maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

Improve allocation order: 

1) For 8-bit registers try to use first the ones that are parts of the
   same register (AL then AH). This way we only alias 2 16/32-bit
   registers after allocating 4 8-bit variables.

2) Move EBX as the last register to allocate. This will cause less
   spills to happen since we will have 8-bit registers available up to
   register excaustion (assuming we use the allocation order). It
   would be nice if we could push all of the 8-bit aliased registers
   towards the end but we much prefer to keep callee saved register to
   the end to avoid saving them on entry and exit of the function.

For example this gives a slight reduction of spills with linear scan
on 164.gzip.

Before:

11221 asm-printer           - Number of machine instrs printed
  975 spiller               - Number of loads added
  675 spiller               - Number of stores added
  398 spiller               - Number of register spills

After:

11182 asm-printer           - Number of machine instrs printed
  952 spiller               - Number of loads added
  652 spiller               - Number of stores added
  386 spiller               - Number of register spills

llvm-svn: 11996
This commit is contained in:
Alkis Evlogimenos 2004-02-29 09:17:01 +00:00
parent ea81b79a97
commit c7fd0770a0
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
llvm/lib/Target/X86/X86RegisterInfo.td
View File

@ -76,8 +76,8 @@ def : RegisterAliases<SP, [ESP]>; def : RegisterAliases<BP, [EBP]>;
// top-level register classes. The order specified in the register list is
// implicitly defined to be the register allocation order.
//
def R8 : RegisterClass<i8, 1, [AL, CL, DL, BL, AH, CH, DH, BH]>;
def R16 : RegisterClass<i16, 2, [AX, CX, DX, BX, SI, DI, BP, SP]> {
def R8 : RegisterClass<i8, 1, [AL, AH, CL, CH, DL, DH, BL, BH]>;
def R16 : RegisterClass<i16, 2, [AX, CX, DX, SI, DI, BX, BP, SP]> {
let Methods = [{
iterator allocation_order_end(MachineFunction &MF) const {
if (hasFP(MF)) // Does the function dedicate EBP to being a frame ptr?
@ -88,7 +88,7 @@ def R16 : RegisterClass<i16, 2, [AX, CX, DX, BX, SI, DI, BP, SP]> {
}];
}
def R32 : RegisterClass<i32, 4, [EAX, ECX, EDX, EBX, ESI, EDI, EBP, ESP]> {
def R32 : RegisterClass<i32, 4, [EAX, ECX, EDX, ESI, EDI, EBX, EBP, ESP]> {
let Methods = [{
iterator allocation_order_end(MachineFunction &MF) const {
if (hasFP(MF)) // Does the function dedicate EBP to being a frame ptr?