gh-886: Added 3 new device attributes and kernel's device-specific attributes #894
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
View rendered docs @ https://intelpython.github.io/dpctl/pulls/894/index.html |
oleksandr-pavlyk
force-pushed
the
gh-886-more-device-attributes
branch
3 times, most recently
from
103e9ee
to
f6369f2
Compare
1 task
These are DPCTLDevice_GetGlobalMemCacheSize, DPCTLDevice_GlobalMemCacheLineSize, and DPCTLDevice_GetGlobalMemCacheType. To support the latter, introduced DPCTLGlobalMemCacheType enum in dpctl_sycl_enum_types.h Tests are added to test_capi target.
oleksandr-pavlyk
force-pushed
the
gh-886-more-device-attributes
branch
from
f6369f2
to
fc0e4b2
Compare
oleksandr-pavlyk
force-pushed
the
gh-886-more-device-attributes
branch
from
6c1b5b8
to
6082c0b
Compare
* dpctl.SyclDevice.global_mem_cache_size * dpctl.SyclDevice.global_mem_cache_line_size * dpctl.SyclDevice.global_mem_cache_type The last property output is a new enum dpctl.global_mem_cache_type which can assume 3 values: none, read_only, and read_write
oleksandr-pavlyk
force-pushed
the
gh-886-more-device-attributes
branch
from
6082c0b
to
bcf1a14
Compare
max_sub_group_size property is currently on hold due to an issue in DPC++ runtime
oleksandr-pavlyk
force-pushed
the
gh-886-more-device-attributes
branch
from
c3a6d4e
to
2870b50
Compare
oleksandr-pavlyk
changed the title
diptorupd
reviewed
Sep 3, 2022
dpctl/_sycl_device.pyx
Outdated
| int: Cache size in bytes | ||
| """ | ||
| cdef uint64_t cache_sz = DPCTLDevice_GetGlobalMemCacheSize( | ||
| self._device_ref) |
There was a problem hiding this comment.
close paren should be on next line.
diptorupd
reviewed
Sep 3, 2022
dpctl/_sycl_device.pyx
Outdated
| int: Cache size in bytes | ||
| """ | ||
| cdef uint64_t cache_line_sz = DPCTLDevice_GetGlobalMemCacheLineSize( | ||
| self._device_ref) |
There was a problem hiding this comment.
same formatting change here
diptorupd
reviewed
Sep 3, 2022
dpctl/program/_program.pyx
Outdated
| a multiple, for executing the kernel on the device it was built for. | ||
| """ | ||
| cdef size_t v = DPCTLKernel_GetPreferredWorkGroupSizeMultiple( | ||
| self._kernel_ref) |
There was a problem hiding this comment.
Also the explanation reads a bit awkward. How about:
Returns a number whose multiple is the preferred work-group size for the specified kernel on the selected device.
diptorupd
reviewed
Sep 4, 2022
| * | ||
| * @param KRef DPCTLSyclKernelRef pointer to an OpenCL | ||
| * @param KRef DPCTLSyclKernelRef pointer to an SYCL |
diptorupd
reviewed
Sep 4, 2022
| * | ||
| * @param KRef DPCTLSyclKernelRef pointer to an OpenCL | ||
| * @param KRef DPCTLSyclKernelRef pointer to an SYCL |
diptorupd
reviewed
Sep 4, 2022
| * | ||
| * @param KRef DPCTLSyclKernelRef pointer to an SYCL | ||
| * interoperability kernel. | ||
| * @return Returns a value, of which work-group size is preferred to be a |
There was a problem hiding this comment.
My reuse the docstring from the Python function corresponding to the C API function.
diptorupd
previously approved these changes
Sep 4, 2022
|
Deleted rendered PR docs from intelpython.github.com/dpctl, latest should be updated shortly. 🤞 |
This was referenced Nov 7, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes #886
These are
DPCTLDevice_GetGlobalMemCacheSize,DPCTLDevice_GlobalMemCacheLineSize,and
DPCTLDevice_GetGlobalMemCacheType.To support the latter, introduced
DPCTLGlobalMemCacheTypeenum in"dpctl_sycl_enum_types.h".Tests are added to test_capi target.