Merge the ND-Range Kernel lesson into Data Parallelism #388
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
The Data Parallelism and ND-Range Kernel lessons duplicated a lot of material, which became confusing when we tried presenting both in a workshop. The duplication is unneccesary and also adds to the maintenance cost. Merge the "ND-Range Kernel" lesson into "Data Parallelism" by moving all the unique slides from the latter into the former and reordering so it has a reasonable flow. The exercises were also similar, where "Data Paralellism" tested sycl::range + sycl::id version and ND-Range tested sycl::range + sycl::item as well as sycl::nd_range + sycl::nd_item. Merge the two by using most of the code from the latter, but using sycl::id instead of sycl::item. Merge the README instructions from both to retain full detail. The top-level README is adjusted to renumber all lessons after the removed one.
DuncanMcBain
approved these changes
Jan 15, 2025
Constant memory is deprecated in SYCL 2020, so remove it from the slides about the SYCL memory model.
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Outdated suggestions cannot be applied.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.
The "Data Parallelism" and "ND-Range Kernel" lessons duplicated a lot of material, which became confusing when we tried presenting both in a workshop. The duplication is unneccesary and also adds to the maintenance cost.
Merge the "ND-Range Kernel" lesson into "Data Parallelism" by moving all the unique slides from the latter into the former and reordering so it has a reasonable flow. The only significant change in the flow is how
sycl::rangekernels are introduced. Previously, the "Data Parallelism" introduced onlysycl::rangeas the SYCL execution model, and only later the "ND-Range Kernel" lesson introducednd_range, extending the knowledge. Now, thend_rangeis introduced first as the SYCL execution model and then, at the end of the same section, thesycl::rangemodel is introduced as a simplified model that is "useful for simple problems which don't require synchronization, local memory and ultimate performance". This is a true reflection of how SYCL is used in real life and sets the right expectations.The exercises were also similar, where "Data Paralellism" tested a
sycl::idkernel and "ND-Range Kernel" testedsycl::itemas well assycl::nd_item. Merge the two by using most of the code from the latter, but usingsycl::idinstead ofsycl::item. Merge the README instructions from both to retain full detail.The top-level README is adjusted to renumber all lessons after the removed one.