llvm-svn: 54148

which is represented in codegen as an 'and' operation. This matches them
with movz instructions, instead of leaving them to be matched by and
instructions with an immediate field.

llvm-svn: 54147

llvm-svn: 54128

parallel its analogue, Value::value_use_iterator. The operator* method
now returns the user, rather than the use.

llvm-svn: 54127

mmx needs its own fancy shuffle logic based on unpack; for now we get correct but awful code.

Also commit Mon Ping's VSETCC patch

llvm-svn: 54039

llvm-svn: 54029

llvm-svn: 54026

Based on patch by Nicolas Capens.

llvm-svn: 53939

llvm-svn: 53933

llvm-svn: 53927

llvm-svn: 53911

llvm-svn: 53873

that include useful information like the name of the
block being viewed and the current phase of compilation.

llvm-svn: 53872

generic SDNode's (nodes with their own constructors
should do sanity checking in the constructor).  Add
sanity checks for BUILD_VECTOR and fix all the places
that were producing bogus BUILD_VECTORs, as found by
"make check".  My favorite is the BUILD_VECTOR with
only two operands that was being used to build a
vector with four elements!

llvm-svn: 53850

Use movaps instead of movups to spill 16-byte vector values when default alignment is >= 16. This fixes some massive performance regressions.

llvm-svn: 53844

into v4i16 vectors.

llvm-svn: 53807

llvm-svn: 53789

llvm-svn: 53788

llvm-svn: 53784

replacement of multiple values. This is slightly more efficient
than doing multiple ReplaceAllUsesOfValueWith calls, and theoretically
could be optimized even further. However, an important property of this
new function is that it handles the case where the source value set and
destination value set overlap. This makes it feasible for isel to use
SelectNodeTo in many very common cases, which is advantageous because
SelectNodeTo avoids a temporary node and it doesn't require CSEMap
updates for users of values that don't change position.

Revamp MorphNodeTo, which is what does all the work of SelectNodeTo, to
handle operand lists more efficiently, and to correctly handle a number
of corner cases to which its new wider use exposes it.

This commit also includes a change to the encoding of post-isel opcodes
in SDNodes; now instead of being sandwiched between the target-independent
pre-isel opcodes and the target-dependent pre-isel opcodes, post-isel
opcodes are now represented as negative values. This makes it possible
to test if an opcode is pre-isel or post-isel without having to know
the size of the current target's post-isel instruction set.

These changes speed up llc overall by 3% and reduce memory usage by 10%
on the InstructionCombining.cpp testcase with -fast and -regalloc=local.

llvm-svn: 53728

llvm-svn: 53722

llvm-svn: 53720

llvm-svn: 53719

atomic operation to the new one.

llvm-svn: 53714

Mark function used by asm block as used, otherwise optimizer may not see the use and may delete the function.

llvm-svn: 53692

llvm-svn: 53688

llvm-svn: 53661

llvm-svn: 53660

has exactly one MachineMemOperand, and change some X86 lowering code to
make use of it.

llvm-svn: 53498

instead of using the frame index for the SVOffset, which was
inconsistent.

llvm-svn: 53486

llvm-svn: 53450

llvm-svn: 53386

constant pool, if relocation model is not static. This directly maps to
the way how GCC works.

llvm-svn: 53370

llvm-svn: 53368

is X86 ATT only.

llvm-svn: 53355

llvm-svn: 53354

llvm-svn: 53350

llvm-svn: 53340

llvm-svn: 53333

llvm-svn: 53324