- Jan 19, 2011
-
-
Bruno Cardoso Lopes authored
llvm-svn: 123837
-
Daniel Dunbar authored
llvm-svn: 123823
-
Andrew Trick authored
of the floating point types less than 64-bits. It's somewhat of a temporary hack but forces more accurate modeling of register pressure and results in fewer spills. llvm-svn: 123811
-
Andrew Trick authored
llvm-svn: 123810
-
Evan Cheng authored
Don't forget to emit the load from indirect symbol when using movw + movt to materialize GA indirect symbols. llvm-svn: 123809
-
- Jan 18, 2011
-
-
Bruno Cardoso Lopes authored
vmrs reg, fpexc vmrs reg, fpsid vmsr fpexc, reg vmsr fpsid, reg llvm-svn: 123783
-
Bruno Cardoso Lopes authored
llvm-svn: 123778
-
Bruno Cardoso Lopes authored
llvm-svn: 123776
-
Bruno Cardoso Lopes authored
llvm-svn: 123772
-
Bruno Cardoso Lopes authored
llvm-svn: 123770
-
Jim Grosbach authored
llvm-svn: 123769
-
Bruno Cardoso Lopes authored
llvm-svn: 123768
-
Bruno Cardoso Lopes authored
llvm-svn: 123766
-
Bruno Cardoso Lopes authored
llvm-svn: 123763
-
Bruno Cardoso Lopes authored
llvm-svn: 123760
-
Chris Lattner authored
llvm-svn: 123752
-
Venkatraman Govindaraju authored
SPARC backend: Modified LowerCall and LowerFormalArguments so that they use CallingConv assignments. llvm-svn: 123749
-
Daniel Dunbar authored
llvm-svn: 123746
-
Daniel Dunbar authored
llvm-svn: 123745
-
Daniel Dunbar authored
llvm-svn: 123744
-
Daniel Dunbar authored
llvm-svn: 123740
-
Daniel Dunbar authored
llvm-svn: 123739
-
Daniel Dunbar authored
McARM: Always keep an offset expression, if used (instead of assuming == 0 if used but not present), and simplify logic. Also, clean up various non-sensicalisms in isMemModeRegThumb() and isMemModeImmThumb(). llvm-svn: 123738
-
Daniel Dunbar authored
llvm-svn: 123737
-
Daniel Dunbar authored
llvm-svn: 123736
-
Daniel Dunbar authored
llvm-svn: 123722
-
Eric Christopher authored
the flags. llvm-svn: 123712
-
Chris Lattner authored
constant but requires a unique address, we can still put it in a readonly section, just not a mergable one. llvm-svn: 123711
-
Jeffrey Yasskin authored
llvm-svn: 123707
-
- Jan 17, 2011
-
-
Douglas Gregor authored
llvm-svn: 123670
-
Kalle Raiskila authored
'rotq*' and 'shlq*' instructions go to the odd pipeline, wheras the inter-vector equivalents 'rot*', 'shl*' go to the even. llvm-svn: 123622
-
Kalle Raiskila authored
llvm-svn: 123620
-
Evan Cheng authored
movw r0, :lower16:(L_foo$non_lazy_ptr-(LPC0_0+4)) movt r0, :upper16:(L_foo$non_lazy_ptr-(LPC0_0+4)) LPC0_0: add r0, pc, r0 It's not yet enabled by default as some tests are failing. I suspect bugs in down stream tools. llvm-svn: 123619
-
- Jan 16, 2011
-
-
Anton Korobeynikov authored
This fixes PR8987 llvm-svn: 123598
-
Anders Carlsson authored
llvm-svn: 123597
-
Rafael Espindola authored
llvm-svn: 123585
-
Chris Lattner authored
into and/shift would cause nodes to move around and a dangling pointer to happen. The code tried to avoid this with a HandleSDNode, but got the details wrong. llvm-svn: 123578
-
Chris Lattner authored
llvm-svn: 123572
-
Evan Cheng authored
llvm-svn: 123567
-
Venkatraman Govindaraju authored
llvm-svn: 123561
-