zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-12-16 01:49:05 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Fix building with CUDA < 12.8 (#2782)
Some checks failed
Build and Test / check_lint (push) Has been cancelled
Details
Build and Test / linux_build_and_test (ubuntu-22.04) (push) Has been cancelled
Details
Build and Test / linux_build_and_test (ubuntu-22.04-arm) (push) Has been cancelled
Details
Build and Test / mac_build_and_test (14.0) (push) Has been cancelled
Details
Build and Test / mac_build_and_test (15.0) (push) Has been cancelled
Details
Build and Test / cuda_build_and_test (cuda-12.6) (push) Has been cancelled
Details
Build and Test / cuda_build_and_test (cuda-12.9) (push) Has been cancelled
Details
Build and Test / build_documentation (push) Has been cancelled
Details
Build and Test / Linux Fedora CPP Build (aarch64) (push) Has been cancelled
Details
Build and Test / Linux Fedora CPP Build (x86_64) (push) Has been cancelled
Details
Nightly Build / build_linux_release (3.10) (push) Has been cancelled
Details
Nightly Build / build_linux_release (3.14) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.11, ubuntu-22.04) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.11, ubuntu-22.04-arm) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.12, ubuntu-22.04) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.12, ubuntu-22.04-arm) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.13, ubuntu-22.04) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.13, ubuntu-22.04-arm) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.14, ubuntu-22.04) (push) Has been cancelled
Details
Nightly Build / build_linux_with_tests (3.14, ubuntu-22.04-arm) (push) Has been cancelled
Details
Nightly Build / build_mac_release (3.10) (push) Has been cancelled
Details
Nightly Build / build_mac_release (3.13) (push) Has been cancelled
Details
Nightly Build / build_cuda_release (push) Has been cancelled
Details

Browse Source
This commit is contained in:
Cheng
2025-11-18 12:55:19 +09:00
committed by GitHub
parent 35f81728f1
commit 940f4c7818
40 changed files with 91 additions and 113 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
mlx/backend/cuda/rms_norm.cu
View File

@@ -190,7 +190,7 @@ void RMSNorm::eval_gpu(
out.copy_shared_buffer(x);
} else {
out.set_data(
cu::malloc_async(x.data_size() * x.itemsize(), encoder.stream()),
cu::malloc_async(x.data_size() * x.itemsize(), encoder),
x.data_size(),
x.strides(),
x.flags());
@@ -274,7 +274,7 @@ void RMSNormVJP::eval_gpu(
gx.copy_shared_buffer(g);
g_in_gx = true;
} else {
gx.set_data(cu::malloc_async(gx.nbytes(), encoder.stream()));
gx.set_data(cu::malloc_async(gx.nbytes(), encoder));
}
if (g_copied && !g_in_gx) {
encoder.add_temporary(g);
@@ -292,7 +292,7 @@ void RMSNormVJP::eval_gpu(
if (!g_in_gx && donate_g) {
gw_temp.copy_shared_buffer(g);
} else {
gw_temp.set_data(cu::malloc_async(gw_temp.nbytes(), encoder.stream()));
gw_temp.set_data(cu::malloc_async(gw_temp.nbytes(), encoder));
encoder.add_temporary(gw_temp);
}
}