This patch makes the Merge Functions pass faster by calculating and comparing
a hash value which captures the essential structure of a function before
performing a full function comparison.
The hash is calculated by hashing the function signature, then walking the basic
blocks of the function in the same order as the main comparison function. The
opcode of each instruction is hashed in sequence, which means that different
functions according to the existing total order cannot have the same hash, as
the comparison requires the opcodes of the two functions to be the same order.
The hash function is a static member of the FunctionComparator class because it
is tightly coupled to the exact comparison function used. For example, functions
which are equivalent modulo a single variant callsite might be merged by a more
aggressive MergeFunctions, and the hash function would need to be insensitive to
these differences in order to exploit this.
The hashing function uses a utility class which accumulates the values into an
internal state using a standard bit-mixing function. Note that this is a different interface
than a regular hashing routine, because the values to be hashed are scattered
amongst the properties of a llvm::Function, not linear in memory. This scheme is
fast because only one word of state needs to be kept, and the mixing function is
a few instructions.
The main runOnModule function first computes the hash of each function, and only
further processes functions which do not have a unique function hash. The hash
is also used to order the sorted function set. If the hashes differ, their
values are used to order the functions, otherwise the full comparison is done.
Both of these are helpful in speeding up MergeFunctions. Together they result in
speedups of 9% for mysqld (a mostly C application with little redundancy), 46%
for libxul in Firefox, and 117% for Chromium. (These are all LTO builds.) In all
three cases, the new speed of MergeFunctions is about half that of the module
verifier, making it relatively inexpensive even for large LTO builds with
hundreds of thousands of functions. The same functions are merged, so this
change is free performance.
Author: jrkoenig
Reviewers: nlewycky, dschuff, jfb
Subscribers: llvm-commits, aemerson
Differential revision: http://reviews.llvm.org/D11923
llvm-svn: 245140
This enables Clang to correctly handle code such as:
struct __declspec(dllexport) S {
int x = 42;
};
where it would otherwise error due to trying to generate the default
constructor before the in-class initializer for x has been parsed.
Differential Revision: http://reviews.llvm.org/D11850
llvm-svn: 245139
This seems to only work some of the time. In some situations,
this seems to use a nonsensical type and isn't actually aware of the
memory being accessed. e.g. if branch condition is an icmp of a pointer,
it checks the addressing mode of i1.
llvm-svn: 245137
True branch instructions do behave as expected with liveness.
Avoid the phrasing "branch decision is based on a value in an SGPR"
because this could be misleading. A VALU compare instruction's
result is still based on an SGPR, even though that condition
may be divergent.
llvm-svn: 245131
Neither of these libraries has been ported to Windows. Eventually
if they are ever ported we can re-enable these tests. But more
immediately what we need to do is add new data formatters for
MSVC's STL implementation. This is tracked in
http://llvm.org/pr24460.
llvm-svn: 245125
Summary: Adding check to emit diagnostic for invalid tag when concept is specified and associated tests.
Reviewers: rsmith, hubert.reinterpretcast, fraggamuffin, faisalv, aaron.ballman
Subscribers: aaron.ballman, cfe-commits
Differential Revision: http://reviews.llvm.org/D11916
llvm-svn: 245123
Summary:
http://reviews.llvm.org/D11212 made Scalar Evolution able to propagate NSW and NUW flags from instructions to SCEVs for add instructions. This patch expands that to sub, mul and shl instructions.
This change makes LSR able to generate pointer induction variables for loops like these, where the index is 32 bit and the pointer is 64 bit:
for (int i = 0; i < numIterations; ++i)
sum += ptr[i - offset];
for (int i = 0; i < numIterations; ++i)
sum += ptr[i * stride];
for (int i = 0; i < numIterations; ++i)
sum += ptr[3 * (i << 7)];
Reviewers: atrick, sanjoy
Subscribers: sanjoy, majnemer, hfinkel, llvm-commits, meheff, jingyue, eliben
Differential Revision: http://reviews.llvm.org/D11860
llvm-svn: 245118
Although targeting CoreCLR is similar to targeting MSVC, there are
certain important differences that the backend must be aware of
(e.g. differences in stack probes, EH, and library calls).
Differential Revision: http://reviews.llvm.org/D11012
llvm-svn: 245115
We canonicalize V64 vectors to V128 through insert_subvector: the other
FMLA/FMLS/FMUL/FMULX patterns match that already, but this one doesn't,
so we'd fail to match fmls and generate fneg+fmla instead.
The vector equivalents are already tested and functional.
llvm-svn: 245107
Add checkers that detect code-level localizability issues for OS X / iOS:
- A path sensitive checker that warns about uses of non-localized
NSStrings passed to UI methods expecting localized strings.
- A syntax checker that warns against not including a comment in
NSLocalizedString macros.
A patch by Kulpreet Chilana!
(This is the second attempt with the compilation issue on Windows and
the random test failures resolved.)
llvm-svn: 245093
Summary:
It always makes NewBB the entry of the region instead of OldBB. This breaks if there are edges from inside the region to OldBB. OldBB is moved out of the region and hence there are exiting edges to OldBB and the region's exit block, contradicting the single-exit condition for regions.
The only use from Polly is going to be removed, hence I propose to remove the function completely.
Reviewers: grosser
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D11873
llvm-svn: 245092
This modifies the order in which Polly passes are executed.
Assuming a function has two scops (A and B), the order before was:
FunctionPassManager
ScopDetection
IndependentBlocks
TempScopInfo for A and B
RegionPassManager
ScopInfo for A
DependenceInfo for A
IslScheduleOptimizer for A
IslAstInfo for A
CodeGeneration for A
ScopInfo for B
DependenceInfo for B
IslScheduleOptimizer for B
IslAstInfo for B
CodeGeneration for B
After this patch:
FunctionPassManager
ScopDetection
IndependentBlocks
RegionPassManager
TempScopInfo for A
ScopInfo for A
DependenceInfo for A
IslScheduleOptimizer for A
IslAstInfo for A
CodeGeneration for A
TempScopInfo for B
ScopInfo for B
DependenceInfo for B
IslScheduleOptimizer for B
IslAstInfo for B
CodeGeneration for B
TempScopInfo for B might store information and references to the IR
that CodeGeneration for A might modify. Changing the order ensures that
the IR is not modified from the analysis of a region until code
generation.
Reviewers: grosser
Differential Revision: http://reviews.llvm.org/D12014
llvm-svn: 245091
Another step towards isolating all language/AST specific code into the files to further abstract specific implementations of parsing types for a given language.
llvm-svn: 245090
It is flaky due to inability to remove files with open handles. We
could paper over it with rm -f, but then the file would still be
present.
This is more evidence to me that we should roll our own 'rm'
implementation in LLVM.
llvm-svn: 245083
This patch makes the Darwin ARM backend take advantage of TargetParser. It
also teaches TargetParser about ARMV7K for the first time. This makes target
triple parsing more consistent across llvm.
Differential Revision: http://reviews.llvm.org/D11996
llvm-svn: 245081
Summary:
MSDN says that fastcall, stdcall, thiscall, and vectorcall are all
accepted but ignored on ARM and X64.
https://msdn.microsoft.com/en-us/library/984x0h58.aspx
MSDN also says cdecl is also accepted and typically ignored
This patch brings ARM in line with how we ignore them for X64
Reviewers: rnk
Subscribers: compnerd, cfe-commits
Differential Revision: http://reviews.llvm.org/D12034
llvm-svn: 245076
This patch fixes the x86 implementation of allowsMisalignedMemoryAccess() to correctly
return the 'Fast' output parameter for 32-byte accesses. To test that, an existing load
merging optimization is changed to use the TLI hook. This exposes a shortcoming in the
current logic and results in the regression test update. Changing other direct users of
the isUnalignedMem32Slow() x86 CPU attribute would be a follow-on patch.
Without the fix in allowsMisalignedMemoryAccesses(), we will infinite loop when targeting
SandyBridge because LowerINSERT_SUBVECTOR() creates 32-byte loads from two 16-byte loads
while PerformLOADCombine() splits them back into 16-byte loads.
Differential Revision: http://reviews.llvm.org/D10662
llvm-svn: 245075
Summary:
This involved various fixes:
- Move a test that uses ulimit to Posix.
- Add a few "REQUIRES: shell" lines to tests using backtick subshell
evaluation.
- The MSVC CRT buffers stdio if the output is a pipe by default. Some
tests need that disabled to avoid interleaving test stdio with asan
output.
- MSVC headers provide _alloca instead of alloca (go figure), so add a
portability macro to the two alloca tests.
- XFAIL tests that rely on accurate symbols, we need to pass more flags
to make that work.
- MSVC's printf implementation of %p uses upper case letters and doesn't
add 0x, so do that manually.
- Accept "SEGV" or "access-violation" reports in crash tests.
Reviewers: samsonov
Subscribers: tberghammer, danalbert, llvm-commits, srhines
Differential Revision: http://reviews.llvm.org/D12019
llvm-svn: 245073
JF Bastien