More work on the constant struct builder. We can now convert the struct to a packed struct when necessary.

llvm-svn: 77038

llvm-svn: 77035

analyzing add recurrences.

llvm-svn: 77034

llvm-svn: 77033

affected after a PHI node has been analyzed, just remove affected
SCEVs from the Scalars map, so that they'll be (lazily) recreated as
needed. This avoids creating SCEV objects that aren't actually needed.

Also, rewrite the associated def-use walking code to be non-recursive
and to continue traversing past Instructions that don't have an
entry in the Scalars map.

llvm-svn: 77032

llvm-svn: 77031

llvm-svn: 77030

getAnalysisIfAvailable<TargetData>().

llvm-svn: 77028

llvm-svn: 77027

llvm-svn: 77026

test for it. The test is by no means complete, but it tests the problem I was
fixing.

llvm-svn: 77025

Before:
      adr r12, #LJTI3_0_0
      ldr pc, [r12, +r0, lsl #2]
LJTI3_0_0:
      .long    LBB3_24
      .long    LBB3_30
      .long    LBB3_31
      .long    LBB3_32

After:
      adr r12, #LJTI3_0_0
      add pc, r12, +r0, lsl #2
LJTI3_0_0:
      b.w    LBB3_24
      b.w    LBB3_30
      b.w    LBB3_31
      b.w    LBB3_32

This has several advantages.
1. This will make it easier to optimize this to a TBB / TBH instruction +
   (smaller) table.
2. This eliminate the need for ugly asm printer hack to force the address
   into thumb addresses (bit 0 is one).
3. Same codegen for pic and non-pic.
4. This eliminate the need to align the table so constantpool island pass
   won't have to over-estimate the size.

Based on my calculation, the later is probably slightly faster as well since
ldr pc with shifter address is very slow. That is, it should be a win as long
as the HW implementation can do a reasonable job of branch predict the second
branch.

llvm-svn: 77024

  Necessary for cases in which the memory is in another process, in a
  file, or on a remote machine.

The primary use for this is the llvm-mc disassemblers, so that they
can be targeted at arbitrary objects, not just in-process memory.

llvm-svn: 77023

llvm-svn: 77022

More work on the constant struct builder. We now try to layout all constant structs but throw away the result.

llvm-svn: 77021

llvm-svn: 77020

 - Some clients which used DOUT have moved to DEBUG. We are deprecating the
   "magic" DOUT behavior which avoided calling printing functions when the
   statement was disabled. In addition to being unnecessary magic, it had the
   downside of leaving code in -Asserts builds, and of hiding potentially
   unnecessary computations.

llvm-svn: 77019

llvm-svn: 77018

llvm-svn: 77017

llvm-svn: 77016

llvm-svn: 77015

llvm-svn: 77014

llvm-svn: 77012

Revert the ConstantInt constructors back to their 2.5 forms where possible, thanks to contexts-on-types.  More to come.

llvm-svn: 77011

 - Chris, please approve.

llvm-svn: 77010

llvm-svn: 77009

llvm-svn: 77007

llvm-svn: 77006

There's still a strict-aliasing violation here, but I don't feel like 
dealing with that right now...

llvm-svn: 77005

llvm-svn: 77004

Note that this also fixes a bug that affects non-template code, where we 
were not treating out-of-line static data members are "file-scope" variables,
and therefore not checking their initializers.

llvm-svn: 77002

llvm-svn: 77001

with type conversion to fix ir-gen crash.

llvm-svn: 77000

llvm-svn: 76999

if the DeclRefExpr is a float, not just either argument.

llvm-svn: 76998

previously virtual) function to being a static function
in the .cpp file.

llvm-svn: 76997

llvm-svn: 76996

Disable my constant island pass optimization (to make use soimm more effectively). It caused infinite looping on lencod.

llvm-svn: 76995

format and add an extract/insert test.

llvm-svn: 76994

It's classifications now include elf-specific discriminators.  Targets
that don't have these features (like darwin and pecoff) simply treat
data.rel like data, etc.

llvm-svn: 76993