Anton Korobeynikov 5cdc3a949a Materialize global addresses via movt/movw pair, this is always better ... than doing the same via constpool: 1. Load from constpool costs 3 cycles on A9, movt/movw pair - just 2. 2. Load from constpool might stall up to 300 cycles due to cache miss. 3. Movt/movw does not use load/store unit. 4. Less constpool entries => better compiler performance. This is only enabled on ELF systems, since darwin does not have needed relocations (yet). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89720 91177308-0d34-0410-b5e6-96231b3b80d8		2009-11-24 00:44:37 +00:00
..
ARMAsmPrinter.cpp	Materialize global addresses via movt/movw pair, this is always better	2009-11-24 00:44:37 +00:00
ARMInstPrinter.cpp	More consistent thumb1 asm printing.	2009-11-19 06:57:41 +00:00
ARMInstPrinter.h	More consistent thumb1 asm printing.	2009-11-19 06:57:41 +00:00
ARMMCInstLower.cpp	It turns out that the testcase in question uncovered subreg-handling bug.	2009-11-07 15:20:32 +00:00
ARMMCInstLower.h	add support for external symbols. The mc instprinter can now handle	2009-10-20 00:56:16 +00:00
CMakeLists.txt	wire up skeletal support for having llc print instructions	2009-10-19 20:20:46 +00:00
Makefile	Normalize makefile comments and sort cmake file lists.	2009-08-31 13:05:24 +00:00