Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

src: gpu: intel: jit: clarify the naming for kernel/thread grids #2792

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

hidefromkgb
Copy link
Contributor

This PR resolves the kernel / threadgroup terminological mixup in the jit:ir generator.

@hidefromkgb hidefromkgb requested a review from a team as a code owner March 1, 2025 00:35
@github-actions github-actions bot added the platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel label Mar 1, 2025
@hidefromkgb hidefromkgb force-pushed the aguskov/jit_ir_kg_tg_refactor branch from 25cd603 to 599ff32 Compare March 4, 2025 09:55
@hidefromkgb hidefromkgb force-pushed the aguskov/jit_ir_kg_tg_refactor branch from fa46623 to 59cd4b6 Compare March 7, 2025 07:38
pvar_tile_t inner;
for (auto &d : conv_index_dims(cfg.prb().prop_kind())) {
dim_t iter = cfg.iter_dim(d);
dim_t tg = cfg.thread_group_dim(d);
dim_t thr = cfg.thread_dim(d);
Copy link
Contributor

Choose a reason for hiding this comment

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

I think tg = cfg.thread_dim(d); still works (tg - thread group size).

Copy link
Contributor Author

@hidefromkgb hidefromkgb Mar 8, 2025

Choose a reason for hiding this comment

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

I still think it's thr, but yes, П(thr) = tg_size — whereas П(tg) = tg_count.

@hidefromkgb hidefromkgb force-pushed the aguskov/jit_ir_kg_tg_refactor branch 2 times, most recently from 2764315 to a78420c Compare March 8, 2025 05:30
@hidefromkgb
Copy link
Contributor Author

make test
set test_scope=NIGHTLY
disable test_device_cpu
disable benchdnn_all
enable benchdnn_conv
enable benchdnn_deconv
enable benchdnn_reorder
enable benchdnn_pool
enable arch_gpu_xe-hpc
enable arch_gpu_xe-hpg-atsm
enable arch_gpu_xe-hpg-dg2
enable arch_gpu_xe-lp
enable arch_gpu_xe-lpg
enable arch_gpu_xe-lpg+
enable arch_gpu_xe2-hpg-bmg
enable arch_gpu_xe2-lpg
enable arch_gpu_xe3-lpg

@hidefromkgb hidefromkgb force-pushed the aguskov/jit_ir_kg_tg_refactor branch from a78420c to edec79e Compare March 13, 2025 07:20
@hidefromkgb hidefromkgb requested a review from a team as a code owner March 13, 2025 07:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants