nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 8 Packages Projects Releases Wiki Activity

ggml: support concat for scalar types at cuda backend (#24011)

Browse Source 
* cuda: support concat for scalar types

* Update concat.cu

* fix metal ci issue
This commit is contained in:
ZihaoMu
2026-06-12 09:32:44 +03:00
committed by GitHub
parent 099ea76fb4
commit 85f99dca8b
4 changed files with 106 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/test-backend-ops.cpp
+5
View File
@@ -8849,7 +8849,12 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
for (int v : { 0, 1, 2, 3 }) {
for (int dim : { 0, 1, 2, 3, }) {
test_cases.emplace_back(new test_concat(GGML_TYPE_F32, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_F16, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_BF16, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_I8, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_I16, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_I32, {11, 12, 13, 14}, 7, dim, v));
test_cases.emplace_back(new test_concat(GGML_TYPE_I64, {11, 12, 13, 14}, 7, dim, v));
}
}