The Pattern isel has some strange memory corruption issues going on. :(
This should have been converted over anyway, but it got forgotten somehow
when switching to the dag isel.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23523 91177308-0d34-0410-b5e6-96231b3b80d8
ReplaceAllUsesWith does not replace scalars SDOperand floating around on
the stack, permitting things to be selected multiple times.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23515 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23505 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23504 91177308-0d34-0410-b5e6-96231b3b80d8
on setjmp calls and restores it on longjmp calls (both of which require syscalls).
This makes the calls REALLY slow. Use _setjmp/_longjmp instead. This speeds up
hexxagon from 120.31s to 15.68s: from 5.53x slower than GCC to 28% faster than GCC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23482 91177308-0d34-0410-b5e6-96231b3b80d8
is performed so it is only at most once per function that contains an invoke
instead of once per invoke in the function. This patch has the following perks:
1. It fixes PR631, which complains about slowness.
2. If fixes PR240, which complains about non-volatile vars being live across
setjmp/longjmps.
3. It improves (but does not fix) the jmpbuf alignment issue on itanium by not
forcing the jmpbufs to always be 8-bytes off the alignment of the structure.
4. It speeds up 253.perlbmk from 338s to 13.70s (a 25x improvement!), making us
now about 4% faster than GCC.
Further improvements are also possible.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23477 91177308-0d34-0410-b5e6-96231b3b80d8
This speeds up isa/dyncast/etc for constants, and also makes them smaller.
For example, the text section of a release build of InstCombine.cpp shrinks
from 230037 bytes to 216363 bytes, a 6% reduction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23467 91177308-0d34-0410-b5e6-96231b3b80d8
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)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23455 91177308-0d34-0410-b5e6-96231b3b80d8
Implement the start of global ctor optimization. It is currently smart
enough to remove the global ctor for cases like this:
struct foo {
foo() {}
} x;
... saving a bit of startup time for the program.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23433 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23403 91177308-0d34-0410-b5e6-96231b3b80d8
not define a value that is used outside of it's block. This catches many
more simplifications, e.g. 854 in 176.gcc, 137 in vpr, etc.
This implements branch-phi-thread.ll:test3.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23397 91177308-0d34-0410-b5e6-96231b3b80d8
control across branches with determined outcomes. More generality to follow.
This triggers a couple thousand times in specint.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23391 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23388 91177308-0d34-0410-b5e6-96231b3b80d8
specified. The various *imm operands defined by PPC are really all i32,
even though the actual immediate is restricted to a smaller value in it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23352 91177308-0d34-0410-b5e6-96231b3b80d8
with incoming arguments instead of the pregs themselves. This fixes
the scheduler from causing problems by moving a copyfromreg for an argument
to after a select_cc node (now it can, and bad things won't happen).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23334 91177308-0d34-0410-b5e6-96231b3b80d8
This is useful for 178.galgel where resolution of dope vectors (by the
optimizer) causes the scales to become apparent.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23328 91177308-0d34-0410-b5e6-96231b3b80d8
if () { store A -> P; } else { store B -> P; }
into a PHI node with one store, in the most trival case. This implements
load.ll:test10.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23324 91177308-0d34-0410-b5e6-96231b3b80d8
load are exactly consequtive. This is picked up by other passes, but this
triggers thousands of times in fortran programs that use static locals
(and is thus a compile-time speedup).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23320 91177308-0d34-0410-b5e6-96231b3b80d8
code for IV uses outside of loops that are not dominated by the latch block.
We should only convert these uses to use the post-inc value if they ARE
dominated by the latch block.
Also use a new LoopInfo method to simplify some code.
This fixes Transforms/LoopStrengthReduce/2005-09-12-UsesOutOutsideOfLoop.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23318 91177308-0d34-0410-b5e6-96231b3b80d8
select (x < y), 1, 0 -> (x < y) incorrectly: the setcc returns i1 but the
select returned i32. Add the zero extend as needed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23301 91177308-0d34-0410-b5e6-96231b3b80d8
with copies, leading to code like this:
lwz r4, 380(r1)
or r10, r4, r4 ;; Last use of r4
By teaching the PPC backend how to fold spills into copies, we now get this
code:
lwz r10, 380(r1)
wow. :)
This reduces a testcase nate sent me from 1505 instructions to 1484.
Note that this could handle FP values but doesn't currently, for reasons
mentioned in the patch
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23298 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23296 91177308-0d34-0410-b5e6-96231b3b80d8
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).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23293 91177308-0d34-0410-b5e6-96231b3b80d8
is zero. This lets the register allocator elide some copies in some cases.
This implements CodeGen/PowerPC/rlwimi-commute.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23292 91177308-0d34-0410-b5e6-96231b3b80d8
1. Add support for defining Pattern's, which can match expressions when there
is no instruction that directly implements something. Instructions usually
implicitly define patterns.
2. Add support for defining SDNodeXForm's, which are node transformations.
This seperates the concept of a node xform out from the existing predicate
support.
Using this new stuff, we add a few instruction patterns, one for testing, and
two for OR/XOR by an arbitrary immediate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23286 91177308-0d34-0410-b5e6-96231b3b80d8
the rest of the instructions, add comment markers to seperate portions of
the file into logical parts
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23277 91177308-0d34-0410-b5e6-96231b3b80d8
This restores all of stanford to being identical with and without the dag
combiner with the add folding turned off in sd.cpp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23258 91177308-0d34-0410-b5e6-96231b3b80d8
'' is not a recognized processor for this target (ignoring processor)
Default to "generic" instead of "" for the default CPU.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23257 91177308-0d34-0410-b5e6-96231b3b80d8
'' is not a recognized processor for this target (ignoring processor)
instead of:
is not a recognized processor for this target (ignoring processor)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23256 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23255 91177308-0d34-0410-b5e6-96231b3b80d8
are allowed to generate 64-bit-only PowerPC instructions for 32 bit hosts,
such as the PowerPC 970.
This speeds up 189.lucas from 81.99 to 32.64 seconds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23250 91177308-0d34-0410-b5e6-96231b3b80d8
I have run so far when run before Legalize. It still needs to pick up the
SetCC folds, and nodes that use SetCC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23243 91177308-0d34-0410-b5e6-96231b3b80d8
Define the PatFrag class which can be used to define subpatterns to match
things with. Define 'not', and use it to define the patterns for andc,
nand, etc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23233 91177308-0d34-0410-b5e6-96231b3b80d8
currently don't do anything. This elides patterns for binary operators
that ping on the carry flag, since we don't model it yet.
This patch also removes PPC::SUB, because it is dead.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23230 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23228 91177308-0d34-0410-b5e6-96231b3b80d8
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!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23221 91177308-0d34-0410-b5e6-96231b3b80d8
in building maximal expressions before simplifying them. In particular, i
cases like this:
X-(A+B+X)
the code would consider A+B+X to be a maximal expression (not understanding
that the single use '-' would be turned into a + later), simplify it (a noop)
then later get simplified again.
Each of these simplify steps is where the cost of reassociation comes from,
so this patch should speed up the already fast pass a bit.
Thanks to Dan for noticing this!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23214 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23205 91177308-0d34-0410-b5e6-96231b3b80d8
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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23196 91177308-0d34-0410-b5e6-96231b3b80d8
platforms. This reduces executable size and makes shark realize the actual
bounds of functions instead of showing each MBB as a function :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23193 91177308-0d34-0410-b5e6-96231b3b80d8
2. Propagate feature "string" to all targets.
3. Implement use of SubtargetFeatures in PowerPCTargetSubtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23192 91177308-0d34-0410-b5e6-96231b3b80d8
is to manage processor specific attributes from the command line. See examples
of use in llc/lli and PowerPCTargetSubtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23191 91177308-0d34-0410-b5e6-96231b3b80d8
directly out of R1 (without using a CopyFromReg, which uses a chain), multiple
allocas were getting CSE'd together, producing bogus code. For this:
int %foo(bool %X, int %A, int %B) {
br bool %X, label %T, label %F
F:
%G = alloca int
%H = alloca int
store int %A, int* %G
store int %B, int* %H
%R = load int* %G
ret int %R
T:
ret int 0
}
We were generating:
_foo:
stwu r1, -16(r1)
stw r31, 4(r1)
or r31, r1, r1
stw r1, 12(r31)
cmpwi cr0, r3, 0
bne cr0, .LBB_foo_2 ; T
.LBB_foo_1: ; F
li r2, 16
subf r2, r2, r1 ;; One alloca
or r1, r2, r2
or r3, r1, r1
or r1, r2, r2
or r2, r1, r1
stw r4, 0(r3)
stw r5, 0(r2)
lwz r3, 0(r3)
lwz r1, 12(r31)
lwz r31, 4(r31)
lwz r1, 0(r1)
blr
.LBB_foo_2: ; T
li r3, 0
lwz r1, 12(r31)
lwz r31, 4(r31)
lwz r1, 0(r1)
blr
Now we generate:
_foo:
stwu r1, -16(r1)
stw r31, 4(r1)
or r31, r1, r1
stw r1, 12(r31)
cmpwi cr0, r3, 0
bne cr0, .LBB_foo_2 ; T
.LBB_foo_1: ; F
or r2, r1, r1
li r3, 16
subf r2, r3, r2 ;; Alloca 1
or r1, r2, r2
or r2, r1, r1
or r6, r1, r1
subf r3, r3, r6 ;; Alloca 2
or r1, r3, r3
or r3, r1, r1
stw r4, 0(r2)
stw r5, 0(r3)
lwz r3, 0(r2)
lwz r1, 12(r31)
lwz r31, 4(r31)
lwz r1, 0(r1)
blr
.LBB_foo_2: ; T
li r3, 0
lwz r1, 12(r31)
lwz r31, 4(r31)
lwz r1, 0(r1)
blr
This fixes Povray and SPASS with the dag isel, the last two failing cases.
Tommorow we will hopefully turn it on by default! :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23190 91177308-0d34-0410-b5e6-96231b3b80d8
could cause a miscompile. Fixing this didn't fix the two programs that fail
though. :(
This also changes the implementation to follow the pattern selector more
closely, causing us to select 0 to li instead of lis.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23189 91177308-0d34-0410-b5e6-96231b3b80d8
