manager. I cannot emphasize enough that this is a WIP. =] I expect it
to change a great deal as things stabilize, but I think its really
important to get *some* functionality here so that the infrastructure
can be tested more traditionally from the commandline.
The current design is looking something like this:
./bin/opt -passes='module(pass_a,pass_b,function(pass_c,pass_d))'
So rather than custom-parsed flags, there is a single flag with a string
argument that is parsed into the pass pipeline structure. This makes it
really easy to have nice structural properties that are very explicit.
There is one obvious and important shortcut. You can start off the
pipeline with a pass, and the minimal context of pass managers will be
built around the entire specified pipeline. This makes the common case
for tests super easy:
./bin/opt -passes=instcombine,sroa,gvn
But this won't introduce any of the complexity of the fully inferred old
system -- we only ever do this for the *entire* argument, and we only
look at the first pass. If the other passes don't fit in the pass
manager selected it is a hard error.
The other interesting aspect here is that I'm not relying on any
registration facilities. Such facilities may be unavoidable for
supporting plugins, but I have alternative ideas for plugins that I'd
like to try first. My plan is essentially to build everything without
registration until we hit an absolute requirement.
Instead of registration of pass names, there will be a library dedicated
to parsing pass names and the pass pipeline strings described above.
Currently, this is directly embedded into opt for simplicity as it is
very early, but I plan to eventually pull this into a library that opt,
bugpoint, and even Clang can depend on. It should end up as a good home
for things like the existing PassManagerBuilder as well.
There are a bunch of FIXMEs in the code for the parts of this that are
just stubbed out to make the patch more incremental. A quick list of
what's coming up directly after this:
- Support for function passes and building the structured nesting.
- Support for printing the pass structure, and FileCheck tests of all of
this code.
- The .def-file based pass name parsing.
- IR priting passes and the corresponding tests.
Some obvious things that I'm not going to do right now, but am
definitely planning on as the pass manager work gets a bit further:
- Pull the parsing into library, including the builders.
- Thread the rest of the target stuff into the new pass manager.
- Wire support for the new pass manager up to llc.
- Plugin support.
Some things that I'd like to have, but are significantly lower on my
priority list. I'll get to these eventually, but they may also be places
where others want to contribute:
- Adding nice error reporting for broken pass pipeline descriptions.
- Typo-correction for pass names.
llvm-svn: 198998
Refactor the parser so that the parser can return arbitrary type of parse
result other than a vector of ExportDesc. Parsers for non-EXPORTS directives
will be implemented in different patches. No functionality change.
llvm-svn: 198993
This patch more cleanly seperates the concepts of Preferred Alignment
and Required Alignment. Most notable that changes to Required Alignment
do *not* impact preferred alignment until late in struct layout. This
is observable when using pragma pack and non-virtual bases and the use
of tail padding when laying them out.
Test cases included.
llvm-svn: 198988
The main goal of this patch is to allow "mach-o encoded as yaml" and "native
encoded as yaml" documents to be intermixed. They are distinguished via
yaml tags at the start of the document. This will enable all mach-o test cases
to be written using yaml instead of checking in object files.
The Registry was extend to allow yaml tag handlers to be registered. The
mach-o Reader adds a yaml tag handler for the tag "!mach-o".
Additionally, this patch fixes some buffer ownership issues. When parsing
mach-o binaries, the mach-o atoms can have pointers back into the memory
mapped .o file. But with yaml encoded mach-o, name and content are ephemeral,
so a copyRefs parameter was added to cause the mach-o atoms to make their
own copy.
llvm-svn: 198986
Use separate callee-save masks for XMM and YMM registers for anyregcc on X86 and
select the proper mask depending on the target cpu we compile for.
llvm-svn: 198985
rules: instead of requiring flexible array members to be POD, require them to
be trivially-destructible. This seems to be the only constraint that actually
matters here (and even then, it's questionable whether this matters).
llvm-svn: 198983
<rdar://problem/15797390>
This new test case will detect this and make sure we don't regress on global name lookups that search all DWARF for everything when we don't need to.
llvm-svn: 198982
symbols correctly. There were a couple of pieces to this.
1) When a breakpoint location finds itself pointing to an Indirect symbol, when the site for it is created
it needs to resolve the symbol and actually set the site at its target.
2) Not all breakpoints want to do this (i.e. a straight address breakpoint should always set itself on the
specified address, so somem machinery was needed to specify that.
3) I added some info to the break list output for indirect symbols so you could see what was happening.
Also I made it clear when we re-route through re-exported symbols.
4) I moved ResolveIndirectFunction from ProcessPosix to Process since it works the exact same way on Mac OS X
and the other posix systems. If we find a platform that doesn't do it this way, they can override the
call in Process.
5) Fixed one bug in RunThreadPlan, if you were trying to run a thread plan after a "running" event had
been broadcast, the event coalescing would cause you to miss the ThreadPlan running event. So I added
a way to override the coalescing.
6) Made DynamicLoaderMacOSXDYLD::GetStepThroughTrampolinePlan handle Indirect & Re-exported symbols.
<rdar://problem/15280639>
llvm-svn: 198976
The presence of a VBPtr suppresses the presence of zero sized
sub-objects in the non-virtual portion of the object in the context of
determining if two base objects need alias-avoidance padding placed
between them.
Test cases included.
llvm-svn: 198975
1- Use the line_iterator class to read profile files.
2- Allow comments in profile file. Lines starting with '#'
are completely ignored while reading the profile.
3- Add parsing support for discriminators and indirect call samples.
Our external profiler can emit more profile information that we are
currently not handling. This patch does not add new functionality to
support this information, but it allows profile files to provide it.
I will add actual support later on (for at least one of these
features, I need support for DWARF discriminators in Clang).
A sample line may contain the following additional information:
Discriminator. This is used if the sampled program was compiled with
DWARF discriminator support
(http://wiki.dwarfstd.org/index.php?title=Path_Discriminators). This
is currently only emitted by GCC and we just ignore it.
Potential call targets and samples. If present, this line contains a
call instruction. This models both direct and indirect calls. Each
called target is listed together with the number of samples. For
example,
130: 7 foo:3 bar:2 baz:7
The above means that at relative line offset 130 there is a call
instruction that calls one of foo(), bar() and baz(). With baz()
being the relatively more frequent call target.
Differential Revision: http://llvm-reviews.chandlerc.com/D2355
4- Simplify format of profile input file.
This implements earlier suggestions to simplify the format of the
sample profile file. The symbol table is not necessary and function
profiles do not need to know the number of samples in advance.
Differential Revision: http://llvm-reviews.chandlerc.com/D2419
llvm-svn: 198973
This adds a propagation heuristic to convert instruction samples
into branch weights. It implements a similar heuristic to the one
implemented by Dehao Chen on GCC.
The propagation proceeds in 3 phases:
1- Assignment of block weights. All the basic blocks in the function
are initial assigned the same weight as their most frequently
executed instruction.
2- Creation of equivalence classes. Since samples may be missing from
blocks, we can fill in the gaps by setting the weights of all the
blocks in the same equivalence class to the same weight. To compute
the concept of equivalence, we use dominance and loop information.
Two blocks B1 and B2 are in the same equivalence class if B1
dominates B2, B2 post-dominates B1 and both are in the same loop.
3- Propagation of block weights into edges. This uses a simple
propagation heuristic. The following rules are applied to every
block B in the CFG:
- If B has a single predecessor/successor, then the weight
of that edge is the weight of the block.
- If all the edges are known except one, and the weight of the
block is already known, the weight of the unknown edge will
be the weight of the block minus the sum of all the known
edges. If the sum of all the known edges is larger than B's weight,
we set the unknown edge weight to zero.
- If there is a self-referential edge, and the weight of the block is
known, the weight for that edge is set to the weight of the block
minus the weight of the other incoming edges to that block (if
known).
Since this propagation is not guaranteed to finalize for every CFG, we
only allow it to proceed for a limited number of iterations (controlled
by -sample-profile-max-propagate-iterations). It currently uses the same
GCC default of 100.
Before propagation starts, the pass builds (for each block) a list of
unique predecessors and successors. This is necessary to handle
identical edges in multiway branches. Since we visit all blocks and all
edges of the CFG, it is cleaner to build these lists once at the start
of the pass.
Finally, the patch fixes the computation of relative line locations.
The profiler emits lines relative to the function header. To discover
it, we traverse the compilation unit looking for the subprogram
corresponding to the function. The line number of that subprogram is the
line where the function begins. That becomes line zero for all the
relative locations.
llvm-svn: 198972
...by synthesizing their body to be "return self->_prop;", with an extra
nudge to RetainCountChecker to still treat the value as +0 if we have no
other information.
This doesn't handle weak properties, but that's mostly correct anyway,
since they can go to nil at any time. This also doesn't apply to properties
whose implementations we can't see, since they may not be backed by an
ivar at all. And finally, this doesn't handle properties of C++ class type,
because we can't invoke the copy constructor. (Sema has actually done this
work already, but the AST it synthesizes is one the analyzer doesn't quite
handle -- it has an rvalue DeclRefExpr.)
Modeling setters is likely to be more difficult (since it requires
handling strong/copy), but not impossible.
<rdar://problem/11956898>
llvm-svn: 198953
which may belong to unrelated classes. It was
primarily intended for miuse of @selector expression.
But warning is too noisy and will be issued when
an actual @selector is used. // rdar://15740134
llvm-svn: 198952
for (i = 0; i < N; ++i)
A[i * Stride1] += B[i * Stride2];
We take loops like this and check that the symbolic strides 'Strided1/2' are one
and drop to the scalar loop if they are not.
This is currently disabled by default and hidden behind the flag
'enable-mem-access-versioning'.
radar://13075509
llvm-svn: 198950
An upcoming loop vectorizer commit will want to replace a SCEVUnknown(Value*)
by a SCEVConstant. This commit modifies the SCEVParameterRewriter to support
this. The SCEVParameterRewriter constructor can optionally specify to follow
this behavior.
llvm-svn: 198949
