llvm-svn: 109956

the jtblock_operand print methods.  This avoids extra newlines in the
disassembler's output.  PR7757.

llvm-svn: 109948

llvm-svn: 109946

Add -disable-shifter-op to disable isel of shifter ops. On Cortex-a9 the shifts cost extra instructions so it might be better to emit them separately to take advantage of dual-issues.

llvm-svn: 109934

reference registers past the end of the NEON register file, and report them
as invalid instead of asserting when trying to print them.  PR7746.

llvm-svn: 109933

beginning on ARM Darwin assembly files so that it won't be placed after
debug sections.  Radar 8252813.

llvm-svn: 109879

have 4 bits per register in the operand encoding), but have undefined
behavior when the operand value is 13 or 15 (SP and PC, respectively).
The trivial coalescer in linear scan sometimes will merge a copy from
SP into a subsequent instruction which uses the copy, and if that
instruction cannot legally reference SP, we get bad code such as:
  mls r0,r9,r0,sp
instead of:
  mov r2, sp
  mls r0, r9, r0, r2

This patch adds a new register class for use by Thumb2 that excludes
the problematic registers (SP and PC) and is used instead of GPR
for those operands which cannot legally reference PC or SP. The
trivial coalescer explicitly requires that the register class
of the destination for the COPY instruction contain the source
register for the COPY to be considered for coalescing. This prevents
errant instructions like that above.

PR7499

llvm-svn: 109842

llvm-svn: 109813

transformations.

llvm-svn: 109800

integers with mov + vdup.  8003375.  This is
currently disabled by default because LICM will
not hoist a VDUP, so it pessimizes the code if
the construct occurs inside a loop (8248029).

llvm-svn: 109799

PR7745.

llvm-svn: 109788

Add intrinsics __builtin_arm_qadd & __builtin_arm_qsub to allow access to the QADD & QSUB instructions.
Behave identically to __qadd & __qsub RealView instruction intrinsics.

llvm-svn: 109770

ARM mode version of r109693. Remove incorrect substitution pattern for UXTB16. It wrongly assumed the input shift was actually a rotate. rdar://8240138

llvm-svn: 109696

Remove incorrect substitution pattern for UXTB16. It wrongly assumed the input shift was actually a rotate. rdar://8240138

llvm-svn: 109693

llvm-svn: 109691

llvm-svn: 109458

llvm-svn: 109456

llvm-svn: 109448

This assumption is not satisfied due to global mergeing.
Workaround the issue by temporary disablinge mergeing of const globals.
Also, ignore LLVM "special" globals. This fixes PR7716

llvm-svn: 109423

llvm-svn: 109421

llvm-svn: 109373

llvm-svn: 109372

llvm-svn: 109359

function live in set. This will give us tGPR for Thumb1 and GPR otherwise,
so the copy will be spillable. rdar://8224931

llvm-svn: 109293

comments explaining why it was wrong.  8225024.

Fix the real problem in 8213383: the code that splits very large
blocks when no other place to put constants can be found was not
considering the case that the block contained a Thumb tablejump.

llvm-svn: 109282

  it's too late to start backing off aggressive latency scheduling when most
  of the registers are in use so the threshold should be a bit tighter.
- Correctly handle live out's and extract_subreg etc.
- Enable register pressure aware scheduling by default for hybrid scheduler.
  For ARM, this is almost always a win on # of instructions. It's runtime
  neutral for most of the tests. But for some kernels with high register
  pressure it can be a huge win. e.g. 464.h264ref reduced number of spills by
  54 and sped up by 20%.

llvm-svn: 109279

ARM/PPC/MSP430-specific code (which are the only targets that
implement the hook) can directly reference their target-specific
instrinfo classes.

llvm-svn: 109171

This is probably not the best way to implement "Force LR to 
be spilled if the Thumb function size is > 2048." do this, 
it should use the branch shortening infrastructure, but I'm
just preserving functionality here.

llvm-svn: 109165

llvm-svn: 109125

llvm-svn: 109091

llvm-svn: 109090

especially on other platforms. Is there a better way to fix this.

llvm-svn: 109084

llvm-svn: 109081

Fix constant island pass's handling of tBR_JTr. The offset of the instruction does not have to be 4-byte aligned. Rather, it's the offset + 2 that must be aligned since the instruction expands into:
        mov     pc, r1
        .align  2
LJTI0_0_0:
        .long    LBB0_14

This fixes rdar://8213383. No test case since it's not possible to come up with a suitable small one.

llvm-svn: 109076

llvm-svn: 109064

rdar://8202967

llvm-svn: 109057

llvm-svn: 109047

llvm-svn: 109009

llvm-svn: 108991

llvm-svn: 108929