Sign in
swiftshader
/
SwiftShader
/
6e992147f2f4c99a8fd574e6a154b97e503e59e7
/
src
/
IceTargetLoweringX8632.cpp
6e99214
Subzero: Try to fix warnings and errors in the Windows build.
by Jim Stichnoth
· 11 years ago
0a45051
Subzero: Add support for SSE4.1 instructions.
by Matt Wala
· 11 years ago
e6e497d
Fix bug when atomic load is fused with an arith op (and not in the entry BB)
by Jan Voung
· 11 years ago
c820ddf
Add a peephole to fuse cmpxchg w/ later cmp+branch.
by Jan Voung
· 11 years ago
839c4ce
A couple of fixes for using Makefile.standalone on Mac.
by Jan Voung
· 11 years ago
78282f6
Subzero: Make Ice::Ostream a typedef for llvm::raw_ostream.
by Jim Stichnoth
· 11 years ago
cfe5146
Use movss to implement insertelement when elements = 4 and index = 0.
by Matt Wala
· 11 years ago
ce0ca8f
Lower the fcmp instruction for <4 x float> operands.
by Matt Wala
· 11 years ago
9cb61e2
Lower the select instruction when the operands are of vector type.
by Matt Wala
· 11 years ago
9a0168a
Lower icmp operations between vector values.
by Matt Wala
· 11 years ago
53c5e60
Fix legalization of source operand to bsr and bsf.
by Matt Wala
· 11 years ago
957c50d
Use lowerCast instead of inlined _movzx, to get legalization, for memset.
by Jan Voung
· 11 years ago
7b34b59
Lower stacksave and restore intrinsics.
by Jan Voung
· 11 years ago
7fa813b
Lower byte swap intrinsic.
by Jan Voung
· 11 years ago
4988923
Lower insertelement and extractelement.
by Matt Wala
· 11 years ago
7fa22d8
Lower the rest of the vector arithmetic operations.
by Matt Wala
· 11 years ago
83b8036
Lower casting operations that involve vector types.
by Matt Wala
· 11 years ago
e4da26f
Lower bitmanip intrinsics, assuming absence of BMI/SSE4.2 for now.
by Jan Voung
· 11 years ago
ad8f726
Various improvements related to legalization code.
by Matt Wala
· 11 years ago
0ecabc8
Fix floating point vector frem lowering.
by Matt Wala
· 11 years ago
a3a01a2
Subzero: lower the rest of the atomic operations.
by Jan Voung
· 11 years ago
8d1072e
Lower vector floating point arithmetic operations.
by Matt Wala
· 11 years ago
ca662e9
Subzero: Fix a regalloc bug involving too-aggressive AllowRegisterOverlap.
by Jim Stichnoth
· 11 years ago
45a0623
Add support for passing and returning vectors in accordance with the x86 calling convention.
by Matt Wala
· 11 years ago
f37fbbe
Add scalar lowering for sqrt intrinsic.
by Jan Voung
· 11 years ago
9559899
Avoid assigning esp (or ebp for framepointer-using frames) in Om1.
by Jan Voung
· 11 years ago
928f129
Add support for vector types.
by Matt Wala
· 11 years ago
de4ca71
Subzero: Partial implementation of global initializers.
by Jim Stichnoth
· 11 years ago
5cd240d
Add atomic load/store, fetch_add, fence, and is-lock-free lowering.
by Jan Voung
· 11 years ago
1ee3416
Bitcast of 64-bit immediates may need to split the immediate, not a var.
by Jan Voung
· 11 years ago
3bd9f1a
Add a few Subzero intrinsics (not the atomic ones yet).
by Jan Voung
· 11 years ago
5a13f45
Add ss/sd suffix to InstX8632Store and legalize FP constants.
by Jan Voung
· 11 years ago
43ff7eb
Use GlobalContext::getConstantZero() to get zero valued constants.
by Matt Wala
· 11 years ago
d8f4a7d
Add support for undef values in ICE IR. Undef values represent an
by Matt Wala
· 11 years ago
bdbe402
Change some tests to be valid PNaCl IR (parameter type from i1 -> i32).
by Jan Voung
· 11 years ago
70d6883
Legalize div/idiv operands to avoid immediates.
by Jan Voung
· 11 years ago
b0e142b
Ignore stack adjustment for ebp-based variables.
by Jim Stichnoth
· 11 years ago
d97c7df
Subzero: Initial O2 lowering
by Jim Stichnoth
· 11 years ago
4376d29
Fix g++ -pedantic warnings.
by Jim Stichnoth
· 11 years ago
f61d5b2
Fix x86 floating-point constant emission.
by Jim Stichnoth
· 11 years ago
5bc2b1d
Add Om1 lowering with no optimizations.
by Jim Stichnoth
· 11 years ago