From da654ee9cb67cf52747adcc11d1b8862dc515ed1 Mon Sep 17 00:00:00 2001 From: Agnes Leroy Date: Thu, 10 Nov 2022 14:12:44 +0100 Subject: [PATCH] chore(core): fix clippy error in cuda backend and fix formatting --- src/bootstrap_wop.cuh | 12 ++++++------ src/device.cu | 15 ++++++++------- 2 files changed, 14 insertions(+), 13 deletions(-) diff --git a/src/bootstrap_wop.cuh b/src/bootstrap_wop.cuh index 6e7e27640..5ead4098d 100644 --- a/src/bootstrap_wop.cuh +++ b/src/bootstrap_wop.cuh @@ -302,8 +302,8 @@ void host_cmux_tree(void *v_stream, Torus *glwe_array_out, Torus *ggsw_in, int ggsw_size = r * polynomial_size * (glwe_dimension + 1) * (glwe_dimension + 1) * level_count; - double2 *d_ggsw_fft_in = - (double2 *)cuda_malloc_async(ggsw_size * sizeof(double), *stream, gpu_index); + double2 *d_ggsw_fft_in = (double2 *)cuda_malloc_async( + ggsw_size * sizeof(double), *stream, gpu_index); batch_fft_ggsw_vector(v_stream, d_ggsw_fft_in, ggsw_in, r, glwe_dimension, @@ -328,10 +328,10 @@ void host_cmux_tree(void *v_stream, Torus *glwe_array_out, Torus *ggsw_in, // Allocate buffers int glwe_size = (glwe_dimension + 1) * polynomial_size; - Torus *d_buffer1 = - (Torus *)cuda_malloc_async(num_lut * glwe_size * sizeof(Torus), *stream, gpu_index); - Torus *d_buffer2 = - (Torus *)cuda_malloc_async(num_lut * glwe_size * sizeof(Torus), *stream, gpu_index); + Torus *d_buffer1 = (Torus *)cuda_malloc_async( + num_lut * glwe_size * sizeof(Torus), *stream, gpu_index); + Torus *d_buffer2 = (Torus *)cuda_malloc_async( + num_lut * glwe_size * sizeof(Torus), *stream, gpu_index); checkCudaErrors(cudaMemcpyAsync(d_buffer1, lut_vector, num_lut * glwe_size * sizeof(Torus), diff --git a/src/device.cu b/src/device.cu index 974d3bd31..e77fbf9e2 100644 --- a/src/device.cu +++ b/src/device.cu @@ -33,14 +33,15 @@ void *cuda_malloc(uint64_t size, uint32_t gpu_index) { /// Allocates a size-byte array at the device memory. Tries to do it /// asynchronously. -void *cuda_malloc_async(uint64_t size, cudaStream_t stream, uint32_t gpu_index) { +void *cuda_malloc_async(uint64_t size, cudaStream_t stream, + uint32_t gpu_index) { void *ptr; int support_async_alloc; - checkCudaErrors(cudaDeviceGetAttribute(&support_async_alloc, cudaDevAttrMemoryPoolsSupported, - gpu_index)); + checkCudaErrors(cudaDeviceGetAttribute( + &support_async_alloc, cudaDevAttrMemoryPoolsSupported, gpu_index)); - if(support_async_alloc) + if (support_async_alloc) checkCudaErrors(cudaMallocAsync((void **)&ptr, size, stream)); else checkCudaErrors(cudaMalloc((void **)&ptr, size)); @@ -160,10 +161,10 @@ int cuda_drop(void *ptr, uint32_t gpu_index) { int cuda_drop_async(void *ptr, cudaStream_t stream, uint32_t gpu_index) { int support_async_alloc; - checkCudaErrors(cudaDeviceGetAttribute(&support_async_alloc, cudaDevAttrMemoryPoolsSupported, - gpu_index)); + checkCudaErrors(cudaDeviceGetAttribute( + &support_async_alloc, cudaDevAttrMemoryPoolsSupported, gpu_index)); - if(support_async_alloc) + if (support_async_alloc) checkCudaErrors(cudaFreeAsync(ptr, stream)); else checkCudaErrors(cudaFree(ptr));