| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Renamed a function as suggested in #175664.
|
|
Build has been broken when OMPTARGET_DEBUG is undefined.
|
|
* Prepare a set of debug types in llvm::offload::debug to be used in
plugin code
* Update debug messages in the plugins
|
|
Add liboffload memory data locking API for libomptarget migration
This PR adds liboffload memory data locking API that needed to make
libomptarget to use liboffload
|
|
Fix the few remaining usages and remove the support for the old REPORT
macro.
|
|
Summary:
This is only really meaningful for the NVPTX target. Not all build
environments support host LTO and these are redundant tests, just clean
this up and make it run faster.
|
|
with other plugins (#172946)
Update information about devices provided by level zero plugin in order
to be more consistent with other plugins.
|
|
This PR refactors how the device image is built so we can expose the
native ELF of the device to DeviceImageTy which solves several issues
regarding symbol look up (as DeviceImageTy expects an ELF). It also
simplifies the module linking code taking into account the latest
changes in the driver (which adds "-library-compilation when necessary).
---------
Co-authored-by: Alexey Sachkov <alexey.sachkov@intel.com>
Co-authored-by: Nick Sarnie <nick.sarnie@intel.com>
Co-authored-by: Joseph Huber <huberjn@outlook.com>
|
|
When looking for the device address of a symbol, we need to also look if
it's a function symbol if not found as global symbol in the device.
---------
Co-authored-by: Alexey Sachkov <alexey.sachkov@intel.com>
Co-authored-by: Nick Sarnie <nick.sarnie@intel.com>
Co-authored-by: Joseph Huber <huberjn@outlook.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
|
|
Support for getDebugLevel was removed as part of the new debug macros
(#165416). This PR updates such usages to use the new ODBG_* macros.
---------
Co-authored-by: Alexey Sachkov <alexey.sachkov@intel.com>
Co-authored-by: Nick Sarnie <nick.sarnie@intel.com>
Co-authored-by: Joseph Huber <huberjn@outlook.com>
|
|
Add a new nextgen plugin that supports GPU devices through the Intel oneAPI Level Zero library. The plugin is not enabled by default and needs to be added to LIBOMPTARGET_PLUGINS_TO_BUILD explicitely.
---------
Co-authored-by: Alexey Sachkov <alexey.sachkov@intel.com>
Co-authored-by: Nick Sarnie <nick.sarnie@intel.com>
Co-authored-by: Joseph Huber <huberjn@outlook.com>
|
|
Update debug messages based on the new method from #170425. Updated the
following files.
- plugins-nextgen/common/include/MemoryManager.h
- plugins-nextgen/common/include/PluginInterface.h
- plugins-nextgen/common/src/GlobalHandler.cpp
- plugins-nextgen/common/src/PluginInterface.cpp
- plugins-nextgen/host/dynamic_ffi/ffi.cpp
|
|
This commit makes the cuLaunchKernel call to pass the total arguments size without tail padding.
|
|
This PR introduces new debug macros that allow a more fined control of
which debug message to output and introduce C++ stream style for debug
messages.
Changing existing messages (except a few that I changed for testing)
will come in subsequent PRs.
I also think that we should make debug enabling OpenMP agnostic but, for
now, I prioritized maintaing the current libomptarget behavior for now,
and we might need more changes further down the line as we we decouple
libomptarget.
|
|
Summary:
We start this thread if the RPC client symbol is detected in the loaded
binary. We should make this sleep if there's no work to avoid the thread
running at high priority when the (scarecely used) RPC call is actually
required. So, right now after 25 microseconds we will assume the server
is inactive and begin sleeping. This resets once we do find work.
AMD supports a more intelligent way to do this. HSA signals can wake a
sleeping thread from the kernel, and signals can be sent from the GPU
side. This would be nice to have and I'm planning on working with it in
the future to make this infrastructure more usable with existing AMD
workloads.
|
|
|
|
Summary:
I made that an unimplemented error, but forgot that it was used for this
environment variable.
|
|
Summary:
This was a lot of code that was only used for upstream LLVM builds of
AMDGPU offloading. We have a generic and fast `malloc` in `libc` now so
just use that. Simplifies code, can be added back if we start providing
alternate forms but I don't think there's a single use-case that would
justify it yet.
|
|
Introduced in OpenMP 6.0, the device UID shall be a unique identifier of
a device on a given system. (Not necessarily a UUID.) Since it is not
guaranteed that the (U)UIDs defined by the device vendor libraries, such
as HSA, do not overlap with those of other vendors, the device UIDs in
offload are always combined with the offload plugin name. In case the
vendor library does not specify any device UID for a given device, we
fall back to the offload-internal device ID.
The device UID can be retrieved using the `llvm-offload-device-info`
tool.
|
|
Adds omp_target_is_accessible routine.
Refactors common code from omp_target_is_present to work for both
routines.
---------
Co-authored-by: Shilei Tian <i@tianshilei.me>
|
|
|
|
This was used for the old interop code. It's dead code after #143491
|
|
Summary:
This check is unnecessarily restrictive and currently incorrectly fires
for any size less than eight bytes. Just remove it, we do sanity checks
elsewhere and at some point need to trust the ABI.
|
|
This implements two pieces to restore the interop functionality (that I
broke) when the 6.0 interfaces were added:
* A set of wrappers that support the old interfaces on top of the new
ones
* The same level of interop support for the CUDA amd AMD plugins
|
|
Co-authored-by: Joseph Huber <huberjn@outlook.com>
|
|
This squishes a warning where the runtime tries to bind a StringRef to
a `%p`.
|
|
|
|
Currently there are two serialization modes for bitstream Remarks:
standalone and separate. The separate mode splits remark metadata (e.g.
the string table) from actual remark data. The metadata is written into
the object file by the AsmPrinter, while the remark data is stored in a
separate remarks file. This means we can't use bitstream remarks with
tools like opt that don't generate an object file. Also, it is confusing
to post-process bitstream remarks files, because only the standalone
files can be read by llvm-remarkutil. We always need to use dsymutil
to convert the separate files to standalone files, which only works for
MachO. It is not possible for clang/opt to directly emit bitstream
remark files in standalone mode, because the string table can only be
serialized after all remarks were emitted.
Therefore, this change completely removes the separate serialization
mode. Instead, the remark string table is now always written to the end
of the remarks file. This requires us to tell the serializer when to
finalize remark serialization. This automatically happens when the
serializer goes out of scope. However, often the remark file goes out of
scope before the serializer is destroyed. To diagnose this, I have added
an assert to alert users that they need to explicitly call
finalizeLLVMOptimizationRemarks.
This change paves the way for further improvements to the remark
infrastructure, including more tooling (e.g. #159784), size optimizations
for bitstream remarks, and more.
Pull Request: https://github.com/llvm/llvm-project/pull/156715
|
|
Summary:
This was originally added in as a hack to work around CUDA's limitation
on allocation. The `libc` implementation now isn't even used for CUDA so
this code is never hit. Even if this case, this code never truly worked.
A true solution would be to use CUDA's virtual memory API instead to
allocate 2MiB slabs independenctly from the normal memory management
done in the stream.
|
|
Summary:
I made the GPU flags accept more of the default LLVM warnings, which
triggered some new cases. Clean those up and fix some other ones while
I'm at it.
|
|
Summary:
This exposes the 'isDeviceCompatible' routine for checking if a binary
*can* be loaded. This is useful if people don't want to consume errors
everywhere when figuring out which image to put to what device.
I don't know if this is a good name, I was thining like `olIsCompatible`
or whatever. Let me know what you think.
Long term I'd like to be able to do something similar to what OpenMP
does where we can conditionally only initialize devices if we need them.
That's going to be support needed if we want this to be more
generic.
|
|
Summary:
Turns out the new CUDA ABI now applies retroactively to all the other
SMs if you upgrade to CUDA 13.0. This patch changes the scheme, keeping
all the SM flags consistent but using an offset.
Fixes: https://github.com/llvm/llvm-project/issues/159088
|
|
Currently get this error
```
offload/plugins-nextgen/common/src/PluginInterface.cpp:859:63: error: member reference type 'StringRef' is not a pointer; did you mean to use '.'?
```
We pass the full image binary now so we can't really print anything
useful here.
Seems introduced in https://github.com/llvm/llvm-project/pull/158748.
---------
Signed-off-by: Sarnie, Nick <nick.sarnie@intel.com>
Co-authored-by: Joseph Huber <huberjn@outlook.com>
|
|
Summary:
Currently we have this `__tgt_device_image` indirection which just takes
a reference to some pointers. This was all find and good when the only
usage of this was from a section of GPU code that came from an ELF
constant section. However, we have expanded beyond that and now need to
worry about managing lifetimes. We have code that references the image
even after it was loaded internally. This patch changes the
implementation to instaed copy the memory buffer and manage it locally.
This PR reworks the JIT and other image handling to directly manage its
own memory. We now don't need to duplicate this behavior externally at
the Offload API level. Also we actually free these if the user unloads
them.
Upside, less likely to crash and burn. Downside, more latency when
loading an image.
|
|
Summary:
This operation is done every time we load a binary, this behavior should
be moved into OpenMP since it concerns an OpenMP specific data struct.
This is a little messy, because ideally we should only be using public
APIs, but more can be extracted later.
|
|
This was added in #154105 , but was not added to the plugin interface's
list of valid modes.
|
|
This is the total number of work items that the device supports (the
equivalent work group properties are for only a single work group).
|
|
Previously, `olDestroyQueue` would not actually destroy the queue,
instead leaving it for the device to clean up when it was destroyed.
Now, the queue is either released immediately if it is complete or put
into a list of "pending" queues if it is not. Whenever we create a new
queue, we check this list to see if any are now completed. If there are
any we release their resources and use them instead of pulling from
the pool.
This prevents long running programs that create and drop many queues
without syncing them from leaking memory all over the place.
|
|
On my system, this will be "Radeon RX 7900 GRE" rather than "gfx1100". For Nvidia, the product name and device name are identical.
|
|
Kernels which are marked as SPMD-No-Loop should be launched with
sufficient number of teams and threads to cover loop iteration space.
No-Loop mode is described in RFC:
https://discourse.llvm.org/t/rfc-no-loop-mode-for-openmp-gpu-kernels/87517/
|
|
|
|
|
|
Implement olMemFill to support filling device memory with arbitrary
length patterns. AMDGPU support will be added in a follow-up PR.
|
|
A simple info query for events that returns whether the event is
complete or not.
|
|
This is equivalent to `cuOccupancyMaxPotentialBlockSize`. It is
currently
only implemented on Cuda; AMDGPU and Host return unsupported.
---------
Co-authored-by: Callum Fare <callum@codeplay.com>
|
|
Add the following properties in Offload device info:
* VENDOR_ID
* NUM_COMPUTE_UNITS
* [SINGLE|DOUBLE|HALF]_FP_CONFIG
* NATIVE_VECTOR_WIDTH_[CHAR|SHORT|INT|LONG|FLOAT|DOUBLE|HALF]
* MAX_CLOCK_FREQUENCY
* MEMORY_CLOCK_RATE
* ADDRESS_BITS
* MAX_MEM_ALLOC_SIZE
* GLOBAL_MEM_SIZE
Add a bitfield option to enumerators, allowing the values to be
bit-shifted instead of incremented. Generate the per-type enums using
`foreach` to reduce code duplication.
Use macros in unit test definitions to reduce code duplication.
|
|
The purpose of this fence is to ensure that any `dataSubmit`s inserted
into a queue before a `dataFence` finish before finish before any
`dataSubmit`s
inserted after it begin.
This is a no-op for most queues, since they are in-order, and by design
any operations inserted into them occur in order.
But the interface is supposed to be functional for out-of-order queues.
The addition of the interface means that any operations that rely on
such ordering (like ATTACH map-type support in #149036) can invoke it,
without worrying about whether the underlying queue is in-order or
out-of-order.
Once a plugin supports out-of-order queues, the plugin can implement
this function, without requiring any change at the libomptarget level.
---------
Co-authored-by: Alex Duran <alejandro.duran@intel.com>
|
|
Add an `olLaunchHostFunction` method that allows enqueueing host work
to the stream.
|
|
Following on from #152304, implement the new query in the CUDA plugin
|
|
Adds a conditional to the error return so that it only returns if there was an error.
|
