Arnold Schwaighofer ae0052f114 ARM cost model: Make some vector integer to float casts cheaper ...

The default logic marks them as too expensive.

For example, before this patch we estimated:
  cost of 16 for instruction:   %r = uitofp <4 x i16> %v0 to <4 x float>

While this translates to:
  vmovl.u16 q8, d16
  vcvt.f32.u32  q8, q8

All other costs are left to the values assigned by the fallback logic. Theses
costs are mostly reasonable in the sense that they get progressively more
expensive as the instruction sequences emitted get longer.

radar://13445992

llvm-svn: 177334

2013-03-18 22:47:09 +00:00

clang

[PCH] Fix assertion hit related to enum decls inside templated funtions.

2013-03-18 22:23:49 +00:00

clang-tools-extra

Don't include outer-most explicit cast in nullptr replacement

2013-03-15 20:18:08 +00:00

compiler-rt

tsan: smaller memory block headers (32b->16b)

2013-03-18 19:47:36 +00:00

debuginfo-tests

Remove IR scenario tests.

2013-03-15 20:52:10 +00:00

libclc

…

libcxx

Removed raw references to __sun__, __FreeBSD__, __GLIBC__ and __linux__; now just check to see if they are defined.

2013-03-18 19:34:07 +00:00

libcxxabi

…

lld

[lld] remove trailing whitespace

2013-03-14 16:09:49 +00:00

lldb

Code cleanup:

2013-03-18 22:34:00 +00:00

llvm

ARM cost model: Make some vector integer to float casts cheaper

2013-03-18 22:47:09 +00:00

polly

Correct function to decide if a SCEV can be ignored

2013-03-18 20:21:13 +00:00

Description

No description provided

intelllvmoneapisycl

5.2 GiB

Languages

LLVM 41.5%

C++ 31.7%

C 13%

Assembly 9.1%

MLIR 1.5%

Other 2.8%