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.

Sign up for GitHub

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

[GPU] Conv and Gemm fix #2290

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
+3 −3
Open

[GPU] Conv and Gemm fix #2290

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
183c6d7
xe: jit: conv: cast mask to s16 before mad
dyoussif Dec 18, 2024
39d01dc
xe: jit: gemm: align kxq lookahead with kxq_load
dyoussif Dec 18, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/gpu/intel/jit/conv/zp_plan.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1268,7 +1268,7 @@ class zp_comp_apply_plan_t : public base_plan_t {
const expr_t &c_buf, const split_dispatcher_t &sd,
int subtile_idx) const {
const auto comp_type = comp_layout_.type();
const auto mask_type = mask_layout_.type();
const auto mask_type = type_t::s16();
const dim_t kw_dim = comp_layout_.dim(comp_kw_idx_);
std::vector<int> comp_off;
std::vector<int> mask_off;
Expand Down
4 changes: 2 additions & 2 deletions src/gpu/intel/jit/gemm/generator/pieces/k_loop.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -960,8 +960,8 @@ void BLASKernelGenerator<hw>::kLoop(KLoop type, const GEMMProblem &problem, GEMM
ls.swapLast2();

// A/B 2D quantization parameter loads.
auto reqLoadAq = every(kaq_load) | lookahead(ka_repackMain);
auto reqLoadBq = every(kbq_load) | lookahead(kb_loadMain);
auto reqLoadAq = every(kaq_load) | lookahead(kaq_load);
auto reqLoadBq = every(kbq_load) | lookahead(kbq_load);
auto reqLoadAqLate = every(kaq_loadLate) | lookahead(kaq_loadLate);
auto reqLoadBqLate = every(kbq_loadLate) | lookahead(kbq_loadLate);

Expand Down
Loading