- May 17, 2006
-
-
Evan Cheng authored
llvm-svn: 28371
-
Evan Cheng authored
llvm-svn: 28370
-
Evan Cheng authored
llvm-svn: 28368
-
Chris Lattner authored
Patch written by Domagoj Babic! llvm-svn: 28367
-
Chris Lattner authored
use getPackedTypeBreakdown at all here. llvm-svn: 28365
-
Chris Lattner authored
llvm-svn: 28364
-
Andrew Lenharth authored
llvm-svn: 28360
-
Andrew Lenharth authored
llvm-svn: 28359
-
Evan Cheng authored
llvm-svn: 28357
-
Evan Cheng authored
llvm-svn: 28356
-
Chris Lattner authored
llvm-svn: 28355
-
Chris Lattner authored
enough to be autogenerated. llvm-svn: 28354
-
Evan Cheng authored
llvm-svn: 28353
-
Evan Cheng authored
VBIT_VECTOR nodes. There were some confusion about the semantics of getPackedTypeBreakdown(). e.g. for <4 x f32> it returns 1 and v4f32, not 4, and f32. llvm-svn: 28352
-
Evan Cheng authored
llvm-svn: 28351
-
Chris Lattner authored
to the new module. Patch provided by Nick Lewycky! llvm-svn: 28349
-
Chris Lattner authored
use UpdateNodeOperands to just update the operands! This is important because getNode will allocate a new node if the node returns a flag and this breaks assumptions in the legalizer that you can legalize some things multiple times and get exactly the same results. This latent bug was exposed by my ppc patch last night, and this fixes gsm/toast. llvm-svn: 28348
-
Chris Lattner authored
change. llvm-svn: 28347
-
Chris Lattner authored
the copyto/fromregs instead of making the PPCISD::CALL selection code create them. This vastly simplifies the selection code, and moves the ABI handling parts into one place. llvm-svn: 28346
-
Chris Lattner authored
1. Rearrange code a bit so that the special case doesn't require indenting lots of code. 2. Add comments describing PPC calling convention. 3. Only round up to 56-bytes of stack space for an outgoing call if the callee is varargs. This saves a bit of stack space. llvm-svn: 28342
-
Chris Lattner authored
llvm-svn: 28341
-
Chris Lattner authored
llvm-svn: 28340
-
Chris Lattner authored
ISD::CALL node, then custom lower that. This means that we only have to handle LEGAL call operands/results, not every possible type. This allows us to simplify the call code, shrinking it by about 1/3. llvm-svn: 28339
-
Chris Lattner authored
produce it. llvm-svn: 28338
-
- May 16, 2006
-
-
Chris Lattner authored
llvm-svn: 28335
-
Chris Lattner authored
llvm-svn: 28334
-
Chris Lattner authored
llvm-svn: 28333
-
Chris Lattner authored
handling. This makes the lower argument code significantly simpler (we only need to handle legal argument types). Incidentally, this also implements support for vector argument registers, so long as they are not on the stack. llvm-svn: 28331
-
Andrew Lenharth authored
llvm-svn: 28330
-
Andrew Lenharth authored
llvm-svn: 28329
-
Chris Lattner authored
arguments at once. llvm-svn: 28327
-
Chris Lattner authored
llvm-svn: 28326
-
Evan Cheng authored
llvm-svn: 28324
-
Chris Lattner authored
llvm-svn: 28321
-
Chris Lattner authored
it doesn't currently use/maintain the chain properly. Also, make the X86ISelLowering.cpp file 80-col clean. llvm-svn: 28320
-
Vladimir Prus authored
can just add lib/Target to TableGen includes. llvm-svn: 28318
-
Chris Lattner authored
This code should be emitted after legalize, so it can't be in sdisel. Note that the EmitFunctionEntryCode hook should be updated to operate on the DAG. The X86 backend is the only one currently using this hook. llvm-svn: 28315
-
Chris Lattner authored
llvm-svn: 28314
-
Chris Lattner authored
for each argument. llvm-svn: 28313
-
Chris Lattner authored
llvm-svn: 28311
-