Silviu Baranga [Mon, 2 Nov 2015 14:41:02 +0000 (14:41 +0000)]
[SCEV][LV] Add SCEV Predicates and use them to re-implement stride versioning
Summary:
SCEV Predicates represent conditions that typically cannot be derived from
static analysis, but can be used to reduce SCEV expressions to forms which are
usable for different optimizers.
ScalarEvolution now has the rewriteUsingPredicate method which can simplify a
SCEV expression using a SCEVPredicateSet. The normal workflow of a pass using
SCEVPredicates would be to hold a SCEVPredicateSet and every time assumptions
need to be made a new SCEV Predicate would be created and added to the set.
Each time after calling getSCEV, the user will call the rewriteUsingPredicate
method.
We add two types of predicates
SCEVPredicateSet - implements a set of predicates
SCEVEqualPredicate - tests for equality between two SCEV expressions
We use the SCEVEqualPredicate to re-implement stride versioning. Every time we
version a stride, we will add a SCEVEqualPredicate to the context.
Instead of adding specific stride checks, LoopVectorize now adds a more
generic SCEV check.
We only need to add support for this in the LoopVectorizer since this is the
only pass that will do stride versioning.
This revision has introduced an issue that only affects bootstrapped compiler
when it is printing the ASM. It turns out that the new code path taken due to
legalizing a scalar_to_vector of i64 -> v2i64 exposes a missing check in a
micro optimization to change a load followed by a scalar_to_vector into a
load and splat instruction on PPC.
Craig Topper [Mon, 2 Nov 2015 07:24:37 +0000 (07:24 +0000)]
[X86] Don't pass a scale value of 0 to scatter/gather intrinsics. This causes the code emitter to throw an assertion if we try to encode it. Need to add a check to fail isel for this, but for now avoid testing it.
[PGO] Value profiling (index format) code cleanup and testing
1. Added a set of public interfaces in InstrProfRecord
class to access (read/write) value profile data.
2. Changed IndexedProfile reader and writer code to
use the newly defined interfaces and hide implementation
details.
3. Added a couple of unittests for value profiling:
- Test new interfaces to get and set value profile data
- Test value profile data merging with various scenarios.
No functional change is expected. The new interfaces will also
make it possible to change on-disk format of value prof data
to be more compact (to be submitted).
Sanjoy Das [Mon, 2 Nov 2015 02:06:01 +0000 (02:06 +0000)]
[SCEV] Fix PR25369
Have `getConstantEvolutionLoopExitValue` work correctly with multiple
entry loops.
As far as I can tell, `getConstantEvolutionLoopExitValue` never did the
right thing for multiple entry loops; and before r249712 it would
silently return an incorrect answer. r249712 changed SCEV to fail an
assert on a multiple entry loop, and this change fixes the underlying
issue.
AVX-512: Optimized SIMD truncate operations for AVX512F set.
Optimized <8 x i32> to <8 x i16>
<4 x i64> to < 4 x i32>
<16 x i16> to <16 x i8>
All these oprtrations use now AVX512F set (KNL). Before this change it was implemented with AVX2 set.
This adds support for COFF I386. This is sufficient for code execution in a
32-bit JIT, though, imported symbols need to custom lowered for the redirection.
Sanjoy Das [Sat, 31 Oct 2015 23:21:40 +0000 (23:21 +0000)]
[SCEV] Don't create SCEV expressions that break LCSSA
Prevent `createNodeFromSelectLikePHI` from creating SCEV expressions
that break LCSSA.
A better fix for the same issue is to teach SCEVExpander to not break
LCSSA by inserting PHI nodes at appropriate places. That's planned for
the future.
Rafael Espindola [Sat, 31 Oct 2015 21:44:42 +0000 (21:44 +0000)]
Don't store a Child to the first regular member.
This is a bit ugly, but has a few advantages:
* Archive is now easy to copy since there is no Archive -> Child -> Archive
loop.
* It makes it clear that we already checked for errors when finding the Child
data.
Davide Italiano [Sat, 31 Oct 2015 20:59:32 +0000 (20:59 +0000)]
[SimplifyLibCalls] Add test to ensure transform is not executed if fast-math
attribute is not present.
During my refactor in r251595 I changed the behavior of optimizeSqrt(),
skipping the transformation if the function wasn't marked with unsafe-fp-math
attribute. This fixed a bug, as confirmed by Sanjay (before the optimization
was silently executed anyway), although it wasn't my primary aim.
This commit adds a test to ensure the code doesn't break again.
Craig Topper [Sat, 31 Oct 2015 17:27:52 +0000 (17:27 +0000)]
[X86] Reduce math for index calculation for inserting and extracting subvectors and elements by exploiting the fact that all supported vector types have a power 2 number of elements.
Add a unittest for SmallDenseMap that tests assigning a SmallDenseMap when it is not small.
This complements CopyConstructorNotSmallTest. If we are testing the copy
constructor in such a way, we should also probably test assignment in the same
way.
Chandler Carruth [Fri, 30 Oct 2015 16:48:08 +0000 (16:48 +0000)]
[FunctionAttrs] Separate another chunk of the logic for functionattrs
from its pass harness by providing a lambda to query for AA results.
This allows the legacy pass to easily provide a lambda that uses the
special helpers to construct function AA results from a legacy CGSCC
pass. With the new pass manager (the next patch) the lambda just
directly wraps the intuitive query API.
Silviu Baranga [Fri, 30 Oct 2015 15:02:28 +0000 (15:02 +0000)]
[SCEV] Generalize the SCEV algorithm for creating expressions for PHI nodes
Summary:
When forming expressions for phi nodes having an incoming value from
outside the loop A and a value coming from the previous iteration B
we were forming an AddRec if:
- B was an AddRec
- the value A was equal to the value for B at iteration -1 (or equal
to the value of B shifted by one iteration, at iteration 0)
In this case, we were computing the expression to be the expression of
B, shifted by one iteration.
This changes generalizes the logic above by removing the restriction that
B needs to be an AddRec. For this we introduce two expression rewriters
that allow us to
- shift an expression by one iteration
- get the value of an expression at iteration 0
This allows us to get SCEV expressions for PHI nodes when these expressions
are not AddRecExprs.
Dehao Chen [Fri, 30 Oct 2015 05:07:15 +0000 (05:07 +0000)]
Recommit r251680 (also need to update clang test)
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Dehao Chen [Fri, 30 Oct 2015 04:29:05 +0000 (04:29 +0000)]
Revert r251680:
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Dehao Chen [Fri, 30 Oct 2015 02:38:29 +0000 (02:38 +0000)]
Update the discriminator assignment algorithm
* If a scope has already been assigned a discriminator, do not reassign a nested discriminator for it.
* If the file and line both match, even if the column does not match, we should assign a new discriminator for the stmt.
original code:
; #1 int foo(int i) {
; #2 if (i == 3 || i == 5) return 100; else return 99;
; #3 }
Alexey Samsonov [Fri, 30 Oct 2015 00:40:20 +0000 (00:40 +0000)]
Let the users of LLVMSymbolizer decide whether they want to symbolize inlined frames.
Introduce LLVMSymbolizer::symbolizeInlinedCode() instead of switching
on PrintInlining option passed to the constructor. This will be needed
once we retrun structured data (instead of std::string) from
LLVMSymbolizer and move printing logic out.
Alexey Samsonov [Thu, 29 Oct 2015 22:21:37 +0000 (22:21 +0000)]
[LLVMSymbolize] Move ModuleInfo into a separate class (SymbolizableModule).
Summary:
This is mostly NFC. It is a first step in cleaning up LLVMSymbolize
library. It removes "ModuleInfo" class which bundles together ObjectFile
and its debug info context in favor of:
* abstract SymbolizableModule in public headers;
* SymbolizableObjectFile subclass in implementation.
Additionally, SymbolizableObjectFile is now created via factory, so we
can properly detect object parsing error at this stage instead of keeping
the broken half-parsed object. As a next step, we would be able to
propagate the error all the way back to the library user.
Further improvements might include:
* factoring out the logic of finding appropriate file with debug info
for a given object file, and caching all parsed object files into a
separate class [A].
* factoring out DILineInfo rendering [B].
This would make what is now a heavyweight "LLVMSymbolizer" a relatively
straightforward class, that calls into [A] to turn filepath into a
SymbolizableModule, delegates actual symbolization to concrete SymbolizableModule
implementation, and lets [C] render the result.
Simon Pilgrim [Thu, 29 Oct 2015 22:11:28 +0000 (22:11 +0000)]
[X86][SSE] Shuffle blends with zero
This patch generalizes the zeroing of vector elements with the BLEND instructions. Currently a zero vector will only blend if the shuffled elements are correctly inline, this patch recognises when a vector input is zero (or zeroable) and modifies a local copy of the shuffle mask to support a blend. As a zeroable vector input may not be all zeroes, the zeroable vector is regenerated if necessary.
Lang Hames [Thu, 29 Oct 2015 22:04:22 +0000 (22:04 +0000)]
[Orc] Teach IndirectStubsManager to manage an expandable pool of stubs, rather
than a pre-allocated slab of stubs. Also add a convenience method for creating a
single stub, rather than a whole block a time.
Teresa Johnson [Thu, 29 Oct 2015 21:24:38 +0000 (21:24 +0000)]
Fix test check label.
Summary:
I noticed when manually modifying this test that it was passing when I
expected it to fail. Looks like the combination of LABEL and NOT on the
check does not work. This can also be seen when running FileCheck with
only that one -check-prefix (removing the additional -check-prefix=B):
/usr/local/google/home/tejohnson/llvm/llvm_11_build/./bin/llvm-link -S -internalize -only-needed /usr/local/google/home/tejohnson/llvm/llvm_11_build/test/Linker/Output/link-flags.ll.tmp.b.bc /usr/local/google/home/tejohnson/llvm/llvm_11_build/test/Linker/Output/link-flags.ll.tmp.c.bc | /usr/local/google/home/tejohnson/llvm/llvm_11_build/./bin/FileCheck /usr/local/google/home/tejohnson/llvm/llvm_11/test/Linker/link-flags.ll -check-prefix=CN
error: no check strings found with prefix 'CN:'
The CN prefix checks don't in fact need "LABEL" so remove that.
Chandler Carruth [Thu, 29 Oct 2015 18:29:15 +0000 (18:29 +0000)]
[FunctionAttrs] Provide a single SCC node set to all of the
transformations in FunctionAttrs rather than building a new one each
time.
This isn't trivial because there are different heuristics from different
passes for exactly what set they want. The primary difference is whether
an *overridable* function completely disables the synthesis of
attributes. I've modeled this by directly testing for overridable, and
using the common set that excludes external and opt-none functions.
This does cause some changes by disabling more optimizations in the face
of opt-none. Specifically, we were still optimizing *calls* to opt-none
functions based on their attributes, just not the bodies. It seems
better to be conservative on both fronts given the intended semanticas
here (best effort to not assume or disturb anything). I've not tried to
test this change as it seems complex, brittle, and not important to the
implicit contract of opt-none. Instead, it seems more like a choice that
should be dictated by the simplified implementation and the change to be
acceptable differences within the space of opt-none.
A big benefit here is that these transformations no longer rely on the
legacy pass manager's SCC types, they just work on generic sets of
function pointers. This will make it easy to re-use their logic in the
new pass manager.
I've also made the transforms static functions instead of members where
trivial while I was touching the signatures.
Jonas Paulsson [Thu, 29 Oct 2015 16:13:55 +0000 (16:13 +0000)]
[SystemZ] Make the CCRegs regclass non-allocatable.
This was discovered to be necessary while running memchr-01.ll with
-verify-machinstrs, because it is not allowed to have a phys reg live
accross block boundaries while on SSA form, if the register is
allocatable (expect in entry block and landing pads).
In this test case, stringRRE pseudos are expanded after isel by adding
a loop block which produces a live out CC register. To make the test
pass, it was also necessary to not say that StringRRELoop pseudo uses
R0L, this is only true for the StringRRE opcode.
-verify-machineinstrs added to memchr-01.ll test.
New test case int-cmp-51.ll to test that MachineCSE can eliminate
an identical compare (which it couldn't do before).
Zoran Jovanovic [Thu, 29 Oct 2015 14:40:19 +0000 (14:40 +0000)]
[mips] wrong opcode for ll/sc instructions on mipsr6 when -integrated-as is used
Summary:
This commit resolves wrong opcodes for ll and sc instructions for r6 architecutres, which were generated in method MipsTargetLowering::emitAtomicBinary.
This patch unify the 39-bit and 42-bit mapping for aarch64 to use only
one instrumentation algorithm. This removes compiler flag
SANITIZER_AARCH64_VMA requirement for MSAN on aarch64.
[mips] Check the register class before replacing materializations of zero with $zero in microMIPS.
Summary:
The microMIPS register class GPRMM16 does not contain the $zero register.
However, MipsSEDAGToDAGISel::replaceUsesWithZeroReg() would replace uses
of the $dst register:
[d]addiu, $dst, $zero, 0
with the $zero register, without checking for membership in the register
class of the target machine operand.
Jonas Paulsson [Thu, 29 Oct 2015 08:28:35 +0000 (08:28 +0000)]
[MachineVerifier] Analyze MachineMemOperands for mem-to-mem moves.
Since the verifier will give false reports if it incorrectly thinks MI is
loading or storing using an FI, it is necessary to scan memoperands and
find out how the FI is used in the instruction. This should be relatively
rare.
Needed to make CodeGen/SystemZ/spill-01.ll pass, which now runs with this flag.
projects / llvm / log