these so I can be less naughty
...
llvm-svn: 25034
2005-12-28 06:29:02 +00:00
e47d9d0e92
HB is *the* code janitor.
...
llvm-svn: 25031
2005-12-28 04:55:42 +00:00
7c3dcb6892
mixed-STL programs are big and nasty :(
...
llvm-svn: 25030
2005-12-28 02:44:35 +00:00
30db2ec59f
allow custom lowering to return null for legal results
...
llvm-svn: 25007
2005-12-25 01:07:37 +00:00
7259426d88
Support Custom lowering of a few more operations.
...
Alpha needs to custom lower *DIV and *REM
llvm-svn: 25006
2005-12-24 23:42:32 +00:00
bdba3e2a46
Remove redundant debug locations.
...
llvm-svn: 24995
2005-12-23 20:08:28 +00:00
c7037abc5b
unbreak the build :-/
...
llvm-svn: 24992
2005-12-23 16:12:20 +00:00
31d15fa093
Allow custom lowering of LOAD, EXTLOAD, ZEXTLOAD, STORE, and TRUNCSTORE. Not
...
currently used.
llvm-svn: 24988
2005-12-23 07:29:34 +00:00
26943b9691
Simplify store(bitconv(x)) to store(x). This allows us to compile this:
...
void bar(double Y, double *X) {
*X = Y;
}
to this:
bar:
save -96, %o6, %o6
st %i1, [%i2+4]
st %i0, [%i2]
restore %g0, %g0, %g0
retl
nop
instead of this:
bar:
save -104, %o6, %o6
st %i1, [%i6+-4]
st %i0, [%i6+-8]
ldd [%i6+-8], %f0
std %f0, [%i2]
restore %g0, %g0, %g0
retl
nop
on sparcv8.
llvm-svn: 24983
2005-12-23 05:48:07 +00:00
54560f6887
fold (conv (load x)) -> (load (conv*)x).
...
This allows us to compile this:
void foo(double);
void bar(double *X) { foo(*X); }
To this:
bar:
save -96, %o6, %o6
ld [%i0+4], %o1
ld [%i0], %o0
call foo
nop
restore %g0, %g0, %g0
retl
nop
instead of this:
bar:
save -104, %o6, %o6
ldd [%i0], %f0
std %f0, [%i6+-8]
ld [%i6+-4], %o1
ld [%i6+-8], %o0
call foo
nop
restore %g0, %g0, %g0
retl
nop
on SparcV8.
llvm-svn: 24982
2005-12-23 05:44:41 +00:00
efbbedbf4a
Fold bitconv(bitconv(x)) -> x. We now compile this:
...
void foo(double);
void bar(double X) { foo(X); }
to this:
bar:
save -96, %o6, %o6
or %g0, %i0, %o0
or %g0, %i1, %o1
call foo
nop
restore %g0, %g0, %g0
retl
nop
instead of this:
bar:
save -112, %o6, %o6
st %i1, [%i6+-4]
st %i0, [%i6+-8]
ldd [%i6+-8], %f0
std %f0, [%i6+-16]
ld [%i6+-12], %o1
ld [%i6+-16], %o0
call foo
nop
restore %g0, %g0, %g0
retl
nop
on V8.
llvm-svn: 24981
2005-12-23 05:37:50 +00:00
a187460552
constant fold bits_convert in getNode and in the dag combiner for fp<->int
...
conversions. This allows V8 to compiles this:
void %test() {
call float %test2( float 1.000000e+00, float 2.000000e+00, double 3.000000e+00, double* null )
ret void
}
into:
test:
save -96, %o6, %o6
sethi 0, %o3
sethi 1049088, %o2
sethi 1048576, %o1
sethi 10403841040384
2005-12-23 05:30:37 +00:00
884eb3adc3
Fix a pasto
...
llvm-svn: 24973
2005-12-23 00:52:30 +00:00
9eae8d5d03
fix a thinko in the bit_convert handling code
...
llvm-svn: 24972
2005-12-23 00:50:25 +00:00
36e663d6e1
add very simple support for the BIT_CONVERT node
...
llvm-svn: 24970
2005-12-23 00:16:34 +00:00
177d7af5d5
remove dead code
...
llvm-svn: 24965
2005-12-22 21:16:08 +00:00
1408c05a8b
The 81st column doesn't like code in it.
...
llvm-svn: 24943
2005-12-22 05:23:45 +00:00
2335fc2f44
Add an eol at the end to shut gcc sup.
...
llvm-svn: 24926
2005-12-22 01:41:00 +00:00
9cdc16c6d3
* Fix a GlobalAddress lowering bug.
...
* Teach DAG combiner about X86ISD::SETCC by adding a TargetLowering hook.
llvm-svn: 24921
2005-12-21 23:05:39 +00:00
9e296bee9a
Disengage DEBUG_LOC from non-PPC targets.
...
llvm-svn: 24919
2005-12-21 20:51:37 +00:00
c1583dbd63
* Added support for X86 RET with an additional operand to specify number of
...
bytes to pop off stack.
* Added support for X86 SETCC.
llvm-svn: 24917
2005-12-21 20:21:51 +00:00
7b52a923b8
Start of Dwarf framework.
...
llvm-svn: 24914
2005-12-21 19:48:16 +00:00
0fab459362
make sure to relegalize all cases
...
llvm-svn: 24911
2005-12-21 19:40:42 +00:00
44c07ed61a
enable the gep isel opt
...
llvm-svn: 24910
2005-12-21 19:36:36 +00:00
ac12f68424
fix a bug I introduced that broke recursive expansion of nodes (e.g. scalarizing vectors)
...
llvm-svn: 24905
2005-12-21 18:02:52 +00:00
803a575616
Lower ConstantAggregateZero into zeros
...
llvm-svn: 24890
2005-12-21 02:43:26 +00:00
434ffe49a9
Don't emit a null terminator, nor anything after it, to the ctor/dtor list
...
llvm-svn: 24887
2005-12-21 01:17:37 +00:00
6af02635a7
Added a hook to print out names of target specific DAG nodes.
...
llvm-svn: 24877
2005-12-20 06:22:03 +00:00
2af3ee4bdd
Fix a nasty latent bug in the legalizer that was triggered by my patch
...
last night, breaking crafty and twolf. Make sure that the newly found
legal nodes are themselves not re-legalized until the next iteration.
Also, since this functionality exists now, we can reduce number of legalizer
iterations by depending on this behavior instead of having to misuse 'do
another iteration' to get the same effect.
llvm-svn: 24875
2005-12-20 00:53:54 +00:00
6fc31046aa
X86 conditional branch support.
...
llvm-svn: 24870
2005-12-19 23:12:38 +00:00
9fd9541367
Print out opcode number if it's an unknown target node.
...
llvm-svn: 24869
2005-12-19 23:11:49 +00:00
50b2d302d5
Fix a case where the DAG Combiner would accidentally CSE flag-producing nodes,
...
creating graphs that cannot be scheduled.
llvm-svn: 24866
2005-12-19 22:21:21 +00:00
9b9688aeb8
Amend comment.
...
llvm-svn: 24861
2005-12-19 16:32:26 +00:00
ce23987e6b
Create a strong dependency for loads following stores. This will leave a
...
latency period between the two.
llvm-svn: 24860
2005-12-19 16:30:13 +00:00
c06da626b4
Make sure to relegalize new nodes
...
llvm-svn: 24843
2005-12-18 23:54:29 +00:00
c7cb351aac
Keep VC++ happy.
...
llvm-svn: 24835
2005-12-18 22:20:05 +00:00
ebcfa0c210
More corrections for flagged copyto/from reg
...
llvm-svn: 24828
2005-12-18 15:36:21 +00:00
e3c67e97c7
legalize copytoreg and copyfromreg nodes that have flag operands correctly.
...
llvm-svn: 24826
2005-12-18 15:27:43 +00:00
c97b7d0be9
Fix a bug Sabre was having where the DAG root was a group. The group dominator
...
needed to be added to the ordering list, not the first member of the group.
llvm-svn: 24816
2005-12-18 04:40:52 +00:00
e220821deb
Groups were not emitted if the dominator node and the node in the ordering list
...
were not the same node. Ultimately the test was bogus.
llvm-svn: 24815
2005-12-18 03:59:21 +00:00
cf12118965
Simplify code
...
llvm-svn: 24806
2005-12-18 01:03:46 +00:00
bf0bd99e03
allow custom expansion of BR_CC
...
llvm-svn: 24804
2005-12-17 23:46:46 +00:00
225a4d0d6d
X86 lowers SELECT to a cmp / test followed by a conditional move.
...
llvm-svn: 24754
2005-12-17 01:21:05 +00:00
7c462768ed
Added source file/line correspondence for dwarf (PowerPC only at this point.)
...
llvm-svn: 24748
2005-12-16 22:45:29 +00:00
83e4407379
Don't create SEXTLOAD/ZEXTLOAD instructions that the target doesn't support
...
if after legalize. This fixes IA64 failures.
llvm-svn: 24725
2005-12-15 19:02:38 +00:00
d39c60fcc8
When folding loads into ops, immediately replace uses of the op with the
...
load. This reduces number of worklist iterations and avoid missing optimizations
depending on folding of things into sext_inreg nodes (which aren't supported by
all targets).
Tested by Regression/CodeGen/X86/extend.ll:test2
llvm-svn: 24712
2005-12-14 19:25:30 +00:00
7dac1083da
Fix the (zext (zextload)) case to trigger, similarly for sign extends.
...
Allow (zext (truncate)) to apply after legalize if the target supports
AND (which all do).
This compiles
short %foo() {
%tmp.0 = load ubyte* %X ; <ubyte> [#uses=1]
%tmp.3 = cast ubyte %tmp.0 to short ; <short> [#uses=1]
ret short %tmp.3
}
to:
_foo:
movzbl _X, %eax
ret
instead of:
_foo:
movzbl _X, %eax
movzbl %al, %eax
ret
thanks to Evan for pointing this out.
llvm-svn: 24709
2005-12-14 19:05:06 +00:00
f753d1a574
Fix a miscompilation in crafty due to a recent patch
...
llvm-svn: 24706
2005-12-14 07:58:38 +00:00
bce7c47306
Fold (zext (load x) to (zextload x).
...
llvm-svn: 24702
2005-12-14 02:19:23 +00:00
5d4e61dd87
Don't lump the filename and working dir together
...
llvm-svn: 24697
2005-12-13 17:40:33 +00:00
f0e9aef954
Add a couple more fields, move ctor init list to .cpp file, add support
...
for emitting the ctor/dtor list for common targets.
llvm-svn: 24694
2005-12-13 06:32:10 +00:00
956aef45c9
Lowering constant pool entries on ppc exposed a bug in the recently added
...
ConstantVec legalizing code, which would return constantpool nodes that
were not of the target's pointer type.
llvm-svn: 24691
2005-12-13 03:03:23 +00:00
9e8b633ec1
Accept and ignore prefetches for now
...
llvm-svn: 24678
2005-12-12 22:51:16 +00:00
b42ce7ca63
Fix CodeGen/Generic/2005-12-12-ExpandSextInreg.ll
...
llvm-svn: 24677
2005-12-12 22:27:43 +00:00
f1a54c0d14
Minor tweak to get isel opt
...
llvm-svn: 24663
2005-12-11 09:05:13 +00:00
4e56db674c
Add support for TargetConstantPool nodes to the dag isel emitter, and use
...
them in the PPC backend, to simplify some logic out of Select and
SelectAddr.
llvm-svn: 24657
2005-12-10 02:36:00 +00:00
dadc1057ac
Added new getNode and getTargetNode variants for X86 stores.
...
llvm-svn: 24653
2005-12-10 00:37:58 +00:00
a6f835f5a0
Avoid emitting two tabs when switching to a named section
...
llvm-svn: 24646
2005-12-09 19:28:49 +00:00
268d457b69
Teach legalize how to promote sext_inreg to fix a problem Andrew pointed
...
out to me.
llvm-svn: 24644
2005-12-09 17:32:47 +00:00
be73d6eece
improve code insertion in two ways:
...
1. Only forward subst offsets into loads and stores, not into arbitrary
things, where it will likely become a load.
2. If the source is a cast from pointer, forward subst the cast as well,
allowing us to fold the cast away (improving cases when the cast is
from an alloca or global).
This hasn't been fully tested, but does appear to further reduce register
pressure and improve code. Lets let the testers grind on it a bit. :)
llvm-svn: 24640
2005-12-08 08:00:12 +00:00
ae89d862f5
Fix a crash where ConstantVec nodes were being generated with the wrong
...
type when the target did not support them. Also teach Legalize how to
expand ConstantVecs.
This allows us to generate
_test:
lwz r2, 12(r3)
lwz r4, 8(r3)
lwz r5, 4(r3)
lwz r6, 0(r3)
addi r2, r2, 4
addi r4, r4, 3
addi r5, r5, 2
addi r6, r6, 1
stw r2, 12(r3)
stw r4, 8(r3)
stw r5, 4(r3)
stw r6, 0(r3)
blr
For:
void %test(%v4i *%P) {
%T = load %v4i* %P
%S = add %v4i %T, <int 1, int 2, int 3, int 4>
store %v4i %S, %v4i * %P
ret void
}
On PowerPC.
llvm-svn: 24633
2005-12-07 19:48:11 +00:00
57c882edf8
Only transform (sext (truncate x)) -> (sextinreg x) if before legalize or
...
if the target supports the resultant sextinreg
llvm-svn: 24632
2005-12-07 18:02:05 +00:00
cbd3d01a43
Teach the dag combiner to turn a truncate/sign_extend pair into a sextinreg
...
when the types match up. This allows the X86 backend to compile:
sbyte %toggle_value(sbyte* %tmp.1) {
%tmp.2 = load sbyte* %tmp.1
ret sbyte %tmp.2
}
to this:
_toggle_value:
mov %EAX, DWORD PTR [%ESP + 4]
movsx %EAX, BYTE PTR [%EAX]
ret
instead of this:
_toggle_value:
mov %EAX, DWORD PTR [%ESP + 4]
movsx %EAX, BYTE PTR [%EAX]
movsx %EAX, %AL
ret
noticed in Shootout/objinst.
-Chris
llvm-svn: 24630
2005-12-07 07:11:03 +00:00
41b1cdc771
Teach the SelectionDAG ISel how to turn ConstantPacked values into
...
constant nodes with vector types. Also teach the asm printer how to print
ConstantPacked constant pool entries. This allows us to generate altivec
code such as the following, which adds a vector constantto a packed float.
LCPI1_0: <4 x float> < float 0.0e+0, float 0.0e+0, float 0.0e+0, float 1.0e+0 >
.space 4
.space 4
.space 4
.long 1065353216 ; float 1
.text
.align 4
.globl _foo
_foo:
lis r2, ha16(LCPI1_0)
la r2, lo16(LCPI1_0)(r2)
li r4, 0
lvx v0, r4, r2
lvx v1, r4, r3
vaddfp v0, v1, v0
stvx v0, r4, r3
blr
For the llvm code:
void %foo(<4 x float> * %a) {
entry:
%tmp1 = load <4 x float> * %a;
%tmp2 = add <4 x float> %tmp1, < float 0.0, float 0.0, float 0.0, float 1.0 >
store <4 x float> %tmp2, <4 x float> *%a
ret void
}
llvm-svn: 24616
2005-12-06 06:18:55 +00:00
3539778883
Fix the #1 code quality problem that I have seen on X86 (and it also affects
...
PPC and other targets). In a particular, consider code like this:
struct Vector3 { double x, y, z; };
struct Matrix3 { Vector3 a, b, c; };
double dot(Vector3 &a, Vector3 &b) {
return a.x * b.x + a.y * b.y + a.z * b.z;
}
Vector3 mul(Vector3 &a, Matrix3 &b) {
Vector3 r;
r.x = dot( a, b.a );
r.y = dot( a, b.b );
r.z = dot( a, b.c );
return r;
}
void transform(Matrix3 &m, Vector3 *x, int n) {
for (int i = 0; i < n; i++)
x[i] = mul( x[i], m );
}
we compile transform to a loop with all of the GEP instructions for indexing
into 'm' pulled out of the loop (9 of them). Because isel occurs a bb at a time
we are unable to fold the constant index into the loads in the loop, leading to
PPC code that looks like this:
LBB3_1: ; no_exit.preheader
li r2, 0
addi r6, r3, 64 ;; 9 values live across the loop body!
addi r7, r3, 56
addi r8, r3, 48
addi r9, r3, 40
addi r10, r3, 32
addi r11, r3, 24
addi r12, r3, 16
addi r30, r3, 8
LBB3_2: ; no_exit
lfd f0, 0(r30)
lfd f1, 8(r4)
fmul f0, f1, f0
lfd f2, 0(r3) ;; no constant indices folded into the loads!
lfd f3, 0(r4)
lfd f4, 0(r10)
lfd f5, 0(r6)
lfd f6, 0(r7)
lfd f7, 0(r8)
lfd f8, 0(r9)
lfd f9, 0(r11)
lfd f10, 0(r12)
lfd f11, 16(r4)
fmadd f0, f3, f2, f0
fmul f2, f1, f4
fmadd f0, f11, f10, f0
fmadd f2, f3, f9, f2
fmul f1, f1, f6
stfd f0, 0(r4)
fmadd f0, f11, f8, f2
fmadd f1, f3, f7, f1
stfd f0, 8(r4)
fmadd f0, f11, f5, f1
addi r29, r4, 24
stfd f0, 16(r4)
addi r2, r2, 1
cmpw cr0, r2, r5
or r4, r29, r29
bne cr0, LBB3_2 ; no_exit
uh, yuck. With this patch, we now sink the constant offsets into the loop, producing
this code:
LBB3_1: ; no_exit.preheader
li r2, 0
LBB3_2: ; no_exit
lfd f0, 8(r3)
lfd f1, 8(r4)
fmul f0, f1, f0
lfd f2, 0(r3)
lfd f3, 0(r4)
lfd f4, 32(r3) ;; much nicer.
lfd f5, 64(r3)
lfd f6, 56(r3)
lfd f7, 48(r3)
lfd f8, 40(r3)
lfd f9, 24(r3)
lfd f10, 16(r3)
lfd f11, 16(r4)
fmadd f0, f3, f2, f0
fmul f2, f1, f4
fmadd f0, f11, f10, f0
fmadd f2, f3, f9, f2
fmul f1, f1, f6
stfd f0, 0(r4)
fmadd f0, f11, f8, f2
fmadd f1, f3, f7, f1
stfd f0, 8(r4)
fmadd f0, f11, f5, f1
addi r6, r4, 24
stfd f0, 16(r4)
addi r2, r2, 1
cmpw cr0, r2, r5
or r4, r6, r6
bne cr0, LBB3_2 ; no_exit
This is much nicer as it reduces register pressure in the loop a lot. On X86,
this takes the function from having 9 spilled registers to 2. This should help
some spec programs on X86 (gzip?)
This is currently only enabled with -enable-gep-isel-opt to allow perf testing
tonight.
llvm-svn: 24606
2005-12-05 07:10:48 +00:00
8782b782cd
dbg.stoppoint returns a value, don't forget to init it
...
llvm-svn: 24583
2005-12-03 18:50:48 +00:00
f9b27d7011
bah, must generate all results
...
llvm-svn: 24574
2005-12-02 06:08:08 +00:00
73420b3795
cycle counter fix
...
llvm-svn: 24573
2005-12-02 04:56:24 +00:00
0142afd6c1
Don't remove two operand, two result nodes from the binary ops map. These
...
should come from the arbitrary ops map.
This fixes Regression/CodeGen/PowerPC/2005-12-01-Crash.ll
llvm-svn: 24571
2005-12-01 23:14:50 +00:00
05b0b4575b
Promote line and column number information for our friendly 64-bit targets.
...
llvm-svn: 24568
2005-12-01 18:21:35 +00:00
9d0d715e83
This is a bugfix for SelectNodeTo. In certain situations, we could be
...
selecting a node and use a mix of getTargetNode() and SelectNodeTo. Because
SelectNodeTo didn't check the CSE maps for a preexisting node and didn't insert
its result into the CSE maps, we would sometimes miss a CSE opportunity.
This is extremely rare, but worth fixing for completeness.
llvm-svn: 24565
2005-12-01 18:00:57 +00:00
006bb04f3a
Support multiple ValueTypes per RegisterClass, needed for upcoming vector
...
work. This change has no effect on generated code.
llvm-svn: 24563
2005-12-01 04:51:06 +00:00
be5dd5da19
Make SelectNodeTo return N
...
llvm-svn: 24548
2005-11-30 22:45:14 +00:00
c174048430
CALLSEQ_START/END nodes don't get memoized, do not add them in when
...
replaceAllUses'ing.
llvm-svn: 24539
2005-11-30 18:20:52 +00:00
6ee8566cae
At long last, you can say that f32 isn't supported for setcc
...
llvm-svn: 24537
2005-11-30 17:12:26 +00:00
1064d6ec43
First chunk of actually generating vector code for packed types. These
...
changes allow us to generate the following code:
_foo:
li r2, 0
lvx v0, r2, r3
vaddfp v0, v0, v0
stvx v0, r2, r3
blr
for this llvm:
void %foo(<4 x float>* %a) {
entry:
%tmp1 = load <4 x float>* %a
%tmp2 = add <4 x float> %tmp1, %tmp1
store <4 x float> %tmp2, <4 x float>* %a
ret void
}
llvm-svn: 24534
2005-11-30 08:22:07 +00:00
8d17c70171
add support for custom lowering SINT_TO_FP
...
llvm-svn: 24531
2005-11-30 06:43:03 +00:00
3fd1b4c9bf
Fix a problem with llvm-ranlib that (on some platforms) caused the archive
...
file to become corrupted due to interactions between mmap'd memory segments
and file descriptors closing. The problem is completely avoiding by using
a third temporary file.
Patch provided by Evan Jones
llvm-svn: 24527
2005-11-30 05:21:10 +00:00
11d61613af
Fixed a bug introduced by my last commit: TargetGlobalValues should key on
...
GlobalValue * and index pair. Update getGlobalAddress() for symmetry.
llvm-svn: 24524
2005-11-30 02:49:21 +00:00
0e0de2f3f0
Added an index field to GlobalAddressSDNode so it can represent X+12, etc.
...
llvm-svn: 24523
2005-11-30 02:04:11 +00:00
435b402e1f
Add support for a new STRING and LOCATION node for line number support, patch
...
contributed by Daniel Berlin, with a few cleanups here and there by me.
llvm-svn: 24515
2005-11-29 06:21:05 +00:00
89b049af90
Add the majority of the vector machien value types we expect to support,
...
and make a few changes to the legalization machinery to support more than
16 types.
llvm-svn: 24511
2005-11-29 05:45:29 +00:00
d37c13154a
Check in code to scalarize arbitrarily wide packed types for some simple
...
vector operations (load, add, sub, mul).
This allows us to codegen:
void %foo(<4 x float> * %a) {
entry:
%tmp1 = load <4 x float> * %a;
%tmp2 = add <4 x float> %tmp1, %tmp1
store <4 x float> %tmp2, <4 x float> *%a
ret void
}
on ppc as:
_foo:
lfs f0, 12(r3)
lfs f1, 8(r3)
lfs f2, 4(r3)
lfs f3, 0(r3)
fadds f0, f0, f0
fadds f1, f1, f1
fadds f2, f2, f2
fadds f3, f3, f3
stfs f0, 12(r3)
stfs f1, 8(r3)
stfs f2, 4(r3)
stfs f3, 0(r3)
blr
llvm-svn: 24484
2005-11-22 18:16:00 +00:00
07890bbec4
Rather than attempting to legalize 1 x float, make sure the SD ISel never
...
generates it. Make MVT::Vector expand-only, and remove the code in
Legalize that attempts to legalize it.
The plan for supporting N x Type is to continually epxand it in ExpandOp
until it gets down to 2 x Type, where it will be scalarized into a pair of
scalars.
llvm-svn: 24482
2005-11-22 01:29:36 +00:00
f28b3bd8b4
I think I know what you meant here, but just to be safe I'll let you
...
do it. :)
<_sabre_> excuses excuses
llvm-svn: 24471
2005-11-21 14:09:40 +00:00
f2991cee1f
Allow target to customize directive used to switch to arbitrary section in SwitchSection,
...
add generic constant pool emitter
llvm-svn: 24464
2005-11-21 08:25:09 +00:00
08adbd13ff
increment the function number in SetupMachineFunction
...
llvm-svn: 24461
2005-11-21 08:13:27 +00:00
bb644e39c0
Adjust to capitalized asmprinter method names
...
llvm-svn: 24457
2005-11-21 07:51:36 +00:00
2ea5c99eca
Add section switching to common code generator code. Add a couple of
...
asserts.
llvm-svn: 24445
2005-11-21 07:06:27 +00:00
44c28c22b7
Legalize MERGE_VALUES, expand READCYCLECOUNTER correctly, so it doesn't
...
break control dependence.
llvm-svn: 24437
2005-11-20 22:56:56 +00:00
627cbd49b1
The first patch of X86 support for read cycle counter
...
llvm-svn: 24429
2005-11-20 21:32:07 +00:00
a8d37d748f
more progress towards bug 291 being finished. Patch by Owen Anderson,
...
HAVE_GV case fixed up by me.
llvm-svn: 24428
2005-11-20 03:45:52 +00:00
19baba67b5
Unbreak codegen of bools. This should fix the llc/jit/llc-beta failures
...
from last night.
llvm-svn: 24427
2005-11-19 18:40:42 +00:00
377bdbff91
Improve Selection DAG printer portability. Patch by Owen Anderson!
...
llvm-svn: 24425
2005-11-19 07:44:09 +00:00
a22eae0163
Teach the graph viewer to handle register operands that are zero.
...
llvm-svn: 24421
2005-11-19 06:58:46 +00:00
301015a703
Silence a bogus warning
...
llvm-svn: 24420
2005-11-19 05:51:46 +00:00
f090f7eb0e
Add some method variants, patch by Evan Cheng
...
llvm-svn: 24418
2005-11-19 01:44:53 +00:00
b2e089c31b
Teach LLVM how to scalarize packed types. Currently, this only works on
...
packed types with an element count of 1, although more generic support is
coming. This allows LLVM to turn the following code:
void %foo(<1 x float> * %a) {
entry:
%tmp1 = load <1 x float> * %a;
%tmp2 = add <1 x float> %tmp1, %tmp1
store <1 x float> %tmp2, <1 x float> *%a
ret void
}
Into:
_foo:
lfs f0, 0(r3)
fadds f0, f0, f0
stfs f0, 0(r3)
blr
llvm-svn: 24416
2005-11-19 00:36:38 +00:00
127321b14c
Split out the shift code from visitBinary.
...
llvm-svn: 24412
2005-11-18 07:42:56 +00:00
45ca1c0194
Allow targets to custom legalize leaf nodes like GlobalAddress.
...
llvm-svn: 24387
2005-11-17 06:41:44 +00:00
4ff65ec745
Teach legalize about targetglobaladdress
...
llvm-svn: 24385
2005-11-17 05:52:24 +00:00
f2b62f317c
when debugging lower dbg intrinsics to calls
...
llvm-svn: 24377
2005-11-16 07:22:30 +00:00
bba9c372c1
Remove extraneous parents around constants when using a constant expr cast.
...
llvm-svn: 24357
2005-11-15 00:03:16 +00:00
dd8eeed096
Teach emitAlignment to handle explicit alignment requests by globals.
...
llvm-svn: 24354
2005-11-14 19:00:06 +00:00
cf1f782a2f
Fix operator precedence bug caught by VC++.
...
llvm-svn: 24318
2005-11-12 00:59:01 +00:00
de1b5d6baa
added a chain output
...
llvm-svn: 24306
2005-11-11 22:48:54 +00:00
01aa56397d
continued readcyclecounter support
...
llvm-svn: 24300
2005-11-11 16:47:30 +00:00
4f827446da
nuke blank line
...
llvm-svn: 24278
2005-11-10 18:49:46 +00:00
c0a1eba0ab
Get rid of casts by #including the right header
...
llvm-svn: 24275
2005-11-10 18:36:17 +00:00
747960d21e
Compile C strings to:
...
l1__2E_str_1: ; '.str_1'
.asciz "foo"
not:
.align 0
l1__2E_str_1: ; '.str_1'
.asciz "foo"
llvm-svn: 24273
2005-11-10 18:09:27 +00:00
55a6d9067b
add support for .asciz, and enable it by default. If your target assemblerdoesn't support .asciz, just set AscizDirective to null in your asmprinter.
...
This compiles C strings to:
l1__2E_str_1: ; '.str_1'
.asciz "foo"
instead of:
l1__2E_str_1: ; '.str_1'
.ascii "foo\000"
llvm-svn: 24272
2005-11-10 18:06:33 +00:00
bf4f233214
Switch the allnodes list from a vector of pointers to an ilist of nodes.This eliminates the vector, allows constant time removal of a node froma graph, and makes iteration over the all nodes list stable when adding
...
nodes to the graph.
llvm-svn: 24263
2005-11-09 23:47:37 +00:00
cd6f0f47f2
Refactor intrinsic lowering stuff out of visitCall
...
llvm-svn: 24261
2005-11-09 19:44:01 +00:00
af3aefa10e
Handle the trivial (but common) two-op case more efficiently
...
llvm-svn: 24259
2005-11-09 18:48:57 +00:00
619dfaa42b
Nuke noop copies.
...
llvm-svn: 24258
2005-11-09 18:22:42 +00:00
41fd6d5d27
Fix CodeGen/X86/shift-folding.ll:test3 on X86
...
llvm-svn: 24256
2005-11-09 16:50:40 +00:00
35ecaa76fa
Disable some overly-aggressive checking code. This speeds up the local
...
allocator from 23s to 11s on kc++ in debug mode.
llvm-svn: 24255
2005-11-09 05:28:45 +00:00
b7cad90e55
Avoid creating a token factor node in trivially redundant cases. This
...
eliminates almost one node per block in common cases.
llvm-svn: 24254
2005-11-09 05:03:03 +00:00
43535a19b1
Handle GEP's a bit more intelligently. Fold constant indices early and
...
turn power-of-two multiplies into shifts early to improve compile time.
llvm-svn: 24253
2005-11-09 04:45:33 +00:00
c4d6050db6
Allocate the right amount of memory for this vector up front.
...
llvm-svn: 24252
2005-11-08 23:32:44 +00:00
88fa11c3d5
Change the ValueList array for each node to be shared instead of individuallyallocated. Further, in the common case where a node has a single value, justreference an element from a small array. This is a small compile-time win.
...
llvm-svn: 24251
2005-11-08 23:30:28 +00:00
7e4b5d33cb
Switch the operandlist/valuelist from being vectors to being just an array.This saves 12 bytes from SDNode, but doesn't speed things up substantially
...
(our graphs apparently already fit within the cache on my g5). In any case
this reduces memory usage.
llvm-svn: 24249
2005-11-08 22:07:03 +00:00
3ba38cba64
Explicitly initialize some instance vars
...
llvm-svn: 24247
2005-11-08 21:54:57 +00:00
aba48dd34c
Clean up RemoveDeadNodes significantly, by eliminating the need for a temporary
...
set and eliminating the need to iterate whenever something is removed (which
can be really slow in some cases). Thx to Jim for pointing out something silly
I was getting stuck on. :)
llvm-svn: 24241
2005-11-08 18:52:27 +00:00
1d2f26adcc
Let's try ignoring resource utilization on the backward pass.
...
llvm-svn: 24231
2005-11-07 19:08:53 +00:00
629ba44e50
Always compute max align.
...
llvm-svn: 24227
2005-11-06 17:43:20 +00:00
3ee3e69556
Add the necessary support to the ISel to allow targets to codegen the new
...
alignment information appropriately. Includes code for PowerPC to support
fixed-size allocas with alignment larger than the stack. Support for
arbitrarily aligned dynamic allocas coming soon.
llvm-svn: 24224
2005-11-06 09:00:38 +00:00
904dbb4a27
Fix logic bug in finding retry slot in tally.
...
llvm-svn: 24188
2005-11-05 00:01:25 +00:00
ded4759d81
Fix a warning
...
llvm-svn: 24187
2005-11-04 18:26:02 +00:00
e682b677c1
Scheduling now uses itinerary data.
...
llvm-svn: 24180
2005-11-04 04:05:35 +00:00
ee065281e8
Fix a crash that Andrew noticed, and add a pair of braces to unfconfuse
...
XCode's indenting.
llvm-svn: 24159
2005-11-02 18:42:59 +00:00
17df608719
Fix a source of undefined behavior when dealing with 64-bit types. This
...
may fix PR652. Thanks to Andrew for tracking down the problem.
llvm-svn: 24145
2005-11-02 01:47:04 +00:00
5ce0538253
1. Embed and not inherit vector for NodeGroup.
...
2. Iterate operands and not uses (performance.)
3. Some long pending comment changes.
llvm-svn: 24119
2005-10-31 12:49:09 +00:00
6871b23d02
Significantly simplify this code and make it more aggressive. Instead of having
...
a special case hack for X86, make the hack more general: if an incoming argument
register is not used in any block other than the entry block, don't copy it to
a vreg. This helps us compile code like this:
%struct.foo = type { int, int, [0 x ubyte] }
int %test(%struct.foo* %X) {
%tmp1 = getelementptr %struct.foo* %X, int 0, uint 2, int 100
%tmp = load ubyte* %tmp1 ; <ubyte> [#uses=1]
%tmp2 = cast ubyte %tmp to int ; <int> [#uses=1]
ret int %tmp2
}
to:
_test:
lbz r3, 108(r3)
blr
instead of:
_test:
lbz r2, 108(r3)
or r3, r2, r2
blr
The (dead) copy emitted to copy r3 into a vreg for extra-block uses was
increasing the live range of r3 past the load, preventing the coallescing.
This implements CodeGen/PowerPC/reg-coallesce-simple.ll
llvm-svn: 24115
2005-10-30 19:42:35 +00:00
dd5663dfa0
Reduce the number of copies emitted as machine instructions by
...
generating results in vregs that will need them. In the case of something
like this: CopyToReg((add X, Y), reg1024), we no longer emit code like
this:
reg1025 = add X, Y
reg1024 = reg 1025
Instead, we emit:
reg1024 = add X, Y
Whoa! :)
llvm-svn: 24111
2005-10-30 18:54:27 +00:00
a70878d4fb
Codegen mul by negative power of two with a shift and negate.
...
This implements test/Regression/CodeGen/PowerPC/mul-neg-power-2.ll,
producing:
_foo:
slwi r2, r3, 1
subfic r3, r2, 63
blr
instead of:
_foo:
mulli r2, r3, -2
addi r3, r2, 63
blr
llvm-svn: 24106
2005-10-30 06:41:49 +00:00
4b6d583d7a
Fix DSE to not nuke dead stores unless they redundant store is the same
...
VT as the killing one. Fix fixes PR491
llvm-svn: 24034
2005-10-27 07:10:34 +00:00
d8c5c066a1
Add a simple xform that is useful for bitfield operations.
...
llvm-svn: 24029
2005-10-27 05:06:38 +00:00
3c7974aade
Fix some spello's pointed out by Gabor Greif
...
llvm-svn: 24019
2005-10-26 18:41:41 +00:00
d8f2a1a0f3
Allow custom lowered FP_TO_SINT ops in the check for whether a larger
...
FP_TO_SINT is preferred to a larger FP_TO_UINT. This seems to be begging
for a TLI.isOperationCustom() helper function.
llvm-svn: 23992
2005-10-25 23:47:25 +00:00
3b409a85eb
Clear a bit in this file that was causing a miscompilation of 178.galgel.
...
llvm-svn: 23980
2005-10-25 18:57:30 +00:00
476b8ddd55
Alkis agrees that that iterative scan allocator isn't going to be worked on
...
in the future, remove it.
llvm-svn: 23952
2005-10-24 04:14:30 +00:00
11e26b52b2
When a function takes a variable number of pointer arguments, with a zero
...
pointer marking the end of the list, the zero *must* be cast to the pointer
type. An un-cast zero is a 32-bit int, and at least on x86_64, gcc will
not extend the zero to 64 bits, thus allowing the upper 32 bits to be
random junk.
The new END_WITH_NULL macro may be used to annotate a such a function
so that GCC (version 4 or newer) will detect the use of un-casted zero
at compile time.
llvm-svn: 23888
2005-10-23 04:37:20 +00:00
4b3932aa89
add TargetExternalSymbol
...
llvm-svn: 23886
2005-10-23 03:40:17 +00:00
9faa5b7a9a
BuildSDIV and BuildUDIV only work for i32/i64, but they don't check that
...
the input is that type, this caused a failure on gs on X86 last night.
Move the hard checks into Build[US]Div since that is where decisions like
this should be made.
llvm-svn: 23881
2005-10-22 18:50:15 +00:00
75ea5b10bf
add a case missing from the dag combiner that exposed the failure on
...
2005-10-21-longlonggtu.ll.
llvm-svn: 23875
2005-10-21 21:23:25 +00:00
e95b5745c0
Make the coallescer a bit smarter, allowing it to join more live ranges.
...
For example, we can now join things like [0-30:0)[31-40:1)[52-59:2)
with [40:60:0) if the 52-59 range is defined by a copy from the 40-60 range.
The resultant range ends up being [0-30:0)[31-60:1).
This fires a lot through-out the test suite (e.g. shrinking bc from
19492 -> 18509 machineinstrs) though most gains are smaller (e.g. about
50 copies eliminated from crafty).
llvm-svn: 23866
2005-10-21 06:49:50 +00:00
76c97afbbc
Fix LiveInterval::getOverlapingRanges to take things in the right order
...
(an unused method).
Fix the merger so that it can merge ranges like this [10:12)[16:40) with
[12:38) into [10:40) instead of bogus ranges. This sort of input will be
possible for the merger coming shortly
llvm-svn: 23865
2005-10-21 06:41:30 +00:00
8f62cd32ad
Fix a typo in the dag combiner, so that this can work on i64 targets
...
llvm-svn: 23856
2005-10-21 01:51:45 +00:00
4dd383120f
Invert the TargetLowering flag that controls divide by consant expansion.
...
Add a new flag to TargetLowering indicating if the target has really cheap
signed division by powers of two, make ppc use it. This will probably go
away in the future.
Implement some more ISD::SDIV folds in the dag combiner
Remove now dead code in the x86 backend.
llvm-svn: 23853
2005-10-21 00:02:42 +00:00
b7b75e1b68
Fix a conditional so we don't access past the end of the range. Thanks to
...
Andrew for bringing this to my attn.
llvm-svn: 23850
2005-10-20 22:50:10 +00:00
7efe53d90b
Fix a couple bugs in the const div stuff where we'd generate MULHS/MULHU
...
for types that aren't legal, and fail a divisor is less than zero
comparison, which would cause us to drop a subtract.
llvm-svn: 23846
2005-10-20 17:45:03 +00:00
a6efeb01f9
don't use llabs with apparently VC++ doesn't have
...
llvm-svn: 23845
2005-10-20 17:01:00 +00:00
35852fc391
Fix order of eval problem from when I refactored this into a function.
...
llvm-svn: 23844
2005-10-20 16:56:40 +00:00
3cf40798ab
add a new method, play around with some code.
...
Fix a *bug* in the extendIntervalEndTo method. In particular, if adding
[2:10) to an interval containing [0:2),[10:30), we produced [0:10),[10,30).
Which is not the most smart thing to do. Now produce [0:30).
llvm-svn: 23841
2005-10-20 07:39:25 +00:00
8816353040
Refactor some code, pulling it out into a function. No functionality change.
...
llvm-svn: 23839
2005-10-20 06:06:30 +00:00
c6f067a8c4
Move the target constant divide optimization up into the dag combiner, so
...
that the nodes can be folded with other nodes, and we can not duplicate
code in every backend. Alpha will probably want this too.
llvm-svn: 23835
2005-10-20 02:15:44 +00:00
5172ce641e
Teach Legalize how to do something with EXTRACT_ELEMENT when the type of
...
the pair of elements is a legal type.
llvm-svn: 23804
2005-10-19 00:06:56 +00:00
78afac2ddd
Add the ability to lower return instructions to TargetLowering. This
...
allows us to lower legal return types to something else, to meet ABI
requirements (such as that i64 be returned in two i32 regs on Darwin/ppc).
llvm-svn: 23802
2005-10-18 23:23:37 +00:00
0a71a9ac86
Fix Generic/2005-10-18-ZeroSizeStackObject.ll by not requesting a zero
...
sized stack object if either the array size or the type size is zero.
llvm-svn: 23801
2005-10-18 22:14:06 +00:00
8396a308a7
remove hack
...
llvm-svn: 23797
2005-10-18 22:11:42 +00:00
6c14c35bd7
Fold (select C, load A, load B) -> load (select C, A, B). This happens quite
...
a lot throughout many programs. In particular, specfp triggers it a bunch for
constant FP nodes when you have code like cond ? 1.0 : -1.0.
If the PPC ISel exposed the loads implicit in pic references to external globals,
we would be able to eliminate a load in cases like this as well:
%X = external global int
%Y = external global int
int* %test4(bool %C) {
%G = select bool %C, int* %X, int* %Y
ret int* %G
}
Note that this breaks things that use SrcValue's (see the fixme), but since nothing
uses them yet, this is ok.
Also, simplify some code to use hasOneUse() on an SDOperand instead of hasNUsesOfValue directly.
llvm-svn: 23781
2005-10-18 06:04:22 +00:00
418c6e4045
Implement some feedback from Chris re: constant canonicalization
...
llvm-svn: 23777
2005-10-18 00:28:13 +00:00
bd5f41a6a6
Legalize BUILD_PAIR appropriately for upcoming 64 bit PowerPC work.
...
llvm-svn: 23776
2005-10-18 00:27:41 +00:00
ec48a1bfbd
fold fmul X, +2.0 -> fadd X, X;
...
llvm-svn: 23774
2005-10-17 20:40:11 +00:00
eeb2bda2fa
add a trivial fold
...
llvm-svn: 23764
2005-10-17 01:07:11 +00:00
e540800d5a
Fix this logic.
...
llvm-svn: 23756
2005-10-15 22:35:40 +00:00
17cc9edd33
Add a case we were missing that was causing us to fail CodeGen/PowerPC/rlwinm.ll:test3
...
llvm-svn: 23755
2005-10-15 22:18:08 +00:00
b986f471be
Use getExtLoad here instead of getNode, as extloads produce two values. This
...
fixes a legalize failure on SPASS for itanium.
llvm-svn: 23747
2005-10-15 20:24:07 +00:00
6e673b24d3
fold sext_in_reg, sext_in_reg where both have the same VT. This was
...
popping up in Fourinarow.
llvm-svn: 23722
2005-10-14 01:29:07 +00:00
d59e5a7abb
Relax the checking on zextload generation a bit, since as sabre pointed out
...
you could be AND'ing with the result of a shift that shifts out all the
bits you care about, in addition to a constant.
Also, move over an add/sub_parts fold from legalize to the dag combiner,
where it works for things other than constants. Woot!
llvm-svn: 23720
2005-10-14 01:12:21 +00:00
b8282987f4
Fix the trunc(load) case, finally allowing crafty and povray to pass
...
llvm-svn: 23718
2005-10-13 22:10:05 +00:00
dbc5ae3109
Fix some bugs in (sext (load x))
...
llvm-svn: 23717
2005-10-13 21:52:31 +00:00
258521d7ea
When ExpandOp'ing a [SZ]EXTLOAD, make sure to remember that the chain
...
is also legal. Add support for ExpandOp'ing raw EXTLOADs too.
llvm-svn: 23716
2005-10-13 21:44:47 +00:00
d23f4b7411
Implement PromoteOp for *EXTLOAD, allowing MallocBench/gs to Legalize
...
llvm-svn: 23715
2005-10-13 20:07:41 +00:00
8e022b3d89
Fix the remaining DAGCombiner issues pointed out by sabre. This should fix
...
the remainder of the failures introduced by my patch last night.
llvm-svn: 23714
2005-10-13 18:34:58 +00:00
a80f1f6e72
Fix a minor bug in the dag combiner that broke pcompress2 and some other
...
tests.
llvm-svn: 23713
2005-10-13 18:16:34 +00:00
c3a89c5259
Add support to Legalize for expanding i64 sextload/zextload into hi and lo
...
parts. This should fix the crafty and signed long long unit test failure
on x86 last night.
llvm-svn: 23711
2005-10-13 17:15:37 +00:00
5d7a50ac44
Inhibit instructions from being pushed before function calls. This will
...
minimize unnecessary spilling.
llvm-svn: 23710
2005-10-13 16:44:00 +00:00
02b23c6065
Move some Legalize functionality over to the DAGCombiner where it belongs.
...
Kill some dead code.
llvm-svn: 23706
2005-10-13 03:11:28 +00:00
70d28c5e32
Fix a potential bug with two combine-to's back to back that chris pointed
...
out, where after the first CombineTo() call, the node the second CombineTo
wishes to replace may no longer exist.
Fix a very real bug with the truncated load optimization on little endian
targets, which do not need a byte offset added to the load.
llvm-svn: 23704
2005-10-12 23:18:53 +00:00
8caf81d617
More cool stuff for the dag combiner. We can now finally handle things
...
like turning:
_foo:
fctiwz f0, f1
stfd f0, -8(r1)
lwz r2, -4(r1)
rlwinm r3, r2, 0, 16, 31
blr
into
_foo:
fctiwz f0,f1
stfd f0,-8(r1)
lhz r3,-2(r1)
blr
Also removed an unncessary constraint from sra -> srl conversion, which
should take care of hte only reason we would ever need to handle sra in
MaskedValueIsZero, AFAIK.
llvm-svn: 23703
2005-10-12 20:40:40 +00:00
63b1419b74
Finally committing to the new scheduler. Still -sched=none by default.
...
llvm-svn: 23702
2005-10-12 18:29:35 +00:00
d00db257c7
Added graphviz/gv support for MF.
...
llvm-svn: 23700
2005-10-12 12:09:05 +00:00
514f058be1
Fix a powerpc crash on CodeGen/Generic/llvm-ct-intrinsics.ll
...
llvm-svn: 23694
2005-10-11 17:56:34 +00:00
c38fb8e2a1
Add a canonicalization that got lost, fixing PowerPC/fold-li.ll:SUB
...
llvm-svn: 23693
2005-10-11 06:07:15 +00:00
cc6e53e6ee
clean up some corner cases
...
llvm-svn: 23692
2005-10-10 23:00:08 +00:00
04c737091f
Implement trivial DSE. If two stores are neighbors and store to the same
...
location, replace them with a new store of the last value. This occurs
in the same neighborhood in 197.parser, speeding it up about 1.5%
llvm-svn: 23691
2005-10-10 22:31:19 +00:00
e260ed8628
Add support for CombineTo, allowing the dag combiner to replace nodes with
...
multiple results.
Use this support to implement trivial store->load forwarding, implementing
CodeGen/PowerPC/store-load-fwd.ll. Though this is the most simple case and
can be extended in the future, it is still useful. For example, it speeds
up 197.parser by 6.2% by avoiding an LSU reject in xalloc:
stw r6, lo16(l5_end_of_array)(r2)
addi r2, r5, -4
stwx r5, r4, r2
- lwzx r5, r4, r2
- rlwinm r5, r5, 0, 0, 30
stwx r5, r4, r2
lwz r2, -4(r4)
ori r2, r2, 1
llvm-svn: 23690
2005-10-10 22:04:48 +00:00
6828ed9bfd
Teach the DAGCombiner several new tricks, teaching it how to turn
...
sext_inreg into zext_inreg based on the signbit (fires a lot), srem into
urem, etc.
llvm-svn: 23688
2005-10-10 21:26:48 +00:00
7730924067
Fix comment
...
llvm-svn: 23686
2005-10-10 16:52:03 +00:00
3d1d4a3d12
Add ISD::ADD to MaskedValueIsZero
...
llvm-svn: 23685
2005-10-10 16:51:40 +00:00
56e44a6da5
This function is now dead
...
llvm-svn: 23684
2005-10-10 16:49:22 +00:00
bcfebebf22
Enable Nate's excellent DAG combiner work by default. This allows the
...
removal of a bunch of ad-hoc and crufty code from SelectionDAG.cpp.
llvm-svn: 23682
2005-10-10 16:47:10 +00:00
6a49b7cabb
add a todo for something I noticed
...
llvm-svn: 23679
2005-10-09 22:59:08 +00:00
1d3dc00674
(X & Y) & C == 0 if either X&C or Y&C are zero
...
llvm-svn: 23678
2005-10-09 22:12:36 +00:00
0832f2635a
When emiting a CopyFromReg and the source is already a vreg, do not bother
...
creating a new vreg and inserting a copy: just use the input vreg directly.
This speeds up the compile (e.g. about 5% on mesa with a debug build of llc)
by not adding a bunch of copies and vregs to be coallesced away. On mesa,
for example, this reduces the number of intervals from 168601 to 129040
going into the coallescer.
llvm-svn: 23671
2005-10-09 05:58:56 +00:00
2042aa5b92
Lo and behold, the last bits of SelectionDAG.cpp have been moved over.
...
llvm-svn: 23665
2005-10-08 00:29:44 +00:00
be4bbca0ba
remove debugging code
...
llvm-svn: 23663
2005-10-07 15:31:26 +00:00
fb12624a3f
implement CodeGen/PowerPC/div-2.ll:test2-4 by propagating zero bits through
...
C-X's
llvm-svn: 23662
2005-10-07 15:30:32 +00:00
b27a4147d3
fix indentation
...
llvm-svn: 23660
2005-10-07 06:37:02 +00:00
5bcd0dd811
Turn sdivs into udivs when we can prove the sign bits are clear. This
...
implements CodeGen/PowerPC/div-2.ll
llvm-svn: 23659
2005-10-07 06:10:46 +00:00
7bf8d06f02
silence a bogus GCC warning
...
llvm-svn: 23646
2005-10-06 17:39:10 +00:00
fabe55f155
Fix the LLC regressions on X86 last night. In particular, when undoing
...
previous copy elisions and we discover we need to reload a register, make
sure to use the regclass of the original register for the reload, not the
class of the current register. This avoid using 16-bit loads to reload 32-bit
values.
llvm-svn: 23645
2005-10-06 17:19:06 +00:00
4bbbb9eed7
Make the legalizer completely non-recursive
...
llvm-svn: 23642
2005-10-06 01:20:27 +00:00
558beb3729
Let the combiner handle more cases
...
llvm-svn: 23641
2005-10-05 21:44:43 +00:00
f8221c5e2c
Remove some bad code from Legalize
...
llvm-svn: 23640
2005-10-05 21:44:10 +00:00
bd7df030d2
Check in some more DAGCombiner pieces
...
llvm-svn: 23639
2005-10-05 21:43:42 +00:00
55149d7835
Fix a bug in the local spiller, where we could take code like this:
...
store r12 -> [ss#2]
R3 = load [ss#1]
use R3
R3 = load [ss#2]
R4 = load [ss#1]
and turn it into this code:
store R12 -> [ss#2]
R3 = load [ss#1]
use R3
R3 = R12
R4 = R3 <- oops!
The problem was that promoting R3 = load[ss#2] to a copy missed the fact that
the instruction invalidated R3 at that point.
llvm-svn: 23638
2005-10-05 18:30:19 +00:00
a49e16fefa
implement visitBR_CC so that PowerPC/inverted-bool-compares.ll passes
...
with the dag combiner. This speeds up espresso by 8%, reaching performance
parity with the dag-combiner-disabled llc.
llvm-svn: 23636
2005-10-05 06:47:48 +00:00
b11d15637a
fix some pastos
...
llvm-svn: 23635
2005-10-05 06:37:22 +00:00
06f1d0f73a
Add a new HandleNode class, which is used to handle (haha) cases in the
...
dead node elim and dag combiner passes where the root is potentially updated.
This fixes a fixme in the dag combiner.
llvm-svn: 23634
2005-10-05 06:35:28 +00:00
a6895d180e
Implement the code for PowerPC/inverted-bool-compares.ll, even though it
...
that testcase still does not pass with the dag combiner. This is because
not all forms of br* are folded yet.
Also, when we combine a node into another one, delete the node immediately
instead of waiting for the node to potentially come up in the future.
llvm-svn: 23632
2005-10-05 06:11:08 +00:00
6bd8fd09b6
make sure that -view-isel-dags is the input to the isel, not the input to
...
the second phase of dag combining
llvm-svn: 23631
2005-10-05 06:09:10 +00:00
746d50a01a
Fix a crash compiling Olden/tsp
...
llvm-svn: 23630
2005-10-05 04:45:43 +00:00
327d4298e1
Reverting to version - until problem isolated.
...
llvm-svn: 23622
2005-10-04 16:41:51 +00:00
5da6908d65
Fix some faulty logic in the libcall inserter.
...
Since calls return more than one value, don't bail if one of their uses
happens to be a node that's not an MVT::Other when following the chain
from CALLSEQ_START to CALLSEQ_END.
Once we've found a CALLSEQ_START, we can just return; there's no need to
tail-recurse further up the graph.
Most importantly, just because something only has one use doesn't mean we
should use it's one use to follow from start to end. This faulty logic
caused us to follow a chain of one-use FP operations back to a much earlier
call, putting a cycle in the graph from a later start to an earlier end.
This is a better fix that reverting to the workaround committed earlier
today.
llvm-svn: 23620
2005-10-04 02:10:55 +00:00
54fb5002e5
Add back a workaround that fixes some breakages from chris's last change.
...
Neither of us have yet figured out why this code is necessary, but stuff
breaks if its not there. Still tracking this down...
llvm-svn: 23617
2005-10-04 00:37:37 +00:00
409a6b204e
Refactor gathering node info and emission.
...
llvm-svn: 23610
2005-10-03 12:30:32 +00:00
57b21f9f10
clean up this code a bit, no functionality change
...
llvm-svn: 23609
2005-10-03 07:22:07 +00:00
5f096e2847
Break the body of the loop out into a new method
...
llvm-svn: 23606
2005-10-03 04:47:08 +00:00
9cfccfb517
Fix a problem where the legalizer would run out of stack space on extremely
...
large basic blocks because it was purely recursive. This switches it to an
iterative/recursive hybrid.
llvm-svn: 23596
2005-10-02 17:49:46 +00:00
7f718e61e8
silence a bogus warning
...
llvm-svn: 23595
2005-10-02 16:30:51 +00:00
704d97f8b2
Add assertions to the trivial scheduler to check that the value types match
...
up between defs and uses.
llvm-svn: 23590
2005-10-02 07:10:55 +00:00
a038d901fb
Codegen CopyFromReg using the regclass that matches the valuetype of the
...
destination vreg.
llvm-svn: 23586
2005-10-02 06:34:16 +00:00
5a7bfe0b72
Add some very paranoid checking for operand/result reg class matchup
...
For instructions that define multiple results, use the right regclass
to define the result, not always the rc of result #0
llvm-svn: 23580
2005-10-01 07:45:09 +00:00
f8a5e5ae6e
Fix VC++ warnings.
...
llvm-svn: 23579
2005-10-01 03:57:14 +00:00
fda6944c5b
add a method
...
llvm-svn: 23575
2005-10-01 00:17:07 +00:00
d3850457a1
typo
...
llvm-svn: 23574
2005-10-01 00:08:23 +00:00
9d96932879
1. Simplify the gathering of node groups.
...
2. Printing node groups when displaying nodes.
llvm-svn: 23573
2005-10-01 00:03:07 +00:00
3fe3841c2a
1. Made things node-centric (from operand).
...
2. Added node groups to handle flagged nodes.
3. Started weaning simple scheduling off existing emitter.
llvm-svn: 23566
2005-09-30 19:15:27 +00:00
2e794c9198
now that we have a reg class to spill with, get this info from the regclass
...
llvm-svn: 23559
2005-09-30 17:19:22 +00:00
51878189c5
Now that we have getCalleeSaveRegClasses() info, use it to pass the register
...
class into the spill/reload methods. Targets can now rely on that argument.
llvm-svn: 23556
2005-09-30 16:59:07 +00:00
5a6199f387
Change this code ot pass register classes into the stack slot spiller/reloader
...
code. PrologEpilogInserter hasn't been updated yet though, so targets cannot
use this info.
llvm-svn: 23536
2005-09-30 01:29:00 +00:00
5b2be1f890
Fix two bugs in my patch earlier today that broke int->fp conversion on X86.
...
llvm-svn: 23522
2005-09-29 06:44:39 +00:00
b01a41a06d
Silence VC++ redeclaration warnings.
...
llvm-svn: 23516
2005-09-29 01:59:49 +00:00
6f3b577ee6
Add FP versions of the binary operators, keeping the int and fp worlds seperate.
...
Though I have done extensive testing, it is possible that this will break
things in configs I can't test. Please let me know if this causes a problem
and I'll fix it ASAP.
llvm-svn: 23504
2005-09-28 22:28:18 +00:00
0fd8f9fbc9
If the target prefers it, use _setjmp/_longjmp should be used instead of setjmp/longjmp for llvm.setjmp/llvm.longjmp.
...
llvm-svn: 23481
2005-09-27 22:15:53 +00:00
63523f98d5
Remove some redundancies.
...
llvm-svn: 23469
2005-09-27 17:32:45 +00:00
5f2443c8a3
Addition of a simple two pass scheduler. This version is currently hacked up
...
for testing and will require target machine info to do a proper scheduling.
The simple scheduler can be turned on using -sched=simple (defaults
to -sched=none)
llvm-svn: 23455
2005-09-26 21:57:04 +00:00
59a05bdde6
Turn (X^C1) == C2 into X == C1^C2 iff X&~C1 = 0 (and move a function)
...
This happens all the time on PPC for bool values, e.g. eliminating a xori
in inverted-bool-compares.ll.
This should be added to the dag combiner as well.
llvm-svn: 23403
2005-09-23 00:55:52 +00:00
b1f8982ff0
Expose the LiveInterval interfaces as public headers.
...
llvm-svn: 23400
2005-09-21 04:19:09 +00:00
c760f80fed
Stub out the rest of the DAG Combiner. Just need to fill in the
...
select_cc bits and then wrap it in a convenience function for use with
regular select.
llvm-svn: 23389
2005-09-19 22:34:01 +00:00
2f838f2192
Teach the local spiller to turn stack slot loads into register-register copies
...
when possible, avoiding the load (and avoiding the copy if the value is already
in the right register).
This patch came about when I noticed code like the following being generated:
store R17 -> [SS1]
...blah...
R4 = load [SS1]
This was causing an LSU reject on the G5. This problem was due to the register
allocator folding spill code into a reg-reg copy (producing the load), which
prevented the spiller from being able to rewrite the load into a copy, despite
the fact that the value was already available in a register. In the case
above, we now rip out the R4 load and replace it with a R4 = R17 copy.
This speeds up several programs on X86 (which spills a lot :) ), e.g.
smg2k from 22.39->20.60s, povray from 12.93->12.66s, 168.wupwise from
68.54->53.83s (!), 197.parser from 7.33->6.62s (!), etc. This may have a larger
impact in some cases on the G5 (by avoiding LSU rejects), though it probably
won't trigger as often (less spilling in general).
Targets that implement folding of loads/stores into copies should implement
the isLoadFromStackSlot hook to get this.
llvm-svn: 23388
2005-09-19 06:56:21 +00:00
24a7eca282
More DAG combining. Still need the branch instructions, and select_cc
...
llvm-svn: 23371
2005-09-16 00:54:12 +00:00
d4382f0afa
If a function has liveins, and if the target requested that they be plopped
...
into particular vregs, emit copies into the entry MBB.
llvm-svn: 23331
2005-09-13 19:30:54 +00:00
2d454bf5be
Allow targets to say they don't support truncstore i1 (which includes a mask
...
when storing to an 8-bit memory location), as most don't.
llvm-svn: 23303
2005-09-10 00:20:18 +00:00
bd39c1a4c6
Add a missing #include, patch courtesy of Baptiste Lepilleur.
...
llvm-svn: 23302
2005-09-09 23:53:39 +00:00
331b311f7b
Fix a problem duraid encountered on itanium where this folding:
...
select (x < y), 1, 0 -> (x < y) incorrectly: the setcc returns i1 but the
select returned i32. Add the zero extend as needed.
llvm-svn: 23301
2005-09-09 23:00:07 +00:00
16e5cb87ba
Fix a crash viewing dags that have target nodes in them
...
llvm-svn: 23300
2005-09-09 22:35:03 +00:00
1410003751
Use continue in the use-processing loop to make it clear what the early exits
...
are, simplify logic, and cause things to not be nested as deeply. This also
uses MRI->areAliases instead of an explicit loop.
No functionality change, just code cleanup.
llvm-svn: 23296
2005-09-09 20:29:51 +00:00
049b748c76
Last round of 2-node folds from SD.cpp. Will move on to 3 node ops such
...
as setcc and select next.
llvm-svn: 23295
2005-09-09 19:49:52 +00:00
ce3662f2a2
remove debugging code *slaps head*
...
llvm-svn: 23294
2005-09-09 19:19:20 +00:00
c9053083eb
When spilling a live range that is used multiple times by one instruction,
...
only add a reload live range once for the instruction. This is one step
towards fixing a regalloc pessimization that Nate notice, but is later undone
by the spiller (so no code is changed).
llvm-svn: 23293
2005-09-09 19:17:47 +00:00
85c1cc4523
Move yet more folds over to the dag combiner from sd.cpp
...
llvm-svn: 23278
2005-09-08 20:18:10 +00:00
2cc2c9a79c
Another round of dag combiner changes. This fixes some missing XOR folds
...
as well as fixing how we replace old values with new values.
llvm-svn: 23260
2005-09-07 23:25:52 +00:00
5d16dbd5bb
Fix a bug that Tzu-Chien Chiu noticed: live interval analysis does NOT
...
preserve livevar
llvm-svn: 23259
2005-09-07 17:34:39 +00:00
6791d63e55
Implement a common missing fold, (add (add x, c1), c2) -> (add x, c1+c2).
...
This restores all of stanford to being identical with and without the dag
combiner with the add folding turned off in sd.cpp.
llvm-svn: 23258
2005-09-07 16:09:19 +00:00
fe883adfd2
Fix a bug nate ran into with replacealluseswith. In the recursive cse case,
...
we were losing a node, causing an assertion to fail. Now we eagerly delete
discovered CSE's, and provide an optional vector to keep track of these
discovered equivalences.
llvm-svn: 23255
2005-09-07 05:37:01 +00:00
007c650699
Add an option to the DAG Combiner to enable it for beta runs, and turn on
...
that option for PowerPC's beta.
llvm-svn: 23253
2005-09-07 00:15:36 +00:00
d23739d020
Next round of DAGCombiner changes. This version now passes all the tests
...
I have run so far when run before Legalize. It still needs to pick up the
SetCC folds, and nodes that use SetCC.
llvm-svn: 23243
2005-09-06 04:43:02 +00:00
821628ff2a
Fix a checking failure in gs
...
llvm-svn: 23235
2005-09-03 01:04:40 +00:00
7cea6ef16e
Next round of DAG Combiner changes. Just need to support multiple return
...
values, and then we should be able to hook it up.
llvm-svn: 23231
2005-09-02 21:18:40 +00:00
1a570f1fe4
Clean up some code from the last checkin
...
llvm-svn: 23229
2005-09-02 20:32:45 +00:00
630226697f
Fix a bug in legalize where it would emit two calls to libcalls that return
...
i64 values on targets that need that expanded to 32-bit registers. This fixes
PowerPC/2005-09-02-LegalizeDuplicatesCalls.ll and speeds up 189.lucas from
taking 122.72s to 81.96s on my desktop.
llvm-svn: 23228
2005-09-02 20:26:58 +00:00
b95b280bee
Make sure to auto-cse nullary ops
...
llvm-svn: 23224
2005-09-02 19:36:17 +00:00
1e89e36dcd
Fix some buggy logic where we would try to remove nodes with two operands
...
from the binary ops map, even if they had multiple results. This latent bug
caused a few failures with the dag isel last night.
To prevent stuff like this from happening in the future, add some really
strict checking to make sure that the CSE maps always match up with reality!
llvm-svn: 23221
2005-09-02 19:15:44 +00:00
b0b4ec5655
Don't create zero sized stack objects even for array allocas with a zero
...
number of elements.
llvm-svn: 23219
2005-09-02 18:41:28 +00:00
b6cde17d29
Fix the release build, noticed by Eric van Riet Paap
...
llvm-svn: 23215
2005-09-02 07:09:28 +00:00
d9af1aab51
Make sure to legalize assert[zs]ext's operand correctly
...
llvm-svn: 23208
2005-09-02 01:15:01 +00:00
7138f91424
Teach live intervals to not crash on dead livein regs
...
llvm-svn: 23206
2005-09-02 00:20:32 +00:00
a66403dbf7
For values that are live across basic blocks and need promotion, use ANY_EXTEND
...
instead of ZERO_EXTEND to eliminate extraneous extensions. This eliminates
dead zero extensions on formal arguments and other cases on PPC, implementing
the newly tightened up test/Regression/CodeGen/PowerPC/small-arguments.ll test.
llvm-svn: 23205
2005-09-02 00:19:37 +00:00
7753f175e6
legalize ANY_EXTEND appropriately
...
llvm-svn: 23204
2005-09-02 00:18:10 +00:00
8c393c218b
Add support for ANY_EXTEND and add a few minor folds for it
...
llvm-svn: 23203
2005-09-02 00:17:32 +00:00
d78d975437
Fix some code in the current node combining code, spotted when it was moved
...
over to DAGCombiner.cpp
1. Don't assume that SetCC returns i1 when folding (xor (setcc) constant)
2. Don't duplicate code in folding AND with AssertZext that is handled by
MaskedValueIsZero
llvm-svn: 23196
2005-09-01 23:25:49 +00:00
2504fe2613
Implement first round of feedback from chris (there's still a couple things
...
left to do).
llvm-svn: 23195
2005-09-01 23:24:04 +00:00
975f5c9f46
It is NDEBUG not _NDEBUG
...
llvm-svn: 23186
2005-09-01 18:44:10 +00:00
e8f78d1aab
Add the rest of the currently implemented visit routines to the switch
...
statement in visit().
llvm-svn: 23185
2005-09-01 00:33:32 +00:00
21158fc485
First pass at the DAG Combiner. It isn't used anywhere yet, but it should
...
be mostly functional. It currently has all folds from SelectionDAG.cpp
that do not involve a condition code.
llvm-svn: 23184
2005-09-01 00:19:25 +00:00
d4d10fff99
If a function has live ins/outs, print them
...
llvm-svn: 23181
2005-08-31 22:34:59 +00:00
8a1a5f2818
Allow targets to custom expand shifts that are too large for their registers
...
llvm-svn: 23173
2005-08-31 19:01:53 +00:00
d8c84e3c7e
Fix VC++ precedence warnings
...
llvm-svn: 23169
2005-08-31 02:47:06 +00:00
539e7c892c
Sigh, not my day. Fix typo.
...
llvm-svn: 23166
2005-08-31 00:43:49 +00:00
d513d8a662
Fix a mistake in my previous patch pointed out by sabre; the AssertZext
...
case in MaskedValueIsZero was wrong.
llvm-svn: 23165
2005-08-31 00:43:08 +00:00
e07bc28cca
Remove some unnecessary casts, and add the AssertZext case to
...
MaskedValueIsZero.
llvm-svn: 23164
2005-08-31 00:27:53 +00:00
5764da422a
Allow physregs to occur in the dag with multiple types. Though I don't likethis, it is a requirement on PPC, which can have an f32 value in r3 at onepoint in a function and a f64 value in r3 at another point. :(
...
This fixes compilation of mesa
llvm-svn: 23161
2005-08-30 22:38:38 +00:00
4d602bed10
When checking the fixed intervals, don't forget to check for register aliases.
...
This fixes PR621 and Regression/CodeGen/X86/2005-08-30-RegAllocAliasProblem.ll
llvm-svn: 23158
2005-08-30 21:03:36 +00:00
61d21b1f3c
Fix FreeBench/fourinarow with the dag isel, by not adding a bogus result
...
to SHIFT_PARTS nodes
llvm-svn: 23151
2005-08-30 17:21:17 +00:00
9a4ad487f0
Fix a miscompile of PtrDist/bc. Sign extending bools is not the right thing,
...
at least tends to expose problems elsewhere.
llvm-svn: 23149
2005-08-30 16:56:19 +00:00
a3da8c4819
Remove a bogus piece of my AssertSext/AssertZext patch. oops.
...
llvm-svn: 23148
2005-08-30 02:54:28 +00:00
43144a2fe0
Add support for AssertSext and AssertZext, folding other extensions with
...
them. This allows for elminination of redundant extends in the entry
blocks of functions on PowerPC.
Add support for i32 x i32 -> i64 multiplies, by recognizing when the inputs
to ISD::MUL in ExpandOp are actually just extended i32 values and not real
i64 values. this allows us to codegen
int mulhs(int a, int b) { return ((long long)a * b) >> 32; }
as:
_mulhs:
mulhw r3, r4, r3
blr
instead of:
_mulhs:
mulhwu r2, r4, r3
srawi r5, r3, 31
mullw r5, r4, r5
add r2, r2, r5
srawi r4, r4, 31
mullw r3, r4, r3
add r3, r2, r3
blr
with a similar improvement on x86.
llvm-svn: 23147
2005-08-30 02:44:00 +00:00
08a1e38730
Name this variable to be what it really is!
...
llvm-svn: 23145
2005-08-30 01:58:51 +00:00
04cb82278a
Handle CopyToReg nodes with flag operands correctly
...
llvm-svn: 23144
2005-08-30 01:57:23 +00:00
f7e5ec84c6
Add a hack to avoid some horrible code in some cases by always emitting
...
token chains first. For this C function:
int test() {
int i;
for (i = 0; i < 100000; ++i)
foo();
}
Instead of emitting this (condition before call)
.LBB_test_1: ; no_exit
addi r30, r30, 1
lis r2, 1
ori r2, r2, 34464
cmpw cr2, r30, r2
bl L_foo$stub
bne cr2, .LBB_test_1 ; no_exit
Emit this:
.LBB_test_1: ; no_exit
bl L_foo$stub
addi r30, r30, 1
lis r2, 1
ori r2, r2, 34464
cmpw cr0, r30, r2
bne cr0, .LBB_test_1 ; no_exit
Which makes it so we don't have to save/restore cr2 in the prolog/epilog of
the function.
This also makes the code much more similar to what the pattern isel produces.
llvm-svn: 23135
2005-08-29 23:21:29 +00:00
c738d000d5
Add a new API for Nate
...
llvm-svn: 23131
2005-08-29 21:59:31 +00:00
835cbb364d
Some of us cared about the the promote path
...
llvm-svn: 23130
2005-08-29 20:46:51 +00:00
dcde1b2b6a
Fix an infinite loop on x86
...
llvm-svn: 23129
2005-08-29 17:30:00 +00:00
46d4c75cd1
Fix a bug in my previous patch that was using the wrong iterator. This fixes
...
Olden/bisort among others.
llvm-svn: 23124
2005-08-29 00:10:46 +00:00
87421c8658
Fix a bug in ReplaceAllUsesWith
...
llvm-svn: 23122
2005-08-28 23:59:36 +00:00
075250bda1
Disable this code, which broke many tests last night
...
llvm-svn: 23114
2005-08-27 16:16:51 +00:00
5ee85e89b6
fix PHI node emission for basic blocks that have select_cc's in them on ppc32
...
llvm-svn: 23113
2005-08-27 00:58:02 +00:00
56ca46ee04
Nate noticed that Andrew never did this. This fixes PR600
...
llvm-svn: 23110
2005-08-26 22:50:40 +00:00
e7a2998064
Don't copy regs that are only used in the entry block into a vreg. This
...
changes the code generated for:
short %test(short %A) {
%B = xor short %A, -32768
ret short %B
}
to:
_test:
xori r2, r3, 32768
xoris r2, r2, 65535
extsh r3, r2
blr
instead of:
_test:
rlwinm r2, r3, 0, 16, 31
xori r2, r3, 32768
xoris r2, r2, 65535
extsh r3, r2
blr
llvm-svn: 23109
2005-08-26 22:49:59 +00:00
d4f43f7967
Make this code safe for when loadRegFromStackSlot inserts multiple instructions.
...
llvm-svn: 23108
2005-08-26 22:18:32 +00:00
4a5ebe94ba
Checking types here is not safe, because multiple types can map to the same
...
register class.
llvm-svn: 23103
2005-08-26 21:39:15 +00:00
13d7c252e5
Call the InsertAtEndOfBasicBlock hook if the usesCustomDAGSchedInserter
...
flag is set on an instruction.
llvm-svn: 23098
2005-08-26 20:54:47 +00:00
373f048a79
Revampt ReplaceAllUsesWith to be more efficient and easier to use.
...
llvm-svn: 23087
2005-08-26 18:36:28 +00:00
c30405e0ee
Change ConstantPoolSDNode to actually hold the Constant itself instead of
...
putting it into the constant pool. This allows the isel machinery to
create constants that it will end up deciding are not needed, without them
ending up in the resultant function constant pool.
llvm-svn: 23081
2005-08-26 17:15:30 +00:00
2091a36631
Fix a huge annoyance: SelectNodeTo took types before the opcode unlike
...
every other SD API. Fix it to take the opcode before the types.
llvm-svn: 23079
2005-08-26 16:36:26 +00:00
c6d481db7a
the 5th operand is the 4th number
...
llvm-svn: 23074
2005-08-26 00:43:46 +00:00
5f573416cd
Add support for targets that want to custom expand select_cc in some cases.
...
llvm-svn: 23071
2005-08-26 00:23:59 +00:00
dff50cadaa
Allow LowerOperation to return a null SDOperand in case it wants to lower
...
some things given to it, but not all.
llvm-svn: 23070
2005-08-26 00:14:16 +00:00
1cb550c603
Fix a nasty bug from a previous patch of mine
...
llvm-svn: 23069
2005-08-26 00:13:12 +00:00
33840c3268
New fold for SELECT_CC
...
llvm-svn: 23058
2005-08-25 20:04:38 +00:00
f9c19157df
Don't auto-cse nodes that return flags
...
llvm-svn: 23055
2005-08-25 19:12:10 +00:00
12756be53b
add printer support for flag operands
...
llvm-svn: 23054
2005-08-25 17:59:23 +00:00
9d28a56d55
simplify the code a bit using isOperationLegal
...
llvm-svn: 23053
2005-08-25 17:54:58 +00:00
8a93f64efa
Add support for flag operands
...
llvm-svn: 23050
2005-08-25 17:48:54 +00:00
407c6415b4
ADd support for TargetConstantPool nodes
...
llvm-svn: 23041
2005-08-25 05:03:06 +00:00
bbe0e7df2c
add a new TargetFrameIndex node
...
llvm-svn: 23035
2005-08-25 00:43:01 +00:00
45e1ce4e28
add a method
...
llvm-svn: 23027
2005-08-24 23:00:29 +00:00
d7ee4d8671
Add ReplaceAllUsesWith that can take a vector of replacement values.
...
Add some foldings to hopefully help the illegal setcc issue, and move some code around.
llvm-svn: 23025
2005-08-24 22:44:39 +00:00
ad9565dfbe
Add support for external symbols, and support for variable arity instructions
...
llvm-svn: 23022
2005-08-24 22:02:41 +00:00
bb8cc0acb2
Fix pasto that prevented VT ndoes from showing up in -view-isel-dags correctly
...
llvm-svn: 23021
2005-08-24 18:30:00 +00:00
86b1658d58
teach selection dag mask tracking about the fact that select_cc operates like
...
select. Also teach it that the bit count instructions can only set the low bits
of the result, depending on the size of the input.
This allows us to compile this:
int %eq0(int %a) {
%tmp.1 = seteq int %a, 0 ; <bool> [#uses=1]
%tmp.2 = cast bool %tmp.1 to int ; <int> [#uses=1]
ret int %tmp.2
}
To this:
_eq0:
cntlzw r2, r3
srwi r3, r2, 5
blr
instead of this:
_eq0:
cntlzw r2, r3
rlwinm r3, r2, 27, 31, 31
blr
when setcc is marked illegal on ppc (which restores parity to non-illegal
setcc). Thanks to Nate for pointing this out.
llvm-svn: 23013
2005-08-24 16:46:55 +00:00
f12eb4d676
Start using isOperationLegal and isTypeLegal to simplify the code
...
llvm-svn: 23012
2005-08-24 16:35:28 +00:00
45bbbb3f11
Teach SelectionDAG how to simplify a few more setcc-equivalent select_cc
...
nodes so that backends don't have to.
llvm-svn: 22999
2005-08-24 04:57:57 +00:00
99282c7b92
Make -view-isel-dags show the dag before instruction selecting, in case
...
the target isel crashes due to unimplemented features like calls :)
llvm-svn: 22997
2005-08-24 00:34:29 +00:00
72eab5dd5c
Fix optimization of select_cc seteq X, 0, 1, 0 -> srl (ctlz X), log2 X size
...
llvm-svn: 22995
2005-08-24 00:21:28 +00:00
eeacce5a60
Implement LiveVariables.h change
...
llvm-svn: 22994
2005-08-24 00:09:33 +00:00
469652752c
adjust to new live variables interface
...
llvm-svn: 22992
2005-08-23 23:42:17 +00:00
774158239b
Simplify this code by using higher-level LiveVariables methods
...
llvm-svn: 22989
2005-08-23 22:51:41 +00:00
22e91cc3b5
Keep track of which registers are related to which other registers.
...
Use this information to avoid doing expensive interval intersections for
registers that could not possible be interesting. This speeds up linscan
on ia64 compiling kc++ in release mode from taking 7.82s to 4.8s(!), total
itanium llc time on this program is 27.3s now. This marginally speeds up
PPC and X86, but they appear to be limited by other parts of linscan, not
this code.
On this program, on itanium, live intervals now takes 41% of llc time.
llvm-svn: 22986
2005-08-23 22:27:31 +00:00
bf8c3939d7
Teach the SelectionDAG how to transform select_cc eq, X, 0, 1, 0 into
...
either seteq X, 0 or srl (ctlz X), size(X-1), depending on what's legal
for the target.
llvm-svn: 22978
2005-08-23 05:41:12 +00:00
987121a61a
Teach Legalize how to turn setcc into select_cc
...
llvm-svn: 22977
2005-08-23 04:29:48 +00:00
834a2316a3
Try to avoid scanning the fixed list. On architectures with a non-stupid
...
number of regs (e.g. most riscs), many functions won't need to use callee
clobbered registers. Do a speculative check to see if we can get a free
register without processing the fixed list (which has all of these). This
saves a lot of time on machines with lots of callee clobbered regs (e.g.
ppc and itanium, also x86).
This reduces ppc llc compile time from 184s -> 172s on kc++. This is probably
worth FAR FAR more on itanium though.
llvm-svn: 22972
2005-08-22 20:59:30 +00:00
95a157ae1a
Move some code in the register assignment case that only needs to happen if
...
we spill out of the fast path. The scan of active_ and the calls to
updateSpillWeights don't need to happen unless a spill occurs. This reduces
debug llc time of kc++ with ppc from 187.3s to 183.2s.
llvm-svn: 22971
2005-08-22 20:20:42 +00:00
7f9e078d11
Fix a problem where constant expr shifts would not have their shift amount
...
promoted to the right type. This fixes: IA64/2005-08-22-LegalizerCrash.ll
llvm-svn: 22969
2005-08-22 17:28:31 +00:00
83b821b584
Speed up this loop a bit, based on some observations that Nate made, and
...
add some comments. This loop really needs to be reevaluated!
llvm-svn: 22966
2005-08-22 16:55:22 +00:00
92626b9bc5
Add a fast-path for register values. Add support for constant pool entries,
...
allowing us to compile this:
float %test2(float* %P) {
%Q = load float* %P
%R = add float %Q, 10.1
ret float %R
}
to this:
_test2:
lfs r2, 0(r3)
lis r3, ha16(.CPI_test2_0)
lfs r3, lo16(.CPI_test2_0)(r3)
fadds f1, r2, r3
blr
llvm-svn: 22962
2005-08-22 01:04:32 +00:00
466fecee19
add anew method
...
llvm-svn: 22957
2005-08-21 22:30:30 +00:00
4866356907
Add support for frame index nodes
...
llvm-svn: 22956
2005-08-21 19:56:04 +00:00
0548f50501
add a method
...
llvm-svn: 22955
2005-08-21 19:48:59 +00:00
707b39fb8c
add a method
...
llvm-svn: 22949
2005-08-21 18:49:33 +00:00
154b2bc59b
Add support for basic blocks, fix a bug in result # computation
...
llvm-svn: 22948
2005-08-21 18:49:29 +00:00
539c3fa863
When legalizing brcond ->brcc or select -> selectcc, make sure to truncate
...
the old condition to a one bit value. The incoming value must have been
promoted, and the top bits are undefined. This causes us to generate:
_test:
rlwinm r2, r3, 0, 31, 31
li r3, 17
cmpwi cr0, r2, 0
bne .LBB_test_2 ;
.LBB_test_1: ;
li r3, 1
.LBB_test_2: ;
blr
instead of:
_test:
rlwinm r2, r3, 0, 31, 31
li r2, 17
cmpwi cr0, r3, 0
bne .LBB_test_2 ;
.LBB_test_1: ;
li r2, 1
.LBB_test_2: ;
or r3, r2, r2
blr
for:
int %test(bool %c) {
%retval = select bool %c, int 17, int 1
ret int %retval
}
llvm-svn: 22947
2005-08-21 18:03:09 +00:00
4b08ba26d8
fix bogus warning
...
llvm-svn: 22943
2005-08-20 18:07:27 +00:00
319e65696d
Add support for global address nodes
...
llvm-svn: 22940
2005-08-19 22:38:24 +00:00
1be7eddecf
Add support for TargetGlobalAddress nodes
...
llvm-svn: 22938
2005-08-19 22:31:04 +00:00
6d7f814b01
Implement CopyFromReg, TokenFactor, and fix a bug in CopyToReg. This allows
...
us to compile stuff like this:
double %test(double %A, double %B, double %C, double %E) {
%F = mul double %A, %A
%G = add double %F, %B
%H = sub double -0.0, %G
%I = mul double %H, %C
%J = add double %I, %E
ret double %J
}
to:
_test:
fnmadd f0, f1, f1, f2
fmadd f1, f0, f3, f4
blr
woot!
llvm-svn: 22937
2005-08-19 21:43:53 +00:00
0875d1ab89
Fix a bug in previous commit
...
llvm-svn: 22936
2005-08-19 21:34:13 +00:00
4990335eb8
Print physreg register nodes with target names (e.g. F1) instead of numbers
...
llvm-svn: 22934
2005-08-19 21:21:16 +00:00
78b200eb74
Before implementing copyfromreg, we'll implement copytoreg correctly.
...
This gets us this for the previous testcase:
_test:
lis r2, 0
ori r3, r2, 65535
blr
Note that we actually write to r3 (the return reg) correctly now :)
llvm-svn: 22933
2005-08-19 20:50:53 +00:00
cc3035e989
Now that we have operand info for machine instructions, use it to create
...
temporary registers for things that define a register. This allows dag->dag
isel to compile this:
int %test() { ret int 65535 }
into:
_test:
lis r2, 0
ori r2, r2, 65535
blr
Next up, getting CopyFromReg to work, allowing arguments and cross-bb values.
llvm-svn: 22932
2005-08-19 20:45:43 +00:00
486e36cfde
Fix VC++ constant truncation warning.
...
llvm-svn: 22907
2005-08-19 16:19:21 +00:00
d1f22b1282
Fix VC++ precedence warning.
...
llvm-svn: 22902
2005-08-19 04:39:48 +00:00
d18beab94c
Fix computation of # operands, add a temporary hack for CopyToReg
...
llvm-svn: 22896
2005-08-19 01:01:34 +00:00
0c8c2c102d
add a new -view-sched-dags option to view dags as they are sent to the scheduler.
...
llvm-svn: 22878
2005-08-18 20:11:49 +00:00
d342de9aaa
Implement the first chunk of a code emitter. This is sophisticated enough to
...
codegen:
_empty:
.LBB_empty_0: ;
blr
but can't do anything more (yet). :)
llvm-svn: 22876
2005-08-18 20:07:59 +00:00
1b4727de7d
new file, obviously just a stub
...
llvm-svn: 22868
2005-08-18 18:45:24 +00:00
1a908c8920
Enable critical edge splitting by default
...
llvm-svn: 22863
2005-08-18 17:35:14 +00:00
19a271a67b
Add support for target DAG nodes that take 4 operands, such as PowerPC's
...
rlwinm.
llvm-svn: 22856
2005-08-18 07:30:15 +00:00
802080d812
Fix printing of VTSDNodes
...
llvm-svn: 22853
2005-08-18 03:31:02 +00:00
d66e616545
Move the code dependency for MathExtras.h from SelectionDAGNodes.h.
...
Added some class dividers in SelectionDAG.cpp.
llvm-svn: 22841
2005-08-17 20:08:02 +00:00
b74c666186
Culling out use of unions for converting FP to bits and vice versa.
...
llvm-svn: 22838
2005-08-17 19:34:49 +00:00
ab0de9d7fc
Fix a bug in RemoveDeadNodes where it would crash when its "optional"
...
argument is not specified.
Implement ReplaceAllUsesWith.
llvm-svn: 22834
2005-08-17 19:00:20 +00:00
686d6a1cb2
Switched to using BitsToDouble for int_to_float to avoid aliasing problem.
...
llvm-svn: 22831
2005-08-17 17:42:52 +00:00
898ba557d0
Change hex float constants for the sake of VC++.
...
llvm-svn: 22828
2005-08-17 09:44:59 +00:00
c9950c11a9
Add a new beta option for critical edge splitting, to avoid a problem that
...
Nate noticed in yacr2 (and I know occurs in other places as well).
This is still rough, as the critical edge blocks are not intelligently placed
but is added to get some idea to see if this improves performance.
llvm-svn: 22825
2005-08-17 06:37:43 +00:00
ba28c2733f
Fix a regression on X86, where FP values can be promoted too.
...
llvm-svn: 22822
2005-08-17 06:06:25 +00:00
f2516a9180
Added generic code expansion for [signed|unsigned] i32 to [f32|f64] casts in the
...
legalizer. PowerPC now uses this expansion instead of ISel version.
Example:
// signed integer to double conversion
double f1(signed x) {
return (double)x;
}
// unsigned integer to double conversion
double f2(unsigned x) {
return (double)x;
}
// signed integer to float conversion
float f3(signed x) {
return (float)x;
}
// unsigned integer to float conversion
float f4(unsigned x) {
return (float)x;
}
Byte Code:
internal fastcc double %_Z2f1i(int %x) {
entry:
%tmp.1 = cast int %x to double ; <double> [#uses=1]
ret double %tmp.1
}
internal fastcc double %_Z2f2j(uint %x) {
entry:
%tmp.1 = cast uint %x to double ; <double> [#uses=1]
ret double %tmp.1
}
internal fastcc float %_Z2f3i(int %x) {
entry:
%tmp.1 = cast int %x to float ; <float> [#uses=1]
ret float %tmp.1
}
internal fastcc float %_Z2f4j(uint %x) {
entry:
%tmp.1 = cast uint %x to float ; <float> [#uses=1]
ret float %tmp.1
}
internal fastcc double %_Z2g1i(int %x) {
entry:
%buffer = alloca [2 x uint] ; <[2 x uint]*> [#uses=3]
%tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0 ; <uint*> [#uses=1]
store uint 1127219200, uint* %tmp.0
%tmp.2 = cast int %x to uint ; <uint> [#uses=1]
%tmp.3 = xor uint %tmp.2, 2147483648 ; <uint> [#uses=1]
%tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1 ; <uint*> [#uses=1]
store uint %tmp.3, uint* %tmp.5
%tmp.9 = cast [2 x uint]* %buffer to double* ; <double*> [#uses=1]
%tmp.10 = load double* %tmp.9 ; <double> [#uses=1]
%tmp.13 = load double* cast (long* %signed_bias to double*) ; <double> [#uses=1]
%tmp.14 = sub double %tmp.10, %tmp.13 ; <double> [#uses=1]
ret double %tmp.14
}
internal fastcc double %_Z2g2j(uint %x) {
entry:
%buffer = alloca [2 x uint] ; <[2 x uint]*> [#uses=3]
%tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0 ; <uint*> [#uses=1]
store uint 1127219200, uint* %tmp.0
%tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1 ; <uint*> [#uses=1]
store uint %x, uint* %tmp.1
%tmp.4 = cast [2 x uint]* %buffer to double* ; <double*> [#uses=1]
%tmp.5 = load double* %tmp.4 ; <double> [#uses=1]
%tmp.8 = load double* cast (long* %unsigned_bias to double*) ; <double> [#uses=1]
%tmp.9 = sub double %tmp.5, %tmp.8 ; <double> [#uses=1]
ret double %tmp.9
}
internal fastcc float %_Z2g3i(int %x) {
entry:
%buffer = alloca [2 x uint] ; <[2 x uint]*> [#uses=3]
%tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0 ; <uint*> [#uses=1]
store uint 1127219200, uint* %tmp.0
%tmp.2 = cast int %x to uint ; <uint> [#uses=1]
%tmp.3 = xor uint %tmp.2, 2147483648 ; <uint> [#uses=1]
%tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1 ; <uint*> [#uses=1]
store uint %tmp.3, uint* %tmp.5
%tmp.9 = cast [2 x uint]* %buffer to double* ; <double*> [#uses=1]
%tmp.10 = load double* %tmp.9 ; <double> [#uses=1]
%tmp.13 = load double* cast (long* %signed_bias to double*) ; <double> [#uses=1]
%tmp.14 = sub double %tmp.10, %tmp.13 ; <double> [#uses=1]
%tmp.16 = cast double %tmp.14 to float ; <float> [#uses=1]
ret float %tmp.16
}
internal fastcc float %_Z2g4j(uint %x) {
entry:
%buffer = alloca [2 x uint] ; <[2 x uint]*> [#uses=3]
%tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0 ; <uint*> [#uses=1]
store uint 1127219200, uint* %tmp.0
%tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1 ; <uint*> [#uses=1]
store uint %x, uint* %tmp.1
%tmp.4 = cast [2 x uint]* %buffer to double* ; <double*> [#uses=1]
%tmp.5 = load double* %tmp.4 ; <double> [#uses=1]
%tmp.8 = load double* cast (long* %unsigned_bias to double*) ; <double> [#uses=1]
%tmp.9 = sub double %tmp.5, %tmp.8 ; <double> [#uses=1]
%tmp.11 = cast double %tmp.9 to float ; <float> [#uses=1]
ret float %tmp.11
}
PowerPC Code:
.machine ppc970
.const
.align 2
.CPIl1__Z2f1i_0: ; float 0x4330000080000000
.long 1501560836 ; float 4.5036e+15
.text
.align 2
.globl l1__Z2f1i
l1__Z2f1i:
.LBBl1__Z2f1i_0: ; entry
xoris r2, r3, 32768
stw r2, -4(r1)
lis r2, 17200
stw r2, -8(r1)
lfd f0, -8(r1)
lis r2, ha16(.CPIl1__Z2f1i_0)
lfs f1, lo16(.CPIl1__Z2f1i_0)(r2)
fsub f1, f0, f1
blr
.const
.align 2
.CPIl2__Z2f2j_0: ; float 0x4330000000000000
.long 1501560832 ; float 4.5036e+15
.text
.align 2
.globl l2__Z2f2j
l2__Z2f2j:
.LBBl2__Z2f2j_0: ; entry
stw r3, -4(r1)
lis r2, 17200
stw r2, -8(r1)
lfd f0, -8(r1)
lis r2, ha16(.CPIl2__Z2f2j_0)
lfs f1, lo16(.CPIl2__Z2f2j_0)(r2)
fsub f1, f0, f1
blr
.const
.align 2
.CPIl3__Z2f3i_0: ; float 0x4330000080000000
.long 1501560836 ; float 4.5036e+15
.text
.align 2
.globl l3__Z2f3i
l3__Z2f3i:
.LBBl3__Z2f3i_0: ; entry
xoris r2, r3, 32768
stw r2, -4(r1)
lis r2, 17200
stw r2, -8(r1)
lfd f0, -8(r1)
lis r2, ha16(.CPIl3__Z2f3i_0)
lfs f1, lo16(.CPIl3__Z2f3i_0)(r2)
fsub f0, f0, f1
frsp f1, f0
blr
.const
.align 2
.CPIl4__Z2f4j_0: ; float 0x4330000000000000
.long 1501560832 ; float 4.5036e+15
.text
.align 2
.globl l4__Z2f4j
l4__Z2f4j:
.LBBl4__Z2f4j_0: ; entry
stw r3, -4(r1)
lis r2, 17200
stw r2, -8(r1)
lfd f0, -8(r1)
lis r2, ha16(.CPIl4__Z2f4j_0)
lfs f1, lo16(.CPIl4__Z2f4j_0)(r2)
fsub f0, f0, f1
frsp f1, f0
blr
llvm-svn: 22814
2005-08-17 00:39:29 +00:00
0d2456e1f0
add a new TargetConstant node
...
llvm-svn: 22813
2005-08-17 00:34:06 +00:00
33182325f5
Eliminate the RegSDNode class, which 3 nodes (CopyFromReg/CopyToReg/ImplicitDef)
...
used to tack a register number onto the node.
Instead of doing this, make a new node, RegisterSDNode, which is a leaf
containing a register number. These three operations just become normal
DAG nodes now, instead of requiring special handling.
Note that with this change, it is no longer correct to make illegal
CopyFromReg/CopyToReg nodes. The legalizer will not touch them, and this
is bad, so don't do it. :)
llvm-svn: 22806
2005-08-16 21:55:35 +00:00
371e49515d
Implement BR_CC and BRTWOWAY_CC. This allows the removal of a rather nasty
...
fixme from the PowerPC backend. Emit slightly better code for legalizing
select_cc.
llvm-svn: 22805
2005-08-16 19:49:35 +00:00
bc89226527
Allow passing a dag into dump and getOperationName. If one is available
...
when printing a node, use it to render target operations with their
target instruction name instead of "<<unknown>>".
llvm-svn: 22804
2005-08-16 18:33:07 +00:00
7e57d18b79
Use a extant helper to do this.
...
llvm-svn: 22802
2005-08-16 18:31:23 +00:00
1973278b38
Add some methods for dag->dag isel.
...
Split RemoveNodeFromCSEMaps out of DeleteNodesIfDead to do it.
llvm-svn: 22801
2005-08-16 18:17:10 +00:00
d5e739dcc2
Fix last night's PPC32 regressions by
...
1. Not selecting the false value of a select_cc in the false arm, which
isn't legal for nested selects.
2. Actually returning the node we created and Legalized in the FP_TO_UINT
Expander.
llvm-svn: 22789
2005-08-14 18:38:32 +00:00
36853ee1fd
Teach the legalizer how to legalize FP_TO_UINT.
...
Teach the legalizer to promote FP_TO_UINT to FP_TO_SINT if the wider
FP_TO_UINT is also illegal. This allows us on PPC to codegen
unsigned short foo(float a) { return a; }
as:
_foo:
.LBB_foo_0: ; entry
fctiwz f0, f1
stfd f0, -8(r1)
lwz r2, -4(r1)
rlwinm r3, r2, 0, 16, 31
blr
instead of:
_foo:
.LBB_foo_0: ; entry
fctiwz f0, f1
stfd f0, -8(r1)
lwz r2, -4(r1)
lis r3, ha16(.CPI_foo_0)
lfs f0, lo16(.CPI_foo_0)(r3)
fcmpu cr0, f1, f0
blt .LBB_foo_2 ; entry
.LBB_foo_1: ; entry
fsubs f0, f1, f0
fctiwz f0, f0
stfd f0, -16(r1)
lwz r2, -12(r1)
xoris r2, r2, 32768
.LBB_foo_2: ; entry
rlwinm r3, r2, 0, 16, 31
blr
llvm-svn: 22785
2005-08-14 01:20:53 +00:00
dc3154ec66
Remove an unncessary argument to SimplifySelectCC and add an additional
...
assert when creating a select_cc node.
llvm-svn: 22780
2005-08-13 06:14:17 +00:00
b6651e81a0
Fix the fabs regression on x86 by abstracting the select_cc optimization
...
out into SimplifySelectCC. This allows both ISD::SELECT and ISD::SELECT_CC
to use the same set of simplifying folds.
llvm-svn: 22779
2005-08-13 06:00:21 +00:00
21381e8424
implement a couple of simple shift foldings.
...
e.g. (X & 7) >> 3 -> 0
llvm-svn: 22774
2005-08-12 23:54:58 +00:00
5c7656fd53
Add a select_cc optimization for recognizing abs(int). This speeds up an
...
integer MPEG encoding loop by a factor of two.
llvm-svn: 22758
2005-08-11 02:18:13 +00:00
180b08897f
Some SELECT_CC cleanups:
...
1. move assertions for node creation to getNode()
2. legalize the values returned in ExpandOp immediately
3. Move select_cc optimizations from SELECT's getNode() to SELECT_CC's,
allowing them to be cleaned up significantly.
This paves the way to pick up additional optimizations on SELECT_CC, such
as sum-of-absolute-differences.
llvm-svn: 22757
2005-08-11 01:12:20 +00:00
e5b86d7442
Add new node, SELECT_CC. This node is for targets that don't natively
...
implement SELECT.
llvm-svn: 22755
2005-08-10 20:51:12 +00:00
21c0fd9e8f
Fix an oversight that may be causing PR617.
...
llvm-svn: 22753
2005-08-10 17:37:53 +00:00
679f5b0b40
Fix spelling, fix some broken canonicalizations by my last patch
...
llvm-svn: 22734
2005-08-09 23:09:05 +00:00
14e060f743
add cc nodes to the AllNodes list so they show up in Graphviz output
...
llvm-svn: 22731
2005-08-09 20:40:02 +00:00
Chris Lattner