2008-07-23 |
Dan Gohman | Enable first-class aggregates support. |
tree | commitdiff |
2008-07-21 |
Dan Gohman | Fix a bug in LSR's dead-PHI cleanup. If a PHI has a... |
tree | commitdiff |
2008-07-18 |
Duncan Sands | Supress a gcc-4.3 warning. |
tree | commitdiff |
2008-07-18 |
Owen Anderson | Make PRE actually handle critical edges (by splitting... |
tree | commitdiff |
2008-07-18 |
Owen Anderson | Reapply r53735. My last patch fixed the failures Dan... |
tree | commitdiff |
2008-07-18 |
Owen Anderson | Add some checks that got lost in the shuffle. This... |
tree | commitdiff |
2008-07-18 |
Dan Gohman | Revert r53735. It broke SPEC 464.h264ref. |
tree | commitdiff |
2008-07-17 |
Owen Anderson | Use MergeBlockIntoPredecessor to simplify some code. |
tree | commitdiff |
2008-07-17 |
Owen Anderson | Make MergeBlockIntoPredecessor more aggressive when... |
tree | commitdiff |
2008-07-17 |
Owen Anderson | Enable PRE. My last batch of changes fixed the miscompile. |
tree | commitdiff |
2008-07-17 |
Matthijs Kooijman | Make GlobalOpt preserve address spaces when scalar... |
tree | commitdiff |
2008-07-17 |
Chris Lattner | Fix PR2553 |
tree | commitdiff |
2008-07-17 |
Evan Cheng | Inliner tweak. Function calls should cost more than... |
tree | commitdiff |
2008-07-17 |
Owen Anderson | Factor MergeBlockIntoPredecessor out into BasicBlockUtils. |
tree | commitdiff |
2008-07-16 |
Owen Anderson | There's no need to iterate block merging and PRE. ... |
tree | commitdiff |
2008-07-16 |
Matthijs Kooijman | Redo InstCombiner::visitExtractValueInst. Instead of... |
tree | commitdiff |
2008-07-16 |
Evan Cheng | Fix PR2296. Do not transform x86_sse2_storel_dq into... |
tree | commitdiff |
2008-07-15 |
Owen Anderson | Revert this, as it seems to still be broken. |
tree | commitdiff |
2008-07-15 |
Owen Anderson | Enable local PRE by default. |
tree | commitdiff |
2008-07-15 |
Owen Anderson | Have GVN do a pre-pass over the CFG that folds away... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Allow deadargelim to change return types even though... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Revert r53606. It turns out that explicitely tracking... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Make deadargelim a bit less smart, so it doesn't choke... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Don't use isa when we can reuse a previous dyn_cast. |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Make DeadArgElim keep liveness of the return value... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Let DAE keep a list of live functions, instead of simpl... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Split DAE::MarkLive into MarkLive and PropagateLiveness. |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Pass around const RetOrArg references instead of copyin... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Simplify debug code by using RetOrArg::getDescription(). |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Fix indentation (intentionally left out of the previous... |
tree | commitdiff |
2008-07-15 |
Matthijs Kooijman | Move the deadargelim code for intrinsically alive funct... |
tree | commitdiff |
2008-07-14 |
Dan Gohman | Fix uninitialized use of the Changed variable. |
tree | commitdiff |
2008-07-14 |
Chris Lattner | Reapply r53540, now with the matching header! |
tree | commitdiff |
2008-07-14 |
Duncan Sands | Revert r53540 - it does not compile. |
tree | commitdiff |
2008-07-14 |
Chris Lattner | If a function calls setjmp, never inline it into other... |
tree | commitdiff |
2008-07-14 |
Chris Lattner | simplify some code, shuffle and insertelt always return... |
tree | commitdiff |
2008-07-14 |
Chris Lattner | Fix PR2506 by being a bit more careful about reverse... |
tree | commitdiff |
2008-07-13 |
Chris Lattner | Fix mishandling of the infinite loop case when merging... |
tree | commitdiff |
2008-07-13 |
Chris Lattner | more refactoring. Use early exits instead of really... |
tree | commitdiff |
2008-07-13 |
Chris Lattner | improve comments. |
tree | commitdiff |
2008-07-13 |
Chris Lattner | factor another large hunk of code out into its own... |
tree | commitdiff |
2008-07-13 |
Chris Lattner | Final bit of simplification for FoldBranchToCommonDest. |
tree | commitdiff |
2008-07-13 |
Chris Lattner | simplify logic a bit |
tree | commitdiff |
2008-07-13 |
Chris Lattner | Refactor some code out into its own helper function... |
tree | commitdiff |
2008-07-12 |
Nick Lewycky | Enhance analysis of srem. |
tree | commitdiff |
2008-07-11 |
Dan Gohman | Use find instead of lower_bound. |
tree | commitdiff |
2008-07-11 |
Owen Anderson | Don't call lookupNumber more than we have to. |
tree | commitdiff |
2008-07-11 |
Nick Lewycky | Document 'mask' in this calculation. |
tree | commitdiff |
2008-07-11 |
Nick Lewycky | Remove misleading constant from comment. |
tree | commitdiff |
2008-07-11 |
Nick Lewycky | Add another optimization from PR2330. Also catch some... |
tree | commitdiff |
2008-07-11 |
Chris Lattner | a missed optimization that Eli spotted |
tree | commitdiff |
2008-07-11 |
Chris Lattner | another bug in the same line. |
tree | commitdiff |
2008-07-11 |
Chris Lattner | fix a bug spotted by Eli's eagle eyes |
tree | commitdiff |
2008-07-11 |
Chris Lattner | simplify and merge a bunch of code. Instead of compari... |
tree | commitdiff |
2008-07-11 |
Chris Lattner | fold away (x <= cst) earlier, allowing us to not have to |
tree | commitdiff |
2008-07-11 |
Chris Lattner | Fix folding of icmp's of i1 where the comparison is... |
tree | commitdiff |
2008-07-11 |
Chris Lattner | Fix a bogus optimization: folding (slt (zext i1 A to... |
tree | commitdiff |
2008-07-10 |
Matthijs Kooijman | Restructure dead argument elimination, try #3 :-) |
tree | commitdiff |
2008-07-10 |
Nick Lewycky | Fix overzealous optimization. Thanks to Duncan Sands... |
tree | commitdiff |
2008-07-09 |
Nick Lewycky | Simplify, suggested by Chris Lattner. |
tree | commitdiff |
2008-07-09 |
Nick Lewycky | Fold (a < 8) && (b < 8) into (a|b) < 8 for unsigned... |
tree | commitdiff |
2008-07-09 |
Nick Lewycky | Fold ((1 << a) & 1) to (a == 0). |
tree | commitdiff |
2008-07-09 |
Nick Lewycky | Reduce x - y to -y when we know the 'x' part will get... |
tree | commitdiff |
2008-07-09 |
Devang Patel | If loop induction variable's start value is less then... |
tree | commitdiff |
2008-07-08 |
Chris Lattner | Fix PR2496, a really nasty bug which involved sinking... |
tree | commitdiff |
2008-07-07 |
Evan Cheng | Fix two serious LSR bugs. |
tree | commitdiff |
2008-07-07 |
Dan Gohman | Make DenseMap's insert return a pair, to more closely... |
tree | commitdiff |
2008-07-05 |
Nick Lewycky | Fix missed optimization opportunity when analyzing... |
tree | commitdiff |
2008-07-03 |
Owen Anderson | Use information already present in the ValueTable to... |
tree | commitdiff |
2008-07-03 |
Devang Patel | Do not try to update dominator info while manipulating... |
tree | commitdiff |
2008-07-03 |
Owen Anderson | Remove the ability for ADCE to remove unreachable block... |
tree | commitdiff |
2008-07-03 |
Bill Wendling | Remove unused function. |
tree | commitdiff |
2008-07-03 |
Devang Patel | Preserve dom info. |
tree | commitdiff |
2008-07-03 |
Devang Patel | Remove extra FIXME |
tree | commitdiff |
2008-07-03 |
Devang Patel | Reconstruct dom info, if loop is unswitched. |
tree | commitdiff |
2008-07-03 |
Devang Patel | LoopUnswitch does not preserve dominator info in all... |
tree | commitdiff |
2008-07-03 |
Devang Patel | Undo previous patch. It is not that simple to fix dom... |
tree | commitdiff |
2008-07-02 |
Devang Patel | Preserve dom info while simplifing loop after the unswitch. |
tree | commitdiff |
2008-07-02 |
Owen Anderson | Use df_ext_iterator to capture the reachable set withou... |
tree | commitdiff |
2008-07-02 |
Owen Anderson | Avoid a redundant call. |
tree | commitdiff |
2008-07-02 |
Owen Anderson | Add support to ADCE for pruning unreachable blocks... |
tree | commitdiff |
2008-07-02 |
Owen Anderson | Use DenseSet rather than SmallPtrSet for the alive... |
tree | commitdiff |
2008-07-02 |
Owen Anderson | A better fix for PR2503 that doesn't pessimize GVN... |
tree | commitdiff |
2008-07-02 |
Devang Patel | reuse vectors. |
tree | commitdiff |
2008-07-02 |
Devang Patel | Fix comment. |
tree | commitdiff |
2008-07-02 |
Devang Patel | Preserve loop data so that it is not fetched everytime... |
tree | commitdiff |
2008-06-30 |
Evan Cheng | - Re-apply 52748 and friends with fix. GetConstantStrin... |
tree | commitdiff |
2008-06-29 |
Anton Korobeynikov | Revert (52748 and friends): |
tree | commitdiff |
2008-06-26 |
Eric Christopher | Remove unused function. |
tree | commitdiff |
2008-06-26 |
Eric Christopher | Move GetConstantStringInfo to lib/Analysis. Remove |
tree | commitdiff |
2008-06-25 |
Evan Cheng | Restore DeadArgElim back to 52570. It's breaking 447... |
tree | commitdiff |
2008-06-25 |
Duncan Sands | Pacify gcc-4.3. |
tree | commitdiff |
2008-06-25 |
Matthijs Kooijman | Fix a (false) warning on darwin. |
tree | commitdiff |
2008-06-25 |
Matthijs Kooijman | Fix some cosmetics in comments. |
tree | commitdiff |
2008-06-25 |
Evan Cheng | - Use O(1) check of basic block size limit. |
tree | commitdiff |
2008-06-25 |
Chris Lattner | Fix PR2488, a case where we deleted stack restores... |
tree | commitdiff |
2008-06-24 |
Dan Gohman | Revert 52645, the loop unroller changes. It caused... |
tree | commitdiff |
2008-06-24 |
Dan Gohman | Fix a typo in a comment. |
tree | commitdiff |
2008-06-24 |
Matthijs Kooijman | Commit the new DeadArgElim pass again, this time with... |
tree | commitdiff |
2008-06-24 |
Matthijs Kooijman | Rename a few variables to be more consistent. |
tree | commitdiff |
next |