Evan Cheng 206d1856ad Added a linearscan register allocation optimization. When the register allocator spill an interval with multiple uses in the same basic block, it creates a different virtual register for each of the reloads. e.g. ... %reg1498<def> = MOV32rm %reg1024, 1, %reg0, 12, %reg0, Mem:LD(4,4) [sunkaddr39 + 0] %reg1506<def> = MOV32rm %reg1024, 1, %reg0, 8, %reg0, Mem:LD(4,4) [sunkaddr42 + 0] %reg1486<def> = MOV32rr %reg1506 %reg1486<def> = XOR32rr %reg1486, %reg1498, %EFLAGS<imp-def,dead> %reg1510<def> = MOV32rm %reg1024, 1, %reg0, 4, %reg0, Mem:LD(4,4) [sunkaddr45 + 0] => %reg1498<def> = MOV32rm %reg2036, 1, %reg0, 12, %reg0, Mem:LD(4,4) [sunkaddr39 + 0] %reg1506<def> = MOV32rm %reg2037, 1, %reg0, 8, %reg0, Mem:LD(4,4) [sunkaddr42 + 0] %reg1486<def> = MOV32rr %reg1506 %reg1486<def> = XOR32rr %reg1486, %reg1498, %EFLAGS<imp-def,dead> %reg1510<def> = MOV32rm %reg2038, 1, %reg0, 4, %reg0, Mem:LD(4,4) [sunkaddr45 + 0] From linearscan's point of view, each of reg2036, 2037, and 2038 are separate registers, each is "killed" after a single use. The reloaded register is available and it's often clobbered right away. e.g. In thise case reg1498 is allocated EAX while reg2036 is allocated RAX. This means we end up with multiple reloads from the same stack slot in the same basic block. Now linearscan recognize there are other reloads from same SS in the same BB. So it'll "downgrade" RAX (and its aliases) after reg2036 is allocated until the next reload (reg2037) is done. This greatly increase the likihood reloads from SS are reused. This speeds up sha1 from OpenSSL by 5.8%. It is also an across the board win for SPEC2000 and 2006. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@69585 91177308-0d34-0410-b5e6-96231b3b80d8		2009-04-20 08:01:12 +00:00
..
Analysis	Add a ScalarEvolution::getCouldNotCompute() function, and use it	2009-04-18 17:58:19 +00:00
Archive	Add the private linkage.	2009-01-15 20:18:42 +00:00
AsmParser	Fix warning in .ll parser, detect and reject available_externally on function	2009-04-13 07:02:02 +00:00
Bitcode	Add a new "available_externally" linkage type. This is intended	2009-04-13 05:44:34 +00:00
CodeGen	Added a linearscan register allocation optimization. When the register allocator spill an interval with multiple uses in the same basic block, it creates a different virtual register for each of the reloads. e.g.	2009-04-20 08:01:12 +00:00
CompilerDriver	-write-graph now can be used with -o.	2009-03-27 12:57:14 +00:00
Debugger	Oops...I committed too much.	2009-03-13 04:39:26 +00:00
ExecutionEngine	Use an AssertingVH to detect the case where the Function was deleted but	2009-04-19 18:32:03 +00:00
Linker	Add a new "available_externally" linkage type. This is intended	2009-04-13 05:44:34 +00:00
Support	Make all raw_ostreams support the tell() function.	2009-04-20 07:34:17 +00:00
System	Call CreateProcess with bInheritHandles = TRUE.	2009-04-14 21:31:36 +00:00
Target	Fixed a few 64 bit cases in X86InstrInfo::commuteInstruction	2009-04-18 05:16:01 +00:00
Transforms	Before trying to introduce/eliminate cast/ext/trunc to make indices type as	2009-04-20 06:05:54 +00:00
VMCore	Limit the number of times we're willing to chase pointers. Removes an O(n^2)	2009-04-15 06:23:41 +00:00
Makefile	Reorganize llvmc code.	2009-03-02 09:01:14 +00:00