- Aug 21, 2009
-
-
Anton Korobeynikov authored
llvm-svn: 79664
-
Owen Anderson authored
This preserves the existing behavior much more closely than my previous attempt. llvm-svn: 79663
-
Anton Korobeynikov authored
llvm-svn: 79659
-
Anton Korobeynikov authored
llvm-svn: 79657
-
Anton Korobeynikov authored
llvm-svn: 79656
-
Daniel Dunbar authored
llvm-svn: 79652
-
Dan Gohman authored
llvm-svn: 79651
-
Bill Wendling authored
llvm-svn: 79649
-
Anton Korobeynikov authored
llvm-svn: 79648
-
Bob Wilson authored
llvm-svn: 79643
-
Duncan Sands authored
warning: comparison is always true due to limited range of data type. llvm-svn: 79642
-
Dan Gohman authored
offset computations. This fixes a truncation bug on targets that don't have 64-bit pointers. llvm-svn: 79639
-
Bob Wilson authored
llvm-svn: 79638
-
Daniel Dunbar authored
llvm-svn: 79636
-
Daniel Dunbar authored
llvm-svn: 79635
-
Anton Korobeynikov authored
llvm-svn: 79634
-
Andreas Neustifter authored
llvm-svn: 79633
-
Sanjiv Gupta authored
leaf functions. This pass will be extended to color other nodes of the call tree as well in future. llvm-svn: 79631
-
Benjamin Kramer authored
llvm-svn: 79629
-
Anton Korobeynikov authored
llvm-svn: 79626
-
Anton Korobeynikov authored
llvm-svn: 79625
-
Anton Korobeynikov authored
llvm-svn: 79624
-
Anton Korobeynikov authored
Add simple cost model to perfect shuffle. Currently we're doing just greedy search for cost instead of (proper) dynamic programming approach. llvm-svn: 79623
-
Anton Korobeynikov authored
llvm-svn: 79622
-
Anton Korobeynikov authored
llvm-svn: 79621
-
Anton Korobeynikov authored
llvm-svn: 79620
-
Anton Korobeynikov authored
Use masks not nodes for vector shuffle predicates. Provide set of 'legal' masks, so legalizer won't infinite cycle llvm-svn: 79619
-
Anton Korobeynikov authored
llvm-svn: 79618
-
Benjamin Kramer authored
llvm-svn: 79617
-
Richard Pennington authored
llvm-svn: 79616
-
Andreas Neustifter authored
llvm-svn: 79615
-
Bill Wendling authored
llvm-svn: 79614
-
Daniel Dunbar authored
- Together these form the (Mach-O) back end of the assembler. - MCAssembler is the actual assembler backend, which is designed to have a reasonable API. This will eventually grow to support multiple object file implementations, but for now its Mach-O/i386 only. - MCMachOStreamer adapts the MCStreamer "actions" API to the MCAssembler API, e.g. converting the various directives into fragments, managing state like the current section, and so on. - llvm-mc will use the new backend via '-filetype=obj', which may eventually be, but is not yet, since I hear that people like assemblers which actually assemble. - The only thing that works at the moment is changing sections. For the time being I have a Python Mach-O dumping tool in test/scripts so this stuff can be easily tested, eventually I expect to replace this with a real LLVM tool. - More doxyments to come. I assume that since this stuff doesn't touch any of the things which are part of 2.6 that it is ok to put this in not so long before the freeze, but if someone objects let me know, I can pull it. llvm-svn: 79612
-
Daniel Dunbar authored
- Add missing flags for various Objective-C sections. - Fix names for [non_]lazy_symbol_pointer (these are misspelled in the manual). - .symbol_stub does not have the self modifying code flag set (this appears to be wrong in the manual?). - Add implicit alignment values; not yet used. Also, call MCStreamer::Finish at the end of a successful parse. llvm-svn: 79611
-
Bill Wendling authored
llvm-svn: 79607
-
Daniel Dunbar authored
llvm-svn: 79604
-
Bill Wendling authored
llvm-svn: 79603
-
Erick Tryzelaar authored
llvm-svn: 79602
-
Bill Wendling authored
llvm-svn: 79601
-
Bill Wendling authored
llvm-svn: 79598
-