Skip to content

Conversation

ClarkChin08
Copy link

@ClarkChin08 ClarkChin08 commented Sep 17, 2025

  1. Generates random data on the host and copies it to the device via syclcompat::memcpy, improving code reuse and supporting a broader range of element types.
  2. An optional bits parameter to control fractional precision, which was not present in the original.

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
@ClarkChin08 ClarkChin08 changed the title refactor the block fill from host Improve Device-Side Random Uniform Filling by Reusing Host Implementation in SYCL Sep 17, 2025
assert(false && "Not supported dtype");
}
typename RealType<Element>::Type min, ///< lower bound for distribution
int bits = -1
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please add some comments for bits.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please add some comments for bits.

Done!

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add copyright changes.

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
int bits = -1 ///< If non-negative, specifies number of fractional bits that
/// are not truncated to zero. Permits reducing precision of
/// data.
) {

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: indent

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: indent

Done!

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
@rolandschulz rolandschulz merged commit 9baca2c into intel:main Sep 18, 2025
2 of 3 checks passed
leonling-lly pushed a commit to leonling-lly/cutlass-sycl that referenced this pull request Sep 19, 2025
…ation in SYCL (intel#515)

1. Generates random data on the host and copies it to the device via
syclcompat::memcpy, improving code reuse and supporting a broader range
of element types.
2. An optional bits parameter to control fractional precision, which was
not present in the original.

---------

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
sunjiweiswift pushed a commit to sunjiweiswift/cutlass-sycl that referenced this pull request Sep 22, 2025
…ation in SYCL (intel#515)

1. Generates random data on the host and copies it to the device via
syclcompat::memcpy, improving code reuse and supporting a broader range
of element types.
2. An optional bits parameter to control fractional precision, which was
not present in the original.

---------

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
nsingh-habana pushed a commit to nsingh-habana/cutlass-sycl that referenced this pull request Sep 22, 2025
…ation in SYCL (intel#515)

1. Generates random data on the host and copies it to the device via
syclcompat::memcpy, improving code reuse and supporting a broader range
of element types.
2. An optional bits parameter to control fractional precision, which was
not present in the original.

---------

Signed-off-by: Chen, Xi2 <xi2.chen@intel.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants