This is an R600 GPU with double support.

Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178929

Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178928

SITargetLowering::analyzeImmediate() was converting the 64-bit values
to 32-bit and then checking if they were an inline immediate.  Some
of these conversions caused this check to succeed and produced
S_MOV instructions with 64-bit immediates, which are illegal.

v2:
  - Clean up logic

Reviewed-by: Christian König <christian.koenig@amd.com>
llvm-svn: 178927

On cores for which we know the misprediction penalty, and we have
the isel instruction, we can profitably perform early if conversion.
This enables us to replace some small branch sequences with selects
and avoid the potential stalls from mispredicting the branches.

Enabling this feature required implementing canInsertSelect and
insertSelect in PPCInstrInfo; isel code in PPCISelLowering was
refactored to use these functions as well.

llvm-svn: 178926

The manual states that there is a minimum of 13 cycles from when the
mispredicted branch is issued to when the correct branch target is
issued.

llvm-svn: 178925

Now we can:
1 - see the return value for functions that return types that use the "ext_vector_size"
2 - dump values that use the vector attributes ("expr $ymm0")
3 - modified the DWARF parser to correctly parse GNU vector types from the DWARF by turning them into clang::Type::ExtVector types instead of just standard arrays

llvm-svn: 178924

more information to the notes.  This information is already present on other
diagnostic messages that involves overloads.

llvm-svn: 178923

This is the counterpart to commit r160637, except it performs the action
in the bottomup portion of the data flow analysis.

llvm-svn: 178922

The normal dataflow sequence in the ARC optimizer consists of the following
states:

    Retain -> CanRelease -> Use -> Release

The optimizer before this patch stored the uses that determine the lifetime of
the retainable object pointer when it bottom up hits a retain or when top down
it hits a release. This is correct for an imprecise lifetime scenario since what
we are trying to do is remove retains/releases while making sure that no
``CanRelease'' (which is usually a call) deallocates the given pointer before we
get to the ``Use'' (since that would cause a segfault).

If we are considering the precise lifetime scenario though, this is not
correct. In such a situation, we *DO* care about the previous sequence, but
additionally, we wish to track the uses resulting from the following incomplete
sequences:

  Retain -> CanRelease -> Release   (TopDown)
  Retain <- Use <- Release          (BottomUp)

*NOTE* This patch looks large but the most of it consists of updating
test cases. Additionally this fix exposed an additional bug. I removed
the test case that expressed said bug and will recommit it with the fix
in a little bit.

llvm-svn: 178921

platform.plugin.darwin-kernel.kext-directories
platform.plugin.darwin-kernel.search-locally-for-kexts

and fix a few FileSpec handling issues for the kext-directories setting.

llvm-svn: 178920

This fixes PEI as previously described, but correctly handles the case where
the instruction defining the virtual register to be scavenged is the first in
the block. Arnold provided me with a bugpoint-reduced test case, but even that
seems too large to use as a regression test. If I'm successful in cleaning it
up then I'll commit that as well.

Original commit message:

    This change fixes a bug that I introduced in r178058. After a register is
    scavenged using one of the available spills slots the instruction defining the
    virtual register needs to be moved to after the spill code. The scavenger has
    already processed the defining instruction so that registers killed by that
    instruction are available for definition in that same instruction. Unfortunately,
    after this, the scavenger needs to iterate through the spill code and then
    visit, again, the instruction that defines the now-scavenged register. In order
    to avoid confusion, the register scavenger needs the ability to 'back up'
    through the spill code so that it can again process the instructions in the
    appropriate order. Prior to this fix, once the scavenger reached the
    just-moved instruction, it would assert if it killed any registers because,
    having already processed the instruction, it believed they were undefined.

    Unfortunately, I don't yet have a small test case. Thanks to Pranav Bhandarkar
    for diagnosing the problem and testing this fix.

llvm-svn: 178919

llvm-svn: 178918

During LTO, the target options on functions within the same Module may
change. This would necessitate resetting some of the back-end. Do this for X86,
because it's a Friday afternoon.

llvm-svn: 178917

Reverting because this breaks one of the LTO builders. Original commit message:

    This change fixes a bug that I introduced in r178058. After a register is
    scavenged using one of the available spills slots the instruction defining the
    virtual register needs to be moved to after the spill code. The scavenger has
    already processed the defining instruction so that registers killed by that
    instruction are available for definition in that same instruction. Unfortunately,
    after this, the scavenger needs to iterate through the spill code and then
    visit, again, the instruction that defines the now-scavenged register. In order
    to avoid confusion, the register scavenger needs the ability to 'back up'
    through the spill code so that it can again process the instructions in the
    appropriate order. Prior to this fix, once the scavenger reached the
    just-moved instruction, it would assert if it killed any registers because,
    having already processed the instruction, it believed they were undefined.

    Unfortunately, I don't yet have a small test case. Thanks to Pranav Bhandarkar
    for diagnosing the problem and testing this fix.

llvm-svn: 178916

llvm-svn: 178915

llvm-svn: 178914

llvm-svn: 178913

This optimization is unstable at this moment; it 
  1) block us on a very important application
  2) PR15200
  3) test6 and test7 in test/Transforms/ScalarRepl/dynamic-vector-gep.ll
     (the CHECK command compare the output against wrong result)

   I personally believe this optimization should not have any impact on the
autovectorized code, as auto-vectorizer is supposed to put gather/scatter
in a "right" way.  Although in theory downstream optimizaters might reveal 
some gather/scatter optimization opportunities, the chance is quite slim.

   For the hand-crafted vectorizing code, in term of redundancy elimination,
load-CSE, copy-propagation and DSE can collectively achieve the same result,
but in much simpler way. On the other hand, these optimizers are able to 
improve the code in a incremental way; in contrast, SROA is sort of all-or-none
approach. However, SROA might slighly win in stack size, as it tries to figure 
out a stretch of memory tightenly cover the area accessed by the dynamic index.

 rdar://13174884
 PR15200

llvm-svn: 178912

rdar://13535645

llvm-svn: 178911

llvm-mips-linux green.

llvm-mips-linux runs on a big endian machine. This test passes if I change 'e'
to 'E' in the target data layout string.

llvm-svn: 178910

It's possible for the lock file to disappear and the owning process to
return before we're able to see the generated file. Spin for a little
while to see if it shows up before failing. 

llvm-svn: 178909

If the directory that will contain the unique file doesn't exist when
we tried to create the file, but another process creates it before we
get a chance to try creating it, we would bail out rather than try to
create the unique file.

llvm-svn: 178908

cast

UseNullptr previously matched the implicit cast to const pointer as well as
the explicit cast within that has an implicit cast to nullptr as a descendant.

-Refactored UseNullptr to avoid special-casing certain kinds of cast sequences
-Added test cases.

llvm-svn: 178907

llvm-svn: 178906

llvm-svn: 178905

llvm-svn: 178904

// rdar://12379114

llvm-svn: 178903

llvm-svn: 178902

With cpp11-migrate core functionality moved to a separate library (for enabling
unit tests) this library contained code that referenced symbols that are still
in the main binary. On some platforms, the shared library build broke as a
result. This revision fixes the dependency problem and is safe for the eventual
lib-ification of the transforms as well.

llvm-svn: 178901

With the lib-ification of cpp11-migrate, real unit tests can be written.
Replacing dummy tests with some simple tests for the Transform public
interface.

llvm-svn: 178900

Now treat AF_None family as impossible in isTrackedFamily()

llvm-svn: 178899

llvm-svn: 178898

Introducing a negative cache for ObjCLanguageRuntime::LookupInCompleteClassCache()
This helps speed up the (common) case of us looking for classes that are hidden deep within Cocoa internals and repeatedly failing at finding type information for them.
In order for this to work, we need to clean this cache whenever debug information is added. A new symbols loaded event is added that is triggered with add-dsym (before modules loaded would be triggered for both adding modules and adding symbols).
Interested parties can register for this event. Internally, we make sure to clean the negative cache whenever symbols are added.
Lastly, ClassDescriptor::IsTagged() has been refactored to GetTaggedPointerInfo() that also (optionally) returns info and value bits. In this way, data formatters can share tagged pointer code instead of duplicating the required arithmetic.

llvm-svn: 178897

These should really be templated like ELF, but this is a start.

llvm-svn: 178896

llvm-svn: 178895

llvm-svn: 178894

llvm-svn: 178893

llvm-svn: 178892

As mentioned in the previous commit message, the use-after-free and
double-free warnings for 'delete' are worth enabling even while the
leak warnings still have false positives.

llvm-svn: 178891

This splits the leak-checking part of alpha.cplusplus.NewDelete into a
separate user-level checker, alpha.cplusplus.NewDeleteLeaks. All the
difficult false positives we've seen with the new/delete checker have been
spurious leak warnings; the use-after-free warnings and mismatched
deallocator warnings, while rare, have always been valid.

<rdar://problem/6194569>

llvm-svn: 178890