twice, in finalization of template argument deduction.
This is a re-commit of r290310 (reverted in r290329); the bug found by the
buildbots was fixed in r290399 (we would sometimes build a deduced template
argument with a bogus type).
llvm-svn: 290403
fail the merge if the arguments have different types (except if one of them was
deduced from an array bound, in which case take the type from the other).
This is correct because (except in the array bound case) the type of the
template argument in each deduction must match the type of the parameter, so at
least one of the two deduced arguments must have a mismatched type.
This is necessary because we would otherwise lose the type information for the
discarded template argument in the merge, and fail to diagnose the mismatch.
In order to power this, we now properly retain the type of a deduced non-type
template argument deduced from a declaration, rather than giving it the type of
the template parameter; we'll convert it to the template parameter type when
checking the deduced arguments.
llvm-svn: 290399
-fno-inline-functions, -O0, and optnone.
These were really, really tangled together:
- We used the noinline LLVM attribute for -fno-inline
- But not for -fno-inline-functions (breaking LTO)
- But we did use it for -finline-hint-functions (yay, LTO is happy!)
- But we didn't for -O0 (LTO is sad yet again...)
- We had weird structuring of CodeGenOpts with both an inlining
enumeration and a boolean. They interacted in weird ways and
needlessly.
- A *lot* of set smashing went on with setting these, and then got worse
when we considered optnone and other inlining-effecting attributes.
- A bunch of inline affecting attributes were managed in a completely
different place from -fno-inline.
- Even with -fno-inline we failed to put the LLVM noinline attribute
onto many generated function definitions because they didn't show up
as AST-level functions.
- If you passed -O0 but -finline-functions we would run the normal
inliner pass in LLVM despite it being in the O0 pipeline, which really
doesn't make much sense.
- Lastly, we used things like '-fno-inline' to manipulate the pass
pipeline which forced the pass pipeline to be much more
parameterizable than it really needs to be. Instead we can *just* use
the optimization level to select a pipeline and control the rest via
attributes.
Sadly, this causes a bunch of churn in tests because we don't run the
optimizer in the tests and check the contents of attribute sets. It
would be awesome if attribute sets were a bit more FileCheck friendly,
but oh well.
I think this is a significant improvement and should remove the semantic
need to change what inliner pass we run in order to comply with the
requested inlining semantics by relying completely on attributes. It
also cleans up tho optnone and related handling a bit.
One unfortunate aspect of this is that for generating alwaysinline
routines like those in OpenMP we end up removing noinline and then
adding alwaysinline. I tried a bunch of other approaches, but because we
recompute function attributes from scratch and don't have a declaration
here I couldn't find anything substantially cleaner than this.
Differential Revision: https://reviews.llvm.org/D28053
llvm-svn: 290398
This patch fixes some ASAN unittest failures on FreeBSD. See the
cfe-commits email thread for r290169 for more on those.
According to the LangRef, the allocsize attribute only tells us about
the number of bytes that exist at the memory location pointed to by the
return value of a function. It does not necessarily mean that the
function will only ever allocate. So, we need to be very careful about
treating functions with allocsize as general allocation functions. This
patch makes us fully conservative in this regard, though I suspect that
we have room to be a bit more aggressive if we want.
This has a FIXME that can be fixed by a relatively straightforward
refactor; I just wanted to keep this patch minimal. If this sticks, I'll
come back and fix it in a few days.
llvm-svn: 290397
Summary:
This change rewrites a core component in the ImplicitNullChecks pass for
greater simplicity since the original design was over-complicated for no
good reason. Please review this as essentially a new pass. The change
is almost NFC and I've added a test case for a scenario that this new
code handles that wasn't handled earlier.
The implicit null check pass, at its core, is a code hoisting transform.
It differs from "normal" code transforms in that it speculates
potentially faulting instructions (by design), but a lot of the usual
hazard detection logic (register read-after-write etc.) still applies.
We previously detected hazards by keeping track of registers defined and
used by machine instructions over an instruction range, but that was
unwieldy and did not actually confer any performance benefits. The
intent was to have linear time complexity over the number of machine
instructions considered, but it ended up being N^2 is practice.
This new version is more obviously O(N^2) (with N capped to 8 by
default) in hazard detection. It does not attempt to be clever in
tracking register uses or defs (the previous cleverness here was a
source of bugs).
Once this is checked in, I'll extract out the `IsSuitableMemoryOp` and
`CanHoistLoadInst` lambda into member functions (they're too complicated
to be inline lambdas) and do some other related NFC cleanups.
Reviewers: reames, anna, atrick
Subscribers: mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D27592
llvm-svn: 290394
Summary: We setup these interceptors twice which hangs test on windows.
Reviewers: eugenis
Subscribers: llvm-commits, kubabrecka
Differential Revision: https://reviews.llvm.org/D28070
llvm-svn: 290393
Much to my surprise, '-disable-llvm-optzns' which I thought was the
magical flag I wanted to get at the raw LLVM IR coming out of Clang
deosn't do that. It still runs some passes over the IR. I don't want
that, I really want the *raw* IR coming out of Clang and I strongly
suspect everyone else using it is in the same camp.
There is actually a flag that does what I want that I didn't know about
called '-disable-llvm-passes'. I suspect many others don't know about it
either. It both does what I want and is much simpler.
This removes the confusing version and makes that spelling of the flag
an alias for '-disable-llvm-passes'. I've also moved everything in Clang
to use the 'passes' spelling as it seems both more accurate (*all* LLVM
passes are disabled, not just optimizations) and much easier to remember
and spell correctly.
This is part of simplifying how Clang drives LLVM to make it cleaner to
wire up to the new pass manager.
Differential Revision: https://reviews.llvm.org/D28047
llvm-svn: 290392
This is necessary for the distribution targets which assume that
each component has an install target. This also moves the CMake
macros into a separate file akin to other LLVM projects.
Differential Revision: https://reviews.llvm.org/D27876
llvm-svn: 290391
This patch adds support for YAML<->DWARF for debug_info sections.
This re-lands r290147, reverted in 290148, re-landed in r290204 after fixing the issue that caused bots to fail (thank you UBSan!), and reverted again in r290209 due to failures on big endian systems.
After adding support for preserving endianness, this should be good now.
llvm-svn: 290386
Use a dummy private function with inline asm calls instead of module
level asm blocks for CFI jumptables.
The main advantage is that now jumptable codegen can be affected by
the function attributes (like target_cpu on ARM). Module level asm
gets the default subtarget based on the target triple, which is often
not good enough.
This change also uses asm constraints/arguments to reference
jumptable targets and aliases directly. We no longer do asm name
mangling in an IR pass.
Differential Revision: https://reviews.llvm.org/D28012
llvm-svn: 290384
We used to not check generic vregs, but that is actually a mistake given
nothing in the GlobalISel pipeline is going to fix the constraints on
target specific instructions. Therefore, the target has to have them
right from the start.
llvm-svn: 290380
Target specific instructions have requirements that are not compatible
with what we want to test here. Namely, target specific instructions
must have their operands properly mapped on register classes.
llvm-svn: 290379
The InstructionSelect pass will not look at target specific instructions
since they are already selected. As a result, the operands of target
specific instructions must be properly constrained, because it is not
going to fix them.
This fixes invalid register classes on call instruction.
llvm-svn: 290377
When generic virtual registers get constrained, because of a use on a
target specific operation for instance, we end up with regular virtual
registers with a type and that's perfectly fine.
llvm-svn: 290376
This is going to be needed to be able to constraint register class on
target specific instruction while the RegBankSelect pass did not run
yet.
llvm-svn: 290375
Move the logic to constraint register from InstructionSelector to a
utility function. It will be required by other passes in the GlobalISel
pipeline.
llvm-svn: 290374
Summary: Make thread local quarantine size an option so it can be turned off to save memory.
Reviewers: eugenis
Patch by Alex Shlyapnikov.
Subscribers: kubabrecka, llvm-commits
Differential Revision: https://reviews.llvm.org/D28027
llvm-svn: 290373
Canonicalize a select with a constant to the false side. This
enables more instruction shrinking opportunities since an
inline immediate can be used for the false side of v_cndmask_b32_e32.
This seems to usually be better but causes some code size regressions
in some tests.
llvm-svn: 290372
Summary:
Warm up ASAN caches in ThreadedQuarantineTest to get more predictable
incremental heap memory usage measurements.
Reviewers: eugenis
Patch by Alex Shlyapnikov.
Subscribers: aemerson, kubabrecka, llvm-commits
Differential Revision: https://reviews.llvm.org/D28061
llvm-svn: 290371
Recommitted after formal approval.
LLVM's JIT is now the foundation of dynamic-compilation features for many languages. Clang also has low-level support for dynamic compilation (ASTImporter and ExternalASTSource, notably). How the compiler is set up for dynamic parsing is generally left up to individual clients, for example LLDB's C/C++/Objective-C expression parser and the ROOT project.
Although this arrangement offers external clients the flexibility to implement dynamic features as they see fit, the lack of an in-tree client means that subtle bugs can be introduced that cause regressions in the external clients but aren't caught by tests (or users) until much later. LLDB for example regularly encounters complicated ODR violation scenarios where it is not immediately clear who is at fault.
Other external clients (notably, Cling) rely on similar functionality, and another goal is to break this functionality up into composable parts so that any client can be built easily on top of Clang without requiring extensive additional code.
I propose that the parts required to build a simple expression parser be added to Clang. Initially, I aim to have the following features:
A piece that looks up external declarations from a variety of sources (e.g., from previous dynamic compilations, from modules, or from DWARF) and uses clear conflict resolution rules to reconcile differences, with easily understood errors. This functionality will be supported by in-tree tests.
A piece that works hand in hand with the LLVM JIT to resolve the locations of external declarations so that e.g. variables can be redeclared and (for high-performance applications like DTrace) external variables can be accessed directly from the registers where they reside.
This commit adds a tester that parses a sequence of source files and then uses them as source data for an expression. External references are resolved using an ExternalASTSource that responds to name queries using an ASTImporter. This is the setup that LLDB uses, and the motivating reason for MinimalImport in ASTImporter. When complete, this tester will implement the first of the above goals.
Differential Revision: https://reviews.llvm.org/D27180
llvm-svn: 290367
This is a succeeding patch of https://reviews.llvm.org/D22840 to address the
issue when a value to be merged into an int64 pair is in a different BB. Redoing
the store splitting in CodeGenPrepare so we can match the pattern across multiple
BBs and move some instructions into the same BB. We still keep the code in dag
combine so that we can catch cases that show up after DAG combining runs.
Differential Revision: https://reviews.llvm.org/D25914
llvm-svn: 290365
program
Offload related code is not quite ready yet, but some simple examples
must not crash the compiler. Patch fixes the problem in offloading code
with exceptions.
llvm-svn: 290364
This is for splitMergedValStore in DAG Combine to share the target query interface
with similar logic in CodeGenPrepare.
Differential Revision: https://reviews.llvm.org/D24707
llvm-svn: 290363
Follow up to D27209 fix, this patch now properly handles single transient
instruction in basic block.
Patch by Aleksandar Beserminji.
Differential Revision: https://reviews.llvm.org/D27856
llvm-svn: 290361
This patch fixes use of incorrect `%zi` to format a plain `int`, and using
`%llu` to format a `uint64_t`. The fix is to use the new typesafe
`llvm::Formatv` based API.
Differential Revision: https://reviews.llvm.org/D28028
Subscribers: lldb-commits
llvm-svn: 290359
COFF has a 2**16 section limit, and on Win64, every COMDAT function
creates at least 3 sections: .text, .pdata, and .xdata. For MSVC, we
enable bigobj on a file-by-file basis, but GCC appears to hit the limit
on different files.
Fixes PR25953
llvm-svn: 290358
When the pipeliner is renaming phi values, it may need to iterate through
the phi operands to check for other phis. However, the pipeliner should
stop once it reaches a phi that is outside the pipelined loop.
Also, when the generateExistingPhis code is unable to reuse an existing
phi, the default code that computes the PhiOp2 is only to be used when
the pipeliner is generating the kernel. Otherwise, the phi may be a value
computed earlier in the same epilog.
Patch by Brendon Cahoon.
llvm-svn: 290355
Summary: This patch attempts to fix test patching-unpatching.cc . The new code flushes the instruction cache after modifying the program at runtime.
Reviewers: dberris, rengolin
Subscribers: llvm-commits, iid_iunknown, aemerson
Differential Revision: https://reviews.llvm.org/D27996
llvm-svn: 290354
Richard Smith