[PM] Simplify how we parse the outer layer of the pass pipeline text and
remove an extra, redundant pass manager wrapping every run.
I had kept seeing these when manually testing, but it was getting really
annoying and was going to cause problems with overly eager invalidation.
The root cause was an overly complex and unnecessary pile of code for
parsing the outer layer of the pass pipeline. We can instead delegate
most of this to the recursive pipeline parsing.
I've added some somewhat more basic and precise tests to catch this.
Craig Topper [Tue, 6 Jan 2015 07:35:50 +0000 (07:35 +0000)]
[X86] Make isel select the 2-byte register form of INC/DEC even in non-64-bit mode. Convert to the 1-byte form in non-64-bit mode as part of MCInst lowering.
Overall this seems simpler. It reduces duplication of patterns between both modes and it simplifies the memory folding/unfolding tables as they don't need to create fake instructions just to keep track of 64-bitness.
The old target DAG combine that allowed for performing int_to_fp(fp_to_int(x))
without a load/store pair is updated here with support for unsigned integers,
and to support single-precision values without a third rounding step, on newer
cores with the appropriate instructions.
[PM] Add a utility pass template that synthesizes the invalidation of
a specific analysis result.
This is quite handy to test things, and will also likely be very useful
for debugging issues. You could narrow down pass validation failures by
walking these invalidate pass runs up and down the pass pipeline, etc.
I've added support to the pass pipeline parsing to be able to create one
of these for any analysis pass desired.
Just adding this class uncovered one latent bug where the
AnalysisManager CRTP base class had a hard-coded Module type rather than
using IRUnitT.
I've also added tests for invalidation and caching of analyses in
a basic way across all the pass managers. These in turn uncovered two
more bugs where we failed to correctly invalidate an analysis -- its
results were invalidated but the key for re-running the pass was never
cleared and so it was never re-run. Quite nasty. I'm very glad to debug
this here rather than with a full system.
Also, yes, the naming here is horrid. I'm going to update some of the
names to be slightly less awful shortly. But really, I've no "good"
ideas for naming. I'll be satisfied if I can get it to "not bad".
[PM] Simplify how we use the registry by including it only once. Still
more verbose than I'd like, but the code really isn't that interesting,
and this still seems vastly simpler than any other solutions I've come
up with. =] Maybe if we get to the 10th IR unit, this will be a problem
in practice.
Craig Topper [Tue, 6 Jan 2015 04:23:57 +0000 (04:23 +0000)]
[X86] Remove 16-bit and 32-bit offset jump instructions from the AsmParser. We always select the 8-bit size and let the assembler backend relax to the larger size.
Craig Topper [Tue, 6 Jan 2015 04:23:53 +0000 (04:23 +0000)]
[X86] Make isel select the shorter form of jump instructions instead of the long form.
The assembler backend will relax to the long form if necessary. This removes a swap from long form to short form in the MCInstLowering code. Selecting the long form used to be required by the old JIT.
[PM] Sink the no-op pass parsing logic into the .def-based registry to
simplify things. This will become more important as I add no-op analyses
that want to re-use the logic we already have for analyses in the
registry. For now, no functionality changed.
[PM] Move the analysis registry into the Passes.cpp file and provide
a normal interface for it in Passes.h.
This gives us essentially a single interface for running pass managers
which are provided from the bottom of the LLVM stack through interfaces
at the top of the LLVM stack that populate them with all of the
different analyses available throughout. It also means there is a single
blob of code that needs to include all of the pass headers and needs to
deal with the registry of passes and parsing names.
No functionality changed intended, should just be cleanup.
[PM] Add a utility to the new pass manager for generating a pass which
is a no-op other than requiring some analysis results be available.
This can be used in real pass pipelines to force the usually lazy
analysis running to eagerly compute something at a specific point, and
it can be used to test the pass manager infrastructure (my primary use
at the moment).
I've also added bit of pipeline parsing magic to support generating
these directly from the opt command so that you can directly use these
when debugging your analysis. The syntax is:
require<analysis-name>
This can be used at any level of the pass manager. For example:
This would produce a no-op function pass requiring my-analysis, followed
by a fully no-op function pass, both of these in a function pass manager
which is nested inside of a bottom-up CGSCC pass manager which is in the
top-level (implicit) module pass manager.
I have zero attachment to the particular syntax I'm using here. Consider
it a straw man for use while I'm testing and fleshing things out.
Suggestions for better syntax welcome, and I'll update everything based
on any consensus that develops.
I've used this new functionality to more directly test the analysis
printing rather than relying on the cgscc pass manager running an
analysis for me. This is still minimally tested because I need to have
analyses to run first! ;] That patch is next, but wanted to keep this
one separate for easier review and discussion.
Now that `LLVMContextImpl` can call `MDNode::dropAllReferences()` to
prevent teardown madness, stop dropping uniquing just because an operand
drops to null.
Frederic Riss [Mon, 5 Jan 2015 21:29:41 +0000 (21:29 +0000)]
Make DIE.h a public CodeGen header.
dsymutil would like to use all the AsmPrinter/MCStreamer infrastructure
to stream out the DWARF. In order to do so, it will reuse the DIE object
and so this header needs to be public.
The interface exposed here has some corners that cannot be used without a
DwarfDebug object, but clients that want to stream Dwarf can just avoid
these.
Frederic Riss [Mon, 5 Jan 2015 21:29:28 +0000 (21:29 +0000)]
[dsymutil] Implement the BinaryHolder object and gain archive support.
This object is meant to own the ObjectFiles and their underlying
MemoryBuffer. It is basically the equivalent of an OwningBinary
except that it efficiently handles Archives. It is optimized for
efficiently providing mappings of members of the same archive when
they are opened successively (which is standard in Darwin debug
maps, objects from the same archive will be contiguous).
Of course, the BinaryHolder will also be used by the DWARF linker
once it is commited, but for now only the debug map parser uses it.
With this change, you can run llvm-dsymutil on your Darwin debug build
of clang and get a complete debug map for it.
Hal Finkel [Mon, 5 Jan 2015 21:10:24 +0000 (21:10 +0000)]
[PowerPC] Fold i1 extensions with other ops
Consider this function from our README.txt file:
int foo(int a, int b) { return (a < b) << 4; }
We now explicitly track CR bits by default, so the comment in the README.txt
about not really having a SETCC is no longer accurate, but we did generate this
somewhat silly code:
cmpw 0, 3, 4
li 3, 0
li 12, 1
isel 3, 12, 3, 0
sldi 3, 3, 4
blr
which generates the zext as a select between 0 and 1, and then shifts the
result by a constant amount. Here we preprocess the DAG in order to fold the
results of operations on an extension of an i1 value into the SELECT_I[48]
pseudo instruction when the resulting constant can be materialized using one
instruction (just like the 0 and 1). This was not implemented as a DAGCombine
because the resulting code would have been anti-canonical and depends on
replacing chained user nodes, which does not fit well into the lowering
paradigm. Now we generate:
cmpw 0, 3, 4
li 3, 0
li 12, 16
isel 3, 12, 3, 0
blr
Hal Finkel [Mon, 5 Jan 2015 18:52:29 +0000 (18:52 +0000)]
[PowerPC] Remove zexts after i32 ctlz
The 64-bit semantics of cntlzw are not special, the 32-bit population count is
stored as a 64-bit value in the range [0,32]. As a result, it is always zero
extended, and it can be added to the PPCISelDAGToDAG peephole optimization as a
frontier instruction for the removal of unnecessary zero extensions.
Hal Finkel [Mon, 5 Jan 2015 18:09:06 +0000 (18:09 +0000)]
[PowerPC] Remove zexts after byte-swapping loads
lhbrx and lwbrx not only load their data with byte swapping, but also clear the
upper 32 bits (at least). As a result, they can be added to the PPCISelDAGToDAG
peephole optimization as frontier instructions for the removal of unnecessary
zero extensions.
The swap implementation for iplist is currently unsupported. Simply splice the
old list into place, which achieves the same purpose. This is needed in order
to thread the -frewrite-map-file frontend option correctly. NFC.
Karthik Bhat [Mon, 5 Jan 2015 13:57:59 +0000 (13:57 +0000)]
Select lower fsub,fabs pattern to fabd on AArch64
This patch lowers patterns such as-
fsub v0.4s, v0.4s, v1.4s
fabs v0.4s, v0.4s
to
fabd v0.4s, v0.4s, v1.4s
on AArch64.
Charlie Turner [Mon, 5 Jan 2015 13:12:17 +0000 (13:12 +0000)]
Emit the build attribute Tag_conformance.
Claim conformance to version 2.09 of the ARM ABI.
This build attribute must be emitted first amongst the build attributes when
written to an object file. This is to simplify conformance detection by
consumers.
[PM] Don't run the machinery of invalidating all the analysis passes
when all are being preserved.
We want to short-circuit this for a couple of reasons. One, I don't
really want passes to grow a dependency on actually receiving their
invalidate call when they've been preserved. I'm thinking about removing
this entirely. But more importantly, preserving everything is likely to
be the common case in a lot of scenarios, and it would be really good to
bypass all of the invalidation and preservation machinery there.
Avoiding calling N opaque functions to try to invalidate things that are
by definition still valid seems important. =]
This wasn't really inpsired by much other than seeing the spam in the
logging for analyses, but it seems better ot get it checked in rather
than forgetting about it.
[PM] Add names and debug logging for analysis passes to the new pass
manager.
This starts to allow us to test analyses more easily, but it's really
only the beginning. Some of the code here is still untestable without
manual changes to create analysis passes, but I wanted to factor it into
a small of chunks as possible.
Next up in order to be able to test things are, in no particular order:
- No-op analyses passes so we don't have to use real ones to exercise
the pass maneger itself.
- Automatic way of generating dummy passes that require an analysis be
run, including a variant that calls a 'print' method on a pass to make
it even easier to print out the results of an analysis.
- Dummy passes that invalidate all analyses for their IR unit so we can
test invalidation and re-runs.
- Automatic way to print each analysis pass as it is re-run.
- Automatic but optional verification of analysis passes everywhere
possible.
I'm not claiming I'll get to all of these immediately, but that's what
is in the pipeline at some stage. I'm fleshing out exactly what I need
and what to prioritize by working on converting analyses and then trying
to test the conversion. =]
The algorithm tries to collect memory access candidates from AliasSetTracker, and then check memory dependences one another. The memory accesses are unique in AliasSetTracker, and a single memory access in AliasSetTracker may map to multiple entries in AccessAnalysis, which could cover both 'read' and 'write'. Originally the algorithm only checked 'write' entry in Accesses if only 'write' exists. This is incorrect and the consequence is it ignored all read access, and finally some RAW and WAR dependence are missed.
For the case given above, if we ignore two reads, the dependence between (1) and (3) would not be able to be captured, and finally this loop will be incorrectly vectorized.
The fix simply inserts a new loop to find all entries in Accesses. Since it will skip most of all other memory accesses by checking the Value pointer at the very beginning of the loop, it should not increase compile-time visibly.
Craig Topper [Mon, 5 Jan 2015 08:19:12 +0000 (08:19 +0000)]
[X86] Remove the predicates from the register forms of the 2-byte inc and dec instructions. Remove the 32-bit mode only versions that existed for the disassembler. Move the patterns out of the instructions so they can still be qualified with predicates.
Craig Topper [Mon, 5 Jan 2015 08:19:05 +0000 (08:19 +0000)]
[X86] Remove a strange fixme referring to a hack that doesn't seem to exist since the code is in a comment. Can't figure out what the body of the 'if' was supposed to be anyway.
Hal Finkel [Mon, 5 Jan 2015 05:24:42 +0000 (05:24 +0000)]
[PowerPC] Enable speculation of cttz/ctlz
PPC has an instruction for ctlz with defined zero behavior, and our lowering of
cttz (provided by DAGCombine) is also efficient and branchless, so speculating
these makes sense.
[SROA] Apply a somewhat heavy and unpleasant hammer to fix PR22093, an
assert out of the new pre-splitting in SROA.
This fix makes the code do what was originally intended -- when we have
a store of a load both dealing in the same alloca, we force them to both
be pre-split with identical offsets. This is really quite hard to do
because we can keep discovering problems as we go along. We have to
track every load over the current alloca which for any resaon becomes
invalid for pre-splitting, and go back to remove all stores of those
loads. I've included a couple of test cases derived from PR22093 that
cover the different ways this can happen. While that PR only really
triggered the first of these two, its the same fundamental issue.
The other challenge here is documented in a FIXME now. We end up being
quite a bit more aggressive for pre-splitting when loads and stores
don't refer to the same alloca. This aggressiveness comes at the cost of
introducing potentially redundant loads. It isn't clear that this is the
right balance. It might be considerably better to require that we only
do pre-splitting when we can presplit every load and store involved in
the entire operation. That would give more consistent if conservative
results. Unfortunately, it requires a non-trivial change to the actual
pre-splitting operation in order to correctly handle cases where we end
up pre-splitting stores out-of-order. And it isn't 100% clear that this
is the right direction, although I'm starting to suspect that it is.
Hal Finkel [Mon, 5 Jan 2015 03:41:38 +0000 (03:41 +0000)]
[PowerPC] Materialize i64 constants using rotation with masking
r225135 added the ability to materialize i64 constants using rotations in order
to reduce the instruction count. Sometimes we can use a rotation only with some
extra masking, so that we take advantage of the fact that generating a bunch of
extra higher-order 1 bits is easy using li/lis.
[PM] Switch the new pass manager to use a reference-based API for IR
units.
This was debated back and forth a bunch, but using references is now
clearly cleaner. Of all the code written using pointers thus far, in
only one place did it really make more sense to have a pointer. In most
cases, this just removes immediate dereferencing from the code. I think
it is much better to get errors on null IR units earlier, potentially
at compile time, than to delay it.
Most notably, the legacy pass manager uses references for its routines
and so as more and more code works with both, the use of pointers was
likely to become really annoying. I noticed this when I ported the
domtree analysis over and wrote the entire thing with references only to
have it fail to compile. =/ It seemed better to switch now than to
delay. We can, of course, revisit this is we learn that references are
really problematic in the API.
[PM] Wire up support for explicitly running the verifier pass.
The required functionality has been there for some time, but I never
managed to actually wire it into the command line registry of passes.
Let's do that.
Hal Finkel [Sun, 4 Jan 2015 15:43:55 +0000 (15:43 +0000)]
[PowerPC] Materialize i64 constants using rotation
Materializing full 64-bit constants on PPC64 can be expensive, requiring up to
5 instructions depending on the locations of the non-zero bits. Sometimes
materializing a rotated constant, and then applying the inverse rotation, requires
fewer instructions than the direct method. If so, do that instead.
In r225132, I added support for forming constants using bit inversion. In
effect, this reverts that commit and replaces it with rotation support. The bit
inversion is useful for turning constants that are mostly ones into ones that
are mostly zeros (thus enabling a more-efficient shift-based materialization),
but the same effect can be obtained by using negative constants and a rotate,
and that is at least as efficient, if not more.
Hal Finkel [Sun, 4 Jan 2015 12:35:03 +0000 (12:35 +0000)]
[PowerPC] Materialize i64 constants using bit inversion
Materializing full 64-bit constants on PPC64 can be expensive, requiring up to
5 instructions depending on the locations of the non-zero bits. Sometimes
materializing the bit-reversed constant, and then flipping the bits, requires
fewer instructions than the direct method. If so, do that instead.
[PM] Split the AssumptionTracker immutable pass into two separate APIs:
a cache of assumptions for a single function, and an immutable pass that
manages those caches.
The motivation for this change is two fold. Immutable analyses are
really hacks around the current pass manager design and don't exist in
the new design. This is usually OK, but it requires that the core logic
of an immutable pass be reasonably partitioned off from the pass logic.
This change does precisely that. As a consequence it also paves the way
for the *many* utility functions that deal in the assumptions to live in
both pass manager worlds by creating an separate non-pass object with
its own independent API that they all rely on. Now, the only bits of the
system that deal with the actual pass mechanics are those that actually
need to deal with the pass mechanics.
Once this separation is made, several simplifications become pretty
obvious in the assumption cache itself. Rather than using a set and
callback value handles, it can just be a vector of weak value handles.
The callers can easily skip the handles that are null, and eventually we
can wrap all of this up behind a filter iterator.
For now, this adds boiler plate to the various passes, but this kind of
boiler plate will end up making it possible to port these passes to the
new pass manager, and so it will end up factored away pretty reasonably.
David Majnemer [Sun, 4 Jan 2015 07:36:02 +0000 (07:36 +0000)]
InstCombine: match can find ConstantExprs, don't assume we have a Value
We assumed the output of a match was a Value, this would cause us to
assert because we would fail a cast<>. Instead, use a helper in the
Operator family to hide the distinction between Value and Constant.
Lang Hames [Sun, 4 Jan 2015 01:20:55 +0000 (01:20 +0000)]
[APFloat][ADT] Fix sign handling logic for FMA results that truncate to zero.
This patch adds a check for underflow when truncating results back to lower
precision at the end of an FMA. The additional sign handling logic in
APFloat::fusedMultiplyAdd should only be performed when the result of the
addition step of the FMA (in full precision) is exactly zero, not when the
result underflows to zero.
Unit tests for this case and related signed zero FMA results are included.
Weak externals are resolved statically, so we can actually generate the tail
call on PE/COFF targets without breaking the requirements. It is questionable
whether we want to propagate the current behaviour for MachO as the requirements
are part of the ARM ELF specifications, and it seems that prior to the SVN
r215890, we would have tail'ed the call. For now, be conservative and only
permit it on PE/COFF where the call will always be fully resolved.
Hal Finkel [Sat, 3 Jan 2015 17:58:24 +0000 (17:58 +0000)]
[PowerPC/BlockPlacement] Allow target to provide a per-loop alignment preference
The existing code provided for specifying a global loop alignment preference.
However, the preferred loop alignment might depend on the loop itself. For
recent POWER cores, loops between 5 and 8 instructions should have 32-byte
alignment (while the others are better with 16-byte alignment) so that the
entire loop will fit in one i-cache line.
To support this, getPrefLoopAlignment has been made virtual, and can be
provided with an optional MachineLoop* so the target can inspect the loop
before answering the query. The default behavior, as before, is to return the
value set with setPrefLoopAlignment. MachineBlockPlacement now queries the
target for each loop instead of only once per function. There should be no
functional change for other targets.
Hal Finkel [Sat, 3 Jan 2015 14:58:25 +0000 (14:58 +0000)]
[PowerPC] Use 16-byte alignment for modern cores for functions/loops
Most modern PowerPC cores prefer that functions and loops start on
16-byte-aligned boundaries (*), so instruct block placement, etc. to make this
happen. The branch selector has also been adjusted so account for the extra
nops that might now be inserted before loop headers.
(*) Some cores actually prefer other alignments for small loops, but that will
be addressed in a follow-up commit.
Craig Topper [Sat, 3 Jan 2015 08:16:34 +0000 (08:16 +0000)]
Minor cleanup to all the switches after MatchInstructionImpl in all the AsmParsers.
Make sure they all have llvm_unreachable on the default path out of the switch. Remove unnecessary "default: break". Remove a 'return' after unreachable. Fix some indentation.
David Majnemer [Sat, 3 Jan 2015 02:33:25 +0000 (02:33 +0000)]
ValueTracking: Make computeKnownBits for Arguments a little more clear
We would sometimes leave the out-param APInts untouched while going
through computeKnownBits. While I don't know of a way to trigger a bug
involving this in practice, it goes against the overall design of
computeKnownBits.
Hal Finkel [Sat, 3 Jan 2015 01:16:37 +0000 (01:16 +0000)]
[PowerPC] Add support for the CMPB instruction
Newer POWER cores, and the A2, support the cmpb instruction. This instruction
compares its operands, treating each of the 8 bytes in the GPRs separately,
returning a 'mask' result of 0 (for false) or -1 (for true) in each byte.
Code generation support is added, in the form of a PPCISelDAGToDAG
DAG-preprocessing routine, that recognizes patterns close to what the
instruction computes (either exactly, or related by a constant masking
operation), and generates the cmpb instruction (along with any necessary
constant masking operation). This can be expanded if use cases arise.
[PM] Add proper documentation for the ModulePassManager and
FunctionPassManager. These never got documented, likely due to the
clutter of this header file. This fixes another problem people noticed
when they started trying to use the new pass manager.
I've also used this to document the aspirational constraints I would
like to hold passes to. I don't really have a better place to document
such things at this point, but eventually will probably create a proper
.rst file and page for the LLVM pass infrastructure that carries such
high-level concerns.
[PM] Lift the majority of the template boilerplate used to implement the
concept-based polymorphism in the pass manager to a separate header.
I got feedback from someone reading the code and trying to use it that
this was really making it hard to dive in and start using these APIs and
that makes a lot of sense.
This only requires a moderate amount of gymnastics to separate in this
way, namely rinsing the PreservedAnalysis object through a template
argument in a few places so that it is dependent and we only examine it
on instantiation.
projects / llvm / log