Added a check for already live regs before claiming HighRegPressure.
Fixed a few cases of checking the wrong number of successors.
Added some tracing until these heuristics are better understood.
llvm-svn: 123892
with an invalid type then split the result and perform the overflow check
normally.
Fixes the 32-bit parts of rdar://8622122 and rdar://8774702.
llvm-svn: 123864
interval after an instruction. The leaveIntvAfter() method only adds liveness
from the instruction's boundary index to the inserted copy.
Ideally, SplitKit should be smarter about this, perhaps by combining useIntv()
and leaveIntvAfter() into one method that guarantees continuity.
llvm-svn: 123858
Region splitting includes loop splitting as a subset, and it is more generic.
The splitting heuristics for variables that are live in more than one block are
now:
1. Try to create a region that covers multiple basic blocks.
2. Try to create a new live range for each block with multiple uses.
3. Spill.
Steps 2 and 3 are similar to what the standard spiller is doing.
llvm-svn: 123853
by indvars through the scev expander.
trunc(add x, y) --> add(trunc x, y). Currently SCEV largely folds the other way
which is probably wrong, but preserved to minimize churn. Instcombine doesn't
do this fold either, demonstrating a missed optz'n opportunity on code doing
add+trunc+add.
llvm-svn: 123838
of the floating point types less than 64-bits. It's somewhat of a temporary
hack but forces more accurate modeling of register pressure and results
in fewer spills.
llvm-svn: 123811
Unfortunately, while this is the "right" thing to do, it breaks some ARM
asm parsing tests because MemMode5 and ThumbMemModeReg are ambiguous. This
is tricky to resolve since neither is a subset of the other.
XFAIL the test for now. The old way was broken in other ways, just ways
we didn't happen to be testing, and our ARM asm parsing is going to require
significant revisiting at a later point anyways.
llvm-svn: 123786
are pointing to the same object, one pointer is accessing the entire
object, and the other is access has a non-zero size. This prevents
TBAA from kicking in and saying NoAlias in such cases.
llvm-svn: 123775
Analyze the live range's behavior entering and leaving basic blocks. Compute an
interference pattern for each allocation candidate, and use SpillPlacement to
find an optimal region where that register can be live.
This code is still not enabled.
llvm-svn: 123774
These were not recommended by my auto-simplifier since they don't fire often enough.
However they do fire from time to time, for example they remove one subtraction from
the final bitcode for 483.xalancbmk.
llvm-svn: 123755
simplification in fully optimized code. It occurs sporadically in the testsuite, and
many times in 403.gcc: the final bitcode has 131 fewer subtractions after this change.
The reason that the multiplies are not eliminated is the same reason that instcombine
did not catch this: they are used by other instructions (instcombine catches this with
a more general transform which in general is only profitable if the operands have only
one use).
llvm-svn: 123754
without whatever this was trying to do. When/if someone has the time to do some empirical
evaluations, it might be worth it to figure out what this code was trying to do and see if
it's worth resurrecting/fixing.
llvm-svn: 123684
checks enabled:
1) Use '<' to compare integers in a comparison function rather than '<='.
2) Use the uniqued set DefBlocks rather than Info.DefiningBlocks to initialize
the priority queue.
The speedup of scalarrepl on test-suite + SPEC2000 + SPEC2006 is a bit less, at
just under 16% rather than 17%.
llvm-svn: 123662
This shaves off 4 popcounts from the hacked 186.crafty source.
This is enabled even when a native popcount instruction is available. The
combined code is one operation longer but it should be faster nevertheless.
llvm-svn: 123621
movw r0, :lower16:(L_foo$non_lazy_ptr-(LPC0_0+4))
movt r0, :upper16:(L_foo$non_lazy_ptr-(LPC0_0+4))
LPC0_0:
add r0, pc, r0
It's not yet enabled by default as some tests are failing. I suspect bugs in
down stream tools.
llvm-svn: 123619
eliminating a potentially quadratic data structure, this also gives a 17%
speedup when running -scalarrepl on test-suite + SPEC2000 + SPEC2006. My initial
experiment gave a greater speedup around 25%, but I moved the dominator tree
level computation from dominator tree construction to PromoteMemToReg.
Since this approach to computing IDFs has a much lower overhead than the old
code using precomputed DFs, it is worth looking at using this new code for the
second scalarrepl pass as well.
llvm-svn: 123609
This fixes the original testcase in PR8927. It also causes a clang
binary built with a patched clang to increase in size by 0.21%.
We can probably get some of the size back by writing a pass that
detects that a global never has its pointer compared and adds
unnamed_addr to it (maybe extend global opt). It is also possible that
there are some other cases clang could add unnamed_addr to.
I will investigate extending globalopt next.
llvm-svn: 123584
into and/shift would cause nodes to move around and a dangling pointer
to happen. The code tried to avoid this with a HandleSDNode, but
got the details wrong.
llvm-svn: 123578
then don't try to decimate it into its individual pieces. This will just make a mess of the
IR and is pointless if none of the elements are individually accessed. This was generating
really terrible code for std::bitset (PR8980) because it happens to be lowered by clang
as an {[8 x i8]} structure instead of {i64}.
The testcase now is optimized to:
define i64 @test2(i64 %X) {
br label %L2
L2: ; preds = %0
ret i64 %X
}
before we generated:
define i64 @test2(i64 %X) {
%sroa.store.elt = lshr i64 %X, 56
%1 = trunc i64 %sroa.store.elt to i8
%sroa.store.elt8 = lshr i64 %X, 48
%2 = trunc i64 %sroa.store.elt8 to i8
%sroa.store.elt9 = lshr i64 %X, 40
%3 = trunc i64 %sroa.store.elt9 to i8
%sroa.store.elt10 = lshr i64 %X, 32
%4 = trunc i64 %sroa.store.elt10 to i8
%sroa.store.elt11 = lshr i64 %X, 24
%5 = trunc i64 %sroa.store.elt11 to i8
%sroa.store.elt12 = lshr i64 %X, 16
%6 = trunc i64 %sroa.store.elt12 to i8
%sroa.store.elt13 = lshr i64 %X, 8
%7 = trunc i64 %sroa.store.elt13 to i8
%8 = trunc i64 %X to i8
br label %L2
L2: ; preds = %0
%9 = zext i8 %1 to i64
%10 = shl i64 %9, 56
%11 = zext i8 %2 to i64
%12 = shl i64 %11, 48
%13 = or i64 %12, %10
%14 = zext i8 %3 to i64
%15 = shl i64 %14, 40
%16 = or i64 %15, %13
%17 = zext i8 %4 to i64
%18 = shl i64 %17, 32
%19 = or i64 %18, %16
%20 = zext i8 %5 to i64
%21 = shl i64 %20, 24
%22 = or i64 %21, %19
%23 = zext i8 %6 to i64
%24 = shl i64 %23, 16
%25 = or i64 %24, %22
%26 = zext i8 %7 to i64
%27 = shl i64 %26, 8
%28 = or i64 %27, %25
%29 = zext i8 %8 to i64
%30 = or i64 %29, %28
ret i64 %30
}
In this case, instcombine was able to eliminate the nonsense, but in PR8980 enough
PHIs are in play that instcombine backs off. It's better to not generate this stuff
in the first place.
llvm-svn: 123571
multiple uses. In some cases, all the uses are the same operation,
so instcombine can go ahead and promote the phi. In the testcase
this pushes an add out of the loop.
llvm-svn: 123568
http://graphics.stanford.edu/~seander/bithacks.html#CountBitsSetParallel
In a silly microbenchmark on a 65 nm core2 this is 1.5x faster than the old
code in 32 bit mode and about 2x faster in 64 bit mode. It's also a lot shorter,
especially when counting 64 bit population on a 32 bit target.
I hope this is fast enough to replace Kernighan-style counting loops even when
the input is rather sparse.
llvm-svn: 123547
half a million non-local queries, each of which would otherwise have triggered a
linear scan over a basic block.
Also fix a fixme for memory intrinsics which dereference pointers. With this,
we prove that a pointer is non-null because it was dereferenced by an intrinsic
112 times in llvm-test.
llvm-svn: 123533
The basic issue is that isel (very reasonably!) expects conditional branches
to be folded, so CGP leaving around a bunch dead computation feeding
conditional branches isn't such a good idea. Just fold branches on constants
into unconditional branches.
llvm-svn: 123526
have objectsize folding recursively simplify away their result when it
folds. It is important to catch this here, because otherwise we won't
eliminate the cross-block values at isel and other times.
llvm-svn: 123524
these would try hard to match constants by inverting the bits
and recursively matching. There are two problems with this:
1) some patterns would match when we didn't want them to (theoretical)
2) this is insanely expensive to do, and most often pointless.
This was apparently useful in just 2 instcombine cases, which I
added code to handle explicitly. This change speeds up 'opt'
time on 176.gcc by 1% and produces bitwise identical code.
llvm-svn: 123518
This is needed to allow an InstAlias for an instruction with an "OptionalDef"
result register (like ARM's cc_out) where you want to set the optional register
to reg0.
llvm-svn: 123490
disabled in this checkin. Sorry for the large diffs due to
refactoring. New functionality is all guarded by EnableSchedCycles.
Scheduling the isel DAG is inherently imprecise, but we give it a best
effort:
- Added MayReduceRegPressure to allow stalled nodes in the queue only
if there is a regpressure need.
- Added BUHasStall to allow checking for either dependence stalls due to
latency or resource stalls due to pipeline hazards.
- Added BUCompareLatency to encapsulate and standardize the heuristics
for minimizing stall cycles (vs. reducing register pressure).
- Modified the bottom-up heuristic (now in BUCompareLatency) to
prioritize nodes by their depth rather than height. As long as it
doesn't stall, height is irrelevant. Depth represents the critical
path to the DAG root.
- Added hybrid_ls_rr_sort::isReady to filter stalled nodes before
adding them to the available queue.
Related Cleanup: most of the register reduction routines do not need
to be templates.
llvm-svn: 123468
simplification present in fully optimized code (I think instcombine fails to
transform some of these when "X-Y" has more than one use). Fires here and
there all over the test-suite, for example it eliminates 8 subtractions in
the final IR for 445.gobmk, 2 subs in 447.dealII, 2 in paq8p etc.
llvm-svn: 123442
threading of shifts over selects and phis while there. This fires here and
there in the testsuite, to not much effect. For example when compiling spirit
it fires 5 times, during early-cse, resulting in 6 more cse simplifications,
and 3 more terminators being folded by jump threading, but the final bitcode
doesn't change in any interesting way: other optimizations would have caught
the opportunity anyway, only later.
llvm-svn: 123441
early in the cleanup code and one late interlaced with the inliner. The second one is
important because inlining and other scalar optzns can unpin allocas, allowing them to
be split up and promoted. While important for performance, this is also relatively
rare, and we would previously force a (non-lazy) computation of DomFrontiers, which
happened even if nothing became unpinned.
With this patch, the first pass of scalarrepl still promotes the vast bulk of allocas
in programs, but hte second pass has changed to use SSAUpdater, which is more "sparse"
and lazy. This speeds up opt -O3 time on kimwitu++ (a c++ app) by about 1%. The
numbers are interesting: the first pass promotes ~17500 allocas. The second pass
promotes about 1600. For non-C++ codes, the compile time win should be greater,
because the second pass of scalarrepl does less.
llvm-svn: 123437
instead of DomTree/DomFrontier. This may be interesting for reducing compile
time. This is currently disabled, but seems to work just fine.
When this is enabled, we eliminate two runs of dominator frontier, one in the
"early per-function" optimizations and one in the "interlaced with inliner"
function passes.
llvm-svn: 123434
- Fixed :upper16: fix up routine. It should be shifting down the top 16 bits first.
- Added support for Thumb2 :lower16: and :upper16: fix up.
- Added :upper16: and :lower16: relocation support to mach-o object writer.
llvm-svn: 123424
most important simplifications, as well as resolving phase ordering issues where instcombine
would inhibit important CSE'ing opportunities, for instance on BitBench/drop3.
llvm-svn: 123418
While there, I noticed that the transform "undef >>a X -> undef" was wrong.
For example if X is 2 then the top two bits must be equal, so the result can
not be anything. I fixed this in the constant folder as well. Also, I made
the transform for "X << undef" stronger: it now folds to undef always, even
though X might be zero. This is in accordance with the LangRef, but I must
admit that it is fairly aggressive. Also, I added "i32 X << 32 -> undef"
following the LangRef and the constant folder, likewise fairly aggressive.
llvm-svn: 123417
Add methods for accessing the (single) entry / exit edge of a region. If no such
edge exists, null is returned. Both accessors return the start block of the
corresponding edge. The edge can finally be formed by utilizing
Region::getEntry() or Region::getExit();
Contributed by: Andreas Simbuerger <simbuerg@fim.uni-passau.de>
llvm-svn: 123410
the symbolic immediate names used for these instructions, fixing their pretty-printers, and
adding proper encoding information for them.
With this, we can properly pretty-print and encode assembly like:
mrc p15, #0, r3, c13, c0, #3
Fixes <rdar://problem/8857858>.
llvm-svn: 123404
set up the source operands. The original instr has an immediate operand that
should be replaced with the frame reg operand rather than just adding the
reg operand. Previously, the instruction ended up with too many operands
causing an assert() when adding the default predicate. rdar://8825456
llvm-svn: 123387
It will still return an iterator that points to the first terminator or end(),
but there may be DBG_VALUE instructions following the first terminator.
llvm-svn: 123384
This is a minor extension of SROA to handle a special case that is
important for some ARM NEON operations. Some of the NEON intrinsics
return multiple values, which are handled as struct types containing
multiple elements of the same vector type. The corresponding return
types declared in the arm_neon.h header have equivalent arrays. We
need SROA to recognize that it can split up those arrays and structs
into separate vectors, even though they are not always accessed with
the same type. SROA already handles loads and stores of an entire
alloca by using insertvalue/extractvalue to access the individual
pieces, and that code works the same regardless of whether the type
is a struct or an array. So, all that needs to be done is to check
for compatible arrays and homogeneous structs.
llvm-svn: 123381
SROA only split up structs and arrays one level at a time, so padding can
only cause trouble if it is located in between the struct or array elements.
llvm-svn: 123380
is "X != 0 -> X" when X is a boolean. This occurs a lot because of the way
llvm-gcc converts gcc's conditional expressions. Add this, and a few other
similar transforms for completeness.
llvm-svn: 123372
in the right direction. It eliminated some hacks and will unblock codegen
work. But it's far from being done. It doesn't reject illegal expressions,
e.g. (FOO - :lower16:BAR). It also doesn't work in Thumb2 mode at all.
llvm-svn: 123369
.code 32 if the TargetMachine's isThumb() boolean does not match. The correct
fix is to switch ARM subtargets at that point and is tracked by rdar://8856789
which is bigger task.
llvm-svn: 123353
that way, unfortunately. If you want to change them to work additively instead
of a one-variant-kind-per-symbolref, that's great and I completely agree it's
worth doing, but it really should be a separate patch. Until then, this isn't
correct."
So I am reverting this bit until a more opportune time.
llvm-svn: 123340
Michael J. Spencer