aboutsummaryrefslogtreecommitdiff
AgeCommit message (Expand)AuthorFilesLines
2004-03-16New testcase for PR293Chris Lattner1-0/+12
2004-03-16Bug fixedChris Lattner1-0/+1
2004-03-16foo Bug 291, bar PR291Chris Lattner1-0/+1
2004-03-16foo bug291 bar PR 291Chris Lattner1-1/+0
2004-03-16PR291Chris Lattner1-0/+1
2004-03-16// Testcase from Bug 291Chris Lattner1-0/+11
2004-03-16Add some missing functions. Make sure to handle calls together in case theChris Lattner2-3/+21
2004-03-16Fix thinko, and PR292Chris Lattner1-1/+1
2004-03-16Fix PR294Chris Lattner1-1/+2
2004-03-15Add LAHF instructionAlkis Evlogimenos1-0/+1
2004-03-15Restore old inlining heuristic. As the comment indicates, this is a nastyChris Lattner1-1/+8
2004-03-15Ok, the assertion was bogus. Calls that do not read/write memory should notChris Lattner1-0/+6
2004-03-15This assertion is bogus now that calls do not necessarily read/write memoryChris Lattner1-2/+0
2004-03-15Add counters for the number of calls elimiantedChris Lattner1-0/+6
2004-03-15Implement CSE of call instructions in the most trivial case. This implementsChris Lattner1-11/+139
2004-03-15New testcase for CSE of call instructionsChris Lattner1-0/+11
2004-03-15Fix a minor bug, implementing GCSE/call_pure_function.llChris Lattner1-3/+6
2004-03-15Make sure that pure calls don't kill loadsChris Lattner1-0/+13
2004-03-15Implement LICM of calls in simple cases. This is sufficient to move aroundChris Lattner1-1/+31
2004-03-15New testcases to test LICM of call instructionsChris Lattner2-0/+30
2004-03-15Don't be COMPLETELY pessimistic in the face of function callsChris Lattner1-4/+20
2004-03-15Tweak argumentChris Lattner1-1/+1
2004-03-15Deinline a couple of methods. Improve comment.Chris Lattner1-12/+7
2004-03-15Deinline some virtual methods, provide better mod/ref answers through theChris Lattner1-5/+21
2004-03-15Pass through the boolean queriesChris Lattner1-0/+12
2004-03-15Teach basicaa about some stdc functions.Chris Lattner1-5/+100
2004-03-15Add two new methods which can be used to enable a bunch of transformationsChris Lattner1-0/+22
2004-03-15Mostly cosmetic improvements. Do fix the bug where a global value was consid...Chris Lattner1-35/+23
2004-03-15Assert that input blocks meet the invariants we expectChris Lattner1-42/+38
2004-03-15New testcase that causes the code extractor to generate bogus code.Chris Lattner1-0/+34
2004-03-15Fix several bugs in the loop extractor. In particular, subloops were neverChris Lattner1-8/+48
2004-03-14No correctness fixes here, just minor qoi fixes:Chris Lattner1-30/+26
2004-03-14Simplify code a bit, and fix bug CodeExtractor/2004-03-14-NoSwitchSupport.llChris Lattner1-62/+34
2004-03-14Simplify the code a bit by making the collection of basic blocks to extractChris Lattner1-57/+39
2004-03-14New testcase that crashes the loop extractorChris Lattner1-0/+27
2004-03-14After reducing a miscompiled program down to the functions which are beingChris Lattner2-28/+110
2004-03-14Fix a minor bug in runPassesOnChris Lattner1-0/+1
2004-03-14Add a new "AutoDebugCrashes" optionChris Lattner2-4/+17
2004-03-14Refactor to use a new methodChris Lattner1-39/+15
2004-03-14Add new methodChris Lattner2-0/+28
2004-03-14Refactor and clean up a bunch more code. No major functionality changes.Chris Lattner6-89/+98
2004-03-14Another API change to MRegisterInfo::foldMemoryOperand. Instead of aAlkis Evlogimenos3-6/+5
2004-03-14Add a method to extract a loopChris Lattner2-1/+42
2004-03-14Split into two passes. Now there is the general loop extractor, usable onChris Lattner1-6/+24
2004-03-14Rename createLoopExtractorPass to createSingleLoopExtractorPassChris Lattner1-63/+64
2004-03-14add a fixmeChris Lattner1-0/+5
2004-03-14Refactor all of the "splitting a module into two pieces" code to avoidChris Lattner5-129/+92
2004-03-14Change MRegisterInfo::foldMemoryOperand to return the foldedAlkis Evlogimenos5-191/+188
2004-03-14Passes don't print stuff!Chris Lattner1-2/+0
2004-03-14Do not create empty basic blocks when the lowerswitch pass expects blocks toChris Lattner1-5/+2