Skip to content

Fix too relaxed check on CUDA "fast copy" (can_be_transposed) condition #17332

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

Open
wants to merge 3 commits into
base: master
Choose a base branch
from
Open

Fix too relaxed check on CUDA "fast copy" (can_be_transposed) condition #17332

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c2bfbbf
Fix too relaxed check on CUDA "fast copy" (can_be_transposed) condition
pwilkin Nov 17, 2025
5e7c26f
Argh.
pwilkin Nov 17, 2025
d51f719
Making CISC happy ;)
pwilkin Nov 17, 2025
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
3 changes: 2 additions & 1 deletion ggml/src/ggml-cuda/cpy.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -384,7 +384,8 @@ void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, gg
char * src1_ddc = (char *) src1->data;

const bool contiguous_srcs = ggml_is_contiguous(src0) && ggml_is_contiguous(src1);
const bool can_be_transposed = nb01 == (int64_t)ggml_element_size(src0) && src0->ne[3] == 1;
const bool can_be_transposed = nb01 == (int64_t)ggml_element_size(src0) &&
src0->ne[3] == 1 && nb02 == ne00 * ne01 * (int64_t)ggml_element_size(src0);

if (src0->type == src1->type && contiguous_srcs) {
GGML_ASSERT(ggml_nbytes(src0) == ggml_nbytes(src1));
Expand Down
32 changes: 32 additions & 0 deletions tests/test-backend-ops.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2800,6 +2800,33 @@ struct test_cont : public test_case {
}
};

struct test_irregular_cont : public test_case {
const ggml_type type;
const std::array<int64_t, 4> ne;

std::string vars() override {
return VARS_TO_STR2(type, ne);
}

test_irregular_cont(ggml_type type = GGML_TYPE_F32,
std::array<int64_t, 4> ne = {1, 4, 2, 1})
: type(type), ne(ne) {}

ggml_tensor * build_graph(ggml_context * ctx) override {
ggml_tensor * src = ggml_new_tensor(ctx, type, 4, ne.data());
ggml_set_param(src);
ggml_set_name(src, "src");

ggml_tensor * view = ggml_view_4d(ctx, src, src->ne[0], 1, src->ne[2], src->ne[3],
src->nb[1], src->nb[2], src->nb[3], src->nb[0] * (src->ne[1] - 1));

ggml_tensor * out = ggml_cont(ctx, view);
ggml_set_name(out, "out");

return out;
}
};

// GGML_OP_ADD
// GGML_OP_SUB
// GGML_OP_MUL
Expand Down Expand Up @@ -6956,6 +6983,11 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_cont(GGML_TYPE_BF16, {2, 1, 3 ,5}));
test_cases.emplace_back(new test_cont(GGML_TYPE_BF16, {2, 3, 5 ,7}));

test_cases.emplace_back(new test_irregular_cont());
test_cases.emplace_back(new test_irregular_cont(GGML_TYPE_F32, {1, 8, 17, 1}));
test_cases.emplace_back(new test_irregular_cont(GGML_TYPE_BF16, {1, 4, 2, 1}));
test_cases.emplace_back(new test_irregular_cont(GGML_TYPE_BF16, {1, 8, 17, 1}));

auto add_test_bin_bcast = [&](ggml_type type, std::array<int64_t, 4> ne, std::array<int, 4> nr) {
for (auto op : {ggml_add, ggml_sub, ggml_mul, ggml_div}) {
test_cases.emplace_back(new test_bin_bcast(op, type, ne, nr));
Expand Down
Loading