- c748412 Update this test case. by Evan Cheng · 17 years ago
- c69d3c9 Don't let globalopt hack on volatile loads or stores. by Chris Lattner · 17 years ago
- 2454a2e eliminate additions of 0.0 when they are obviously dead. This has to be careful to by Chris Lattner · 17 years ago
- cca1867 Add a testcase for eliminating memcpy's at the end of functions. Forgot to commit this with my last commit. by Owen Anderson · 17 years ago
- 7cf9abf Add target triples to these so they don't fail on linux. by Chris Lattner · 17 years ago
- b7cbab8 This would be better done as an executable test. by Duncan Sands · 17 years ago
- 3cecb51 After recent changes we fail to optimize this test by Duncan Sands · 17 years ago
- c840da1 Filter loops that subtract induction variables. These loops are not yet handled. by Devang Patel · 17 years ago
- 053c1da Overhaul Cell SPU's addressing mode internals so that there are now by Scott Michel · 17 years ago
- 2ff48bc New test. by Devang Patel · 17 years ago
- fcd06cd Add test to make sure that #pragma mark/error doesn't error if there are by Bill Wendling · 17 years ago
- 07d403e Pure/const functions with ByVal parameters cannot be marked readonly either. by Duncan Sands · 17 years ago
- 349c72a this test is now compiled into the right thing. by Chris Lattner · 17 years ago
- 1f03154 Make this more likely to be passed byval. by Duncan Sands · 17 years ago
- 4189a53 Handle some more combinations of extend and icmp. Fixes PR1940. by Nick Lewycky · 17 years ago
- 25feae5 Fix PR1932 by disabling an xform invalid for fdiv. by Chris Lattner · 17 years ago
- 05bb789 Fix PR1938 by forcing the code that uses an undefined value to branch one by Chris Lattner · 17 years ago
- 15c2351 Update this test. Due to dag combiner improvements, we now compile f7/f11 to: by Chris Lattner · 17 years ago
- c837abe Be more careful modifying the use_list while also iterating through it. by Nick Lewycky · 17 years ago
- d82375c Revert r46393: readonly/readnone functions are no by Duncan Sands · 17 years ago
- 3bd39d4 Implement some dag combines that allow doing fneg/fabs/fcopysign in integer by Chris Lattner · 17 years ago
- 67fed1c The CorrelatedExpressions pass is now no more. by Bill Wendling · 17 years ago
- b753065 Fold fptrunc(add (fpextend x), (fpextend y)) -> add(x,y), as GCC does. by Chris Lattner · 17 years ago
- b8f4387 New test to verify that "merging 4 loads into a vec load" continues to work and by Chris Lattner · 17 years ago
- 1329cb8 Infer alignment of loads and increase their alignment when we can tell they are by Chris Lattner · 17 years ago
- 216f3f6 remove a useless xfailed test. by Chris Lattner · 17 years ago
- 63b2c2d Invert this test, because it is wrong if we allow by Duncan Sands · 17 years ago
- 824a721 If there's no instructions being emitted on X86 for a function, emit a by Bill Wendling · 17 years ago
- c79348d Need to convert to LLVM code and not C. by Bill Wendling · 17 years ago
- cd76af4 Rename the .c to .ll by Bill Wendling · 17 years ago
- f7214bc Move testcase to the code gen directory. by Bill Wendling · 17 years ago
- 418ab37 Create an explicit copy for byval parameters even by Duncan Sands · 17 years ago
- f69d5d7 If we have a function like this: by Bill Wendling · 17 years ago
- 15b86db Add another testcase. by Devang Patel · 17 years ago
- 5eee427 Fix some bugs in SimplifyNodeWithTwoResults where it would call deletenode to by Chris Lattner · 17 years ago
- 6bccafd New test case. by Evan Cheng · 17 years ago
- 0298dbf add a testcase for a bug Duncan pointed out. by Chris Lattner · 17 years ago
- a699574 Test for PR1942. by Duncan Sands · 17 years ago
- e3c36f6 DeadStoreElimination can treat byval parameters as if there were alloca's for the purpose of removing end-of-function stores. by Owen Anderson · 17 years ago
- af723b9 Add target-specific dag combines for FAND(x,0) and FOR(x,0). This allows by Chris Lattner · 17 years ago
- e205fef New test. by Devang Patel · 17 years ago
- fd68750 Teach basicaa that 'byval' arguments define a new memory location that by Chris Lattner · 17 years ago
- d43d00c Significantly simplify and improve handling of FP function results on x86-32. by Chris Lattner · 17 years ago
- 7323999 take these with a pr # by Chris Lattner · 17 years ago
- 2928650 Let each target decide byval alignment. For X86, it's 4-byte unless the aggregare contains SSE vector(s). For x86-64, it's max of 8 or alignment of the type. by Evan Cheng · 17 years ago
- 2cbdd27 SSE varargs arguments are passed in memory. by Evan Cheng · 17 years ago
- 3b8ea6e update this test to pass with duncan's change. by Chris Lattner · 17 years ago
- e6b0c00 Multiply can be evaluated in a different type, so long as the target type has by Nick Lewycky · 17 years ago
- 5768c96 New test. by Devang Patel · 17 years ago
- e5d53f5 New test. by Devang Patel · 17 years ago
- 5c5eb80 Implement flt_rounds for PowerPC. by Dale Johannesen · 17 years ago
- 994d6cf remove extraneous &&'s from tests, as Scott is apparently not going to. by Chris Lattner · 17 years ago
- 9929207 Test is correct again for the moment. by Dale Johannesen · 17 years ago
- a7a02fb Fix a latent bug exposed by my truncstore patch. We compiled stfiwx-2.ll to: by Chris Lattner · 17 years ago
- 58c5818 Forward progress: crtbegin.c now compiles successfully! by Scott Michel · 17 years ago
- ddf8956 This commit changes: by Chris Lattner · 17 years ago
- 41c5a39 new testcase. by Chris Lattner · 17 years ago
- e6d5c77 Test case for varargs parameter attribute issue I just fixed. by Evan Cheng · 17 years ago
- f51a9d4 add testcase that has been sitting in my tree for awhile. by Chris Lattner · 17 years ago
- 839b759 When a live virtual register is being clobbered by an implicit def, it is spilled by Evan Cheng · 17 years ago
- ab04e13 Fix arg promotion to propagate the correct attrs on the calls to by Chris Lattner · 17 years ago
- 02c4285 Fixes a nasty dag combiner bug that causes a bunch of tests to fail at -O0. by Evan Cheng · 17 years ago
- 339e14f Trampoline support for x86-64. This looks like by Duncan Sands · 17 years ago
- b8b9221 add testcase for regression by Chris Lattner · 17 years ago
- f763288 make sure to use a cpu that has sse. by Chris Lattner · 17 years ago
- 7ff79c2 My previous commit had an incomplete message, it should have been: by Chris Lattner · 17 years ago
- 8c231e5 Do not strip llvm.used values. by Devang Patel · 17 years ago
- beb1689 add a test to ensure that argpromote of one argument doesn't by Chris Lattner · 17 years ago
- e07704f verify x86 generates ud2 for llvm.trap by Chris Lattner · 17 years ago
- 9a856b0 new testcase for llvm.trap. by Chris Lattner · 17 years ago
- fdd7512 Testcase for gimplify_expr crash caused by an unexpected placeholder_expr. by Duncan Sands · 17 years ago
- b0c9b93 I noticed that the trampoline straightening transformation could by Duncan Sands · 17 years ago
- 02e30d7 This test is now the same as byval-1.ll, so remove it. by Duncan Sands · 17 years ago
- 559ad3c Test that byval cannot be used with pointers to types with no size. by Duncan Sands · 17 years ago
- d6a844c We now allow byval on fairly general pointer types. by Duncan Sands · 17 years ago
- 941db49 Fix the miscompilation of MiBench/consumer-lame that was exposed by Evan's by Chris Lattner · 17 years ago
- 37ac608 Turn a memcpy from a double* into a load/store of double instead of by Chris Lattner · 17 years ago
- ee2b7a4 Fix PR1907, a nasty miscompilation because instcombine didn't by Chris Lattner · 17 years ago
- 1fac17f Check that nested functions don't get pointless static chains. by Duncan Sands · 17 years ago
- e1e520f When turning a call to a bitcast function into a direct call, by Duncan Sands · 17 years ago
- 4db8f85 new testcase for rdar://5685492 by Chris Lattner · 17 years ago
- 92938b7 we don't have to make an explicit copy of a byval argument when by Chris Lattner · 17 years ago
- cfad1b4 Be more liberal in what parameter attributes are by Duncan Sands · 17 years ago
- a3554a4 this actually does pass with 4.0 by Chris Lattner · 17 years ago
- bfc5ae6 When DAE drops the varargs part of a function, ensure any by Duncan Sands · 17 years ago
- 48e1dce llvm-g++ 4.0 has completely different code for this warning, just xfail it. by Chris Lattner · 17 years ago
- 28123d7 new testcase by Chris Lattner · 17 years ago
- 623a389 Do not allow attributes beyond a function's last by Duncan Sands · 17 years ago
- 10603e0 Teach argpromote to ruthlessly hack small byval structs when it can by Chris Lattner · 17 years ago
- 8bc16f0 Two occurrences on one line count as one... by Duncan Sands · 17 years ago
- eb82470 If there are attributes on the varargs part of a call, don't discard them. by Duncan Sands · 17 years ago
- 497e888 More CellSPU refinements: by Scott Michel · 17 years ago
- ef68e75 Disable for now. by Dale Johannesen · 17 years ago
- c93adca When inlining a functino with a byval argument, make an explicit by Chris Lattner · 17 years ago
- 9de5d0d More CellSPU refinement and progress: by Scott Michel · 17 years ago
- 007f984 Output sinl for a long double FSIN node, not sin. by Duncan Sands · 17 years ago
- 9c9cec4 Codegen improvement has reduced one spill. by Evan Cheng · 17 years ago
- 314286f new testcase for PR1845 by Chris Lattner · 17 years ago
- 1910756 Special copy SUnit's do not have SDNode's. by Evan Cheng · 17 years ago
- 92b7c1d Fix sse2.psrl.w and sse2.psrl.q definitions. by Evan Cheng · 17 years ago