summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Target/PowerPC/InstPrinter
diff options
context:
space:
mode:
authorHal Finkel <hfinkel@anl.gov>2014-03-13 07:58:58 +0000
committerHal Finkel <hfinkel@anl.gov>2014-03-13 07:58:58 +0000
commit27774d9274fb4037322397e7cd0e6976c5bf57ed (patch)
tree8d41282ca124a9e0ebcacce670d9154bc78d82e3 /llvm/lib/Target/PowerPC/InstPrinter
parent5457bd08cb93c9f44ca5379cef87d2a9250e685f (diff)
downloadbcm5719-llvm-27774d9274fb4037322397e7cd0e6976c5bf57ed.tar.gz
bcm5719-llvm-27774d9274fb4037322397e7cd0e6976c5bf57ed.zip
[PowerPC] Initial support for the VSX instruction set
VSX is an ISA extension supported on the POWER7 and later cores that enhances floating-point vector and scalar capabilities. Among other things, this adds <2 x double> support and generally helps to reduce register pressure. The interesting part of this ISA feature is the register configuration: there are 64 new 128-bit vector registers, the 32 of which are super-registers of the existing 32 scalar floating-point registers, and the second 32 of which overlap with the 32 Altivec vector registers. This makes things like vector insertion and extraction tricky: this can be free but only if we force a restriction to the right register subclass when needed. A new "minipass" PPCVSXCopy takes care of this (although it could do a more-optimal job of it; see the comment about unnecessary copies below). Please note that, currently, VSX is not enabled by default when targeting anything because it is not yet ready for that. The assembler and disassembler are fully implemented and tested. However: - CodeGen support causes miscompiles; test-suite runtime failures: MultiSource/Benchmarks/FreeBench/distray/distray MultiSource/Benchmarks/McCat/08-main/main MultiSource/Benchmarks/Olden/voronoi/voronoi MultiSource/Benchmarks/mafft/pairlocalalign MultiSource/Benchmarks/tramp3d-v4/tramp3d-v4 SingleSource/Benchmarks/CoyoteBench/almabench SingleSource/Benchmarks/Misc/matmul_f64_4x4 - The lowering currently falls back to using Altivec instructions far more than it should. Worse, there are some things that are scalarized through the stack that shouldn't be. - A lot of unnecessary copies make it past the optimizers, and this needs to be fixed. - Many more regression tests are needed. Normally, I'd fix these things prior to committing, but there are some students and other contributors who would like to work this, and so it makes sense to move this development process upstream where it can be subject to the regular code-review procedures. llvm-svn: 203768
Diffstat (limited to 'llvm/lib/Target/PowerPC/InstPrinter')
-rw-r--r--llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.cpp12
-rw-r--r--llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.h2
2 files changed, 12 insertions, 2 deletions
diff --git a/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.cpp b/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.cpp
index b1c480b4aa3..dc54b52a248 100644
--- a/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.cpp
+++ b/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.cpp
@@ -199,6 +199,13 @@ void PPCInstPrinter::printPredicateOperand(const MCInst *MI, unsigned OpNo,
printOperand(MI, OpNo+1, O);
}
+void PPCInstPrinter::printU2ImmOperand(const MCInst *MI, unsigned OpNo,
+ raw_ostream &O) {
+ unsigned int Value = MI->getOperand(OpNo).getImm();
+ assert(Value <= 3 && "Invalid u2imm argument!");
+ O << (unsigned int)Value;
+}
+
void PPCInstPrinter::printS5ImmOperand(const MCInst *MI, unsigned OpNo,
raw_ostream &O) {
int Value = MI->getOperand(OpNo).getImm();
@@ -316,7 +323,10 @@ static const char *stripRegisterPrefix(const char *RegName) {
switch (RegName[0]) {
case 'r':
case 'f':
- case 'v': return RegName + 1;
+ case 'v':
+ if (RegName[1] == 's')
+ return RegName + 2;
+ return RegName + 1;
case 'c': if (RegName[1] == 'r') return RegName + 2;
}
diff --git a/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.h b/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.h
index 8a4c03d6454..4d1df78ee48 100644
--- a/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.h
+++ b/llvm/lib/Target/PowerPC/InstPrinter/PPCInstPrinter.h
@@ -43,7 +43,7 @@ public:
void printPredicateOperand(const MCInst *MI, unsigned OpNo,
raw_ostream &O, const char *Modifier = 0);
-
+ void printU2ImmOperand(const MCInst *MI, unsigned OpNo, raw_ostream &O);
void printS5ImmOperand(const MCInst *MI, unsigned OpNo, raw_ostream &O);
void printU5ImmOperand(const MCInst *MI, unsigned OpNo, raw_ostream &O);
void printU6ImmOperand(const MCInst *MI, unsigned OpNo, raw_ostream &O);
OpenPOWER on IntegriCloud