Skip to content

Conversation

@bernhardmgruber
Copy link
Contributor

@bernhardmgruber bernhardmgruber commented Oct 7, 2025

This avoids needing to specify the cache modifier in two places and eases integration of #6077

No relevant SASS changes on sm86 and sm120 for cub.cpp17.test.device_merge.lid_0. Just a few symbol changes and constants like:

<         /*0110*/                   HFMA2 R8, -RZ, RZ, 0, 6.258487701416015625e-06 ;           /* 0x00000069ff087431 */
---
>         /*0110*/                   HFMA2 R8, -RZ, RZ, 0, 5.900859832763671875e-06 ;           /* 0x00000063ff087431 */

This avoids needing to specify the cache modifier in two places and eases integration of NVIDIA#6077
@bernhardmgruber bernhardmgruber requested a review from a team as a code owner October 7, 2025 11:06
@github-project-automation github-project-automation bot moved this to Todo in CCCL Oct 7, 2025
@cccl-authenticator-app cccl-authenticator-app bot moved this from Todo to In Review in CCCL Oct 7, 2025
@github-actions
Copy link
Contributor

github-actions bot commented Oct 7, 2025

🥳 CI Workflow Results

🟩 Finished in 4h 43m: Pass: 100%/185 | Total: 6d 20h | Max: 4h 42m | Hits: 79%/190024

See results here.

merge_sort::reg_to_shared<threads_per_block>(&storage.keys_shared[0], keys_loc);
__syncthreads();
{
auto keys1_in_cm = try_make_cache_modified_iterator<Policy::LOAD_MODIFIER>(keys1_in);
Copy link
Contributor

Choose a reason for hiding this comment

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

nit. cm for "cache modifier` is a really short abbreviation and not very intuitive

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I agree, but the use of the variable is so local that I didn't bother with a long and descriptive name.

@bernhardmgruber bernhardmgruber merged commit 66938f1 into NVIDIA:main Oct 7, 2025
197 of 198 checks passed
@github-project-automation github-project-automation bot moved this from In Review to Done in CCCL Oct 7, 2025
@bernhardmgruber bernhardmgruber deleted the ref_cm_iter_merge branch October 7, 2025 22:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

2 participants