By default, store all local variables in dynamic alloca instead of
static one. It reduces the stack space usage in use-after-return mode
(dynamic alloca will not be called if the local variables are stored
in a fake stack), and improves the debug info quality for local
variables (they will not be described relatively to %rbp/%rsp, which
are assumed to be clobbered by function calls).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228336 91177308-0d34-0410-b5e6-96231b3b80d8
PassManager instance. In one case we can make the determination
from the Triple, in the other (execution dependency pass) the
pass will avoid running if we don't have any code that uses that
register class so go ahead and add it to the pipeline.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228334 91177308-0d34-0410-b5e6-96231b3b80d8
dealing with module level emission. Currently this is using
the Triple to determine, but eventually the logic should
probably migrate to TLOF.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228332 91177308-0d34-0410-b5e6-96231b3b80d8
PowerPC supports pre-increment load/store instructions (except for Altivec/VSX
vector load/stores). Using these on embedded cores can be very important, but
most loops are not naturally set up to use them. We can often change that,
however, by placing loops into a non-canonical form. Generically, this means
transforming loops like this:
for (int i = 0; i < n; ++i)
array[i] = c;
to look like this:
T *p = array[-1];
for (int i = 0; i < n; ++i)
*++p = c;
the key point is that addresses accessed are pulled into dedicated PHIs and
"pre-decremented" in the loop preheader. This allows the use of pre-increment
load/store instructions without loop peeling.
A target-specific late IR-level pass (running post-LSR), PPCLoopPreIncPrep, is
introduced to perform this transformation. I've used this code out-of-tree for
generating code for the PPC A2 for over a year. Somewhat to my surprise,
running the test suite + externals on a P7 with this transformation enabled
showed no performance regressions, and one speedup:
External/SPEC/CINT2006/483.xalancbmk/483.xalancbmk
-2.32514% +/- 1.03736%
So I'm going to enable it on everything for now. I was surprised by this
because, on the POWER cores, these pre-increment load/store instructions are
cracked (and, thus, harder to schedule effectively). But seeing no regressions,
and feeling that it is generally easier to split instructions apart late than
it is to combine them late, this might be the better approach regardless.
In the future, we might want to integrate this functionality into LSR (but
currently LSR does not create new PHI nodes, so (for that and other reasons)
significant work would need to be done).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228328 91177308-0d34-0410-b5e6-96231b3b80d8
PowerPC supports pre-increment floating-point load/store instructions, both r+r
and r+i, and we had patterns for them, but they were not marked as legal. Mark
them as legal (and add a test case).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228327 91177308-0d34-0410-b5e6-96231b3b80d8
The combine that forms extloads used to be disabled on vector types,
because "None of the supported targets knows how to perform load and
sign extend on vectors in one instruction."
That's not entirely true, since at least SSE4.1 X86 knows how to do
those sextloads/zextloads (with PMOVS/ZX).
But there are several aspects to getting this right.
First, vector extloads are controlled by a profitability callback.
For instance, on ARM, several instructions have folded extload forms,
so it's not always beneficial to create an extload node (and trying to
match extloads is a whole 'nother can of worms).
The interesting optimization enables folding of s/zextloads to illegal
(splittable) vector types, expanding them into smaller legal extloads.
It's not ideal (it introduces some legalization-like behavior in the
combine) but it's better than the obvious alternative: form illegal
extloads, and later try to split them up. If you do that, you might
generate extloads that can't be split up, but have a valid ext+load
expansion. At vector-op legalization time, it's too late to generate
this kind of code, so you end up forced to scalarize. It's better to
just avoid creating egregiously illegal nodes.
This optimization is enabled unconditionally on X86.
Note that the splitting combine is happy with "custom" extloads. As
is, this bypasses the actual custom lowering, and just unrolls the
extload. But from what I've seen, this is still much better than the
current custom lowering, which does some kind of unrolling at the end
anyway (see for instance load_sext_4i8_to_4i64 on SSE2, and the added
FIXME).
Also note that the existing combine that forms extloads is now also
enabled on legal vectors. This doesn't have a big effect on X86
(because sext+load is usually combined to sext_inreg+aextload).
On ARM it fires on some rare occasions; that's for a separate commit.
Differential Revision: http://reviews.llvm.org/D6904
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228325 91177308-0d34-0410-b5e6-96231b3b80d8
The return value's address must be returned in %rax.
i.e. the callee needs to copy the sret argument (%rdi)
into the return value (%rax).
This probably won't manifest as a bug when the caller is LLVM-compiled
code. But it is an ABI guarantee and tools expect it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228321 91177308-0d34-0410-b5e6-96231b3b80d8
We should be setting UnrollingPreferences::MaxCount to MAX_UINT instead
of UnrollingPreferences::Count.
Count is a 'forced unrolling factor', while MaxCount sets an upper
limit to the unrolling factor.
Setting Count to MAX_UINT was causing the loop in the testcase to be
unrolled 15 times, when it only had a maximum of 4 iterations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228303 91177308-0d34-0410-b5e6-96231b3b80d8
The llvm.SI.end.cf intrinsic is used to mark the end of if-then blocks,
if-then-else blocks, and loops. It is responsible for updating the
exec mask to re-enable threads that had been masked during the preceding
control flow block. For example:
s_mov_b64 exec, 0x3 ; Initial exec mask
s_mov_b64 s[0:1], exec ; Saved exec mask
v_cmpx_gt_u32 exec, s[2:3], v0, 0 ; llvm.SI.if
do_stuff()
s_or_b64 exec, exec, s[0:1] ; llvm.SI.end.cf
The bug fixed by this patch was one where the llvm.SI.end.cf intrinsic
was being inserted into the header of loops. This would happen when
an if block terminated in a loop header and we would end up with
code like this:
s_mov_b64 exec, 0x3 ; Initial exec mask
s_mov_b64 s[0:1], exec ; Saved exec mask
v_cmpx_gt_u32 exec, s[2:3], v0, 0 ; llvm.SI.if
do_stuff()
LOOP: ; Start of loop header
s_or_b64 exec, exec, s[0:1] ; llvm.SI.end.cf <-BUG: The exec mask has the
same value at the beginning of each loop
iteration.
do_stuff();
s_cbranch_execnz LOOP
The fix is to create a new basic block before the loop and insert the
llvm.SI.end.cf there. This way the exec mask is restored before the
start of the loop instead of at the beginning of each iteration.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228302 91177308-0d34-0410-b5e6-96231b3b80d8
Patch by Kit Barton.
Add the vector count leading zeros instruction for byte, halfword,
word, and doubleword sizes. This is a fairly straightforward addition
after the changes made for vpopcnt:
1. Add the correct definitions for the various instructions in
PPCInstrAltivec.td
2. Make the CTLZ operation legal on vector types when using P8Altivec
in PPCISelLowering.cpp
Test Plan
Created new test case in test/CodeGen/PowerPC/vec_clz.ll to check the
instructions are being generated when the CTLZ operation is used in
LLVM.
Check the encoding and decoding in test/MC/PowerPC/ppc_encoding_vmx.s
and test/Disassembler/PowerPC/ppc_encoding_vmx.txt respectively.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228301 91177308-0d34-0410-b5e6-96231b3b80d8
Implement a BITCAST dag combine to transform i32->mmx conversion patterns
into a X86 specific node (MMX_MOVW2D) and guarantee that moves between
i32 and x86mmx are better handled, i.e., don't use store-load to do the
conversion..
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228293 91177308-0d34-0410-b5e6-96231b3b80d8
Complete loop unrolling can make some loads constant, thus enabling a
lot of other optimizations. To catch such cases, we look for loads that
might become constants and estimate number of instructions that would be
simplified or become dead after substitution.
Example:
Suppose we have:
int a[] = {0, 1, 0};
v = 0;
for (i = 0; i < 3; i ++)
v += b[i]*a[i];
If we completely unroll the loop, we would get:
v = b[0]*a[0] + b[1]*a[1] + b[2]*a[2]
Which then will be simplified to:
v = b[0]* 0 + b[1]* 1 + b[2]* 0
And finally:
v = b[1]
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228265 91177308-0d34-0410-b5e6-96231b3b80d8
Summary: When evaluating floating point instructions in the inliner, ask the TTI whether it is an expensive operation. By default, it's not an expensive operation. This keeps the default behavior the same as before. The ARM TTI has been updated to return back TCC_Expensive for targets which don't have hardware floating point.
Reviewers: chandlerc, echristo
Reviewed By: echristo
Subscribers: t.p.northover, aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D6936
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228263 91177308-0d34-0410-b5e6-96231b3b80d8
Split out `LineField`, which restricts the legal line numbers. This
will make it easier to be consistent between different node parsers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228226 91177308-0d34-0410-b5e6-96231b3b80d8
Minimize the boilerplate required for the `MDNode` subclass
`DenseMapInfo<>` overrides in `LLVMContextImpl`.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228212 91177308-0d34-0410-b5e6-96231b3b80d8
