This patch canonicalizes getelementptr instructions with constant
indices to use the `i8` source element type. This makes it easier for
optimizations to recognize that two GEPs are identical, because they
don't need to see past many different ways to express the same offset.
This is a first step towards
https://discourse.llvm.org/t/rfc-replacing-getelementptr-with-ptradd/68699.
This is limited to constant GEPs only for now, as they have a clear
canonical form, while we're not yet sure how exactly to deal with
variable indices.
The test llvm/test/Transforms/PhaseOrdering/switch_with_geps.ll gives
two representative examples of the kind of optimization improvement we
expect from this change. In the first test SimplifyCFG can now realize
that all switch branches are actually the same. In the second test it
can convert it into simple arithmetic. These are representative of
common optimization failures we see in Rust.
Fixes https://github.com/llvm/llvm-project/issues/69841.
This flag forces the compiler to generate code for OpenMP target regions
as if the user specified the #pragma omp requires unified_shared_memory
in each source file.
The option does not have a -fno-* friend since OpenMP requires the
unified_shared_memory clause to be present in all source files. Since
this flag does no harm if the clause is present, it can be used in
conjunction. My understanding is that USM should not be turned off
selectively, hence, no -fno- version.
This adds a basic test to check the correct generation of double
indirect access to declare target globals in USM mode vs non-USM mode.
Which I think is the only difference observable in code generation.
This runtime test checks for the (non-)occurence of data movement between host
and device. It does one run without the flag and one with the flag to
also see that both versions behave as expected. In the case w/o the new
flag data movement between host and device is expected. In the case with
the flag such data movement should not be present / reported.
Set the writable and dead_on_unwind attributes for sret arguments. These
indicate that the argument points to writable memory (and it's legal to
introduce spurious writes to it on entry to the function) and that the
argument memory will not be used if the call unwinds.
This enables additional MemCpyOpt/DSE/LICM optimizations.
Changes uploaded to the phabricator on Dec 16th are lost because the
phabricator is down. Hence re-uploading it to the github.com.
Changes to be committed:
modified: clang/include/clang/Sema/Sema.h
modified: clang/lib/Sema/SemaOpenMP.cpp
modified: clang/test/OpenMP/generic_loop_ast_print.cpp
modified: clang/test/OpenMP/loop_bind_messages.cpp
modified: clang/test/PCH/pragma-loop.cpp
---------
Co-authored-by: Sunil Kuravinakop
This is a continuation of https://reviews.llvm.org/D123235 ([OpenMP]
atomic compare fail : Parser & AST support). In this branch Support for
codegen support for atomic compare fail is being added.
---------
Co-authored-by: Sunil Kuravinakop
This patch makes `num_teams` and `thread_limit` mandatory for bare
kernels,
similar to a reguar kernel language that when launching a kernel, the
grid size
has to be set explicitly.
Fix mapping of structs to device.
The following example fails:
```
#include <stdio.h>
#include <stdlib.h>
struct Descriptor {
int *datum;
long int x;
int xi;
long int arr[1][30];
};
int main() {
Descriptor dat = Descriptor();
dat.datum = (int *)malloc(sizeof(int)*10);
dat.xi = 3;
dat.arr[0][0] = 1;
#pragma omp target enter data map(to: dat.datum[:10]) map(to: dat)
#pragma omp target
{
dat.xi = 4;
dat.datum[dat.arr[0][0]] = dat.xi;
}
#pragma omp target exit data map(from: dat)
return 0;
}
```
This is a rework of the previous attempt:
https://github.com/llvm/llvm-project/pull/72410
Currently we are missing set up-boundary address for FinalArraySection
as highests elements in partial struct data.
Currently for:
\#pragma omp target map(D.a) map(D.b[:2])
The size is:
%a = getelementptr inbounds %struct.DataTy, ptr %D, i32 0, i32 0
%b = getelementptr inbounds %struct.DataTy, ptr %D, i32 0, i32 1
%arrayidx = getelementptr inbounds [2 x float], ptr %b, i64 0, i64 0
%2 = getelementptr float, ptr %arrayidx, i32 1
%3 = ptrtoint ptr %2 to i64
%4 = ptrtoint ptr %a to i64
%5 = sub i64 %3, %4
%6 = sdiv exact i64 %5, ptrtoint (ptr getelementptr (i8, ptr null, i32
1) to i64)
Where %2 is wrong for (D.b[:2]) is pointer to first element of array
section. It should pointe to last element of array section.
The fix is to emit the pointer to the last element of array section and
use this pointer as the highest element in partial struct data.
After change IR:
%a = getelementptr inbounds %struct.DataTy, ptr %D, i32 0, i32 0
%b = getelementptr inbounds %struct.DataTy, ptr %D, i32 0, i32 1
%arrayidx = getelementptr inbounds [2 x float], ptr %b, i64 0, i64 0
%b1 = getelementptr inbounds %struct.DataTy, ptr %D, i32 0, i32 1
%arrayidx2 = getelementptr inbounds [2 x float], ptr %b1, i64 0, i64 1
%1 = getelementptr float, ptr %arrayidx2, i32 1
%2 = ptrtoint ptr %1 to i64
%3 = ptrtoint ptr %a to i64
%4 = sub i64 %2, %3
%5 = sdiv exact i64 %4, ptrtoint (ptr getelementptr (i8, ptr null, i32
1) to i64)
Fix#69214
In `emitOMPSimdRegion`, the `EmitOMPPrivateLoopCounters` should be after
`EmitOMPPrivateClause`.
If not, the private variables will be registered too early, which is not
allowed by `EmitOMPPrivateClause`.
Update all callers to pass through the Address.
For the older builtins such as `__sync_*` and MSVC `_Interlocked*`,
natural alignment of the atomic access is _assumed_. This change
preserves that behavior. It will pass through greater-than-required
alignments, however.
- `nothing` directive was effecting the `if` block structure which it
should not. So return an empty statement instead of an error statement
while parsing to avoid this.
Currently PresentModifierLocs defined with size DefaultmapKindNum; where
DefaultmapKindNum = OMPC_DEFAULTMAP_pointer + 1
Before 5.0 variable-category can not be omitted. For the test like
\#pragma omp target map(tofrom: errors) defaultmap(present)
error would be mitted.
After 5.0 that is allowd.
When try to:
PresentModifierLocs[DMC->getDefaultmapKind()] =
DMC->getDefaultmapModifierLoc();
It is accessed beyond array end.
To fix this using OMPC_DEFAULTMAP_unknow instead OMPC_DEFAULTMAP_poiner.
Hello!
This PR fixes#63871. Clang should no longer crash and instead emits an
error message.
Below is an example of the new error message:
```
~/dev/fork-llvm-project omp_dispatch_unimpl
❯ ./install/bin/clang -fopenmp -c -emit-llvm -Xclang -disable-llvm-passes test.c
test.c:6:5: error: cannot compile this OpenMP dispatch directive yet
6 | #pragma omp dispatch
| ^~~~~~~~~~~~~~~~~~~~
1 error generated.
```
This reverts commit fe5c360a9aae61db37886c0c795c409b5129905f.
The commit causes the tests below to fail on many buildbots, e.g.
https://lab.llvm.org/buildbot/#/builders/245/builds/17047
Clang :: CodeGen/aapcs-align.cpp
Clang :: CodeGen/aapcs64-align.cpp
Summary:
This patch provides the initial support to allow handling the new
driver's offloading entries. Normally, the ELF target can emit varibles
at C-identifier named sections and the linker will provide a pointer to
the section. For COFF target, instead the linker merges sections
containing a `$` in alphabetical order. We thus can emit these variables
at sections and then emit two variables that are guaranteed to be sorted
before and after the others to traverse it. Previous patches
consolidated the handling of offloading entries so that this patch more
easily can handle mapping them to the appropriate section.
Ideally, the only remaining step to allow the new driver to run on
Windows targets is to accurately map the following `ld.lld` arguments to
their `llvm-link` equivalents. These are used inside the linker-wrapper,
so we should simply need to remap the arguments to the same
functionality if possible.
```
-o, -output
-l, --library
-L, --library-path
-v, --version
-rpath
-whole-archive, -no-whole-archive
```
I have not tested this at runtime as I do not have access to a windows
machine.
This patch was adapted from some initial efforts in
https://reviews.llvm.org/D137470.
This reverts commit edd675ac283909397880f85ba68d0d5f99dc1be2.
This breaks clang build where every component is a shared library.
The file clang/lib/Basic/OpenMPKinds.cpp, which is a part of
libclangBasic.so, uses `getOpenMPClauseName` which isn't:
/usr/bin/ld: CMakeFiles/obj.clangBasic.dir/OpenMPKinds.cpp.o: in functio
n `clang ::getOpenMPSimpleClauseTypeName(llvm::omp::Clause, unsigned int
)':
OpenMPKinds.cpp:(.text._ZN5clang29getOpenMPSimpleClauseTypeNameEN4llvm3o
mp6ClauseEj+0x9b): undefined reference to `llvm::omp::getOpenMPClauseNam
e(llvm::omp::Clause)'
In Clang 16, we implemented the ability to add a label at the end of a
compound statement. These changes complete the implementation by
allowing a label to be followed by a declaration in C.
Note, this seems to have fixed an issue with some OpenMP stand-alone
directives not being properly diagnosed as per:
https://www.openmp.org/spec-html/5.1/openmpsu19.html#x34-330002.1.3
(The same requirement exists in OpenMP 5.2 as well.)
Emission of `mustprogress` attribute previously occurred only within
`EmitFunctionBody`, after generating the function body. Other routines
for function body creation may lack the attribute, potentially leading
to suboptimal optimizations later in the pipeline. Attribute emission
is now anticipated prior to generating the function body.
Fixes: https://github.com/llvm/llvm-project/issues/69833.
If we have more than one reduction variable we need to be consistent
wrt. indexing. In 3de645efe30b83ba1b6d7e500486c4f441a17a61 we broke this
as the buffer type was reduced to a singleton but the index computation
was not adjusted to account for that offset. This fixes it by
interleaving the reduction variables properly in a array-of-struct
style. We can revert it back to struct-of-array in a follow up if turns
out to be a problem. I doubt it since half the accesses should benefit
from the locallity this layout offers and only the other half were
consecutive before.
Summary:
This patch reworks how we handle global constructors in OpenMP.
Previously, we emitted individual kernels that were all registered and
called individually. In order to provide more generic support, this
patch moves all handling of this to the target backend and the runtime
plugin. This has the benefit of supporting the GNU extensions for
constructors an destructors, removing a class of failures related to
shared library destruction order, and allows targets other than OpenMP
to use the same support without needing to change the frontend.
This is primarily done by calling kernels that the backend emits to
iterate a list of ctor / dtor functions. For x64, this is automatic and
we get it for free with the standard `dlopen` handling. For AMDGPU, we
emit `amdgcn.device.init` and `amdgcn.device.fini` functions which
handle everything atuomatically and simply need to be called. For NVPTX,
a patch https://github.com/llvm/llvm-project/pull/71549 provides the
kernels to call, but the runtime needs to set up the array manually by
pulling out all the known constructor / destructor functions.
One concession that this patch requires is the change that for GPU
targets in OpenMP offloading we will use `llvm.global_dtors` instead of
using `atexit`. This is because `atexit` is a separate runtime function
that does not mesh well with the handling we're trying to do here. This
should be equivalent in all cases except for cases where we would need
to destruct manually such as:
```
struct S { ~S() { foo(); } };
void foo() {
static S s;
}
```
However this is broken in many other ways on the GPU, so it is not
regressing any support, simply increasing the scope of what we can
handle.
This changes the handling of ctors / dtors. This patch now outputs a
information message regarding the deprecation if the old format is used.
This will be completely removed in a later release.
Depends on: https://github.com/llvm/llvm-project/pull/71549
This is a support for " #pragma omp atomic compare fail ". It has Parser & AST support for now.
Reviewed By: tianshilei1992, ABataev
Differential Revision: https://reviews.llvm.org/D123235
Before we tracked the size of the teams reduction buffer in order to
allocate it at runtime per kernel launch. This patch splits the number
into two parts, the size of the reduction data (=all reduction
variables) and the (maximal) length of the buffer. This will allow us to
allocate less if we need less, e.g., if we have less teams than the
maximal length. It also allows us to move code from clangs codegen into
the runtime as we now know how large the reduction data is.
The alignment did likely not help much but increases the memory
requirement. Note that half of the affected accesses are all performed
by a single thread in each block. The reads are by consecutive threads
in a single block.
We used to perform team reduction on global memory allocated in the
runtime and by clang. This was racy as multiple instances of a kernel,
or different kernels with team reductions, would use the same locations.
Since we now have the kernel launch environment, we can allocate dynamic
memory per-launch, allowing us to move all the state into a non-racy
place.
Fixes: https://github.com/llvm/llvm-project/issues/70249
The KernelEnvironment is for compile time information about a kernel. It
allows the compiler to feed information to the runtime. The
KernelLaunchEnvironment is for dynamic information *per* kernel launch.
It allows the rutime to feed information to the kernel that is not
shared with other invocations of the kernel. The first use case is to
replace the globals that synchronize teams reductions with per-launch
versions. This allows concurrent teams reductions. More uses cases will
follow, e.g., per launch memory pools.
Fixes: https://github.com/llvm/llvm-project/issues/70249
By associating the kernel environment with the generic kernel we can
access middle-end information easily, including the launch bounds ranges
that are acceptable. By constraining the number of threads accordingly,
we now obey the user-provided bounds that were passed via attributes.
We used to pass the min/max threads/teams values through different paths
from the frontend to the middle end. This simplifies the situation by
passing the values once, only when we will create the KernelEnvironment,
which contains the values. At that point we also manifest the metadata,
as appropriate. Some footguns have also been removed, e.g., our target
check is now triple-based, not calling convention-based, as the latter
is dependent on the ordering of operations. The types of the values have
been unified to int32_t.
This reverts commit 86bfeb906e3a95ae428f3e97d78d3d22a7c839f3.
This is a long time coming re-application that was originally reverted due to
regressions, unrelated to the actual inlining change. These regressions have since
been fixed due to another long-in-the-making change of a66051c6 landing.
Original commit message for reference:
---
We have several situations where it's beneficial for code size to ensure that every
call to always-inline functions are inlined before normal inlining decisions are
made. While the normal inliner runs in a "MandatoryOnly" mode to try to do this,
it only does it on a per-SCC basis, rather than the whole module. Ensuring that
all mandatory inlinings are done before any heuristic based decisions are made
just makes sense.
Despite being referred to the "legacy" AlwaysInliner pass, it's already necessary
for -O0 because the CGSCC inliner is too expensive in compile time to run at -O0.
This also fixes an exponential compile time blow up in
https://github.com/llvm/llvm-project/issues/59126
Differential Revision: https://reviews.llvm.org/D143624
---
We now provide the information about the min/max thread and team count
from to the OMPIRBuilder, no matter what the source was. That means we
unify `thread_limit`, `num_teams`, `num_threads` handling with the
target specific attriutes (`__launch_bounds__` and
`amdgpu_flat_work_group_size`). This is in preparation to pass the
values to the runtime, and to allow the middle-end (OpenMP-opt) to
tighten the values if it seems appropriate. There is no "real" change
after this commit.
Fixed assertion failure
Basic Block in function 'main' does not have terminator!
label %land.end
caused by premature setting of CodeGenIP upon entry to
emitTargetDataCalls, where subsequent evaluation of logical expression
created new basic blocks, leaving CodeGenIP pointing to the wrong basic
block. CodeGenIP is now set near the end of the function, just prior to
generating a comparison of the logical expression result (from the if
clause) which uses CodeGenIP to insert new IR.
These are an artifact of how types are structured but serve little
purpose, merely showing that the type is sugared in some way. For
example, ElaboratedType's existence means struct S gets printed as
'struct S':'struct S' in the AST, which is unnecessary visual clutter.
Note that skipping the second print when the types have the same string
matches what we do for diagnostics, where the aka will be skipped.
