Fix a debug location handling bug in GVN.
Caught by the famous "DebugLoc describes the currect SubProgram" assertion.
When GVN is removing a nonlocal load it updates the debug location of the
SSA value it replaced the load with with the one of the load. In the
testcase this actually overwrites a valid debug location with an empty one.
In reality GVN has to make an arbitrary choice between two equally valid
debug locations. This patch changes to behavior to only update the
location if the value doesn't already have a debug location.
[PowerPC] Recurse through constants when looking for TLS globals
We cannot form ctr-based loops around function calls, including calls to
__tls_get_addr used for PIC TLS variables. References to such TLS variables,
however, might be buried within constant expressions, and so we need to search
the entire constant expression to be sure that no references to such TLS
variables exist.
Fixes PR25256, reported by Eric Schweitz. This is a slightly-modified version
of the patch suggested by Eric in the bug report, and a test case I created.
[RemoveDuplicatePHINodes] Start over after removing a PHI.
This makes RemoveDuplicatePHINodes more effective and fixes an assertion
failure. Triggering the assertions requires a DenseSet reallocation
so this change only contains a constructive test.
I'll explain the issue with a small example. In the following function
there's a duplicate PHI, %4 and %5 are identical. When this is found
the DenseSet in RemoveDuplicatePHINodes contains %2, %3 and %4.
after RemoveDuplicatePHINodes runs the function looks like this. %3 has
changed and is now identical to %2, but RemoveDuplicatePHINodes never
saw this.
If the DenseSet does a reallocation now it will reinsert all
keys and stumble over %3 now having a different hash value than it had
when inserted into the map for the first time. This change clears the
set whenever a PHI is deleted and starts the progress from the
beginning, allowing %3 to be deleted and avoiding inconsistent DenseSet
state. This potentially has a negative performance impact because
it rescans all PHIs, but I don't think that this ever makes a difference
in practice.
On the code path in ExpandUnalignedLoad which expands an unaligned vector/fp
value in terms of a legal integer load of the same size, the ChainResult needs
to be the chain result of the integer load.
[PPC64LE] Fix PR24546 - Swap optimization and debug values
This patch fixes PR24546, which demonstrates a segfault during the VSX
swap removal pass. The problem is that debug value instructions were
not excluded from the list of instructions to be analyzed for webs of
related computation. I've added the test case from the PR as a crash
test in test/CodeGen/PowerPC.
Require Dominator Tree For SROA, improve compile-time
TL-DR: SROA is followed by EarlyCSE which requires the DominatorTree.
There is no reason not to require it up-front for SROA.
Some history is necessary to understand why we ended-up here.
r123437 switched the second (Legacy)SROA in the optimizer pipeline to
use SSAUpdater in order to avoid recomputing the costly
DominanceFrontier. The purpose was to speed-up the compile-time.
Later r123609 removed the need for the DominanceFrontier in
(Legacy)SROA.
Right after, some cleanup was made in r123724 to remove any reference
to the DominanceFrontier. SROA existed in two flavors: SROA_SSAUp and
SROA_DT (the latter replacing SROA_DF).
The second argument of `createScalarReplAggregatesPass` was renamed
from `UseDomFrontier` to `UseDomTree`.
I believe this is were a mistake was made. The pipeline was not
updated and the call site was still:
PM->add(createScalarReplAggregatesPass(-1, false));
At that time, SROA was immediately followed in the pipeline by
EarlyCSE which required alread the DominatorTree. Not requiring
the DominatorTree in SROA didn't save anything, but unfortunately
it was lost at this point.
When the new SROA Pass was introduced in r163965, I believe the goal
was to have an exact replacement of the existing SROA, this bug
slipped through.
You can see currently:
$ echo "" | clang -x c++ -O3 -c - -mllvm -debug-pass=Structure
...
...
FunctionPass Manager
SROA
Dominator Tree Construction
Early CSE
After this patch:
$ echo "" | clang -x c++ -O3 -c - -mllvm -debug-pass=Structure
...
...
FunctionPass Manager
Dominator Tree Construction
SROA
Early CSE
This improves the compile time from 88s to 23s for PR17855.
https://llvm.org/bugs/show_bug.cgi?id=17855
And from 113s to 12s for PR16756
https://llvm.org/bugs/show_bug.cgi?id=16756
[x86] Fix wrong lowering of vsetcc nodes (PR25080).
Function LowerVSETCC (in X86ISelLowering.cpp) worked under the wrong
assumption that for non-AVX512 targets, the source type and destination type
of a type-legalized setcc node were always the same type.
This assumption was unfortunately incorrect; the type legalizer is not always
able to promote the return type of a setcc to the same type as the first
operand of a setcc.
In the case of a vsetcc node, the legalizer firstly checks if the first input
operand has a legal type. If so, then it promotes the return type of the vsetcc
to that same type. Otherwise, the return type is promoted to the 'next legal
type', which, for vectors of MVT::i1 is always a 128-bit integer vector type.
Example (-mattr=+avx):
%0 = trunc <8 x i32> %a to <8 x i23>
%1 = icmp eq <8 x i23> %0, zeroinitializer
The type legalizer would firstly check if 't5' has a legal type. If so, then it
would reuse that same type to promote the return type of the setcc node.
Unfortunately 't5' is of illegal type v8i23, and therefore it cannot be used to
promote the return type of the setcc node. Consequently, the setcc return type
is promoted to v8i16. Later on, 't5' is promoted to v8i32 thus leading to the
following dag node:
v8i16 = setcc t32, t25, seteq:ch
where t32 and t25 are now values of type v8i32.
Before this patch, function LowerVSETCC would have wrongly expanded the setcc
to a single X86ISD::PCMPEQ. Surprisingly, ISel was still able to match an
instruction. In our case, ISel would have matched a VPCMPEQWrr:
t37: v8i16 = X86ISD::VPCMPEQWrr t36, t25
However, t36 and t25 are both VR256, while the result type is instead of class
VR128. This inconsistency ended up causing the insertion of COPY instructions
like this:
%vreg7<def> = COPY %vreg3; VR128:%vreg7 VR256:%vreg3
Which is an invalid full copy (not a sub register copy).
Eventually, the backend would have hit an UNREACHABLE "Cannot emit physreg copy
instruction" in the attempt to expand the malformed pseudo COPY instructions.
This patch fixes the problem adding the missing logic in LowerVSETCC to handle
the corner case of a setcc with 128-bit return type and 256-bit operand type.
This problem was originally reported by Dimitry as PR25080. It has been latent
for a very long time. I have added the minimal reproducible from that bugzilla
as test setcc-lowering.ll.
PR25157 identifies a bug where a load plus a vector shuffle is
incorrectly converted into an LXVDSX instruction. That optimization
is only valid if the load is of a doubleword, and in the noted case,
it was not. This corrects that problem.
Joint patch with Eric Schweitz, who provided the bugpoint-reduced test
case.
To commute a trivial rlwimi instructions (meaning one with a full mask and zero
shift), we'd need to ability to form an all-zero mask (instead of an all-one
mask) using rlwimi. We can't represent this, however, and we'll miscompile code
if we try.
The code quality problem that this highlights (that SDAG simplification can
lead to us generating an ISD::OR node with a constant zero LHS) will be fixed
as a follow-up.
PPCISelDAGToDAG has a transformation that generates a rlwimi instruction from
an input pattern that looks like this:
and(or(x, c1), c2)
but the associated logic does not work if there are bits that are 1 in c1 but 0
in c2 (these are normally canonicalized away, but that can't happen if the 'or'
has other users. Make sure we abort the transformation if such bits are
discovered.
[PowerPC] Don't always consider P8Altivec-only masks in LowerVECTOR_SHUFFLE
LowerVECTOR_SHUFFLE needs to decide whether to pass a vector shuffle off to the
TableGen-generated matching code, and it does this by testing the same
predicates used by the TableGen files. Unfortunately, when we added new
P8Altivec-only predicates, we started universally testing them in
LowerVECTOR_SHUFFLE, and if then matched when targeting a system prior to a P8,
we'd end up with a selection failure.
[PowerPC] Fixup SELECT_CC (and SETCC) patterns with i1 comparison operands
There were really two problems here. The first was that we had the truth tables
for signed i1 comparisons backward. I imagine these are not very common, but if
you have:
setcc i1 x, y, LT
this has the '0 1' and the '1 0' results flipped compared to:
setcc i1 x, y, ULT
because, in the signed case, '1 0' is really '-1 0', and the answer is not the
same as in the unsigned case.
The second problem was that we did not have patterns (at all) for the unsigned
comparisons select_cc nodes for i1 comparison operands. This was the specific
cause of PR24552. These had to be added (and a missing Altivec promotion added
as well) to make sure these function for all types. I've added a bunch more
test cases for these patterns, and there are a few FIXMEs in the test case
regarding code-quality.
[PowerPC] Don't assume ADDISdtprelHA's source is r3
Even through ADDISdtprelHA generally has r3 as its source register, it is
possible for the instruction scheduler to move things around such that some
other register is the source. We need to print the actual source register, not
always r3. Fixes PR24394.
The test case will come in a follow-up commit because it depends on MIR
target-flags parsing.
[PowerPC] PPCVSXFMAMutate should ignore trivial-copy addends
We might end up with a trivial copy as the addend, and if so, we should ignore
the corresponding FMA instruction. The trivial copy can be coalesced away later,
so there's nothing to do here. We should not, however, assert. Fixes PR24544.
[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.
[PowerPC] PPCVSXFMAMutate should not segfault on undef input registers
When PPCVSXFMAMutate would look at the input addend register, it would get its
input value number. This would fail, however, if the register was undef,
causing a segfault. Don't segfault (just skip such FMA instructions).
Fixes the test case from PR24542 (although that may have been over-reduced).
[bpf] Do not expand UNDEF SDNode during insn selection lowering
o Before this patch, BPF backend will expand UNDEF node
to i64 constant 0.
o For second pass of dag combiner, legalizer will run through
each to-be-processed dag node.
o If any new SDNode is generated and has an undef operand,
dag combiner will put undef node, newly-generated constant-0 node,
and any node which uses these nodes in the working list.
o During this process, it is possible undef operand is
generated again, and this will form an infinite loop
for dag combiner pass2.
o This patch allows UNDEF to be a legal type.
Signed-off-by: Yonghong Song <yhs@plumgrid.com> Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
------------------------------------------------------------------------
[bpf] Avoid extra pointer arithmetic for stack access
For the program like below
struct key_t {
int pid;
char name[16];
};
extern void test1(char *);
int test() {
struct key_t key = {};
test1(key.name);
return 0;
}
For key.name, the llc/bpf may generate the below code:
R1 = R10 // R10 is the frame pointer
R1 += -24 // framepointer adjustment
R1 |= 4 // R1 is then used as the first parameter of test1
OR operation is not recognized by in-kernel verifier.
This patch introduces an intermediate FI_ri instruction and
generates the following code that can be properly verified:
R1 = R10
R1 += -20
Patch by Yonghong Song <yhs@plumgrid.com>
------------------------------------------------------------------------
Fix vector splitting for extract_vector_elt and vector elements of <8-bits.
Summary:
One of the vector splitting paths for extract_vector_elt tries to lower:
define i1 @via_stack_bug(i8 signext %idx) {
%1 = extractelement <2 x i1> <i1 false, i1 true>, i8 %idx
ret i1 %1
}
to:
define i1 @via_stack_bug(i8 signext %idx) {
%base = alloca <2 x i1>
store <2 x i1> <i1 false, i1 true>, <2 x i1>* %base
%2 = getelementptr <2 x i1>, <2 x i1>* %base, i32 %idx
%3 = load i1, i1* %2
ret i1 %3
}
However, the elements of <2 x i1> are not byte-addressible. The result of this
is that the getelementptr expands to '%base + %idx * (1 / 8)' which simplifies
to '%base + %idx * 0', and then simply '%base' causing all values of %idx to
extract element zero.
This commit fixes this by promoting the vector elements of <8-bits to i8 before
splitting the vector.
Fix embarrassing bugs I introduced to the `SlotTracker` in or around
r235785. I had us iterating through every instruction in a function
(and hitting a map in the LLVMContext) for every basic block in the
function.
While there, completely avoid the call to
`SlotTracker::processFunctionMetadata()` from
`SlotTracker::processFunction()` if we've speculatively done this
already in `SlotTracker::processModule()` by checking
`ShouldInitializeAllMetadata` (this wasn't an algorithmic problem, but
it's touching the same line of code).
Michael Wong [Mon, 24 Aug 2015 14:50:26 +0000 (14:50 +0000)]
Update CREDITS.TXT with Clang OpenMP implementation + test suite contributors from AMD, Argonne National Lab., IBM, Intel, Texas Instruments, University of Houston and many others.
Renato Golin [Thu, 20 Aug 2015 16:41:22 +0000 (16:41 +0000)]
Merge r245577 into branch_37
[ARM] Don't try and custom lower a vNi64 SETCC.
It won't go well. We've already marked 64-bit SETCCs as non-Custom, but it's
just possible that a SETCC has a legal result type but an illegal operand
type. If this happens, bail out before we create unselectable nodes.
Fixes PR24292. I tried to create a testcase but in 99% of cases we can't
trigger this - not surprising that this bug has been latent since 2009.
[PowerPC] Fix the int2fp(fp2int(x)) DAGCombine to ignore ppc_fp128
This DAGCombine was creating custom SDAG nodes with an illegal ppc_fp128
operand type because it was triggering on f64/f32 int2fp(fp2int(ppc_fp128 x)),
but shouldn't (it should only apply to f32/f64 types). The result was a crash.
------------------------------------------------------------------------
Hans Wennborg [Thu, 20 Aug 2015 15:56:38 +0000 (15:56 +0000)]
Merging r245365 and r245369:
------------------------------------------------------------------------
r245365 | majnemer | 2015-08-18 15:07:25 -0700 (Tue, 18 Aug 2015) | 4 lines
[InstSimplify] Don't assume getAggregateElement will succeed
It isn't always possible to get a value from getAggregateElement.
This fixes PR24488.
------------------------------------------------------------------------
Renato Golin [Thu, 20 Aug 2015 15:05:48 +0000 (15:05 +0000)]
Revert "[SimplifyCFG] Be more aggressive" on branch_37
This reverts commit r229099 in branch 37 only, because it caused PR24292.
I'll continue investigating and will fix on trunk, but being an optimization
change, we can let the rest of the release go without this one.
Prevent the scalarizer from caching incorrect entries
The scalarizer can cache incorrect entries when walking up a chain of
insertelement instructions. This occurs when it encounters more than one
instruction that it is not actively searching for, as it unconditionally caches
every element it finds. The fix is to only cache the first element that it
isn't searching for so we don't overwrite correct entries.
This change adds RTTI and Exception flags to llvm-config's cxxflags. This solution is a minimal patch to solve the issue, and is recommended for the 3.7 release branch. Tom Stellard's outstanding work is the longer term solution.
Patch By: David Wiberg
------------------------------------------------------------------------
Daniel Sanders [Fri, 14 Aug 2015 10:18:40 +0000 (10:18 +0000)]
[mips] Drop the claim that ubsan works since r243384 and r244646 are not yet merged.
The timezone difference between myself, the code-owner, and release manager means
it's sensible to update the release notes on the assumption that they won't be
merged. If we do merge them, then we can revert this release notes change.
[SystemZ] Support large LLVM IR struct return values
Recent mesa/llvmpipe crashes on SystemZ due to a failed assertion when
attempting to compile a routine with a return type of
{ <4 x float>, <4 x float>, <4 x float>, <4 x float> }
on a system without vector instruction support.
This is because after legalizing the vector type, we get a return value
consisting of 16 floats, which cannot all be returned in registers.
Usually, what should happen in this case is that the target's CanLowerReturn
routine rejects the return type, in which case SelectionDAG falls back to
implementing a structure return in memory via implicit reference.
However, the SystemZ target never actually implemented any CanLowerReturn
routine, and thus would accept any struct return type.
This patch fixes the crash by implementing CanLowerReturn. As a side effect,
this also handles fp128 return values, fixing a todo that was noted in
SystemZCallingConv.td.
`InstCombiner::OptimizeOverflowCheck` was asserting an
invariant (operands to binary operations are ordered by decreasing
complexity) that wasn't really an invariant. Fix this by instead having
`InstCombiner::OptimizeOverflowCheck` establish the invariant if it does
not hold.
------------------------------------------------------------------------
[PHITransAddr] Don't assume that instruction operands are translatable
We can only PHI translate instructions. In our attempt to PHI translate
a bitcast, we attempt to translate its operand; however, the operand
might be an argument or a global instead of an instruction. Benignly
bail out when this happens.
Kai Nacke [Thu, 6 Aug 2015 19:44:38 +0000 (19:44 +0000)]
diff --git a/docs/ReleaseNotes.rst b/docs/ReleaseNotes.rst
index 96461e5..f669a1f 100644
--- a/docs/ReleaseNotes.rst
+++ b/docs/ReleaseNotes.rst
@@ -206,6 +206,21 @@ Jade project is hosted as part of the Open RVC-CAL Compiler
(`Orcc <http://orcc.sf.net>`_) and requires it to translate the RVC-CAL standard
library of video coding tools into an LLVM assembly code.
LDC - the LLVM-based D compiler
-------------------------------
`D <http://dlang.org>`_ is a language with C-like syntax and static typing. It
pragmatically combines efficiency, control, and modeling power, with safety and
programmer productivity. D supports powerful concepts like Compile-Time Function
Execution (CTFE) and Template Meta-Programming, provides an innovative approach
to concurrency and offers many classical paradigms.
`LDC <http://wiki.dlang.org/LDC>`_ uses the frontend from the reference compiler
combined with LLVM as backend to produce efficient native code. LDC targets
x86/x86_64 systems like Linux, OS X and Windows and also PowerPC (32/64 bit).
Ports to other architectures like ARM, AArch64 and MIPS64 are underway.
Hans Wennborg [Thu, 6 Aug 2015 16:08:01 +0000 (16:08 +0000)]
Merging r244123:
------------------------------------------------------------------------
r244123 | pete | 2015-08-05 13:55:53 -0700 (Wed, 05 Aug 2015) | 7 lines
Update GettingStarted docs list of LLVM_TARGETS_TO_BUILD to match cmake.
Since the docs were written, we've added the BPF backend to the list.
Updating the docs to take this in to account. Also sorted them to
match cmake while I was changing these lines.
Reviewed by Chris B.
------------------------------------------------------------------------
Hans Wennborg [Thu, 6 Aug 2015 16:02:17 +0000 (16:02 +0000)]
Merging r243927, r243932, and r243934:
------------------------------------------------------------------------
r243927 | chandlerc | 2015-08-03 17:44:07 -0700 (Mon, 03 Aug 2015) | 11 lines
[UB] Fix a nasty place where we would pass null pointers to memcpy.
This happens to work, but is not guaranteed to work. Indeed, most memcpy
interfaces in Linux-land annotate these arguments as nonnull, and GCC
and LLVM both can and do optimized based upon that. When they do so,
they might legitimately have miscompiled code calling this routine with
two valid iterators, 'nullptr' and 'nullptr'. There was even code doing
precisely this because StringRef().begin() and StringRef().end() both
produce null pointers.
This was found by UBSan.
------------------------------------------------------------------------
projects / llvm / log