llvm-svn: 19857

llvm-svn: 19856

llvm-svn: 19855

llvm-svn: 19854

llvm-svn: 19853

llvm-svn: 19852

llvm-svn: 19851

llvm-svn: 19850

llvm-svn: 19849

llvm-svn: 19848

llvm-svn: 19847

llvm-svn: 19846

llvm-svn: 19845

llvm-svn: 19844

llvm-svn: 19843

llvm-svn: 19835

of FP ops.

llvm-svn: 19834

llvm-svn: 19833

automatically.

llvm-svn: 19828

llvm-svn: 19822

llvm-svn: 19821

llvm-svn: 19818

llvm-svn: 19817

llvm-svn: 19816

llvm-svn: 19815

llvm-svn: 19814

llvm-svn: 19813

llvm-svn: 19811

llvm-svn: 19810

llvm-svn: 19809

llvm-svn: 19808

labels) even though they are concrete.  This fixes the DSA regressions from
last night.

llvm-svn: 19807

large nested types over and over again to determine if they are sized or not.
Now, isSized() is able to make snap decisions about all concrete types, which
are a common occurance (and includes all primitives).

On 177.mesa, this speeds up DSE from 39.5s -> 21.3s and GCSE from
13.2s -> 11.3s, reducing gccas time from 80s -> 61s (this is a debug build).

DSE and GCSE are still too slow on this testcase, but this is a simple
improvement.

llvm-svn: 19800

llvm-svn: 19799

llvm-svn: 19798

llvm-svn: 19797

all.  This should speed up the X86 backend fairly significantly on integer
codes.  Now if only we didn't have to compute livevar still... ;-)

llvm-svn: 19796

registers.  This information is computed directly by the register allocator
now.

llvm-svn: 19795

llvm-svn: 19794

llvm-svn: 19793