Benjamin Kramer 9b108a338d Teach DAGCombine to fold fold (sra (trunc (sr x, c1)), c2) -> (trunc (sra x, c1+c2) when c1 equals the amount of bits that are truncated off. ... This happens all the time when a smul is promoted to a larger type. On x86-64 we now compile "int test(int x) { return x/10; }" into movslq %edi, %rax imulq $1717986919, %rax, %rax movq %rax, %rcx shrq $63, %rcx sarq $34, %rax <- used to be "shrq $32, %rax; sarl $2, %eax" addl %ecx, %eax This fires 96 times in gcc.c on x86-64. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124559 91177308-0d34-0410-b5e6-96231b3b80d8		2011-01-30 16:38:43 +00:00
..
Alpha	If dbg_declare() or dbg_value() is not lowered by isel then emit DEBUG message instead of creating DBG_VALUE for undefined value in reg0.	2010-12-06 22:39:26 +00:00
ARM	Add a testcase for my last checkin.	2011-01-27 06:01:17 +00:00
Blackfin
CBackend
CellSPU	Allow sign-extending of i8 and i16 to i128 on SPU.	2011-01-20 15:49:06 +00:00
CPP
Generic	fix rdar://8878965, a regression I introduced with the recent	2011-01-18 20:53:04 +00:00
MBlaze	Lower the MBlaze target specific calling conventions for "interrupt_handler"	2010-12-15 20:27:28 +00:00
Mips	Add support for mips32 madd and msub instructions. Patch by Akira Hatanaka	2011-01-18 19:29:17 +00:00
MSP430	If dbg_declare() or dbg_value() is not lowered by isel then emit DEBUG message instead of creating DBG_VALUE for undefined value in reg0.	2010-12-06 22:39:26 +00:00
PowerPC	Restore the behavior of frame lowering before my refactoring.	2010-12-18 19:53:14 +00:00
PTX	ptx: remove reg-reg addressing mode and st.const	2011-01-01 11:58:58 +00:00
SPARC	Pass sret arguments through the stack instead of through registers in Sparc backend. It makes the code generated more compliant with the sparc32 ABI.	2011-01-22 13:05:16 +00:00
SystemZ	If dbg_declare() or dbg_value() is not lowered by isel then emit DEBUG message instead of creating DBG_VALUE for undefined value in reg0.	2010-12-06 22:39:26 +00:00
Thumb	Sorry, several patches in one.	2011-01-20 08:34:58 +00:00
Thumb2	Last round of fixes for movw + movt global address codegen.	2011-01-21 18:55:51 +00:00
X86	Teach DAGCombine to fold fold (sra (trunc (sr x, c1)), c2) -> (trunc (sra x, c1+c2) when c1 equals the amount of bits that are truncated off.	2011-01-30 16:38:43 +00:00
XCore	Update tests.	2011-01-16 18:02:57 +00:00
thumb2-mul.ll