interest for this, as it currently reserves a register rather than using
the scavenger for matierializing constants as needed.
Instead of scavenging registers on the fly while eliminating frame indices,
new virtual registers are created, and then a scavenged collectively in a
post-pass over the function. This isolates the bits that need to interact
with the scavenger, and sets the stage for more intelligent use, and reuse,
of scavenged registers.
For the time being, this is disabled by default. Once the bugs are worked out,
the current scavenging calls in replaceFrameIndices() will be removed and
the post-pass scavenging will be the default. Until then,
-enable-frame-index-scavenging enables the new code. Currently, only the
Thumb1 back end is set up to use it.
llvm-svn: 82734
first implementation recognizes when a function declaration is an
explicit function template specialization (based on the presence of a
template<> header), performs template argument deduction + ambiguity
resolution to determine which template is being specialized, and hooks
There are many caveats here:
- We completely and totally drop any explicitly-specified template
arguments on the floor
- We don't diagnose any of the extra semantic things that we should
diagnose.
- I haven't looked to see that we're getting the right linkage for
explicit specializations
On a happy note, this silences a bunch of errors that show up in
libstdc++'s <iostream>, although Clang still can't get through the
entire header.
llvm-svn: 82728
CHECK strings, instead of canonicalizing the patterns directly. This allows
Pattern to just contain a StringRef instead of std::string.
llvm-svn: 82713
Type hierarchy. Demote 'volatile' to extended-qualifier status. Audit our
use of qualifiers and fix a few places that weren't dealing with qualifiers
quite right; many more remain.
llvm-svn: 82705
have the iterators and operator[] handle the traversal of statements, as they
are stored in reverse order. Tests show this has no real performance impact, but
it does simply the CFG construction logic and will make it slightly easier to
change the allocation strategy for CFGBlocks (as we have fewer copies).
llvm-svn: 82702
LocalAreaOffset. (We don't have any of those right now.)
PEI::calculateFrameObjectOffsets includes the absolute value of the
LocalAreaOffset in the cumulative offset value used to calculate the
stack frame size. It then adds the raw value of the LocalAreaOffset
to the stack size. For a StackGrowsDown target, that raw value is negative
and has the effect of cancelling out the absolute value that was added
earlier, but that obviously won't work for a StackGrowsUp target. Change
to subtract the absolute value of the LocalAreaOffset.
llvm-svn: 82693
handled correctly.
- <rdar://problem/7247671> Function arguments incorrect when function returns a
struct on i386 w/ llvm-g++ and clang
llvm-svn: 82681
The issue was a discrepancy between how RegionStoreManager::Bind() and
RegionStoreManager::Retrieve() derived the "key" for the first element
of a symbolic region.
llvm-svn: 82680
LiveVariables add implicit kills to correctly track partial register kills. This works well enough and is fairly accurate. But coalescer can make it impossible to maintain these markers. e.g.
BL <ga:sss1>, %R0<kill,undef>, %S0<kill>, %R0<imp-def>, %R1<imp-def,dead>, %R2<imp-def,dead>, %R3<imp-def,dead>, %R12<imp-def,dead>, %LR<imp-def,dead>, %D0<imp-def>, ...
...
%reg1031<def> = FLDS <cp#1>, 0, 14, %reg0, Mem:LD4[ConstantPool]
...
%S0<def> = FCPYS %reg1031<kill>, 14, %reg0, %D0<imp-use,kill>
When reg1031 and S0 are coalesced, the copy (FCPYS) will be eliminated the the implicit-kill of D0 is lost. In this case it's possible to move the marker to the FLDS. But in many cases, this is not possible. Suppose
%reg1031<def> = FOO <cp#1>, %D0<imp-def>
...
%S0<def> = FCPYS %reg1031<kill>, 14, %reg0, %D0<imp-use,kill>
When FCPYS goes away, the definition of S0 is the "FOO" instruction. However, transferring the D0 implicit-kill to FOO doesn't work since it is the def of D0 itself. We need to fix this in another time by introducing a "kill" pseudo instruction to track liveness.
Disabling the assertion is not ideal, but machine verifier is doing that job now. It's important to know double-def is not a miscomputation since it means a register should be free but it's not tracked as free. It's a performance issue instead.
llvm-svn: 82677
Jim Grosbach