- Dec 09, 2009
-
-
Bob Wilson authored
llvm-svn: 90975
-
- Dec 08, 2009
-
-
Bob Wilson authored
llvm-svn: 90866
-
Bob Wilson authored
I'm not aware that this does anything significant on its own, but it's needed for another patch that I'm working on. llvm-svn: 90864
-
- Dec 04, 2009
-
-
Bob Wilson authored
llvm-svn: 90603
-
Bob Wilson authored
llvm-svn: 90601
-
- Nov 29, 2009
-
-
Benjamin Kramer authored
llvm-svn: 90097
-
Benjamin Kramer authored
llvm-svn: 90089
-
- Nov 27, 2009
-
-
Chris Lattner authored
array indexes. The "complex" case of SRoA still handles them, and correctly. This fixes a weirdness where we'd correctly avoid transforming A[0][42] if the 42 was too large, but we'd only do it if it was one gep, not two separate ones. llvm-svn: 90007
-
- Nov 23, 2009
-
-
Nick Lewycky authored
llvm-svn: 89645
-
- Oct 23, 2009
-
-
Victor Hernandez authored
Remove AllocationInst. Since MallocInst went away, AllocaInst is the only subclass of AllocationInst, so it no longer is necessary. llvm-svn: 84969
-
- Oct 05, 2009
-
-
Chris Lattner authored
the new predicates I added) instead of going through a context and doing a pointer comparison. Besides being cheaper, this allows a smart compiler to turn the if sequence into a switch. llvm-svn: 83297
-
- Sep 15, 2009
-
-
Nick Lewycky authored
llvm-svn: 81851
-
Chris Lattner authored
llvm-svn: 81840
-
- Sep 02, 2009
-
-
Chris Lattner authored
llvm-svn: 80766
-
- Aug 23, 2009
-
-
Chris Lattner authored
This required converting a bunch of stuff off DOUT and other cleanups. llvm-svn: 79819
-
- Aug 19, 2009
-
-
Dan Gohman authored
available. This is very conservative for now. llvm-svn: 79442
-
- Aug 17, 2009
-
-
Nick Lewycky authored
llvm-svn: 79226
-
- Aug 13, 2009
-
-
Owen Anderson authored
llvm-svn: 78948
-
- Jul 31, 2009
-
-
Owen Anderson authored
metadata related, which I'm waiting on to avoid conflicting with Devang. llvm-svn: 77721
-
Owen Anderson authored
llvm-svn: 77635
-
- Jul 30, 2009
-
-
Daniel Dunbar authored
llvm-svn: 77605
-
Daniel Dunbar authored
a Twine, e.g., for names). - I am a little ambivalent about this; we don't want the string conversion of utostr, but using overload '+' mixed with string and integer arguments is sketchy. On the other hand, this particular usage is something of an idiom. llvm-svn: 77579
-
Owen Anderson authored
llvm-svn: 77516
-
- Jul 29, 2009
-
-
Owen Anderson authored
llvm-svn: 77494
-
- Jul 28, 2009
-
-
Owen Anderson authored
llvm-svn: 77366
-
- Jul 25, 2009
-
-
Daniel Dunbar authored
- The only meat here is in Value.{h,cpp} the rest is essential 'const std::string &' -> 'const Twine &'. llvm-svn: 77048
-
Owen Anderson authored
Revert the ConstantInt constructors back to their 2.5 forms where possible, thanks to contexts-on-types. More to come. llvm-svn: 77011
-
- Jul 22, 2009
-
-
Owen Anderson authored
llvm-svn: 76702
-
- Jul 16, 2009
-
-
Owen Anderson authored
llvm-svn: 75863
-
- Jul 15, 2009
-
-
Owen Anderson authored
llvm-svn: 75703
-
- Jul 14, 2009
-
-
Torok Edwin authored
This adds location info for all llvm_unreachable calls (which is a macro now) in !NDEBUG builds. In NDEBUG builds location info and the message is off (it only prints "UREACHABLE executed"). llvm-svn: 75640
-
- Jul 11, 2009
-
-
Torok Edwin authored
Make llvm_unreachable take an optional string, thus moving the cerr<< out of line. LLVM_UNREACHABLE is now a simple wrapper that makes the message go away for NDEBUG builds. llvm-svn: 75379
-
Torok Edwin authored
and abort()/exit() -> llvm_report_error(). llvm-svn: 75363
-
- Jul 10, 2009
-
-
Owen Anderson authored
This started as a small change, I swear. Unfortunately, lots of things call the [I|F]CmpInst constructors. Who knew!? llvm-svn: 75200
-
- Jul 07, 2009
-
-
Owen Anderson authored
llvm-svn: 74878
-
- Jul 06, 2009
-
-
Owen Anderson authored
llvm-svn: 74807
-
- Jul 03, 2009
-
-
Owen Anderson authored
llvm-svn: 74781
-
- Jun 16, 2009
-
-
Dan Gohman authored
llvm-svn: 73451
-
- Jun 06, 2009
-
-
Jay Foad authored
Instructions. llvm-svn: 73002
-
- Jun 01, 2009
-
-
Eli Friedman authored
RewriteStoreUserOfWholeAlloca deal with tail padding because isSafeUseOfBitCastedAllocation expects them to. Otherwise, we crash trying to erase the bitcast. llvm-svn: 72688
-