Now that we have a first-class way to represent unaligned loads, the unaligned
load intrinsics are superfluous.
First part of <rdar://problem/8460511>.
llvm-svn: 129401
regs. This is the only change in this checkin that may affects the
default scheduler. With better register tracking and heuristics, it
doesn't make sense to artificially lower the register limit so much.
Added -sched-high-latency-cycles and X86InstrInfo::isHighLatencyDef to
give the scheduler a way to account for div and sqrt on targets that
don't have an itinerary. It is currently defaults to 10 (the actual
number doesn't matter much), but only takes effect on non-default
schedulers: list-hybrid and list-ilp.
Added several heuristics that can be individually disabled for the
non-default sched=list-ilp mode. This helps us determine how much
better we can do on a given benchmark than the default
scheduler. Certain compute intensive loops run much faster in this
mode with the right set of heuristics, and it doesn't seem to have
much negative impact elsewhere. Not all of the heuristics are needed,
but we still need to experiment to decide which should be disabled by
default for sched=list-ilp.
llvm-svn: 127067
"long latency" enough to hoist even if it may increase spilling. Reloading
a value from spill slot is often cheaper than performing an expensive
computation in the loop. For X86, that means machine LICM will hoist
SQRT, DIV, etc. ARM will be somewhat aggressive with VFP and NEON
instructions.
- Enable register pressure aware machine LICM by default.
llvm-svn: 116781
The reg-reg copies were no longer being generated since copyPhysReg copies
physical registers only.
The loads and stores are not necessary - The TC constraint is imposed by the
TAILJMP and TCRETURN instructions, there should be no need for constrained loads
and stores.
llvm-svn: 116314
reapply: reimplement the second half of the or/add optimization. We should now
with no changes. Turns out that one missing "Defs = [EFLAGS]" can upset things
a bit.
llvm-svn: 116040
only end up emitting LEA instead of OR. If we aren't able to promote
something into an LEA, we should never be emitting it as an ADD.
Add some testcases that we emit "or" in cases where we used to produce
an "add".
llvm-svn: 116026
is general goodness because it allows ORs to be converted to LEA to avoid
inserting copies. However, this is bad because it makes the generated .s
file less obvious and gives valgrind heartburn (tons of false positives in
bitfield code).
While the general fix should be in valgrind, we can at least try to avoid
emitting ADD instructions that *don't* get promoted to LEA. This is more
work because it requires introducing pseudo instructions to represents
"add that knows the bits are disjoint", but hey, people really love valgrind.
This fixes this testcase:
https://bugs.kde.org/show_bug.cgi?id=242137#c20
the add r/i cases are coming next.
llvm-svn: 116007
operands.
With this done, we can remove the _Int suffixes from the round instructions
without the disassembler blowing up. This allows the assembler to support
them, implementing rdar://8456376 - llvm-mc rejects 'roundss'
llvm-svn: 115019
- Make foldMemoryOperandImpl aware of 256-bit zero vectors folding and support the 128-bit counterparts of AVX too.
- Make sure MOV[AU]PS instructions are only selected when SSE1 is enabled, and duplicate the patterns to match AVX.
- Add a testcase for a simple 128-bit zero vector creation.
llvm-svn: 110946
When a register is defined by a partial load:
%reg1234:sub_32 = MOV32mr <fi#-1>; GR64:%reg1234
That load cannot be folded into an instruction using the full 64-bit register.
It would become a 64-bit load.
This is related to the recent change to have isLoadFromStackSlot return false on
a sub-register load.
llvm-svn: 110874
We do sometimes load from a too small stack slot when dealing with x86 arguments
(varargs and smaller-than-32-bit args). It looks like we know what we are doing
in those cases, so I am going to remove the assert instead of artifically
enlarging stack slot sizes.
The assert in storeRegToStackSlot stays in. We don't want to write beyond the
bounds of a stack slot.
llvm-svn: 109764
subregister operands like this:
%reg1040:sub_32bit<def> = MOV32rm <fi#-2>, 1, %reg0, 0, %reg0, %reg1040<imp-def>; mem:LD4[FixedStack-2](align=8)
Make them return false when subreg operands are present. VirtRegRewriter is
making bad assumptions otherwise.
This fixes PR7713.
llvm-svn: 109489
rip out the implementation of X86InstrInfo::GetInstSizeInBytes.
The code being ripped out just implemented a copy and hacked up
version of the (old) instruction encoder, and is buggy and
terrible in other ways. Since "GetInstSizeInBytes" is really
only there to support the JIT's "NeedsExactSize" hook (which
noone is using), just rip out the code. I will rip out the
NeedsExactSize hook next.
This resolves rdar://7617809 - switch X86InstrInfo::GetInstSizeInBytes to use X86MCCodeEmitter
llvm-svn: 109149
1) all registers were spilled as xmm, regardless of actual size
2) win64 abi doesn't do the varargs-size-in-%al thing
Still to look into:
xmm6-15 are marked as clobbered by call instructions on win64 even though they aren't.
llvm-svn: 109035
getMinimalPhysRegClass. It was used to produce spills, and it is better to
use the most specific class if possible.
Update getLoadStoreRegOpcode to handle GR32_AD.
llvm-svn: 108115
Based on a patch by Rafael Espíndola.
Attempt to make the FpSET_ST1 hack more robust, but we are still relying on
FpSET_ST0 preceeding it. This is only for supporting really weird x87 inline
asm.
We support:
FpSET_ST0
INLINEASM
FpSET_ST0
FpSET_ST1
INLINEASM
with and without kills on the arguments. We don't support:
FpSET_ST1
FpSET_ST0
INLINEASM
nor
FpSET_ST1
INLINEASM
Just Don't Do It!
llvm-svn: 108047
- Check getBytesToPopOnReturn().
- Eschew ST0 and ST1 for return values.
- Fix the PIC base register initialization so that it doesn't ever
fail to end up the top of the entry block.
llvm-svn: 108039
notes:
- The instructions are being added with dummy placeholder patterns using some 256
specifiers, this is not meant to work now, but since there are some multiclasses
generic enough to accept them, when we go for codegen, the stuff will be already
there.
- Add VEX encoding bits to support YMM
- Add MOVUPS and MOVAPS in the first round
- Use "Y" as suffix for those Instructions: MOVUPSYrr, ...
- All AVX instructions in X86InstrSSE.td will move soon to a new X86InstrAVX
file.
llvm-svn: 107996
like all other instructions, even though a segment is not
allowed. This resolves a bunch of gross hacks in the
encoder and makes LEA more consistent with the rest of the
instruction set.
No functionality change.
llvm-svn: 107934
EXTRACT_SUBREG no longer appears as a machine instruction. Use COPY instead.
Add isCopy() checks in many places using isMoveInstr() and isExtractSubreg().
The isMoveInstr hook will be removed later.
llvm-svn: 107879
- X86 unfolding should check if the instructions being unfolded has memoperands.
If there is no memoperands, then it must assume conservative alignment. If this
would introduce an expensive sse unaligned load / store, then unfoldMemoryOperand
etc. should not unfold the instruction.
llvm-svn: 107509
addresses a longstanding deficiency noted in many FIXMEs scattered
across all the targets.
This effectively moves the problem up one level, replacing eleven
FIXMEs in the targets with eight FIXMEs in CodeGen, plus one path
through FastISel where we actually supply a DebugLoc, fixing Radar
7421831.
llvm-svn: 106243
instruction defines subregisters.
Any existing subreg indices on the original instruction are preserved or
composed with the new subreg index.
Also substitute multiple operands mentioning the original register by using the
new MachineInstr::substituteRegister() function. This is necessary because there
will soon be <imp-def> operands added to non read-modify-write partial
definitions. This instruction:
%reg1234:foo = FLAP %reg1234<imp-def>
will reMaterialize(%reg3333, bar) like this:
%reg3333:bar-foo = FLAP %reg333:bar<imp-def>
Finally, replace the TargetRegisterInfo pointer argument with a reference to
indicate that it cannot be NULL.
llvm-svn: 105358
This is the beginning of purely symbolic subregister indices, but we need a bit
of jiggling before the explicit numeric indices can be completely removed.
llvm-svn: 104492
otherwise labels get incorrectly merged. We handled this by emitting a
".byte 0", but this isn't correct on thumb/arm targets where the text segment
needs to be a multiple of 2/4 bytes. Handle this by emitting a noop. This
is more gross than it should be because arm/ppc are not fully mc'ized yet.
This fixes rdar://7908505
llvm-svn: 102400
If we have this situation:
jCC L1
jmp L2
L1:
...
L2:
...
We can get a small performance boost by emitting this instead:
jnCC L2
L1:
...
L2:
...
This testcase shows an example of this:
float func(float x, float y) {
double product = (double)x * y;
if (product == 0.0)
return product;
return product - 1.0;
}
llvm-svn: 101075
SSEDomainFix will collapse to the domain with the lower number when it has a
choice. The SSEPackedSingle domain often has smaller instructions, so prefer
that.
llvm-svn: 99952
On Nehalem and newer CPUs there is a 2 cycle latency penalty on using a register
in a different domain than where it was defined. Some instructions have
equvivalents for different domains, like por/orps/orpd.
The SSEDomainFix pass tries to minimize the number of domain crossings by
changing between equvivalent opcodes where possible.
This is a work in progress, in particular the pass doesn't do anything yet. SSE
instructions are tagged with their execution domain in TableGen using the last
two bits of TSFlags. Note that not all instructions are tagged correctly. Life
just isn't that simple.
The SSE execution domain issue is very similar to the ARM NEON/VFP pipeline
issue handled by NEONMoveFixPass. This pass may become target independent to
handle both.
llvm-svn: 99524
This is work in progress. So far, SSE execution domain tables are added to
X86InstrInfo, and a skeleton pass is enabled with -sse-domain-fix.
llvm-svn: 99345
This code:
float floatingPointComparison(float x, float y) {
double product = (double)x * y;
if (product == 0.0)
return product;
return product - 1.0;
}
produces this:
_floatingPointComparison:
0000000000000000 cvtss2sd %xmm1,%xmm1
0000000000000004 cvtss2sd %xmm0,%xmm0
0000000000000008 mulsd %xmm1,%xmm0
000000000000000c pxor %xmm1,%xmm1
0000000000000010 ucomisd %xmm1,%xmm0
0000000000000014 jne 0x00000004
0000000000000016 jp 0x00000002
0000000000000018 jmp 0x00000008
000000000000001a addsd 0x00000006(%rip),%xmm0
0000000000000022 cvtsd2ss %xmm0,%xmm0
0000000000000026 ret
The "jne/jp/jmp" sequence can be reduced to this instead:
_floatingPointComparison:
0000000000000000 cvtss2sd %xmm1,%xmm1
0000000000000004 cvtss2sd %xmm0,%xmm0
0000000000000008 mulsd %xmm1,%xmm0
000000000000000c pxor %xmm1,%xmm1
0000000000000010 ucomisd %xmm1,%xmm0
0000000000000014 jp 0x00000002
0000000000000016 je 0x00000008
0000000000000018 addsd 0x00000006(%rip),%xmm0
0000000000000020 cvtsd2ss %xmm0,%xmm0
0000000000000024 ret
for a savings of 2 bytes.
This xform can happen when we recognize that jne and jp jump to the same "true"
MBB, the unconditional jump would jump to the "false" MBB, and the "true" branch
is the fall-through MBB.
llvm-svn: 97766
Extracting the low element of a vector is now done with EXTRACT_SUBREG,
and the zero-extension performed by load movss is now modeled with
SUBREG_TO_REG, and so on.
Register-to-register movss and movsd are no longer considered copies;
they are two-address instructions which insert a scalar into a vector.
llvm-svn: 97354
into TargetOpcodes.h. #include the new TargetOpcodes.h
into MachineInstr. Add new inline accessors (like isPHI())
to MachineInstr, and start using them throughout the
codebase.
llvm-svn: 95687
function can support dynamic stack realignment. That's a much easier question
to answer at instruction selection stage than whether the function actually
will have dynamic alignment prologue. This allows the removal of the
stack alignment heuristic pass, and improves code quality for cases where
the heuristic would result in dynamic alignment code being generated when
it was not strictly necessary.
llvm-svn: 93885
For now, this pass is fairly conservative. It only perform the replacement when both the pre- and post- extension values are used in the block. It will miss cases where the post-extension values are live, but not used.
llvm-svn: 93278
instruction is copy like where the source and destination registers can
overlap. This is to be used by the coalescable to coalesce the source and
destination registers of instructions like X86::MOVSX64rr32. Apparently
some crazy people believe the coalescer is too simple.
llvm-svn: 93210
new AsmPrinter. This is perhaps less elegant than describing them
in terms of MOV32r0 and subreg operations, but it allows the
current register to rematerialize them.
llvm-svn: 93158
be non-optimal. To be precise, we should avoid folding loads if the instructions
only update part of the destination register, and the non-updated part is not
needed. e.g. cvtss2sd, sqrtss. Unfolding the load from these instructions breaks
the partial register dependency and it can improve performance. e.g.
movss (%rdi), %xmm0
cvtss2sd %xmm0, %xmm0
instead of
cvtss2sd (%rdi), %xmm0
An alternative method to break dependency is to clear the register first. e.g.
xorps %xmm0, %xmm0
cvtss2sd (%rdi), %xmm0
llvm-svn: 91672
Note that "hasDotLocAndDotFile"-style debug info was already broken;
people wanting this functionality should implement it in the
AsmPrinter/DwarfWriter code.
llvm-svn: 89711
Provide special isLoadFromStackSlotPostFE and isStoreToStackSlotPostFE
interfaces to explicitly request checking for post-frame ptr elimination
operands. This uses a heuristic so it isn't reliable for correctness.
llvm-svn: 87047
machine instruction loads or stores from/to a stack slot. Unlike
isLoadFromStackSlot and isStoreFromStackSlot, the instruction may be
something other than a pure load/store (e.g. it may be an arithmetic
operation with a memory operand). This helps AsmPrinter determine when
to print a spill/reload comment.
This is only a hint since we may not be able to figure this out in all
cases. As such, it should not be relied upon for correctness.
Implement for X86. Return false by default for other architectures.
llvm-svn: 87026
This patch forbids implicit conversion of DenseMap::const_iterator to
DenseMap::iterator which was possible because DenseMapIterator inherited
(publicly) from DenseMapConstIterator. Conversion the other way around is now
allowed as one may expect.
The template DenseMapConstIterator is removed and the template parameter
IsConst which specifies whether the iterator is constant is added to
DenseMapIterator.
Actually IsConst parameter is not necessary since the constness can be
determined from KeyT but this is not relevant to the fix and can be addressed
later.
Patch by Victor Zverovich!
llvm-svn: 86636
unfolding loads for hoisting. getOpcodeAfterMemoryUnfold returns the
opcode of the original operation without the load, not the load
itself, MachineLICM needs to know the operand index in order to get
the correct register class. Extend getOpcodeAfterMemoryUnfold to
return this information.
llvm-svn: 85622
(for uses marked kill and defs marked dead) a few instructions in
addition to forwards. Also, increase the maximum number of instructions
to scan, as it appears to help in a fair number of cases.
llvm-svn: 84061
implementations with a new MachineInstr::isInvariantLoad, which uses
MachineMemOperands and is target-independent. This brings MachineLICM
and other functionality to targets which previously lacked an
isInvariantLoad implementation.
llvm-svn: 83475
unused DECLARE instruction.
KILL is not yet used anywhere, it will replace TargetInstrInfo::IMPLICIT_DEF
in the places where IMPLICIT_DEF is just used to alter liveness of physical
registers.
llvm-svn: 83006
- Allocate MachineMemOperands and MachineMemOperand lists in MachineFunctions.
This eliminates MachineInstr's std::list member and allows the data to be
created by isel and live for the remainder of codegen, avoiding a lot of
copying and unnecessary translation. This also shrinks MemSDNode.
- Delete MemOperandSDNode. Introduce MachineSDNode which has dedicated
fields for MachineMemOperands.
- Change MemSDNode to have a MachineMemOperand member instead of its own
fields with the same information. This introduces some redundancy, but
it's more consistent with what MachineInstr will eventually want.
- Ignore alignment when searching for redundant loads for CSE, but remember
the greatest alignment.
Target-specific code which previously used MemOperandSDNodes with generic
SDNodes now use MemIntrinsicSDNodes, with opcodes in a designated range
so that the SelectionDAG framework knows that MachineMemOperand information
is available.
llvm-svn: 82794
naming scheme used in SelectionDAG, where there are multiple kinds
of "target" nodes, but "machine" nodes are nodes which represent
a MachineInstr.
llvm-svn: 82790
safe. This can happen we a subreg_to_reg 0 has been coalesced. One
exception is when the instruction that folds the load is a move, then we
can simply turn it into a 32-bit load from the stack slot.
rdar://7170444
llvm-svn: 81494
MachineInstr and MachineOperand. This required eliminating a
bunch of stuff that was using DOUT, I hope that bill doesn't
mind me stealing his fun. ;-)
llvm-svn: 79813
the only real caller (GetFunctionSizeInBytes) uses it.
The custom ARM implementation of this is basically reimplementing
an assembler poorly for negligible gain. It should be removed
IMNSHO, but I'll leave that to ARMish folks to decide.
llvm-svn: 77877
and convert code to using it, instead of having lots of things
poke the isLookupPtrRegClass() method directly.
2. Make PointerLikeRegClass contain a 'kind' int, and store it in
the existing regclass field of TargetOperandInfo when the
isLookupPtrRegClass() predicate is set. Make getRegClass pass
this into TargetRegisterInfo::getPointerRegClass(), allowing
targets to have multiple ptr_rc things.
llvm-svn: 77504
This adds location info for all llvm_unreachable calls (which is a macro now) in
!NDEBUG builds.
In NDEBUG builds location info and the message is off (it only prints
"UREACHABLE executed").
llvm-svn: 75640
This involves temporarily hard wiring some parts to use the global context. This isn't ideal, but it's
the only way I could figure out to make this process vaguely incremental.
llvm-svn: 75445
Make llvm_unreachable take an optional string, thus moving the cerr<< out of
line.
LLVM_UNREACHABLE is now a simple wrapper that makes the message go away for
NDEBUG builds.
llvm-svn: 75379
registers based on dynamic conditions. For example, X86 EBP/RBP, when used as
frame register has to be spilled in the first fixed object. It should inform
PEI this so it doesn't get allocated another stack object. Also, it should not
be spilled as other callee-saved registers but rather its spilling and restoring
are being handled by emitPrologue and emitEpilogue. Avoid spilling it twice.
llvm-svn: 75116
The register allocator, when it allocates a register to a virtual register defined by an implicit_def, can allocate any physical register without worrying about overlapping live ranges. It should mark all of operands of the said virtual register so later passes will do the right thing.
This is not the best solution. But it should be a lot less fragile to having the scavenger try to track what is defined by implicit_def.
llvm-svn: 74518
implementation primarily differs from the former in that the asmprinter
doesn't make a zillion decisions about whether or not something will be
RIP relative or not. Instead, those decisions are made by isel lowering
and propagated through to the asm printer. To achieve this, we:
1. Represent RIP relative addresses by setting the base of the X86 addr
mode to X86::RIP.
2. When ISel Lowering decides that it is safe to use RIP, it lowers to
X86ISD::WrapperRIP. When it is unsafe to use RIP, it lowers to
X86ISD::Wrapper as before.
3. This removes isRIPRel from X86ISelAddressMode, representing it with
a basereg of RIP instead.
4. The addressing mode matching logic in isel is greatly simplified.
5. The asmprinter is greatly simplified, notably the "NotRIPRel" predicate
passed through various printoperand routines is gone now.
6. The various symbol printing routines in asmprinter now no longer infer
when to emit (%rip), they just print the symbol.
I think this is a big improvement over the previous situation. It does have
two small caveats though: 1. I implemented a horrible "no-rip" modifier for
the inline asm "P" constraint modifier. This is a short term hack, there is
a much better, but more involved, solution. 2. I had to xfail an
-aggressive-remat testcase because it isn't handling the use of RIP in the
constant-pool reading instruction. This specific test is easy to fix without
-aggressive-remat, which I intend to do next.
llvm-svn: 74372
decoding. Essentially, they both map to the same column in the "opcode
extensions for one- and two-byte opcodes" table in the x86 manual. The RawFrm
complicates decoding this.
Instead, use opcode 0x01, prefix 0x01, and form MRM1r. Then have the code
emitter special case these, a la [SML]FENCE.
llvm-svn: 72556
booleans. This gives a better indication of what the "addReg()" is
doing. Remembering what all of those booleans mean isn't easy, especially if you
aren't spending all of your time in that code.
I took Jakob's suggestion and made it illegal to pass in "true" for the
flag. This should hopefully prevent any unintended misuse of this (by reverting
to the old way of using addReg()).
llvm-svn: 71722
to precisely describe the h-register subreg register classes.
Thanks to Jakob Stoklund Olesen for spotting this and for the
initial patch!
Also, make getStoreRegOpcode and getLoadRegOpcode aware of the
needs of h registers.
llvm-svn: 70211
- Add patterns for h-register extract, which avoids a shift and mask,
and in some cases a temporary register.
- Add address-mode matching for turning (X>>(8-n))&(255<<n), where
n is a valid address-mode scale value, into an h-register extract
and a scaled-offset address.
- Replace X86's MOV32to32_ and related instructions with the new
target-independent COPY_TO_SUBREG instruction.
On x86-64 there are complicated constraints on h registers, and
CodeGen doesn't currently provide a high-level way to express all of them,
so they are handled with a bunch of special code. This code currently only
supports extracts where the result is used by a zero-extend or a store,
though these are fairly common.
These transformations are not always beneficial; since there are only
4 h registers, they sometimes require extra move instructions, and
this sometimes increases register pressure because it can force out
values that would otherwise be in one of those registers. However,
this appears to be relatively uncommon.
llvm-svn: 68962
builds.
--- Reverse-merging (from foreign repository) r68552 into '.':
U test/CodeGen/X86/tls8.ll
U test/CodeGen/X86/tls10.ll
U test/CodeGen/X86/tls2.ll
U test/CodeGen/X86/tls6.ll
U lib/Target/X86/X86Instr64bit.td
U lib/Target/X86/X86InstrSSE.td
U lib/Target/X86/X86InstrInfo.td
U lib/Target/X86/X86RegisterInfo.cpp
U lib/Target/X86/X86ISelLowering.cpp
U lib/Target/X86/X86CodeEmitter.cpp
U lib/Target/X86/X86FastISel.cpp
U lib/Target/X86/X86InstrInfo.h
U lib/Target/X86/X86ISelDAGToDAG.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.h
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.h
U lib/Target/X86/X86ISelLowering.h
U lib/Target/X86/X86InstrInfo.cpp
U lib/Target/X86/X86InstrBuilder.h
U lib/Target/X86/X86RegisterInfo.td
llvm-svn: 68560
This introduces a small regression on the generated code
quality in the case we are just computing addresses, not
loading values.
Will work on it and on X86-64 support.
llvm-svn: 68552
1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants.
2. MachineConstantPool alignment field is also a log2 value.
3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values.
4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries.
5. Asm printer uses expensive data structure multimap to track constant pool entries by sections.
6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic.
Solutions:
1. ConstantPoolSDNode alignment field is changed to keep non-log2 value.
2. MachineConstantPool alignment field is also changed to keep non-log2 value.
3. Functions that create ConstantPool nodes are passing in non-log2 alignments.
4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT.
5. Asm printer uses cheaper data structure to group constant pool entries.
6. Asm printer compute entry offsets after grouping is done.
7. Change JIT code to compute entry offsets on the fly.
llvm-svn: 66875
of MachineInstr def operands must be subtracted out. This bug
was uncovered by the recent x86 EFLAGS optimization. Before
that, the only instructions that ever needed unfolding were
things like CMP32rm, where NumDefs is zero.
llvm-svn: 66056
suprise to some callers, e.g. register coalescer. For now, add an parameter
that tells AnalyzeBranch whether it's safe to modify the mbb. A better
solution is out there, but I don't have time to deal with it right now.
llvm-svn: 64124