vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 01:24:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

vulkan: Support FA with any multiple of 8 head sizes (#15537)

Browse source 
The scalar FA shader already handled multiples of 8. The coopmat1 FA
shader assumed 16x16x16 and the shared memory allocations need the HSK
dimensions padded to a multiple of 16. NVIDIA's coopmat2 implementation
requires multiples of 16 for N and K, and needs the matrix dimensions
padded and loads clamped.

Store the FA pipelines in a map, indexed by the pipeline state.
This commit is contained in:
Jeff Bolz 2025-08-24 04:24:25 -05:00 • committed by GitHub
parent a9c6ffcbfa
commit c9a24fb932
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 143 additions and 137 deletions
Download patch file Download diff file Expand all files Collapse all files

4
tests/test-backend-ops.cpp
View file

@ -6239,8 +6239,8 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
test_cases.emplace_back(new test_timestep_embedding());
test_cases.emplace_back(new test_leaky_relu());
for (int hsk : { 64, 80, 128, 192, 256, 576 }) {
for (int hsv : { 64, 80, 128, 192, 256, 512 }) {
for (int hsk : { 40, 64, 80, 128, 192, 256, 576 }) {
for (int hsv : { 40, 64, 80, 128, 192, 256, 512 }) {
if (hsk != 192 && hsk != 576 && hsk != hsv) continue;
if (hsk == 192 && (hsv != 128 && hsv != 192)) continue;
if (hsk == 576 && hsv != 512) continue; // DeepSeek MLA