2008-01-29 |
Duncan Sands | When expanding an operand, it's not the result value
|
commit | commitdiff | tree |
2008-01-29 |
Duncan Sands | Use getPreferredAlignmentLog or getPreferredAlignment
|
commit | commitdiff | tree |
2008-01-29 |
Duncan Sands | This would be better done as an executable test.
|
commit | commitdiff | tree |
2008-01-29 |
Duncan Sands | After recent changes we fail to optimize this test
|
commit | commitdiff | tree |
2008-01-28 |
Duncan Sands | Pure/const functions with ByVal parameters cannot
|
commit | commitdiff | tree |
2008-01-28 |
Duncan Sands | Unbreak builds with differing object and source
|
commit | commitdiff | tree |
2008-01-28 |
Duncan Sands | Make this more likely to be passed byval.
|
commit | commitdiff | tree |
2008-01-27 |
Duncan Sands | Revert r46393: readonly/readnone functions are no
|
commit | commitdiff | tree |
2008-01-26 |
Duncan Sands | Invert this test, because it is wrong if we allow
|
commit | commitdiff | tree |
2008-01-26 |
Duncan Sands | Create an explicit copy for byval parameters even
|
commit | commitdiff | tree |
2008-01-25 |
Duncan Sands | Do this more neatly.
|
commit | commitdiff | tree |
2008-01-25 |
Duncan Sands | Test for PR1942.
|
commit | commitdiff | tree |
2008-01-25 |
Duncan Sands | Add more assertions to catch accesses outside of
|
commit | commitdiff | tree |
2008-01-23 |
Duncan Sands | The last pieces needed for loading arbitrary
|
commit | commitdiff | tree |
2008-01-22 |
Duncan Sands | The final piece needed for storing arbitrary precision
|
commit | commitdiff | tree |
2008-01-21 |
Duncan Sands | It turns out that in C++ it is legal to declare functions
|
commit | commitdiff | tree |
2008-01-21 |
Duncan Sands | Check that sret is only used on pointers to types
|
commit | commitdiff | tree |
2008-01-21 |
Duncan Sands | Be consistent with other attribute methods, and
|
commit | commitdiff | tree |
2008-01-20 |
Duncan Sands | Make sure the caller doesn't use freed memory.
|
commit | commitdiff | tree |
2008-01-20 |
Duncan Sands | Initializing an unsigned with ~0UL causes the compiler
|
commit | commitdiff | tree |
2008-01-19 |
Duncan Sands | Need to handle any 'nest' parameter before integer
|
commit | commitdiff | tree |
2008-01-16 |
Duncan Sands | Trampoline support for x86-64. This looks like
|
commit | commitdiff | tree |
2008-01-15 |
Duncan Sands | Testcase for gimplify_expr crash caused by an
|
commit | commitdiff | tree |
2008-01-14 |
Duncan Sands | Simplify CallInst::hasByValArgument using a new method.
|
commit | commitdiff | tree |
2008-01-14 |
Duncan Sands | I noticed that the trampoline straightening transformation...
|
commit | commitdiff | tree |
2008-01-14 |
Duncan Sands | This test is now the same as byval-1.ll, so remove it.
|
commit | commitdiff | tree |
2008-01-14 |
Duncan Sands | Test that byval cannot be used with pointers to
|
commit | commitdiff | tree |
2008-01-14 |
Duncan Sands | We now allow byval on fairly general pointer types.
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | Whitespace tweak.
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | Remove the assumption that byval has been applied to
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | Allow the byval attribute for pointers to any type...
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | Check that nested functions don't get pointless
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | Small simplification.
|
commit | commitdiff | tree |
2008-01-13 |
Duncan Sands | When turning a call to a bitcast function into a direct...
|
commit | commitdiff | tree |
2008-01-12 |
Duncan Sands | Be more liberal in what parameter attributes are
|
commit | commitdiff | tree |
2008-01-11 |
Duncan Sands | When DAE drops the varargs part of a function, ensure any
|
commit | commitdiff | tree |
2008-01-11 |
Duncan Sands | Do not allow attributes beyond a function's last
|
commit | commitdiff | tree |
2008-01-11 |
Duncan Sands | Two occurrences on one line count as one...
|
commit | commitdiff | tree |
2008-01-11 |
Duncan Sands | If there are attributes on the varargs part of a
|
commit | commitdiff | tree |
2008-01-10 |
Duncan Sands | Output sinl for a long double FSIN node, not sin.
|
commit | commitdiff | tree |
2008-01-09 |
Duncan Sands | Fix compile failures with g++-4.3.
|
commit | commitdiff | tree |
2008-01-08 |
Duncan Sands | Crashes llc when using Chris's new legalization logic.
|
commit | commitdiff | tree |
2008-01-08 |
Duncan Sands | Use size_t to store Pos, avoid truncating value
|
commit | commitdiff | tree |
2008-01-07 |
Duncan Sands | I doubt the address of the Error string was intended
|
commit | commitdiff | tree |
2008-01-07 |
Duncan Sands | Add missing newline at EOF.
|
commit | commitdiff | tree |
2008-01-07 |
Duncan Sands | Small cleanup for handling of type/parameter attribute
|
commit | commitdiff | tree |
2008-01-07 |
Duncan Sands | Unbreak x86-32 darwin long double!
|
commit | commitdiff | tree |
2008-01-07 |
Duncan Sands | Fix long double support on x86-32 linux.
|
commit | commitdiff | tree |
2008-01-06 |
Duncan Sands | The transform that tries to turn calls to bitcast functions...
|
commit | commitdiff | tree |
2008-01-06 |
Duncan Sands | When transforming a call to a bitcast function into
|
commit | commitdiff | tree |
2008-01-06 |
Duncan Sands | Remove an unused variable.
|
commit | commitdiff | tree |
2008-01-04 |
Duncan Sands | Testcase with non-integer "bitfields" (in quotes,
|
commit | commitdiff | tree |
2008-01-04 |
Duncan Sands | Testcase for PR1386.
|
commit | commitdiff | tree |
2008-01-03 |
Duncan Sands | Test for handling of large bit offset from a
|
commit | commitdiff | tree |
2008-01-03 |
Duncan Sands | An example for which the TYPE_SIZE was being set from
|
commit | commitdiff | tree |
2008-01-03 |
Duncan Sands | Test handling of records for which the fields are
|
commit | commitdiff | tree |
2007-12-31 |
Duncan Sands | Fix PR1833 - eh.exception and eh.selector return two
|
commit | commitdiff | tree |
2007-12-21 |
Duncan Sands | Fix a brain fart by our beloved leader (the content
|
commit | commitdiff | tree |
2007-12-21 |
Duncan Sands | Get the verifier to check attributes on calls as well
|
commit | commitdiff | tree |
2007-12-21 |
Duncan Sands | Make DAE not wipe out attributes on calls, and not...
|
commit | commitdiff | tree |
2007-12-19 |
Duncan Sands | When inlining through an 'nounwind' call, mark inlined
|
commit | commitdiff | tree |
2007-12-19 |
Duncan Sands | Simplify LowerCallTo by using a callsite.
|
commit | commitdiff | tree |
2007-12-19 |
Duncan Sands | The C++ exception handling personality function wants
|
commit | commitdiff | tree |
2007-12-18 |
Duncan Sands | Rename isNoReturn to doesNotReturn, and isNoUnwind to
|
commit | commitdiff | tree |
2007-12-17 |
Duncan Sands | Make invokes of inline asm legal. Teach codegen
|
commit | commitdiff | tree |
2007-12-16 |
Duncan Sands | Revert this part of r45073 until the verifier is
|
commit | commitdiff | tree |
2007-12-16 |
Duncan Sands | Make instcombine promote inline asm calls to 'nounwind'
|
commit | commitdiff | tree |
2007-12-15 |
Duncan Sands | These are more correctly called signaling NaNs.
|
commit | commitdiff | tree |
2007-12-14 |
Duncan Sands | Teach the interpreter to read and write memory in the
|
commit | commitdiff | tree |
2007-12-12 |
Duncan Sands | Remove host endianness info from TargetData and
|
commit | commitdiff | tree |
2007-12-12 |
Duncan Sands | Revert r44626, which turned off the use of readonly
|
commit | commitdiff | tree |
2007-12-11 |
Duncan Sands | Fix compilation.
|
commit | commitdiff | tree |
2007-12-10 |
Duncan Sands | Make PruneEH update the nounwind/noreturn attributes
|
commit | commitdiff | tree |
2007-12-10 |
Duncan Sands | Fix PR1836: in the interpreter, read and write apints
|
commit | commitdiff | tree |
2007-12-10 |
Duncan Sands | It looks like this has been broken for some time -
|
commit | commitdiff | tree |
2007-12-05 |
Duncan Sands | Commit 44487 broke bootstrap of llvm-gcc-4.2. It is
|
commit | commitdiff | tree |
2007-12-03 |
Duncan Sands | Rather than having special rules like "intrinsics cannot
|
commit | commitdiff | tree |
2007-12-01 |
Duncan Sands | Integrate the readonly/readnone logic more deeply
|
commit | commitdiff | tree |
2007-11-30 |
Duncan Sands | Small optimization of parameter attribute lookup.
|
commit | commitdiff | tree |
2007-11-30 |
Duncan Sands | Add a convenience method for modifying parameter
|
commit | commitdiff | tree |
2007-11-30 |
Duncan Sands | Check that there are not more attributes than
|
commit | commitdiff | tree |
2007-11-29 |
Duncan Sands | Small parameter attributes cleanup.
|
commit | commitdiff | tree |
2007-11-28 |
Duncan Sands | Add some convenience methods for querying attributes...
|
commit | commitdiff | tree |
2007-11-28 |
Duncan Sands | Revert previous "fix" - the breakage was due to some
|
commit | commitdiff | tree |
2007-11-28 |
Duncan Sands | This test somehow got an extra "not" during the
|
commit | commitdiff | tree |
2007-11-28 |
Duncan Sands | My compiler complains that "x always evaluates to true"
|
commit | commitdiff | tree |
2007-11-28 |
Duncan Sands | Add missing newlines at EOF.
|
commit | commitdiff | tree |
2007-11-27 |
Duncan Sands | Fix PR1146: parameter attributes are longer part of
|
commit | commitdiff | tree |
2007-11-25 |
Duncan Sands | Fix PR1816. If a bitcast of a function only exists...
|
commit | commitdiff | tree |
2007-11-23 |
Duncan Sands | Ding dong, the DoesntAccessMemoryFns and
|
commit | commitdiff | tree |
2007-11-23 |
Duncan Sands | Remove some logic I thoughtlessly copied over
|
commit | commitdiff | tree |
2007-11-22 |
Duncan Sands | Turn invokes of nounwind functions into ordinary calls.
|
commit | commitdiff | tree |
2007-11-22 |
Duncan Sands | Teach alias analysis about readnone/readonly functions.
|
commit | commitdiff | tree |
2007-11-22 |
Duncan Sands | Readonly/readnone functions are allowed to throw
|
commit | commitdiff | tree |
2007-11-22 |
Duncan Sands | Rename the 'const' parameter attribute to 'readnone',
|
commit | commitdiff | tree |
2007-11-21 |
Duncan Sands | Fix a bug in which node A is replaced by node B, but...
|
commit | commitdiff | tree |
2007-11-20 |
Duncan Sands | In order for parameter attribute uniquing to make
|
commit | commitdiff | tree |
2007-11-15 |
Duncan Sands | This assertion was bogus.
|
commit | commitdiff | tree |
2007-11-14 |
Duncan Sands | I discover array_lengthof, thanks to gabor on #llvm.
|
commit | commitdiff | tree |
2007-11-14 |
Duncan Sands | Tweak the descriptions of the new 'const' and 'pure...
|
commit | commitdiff | tree |
next |