2020-04-21 02:28:06

by 赵军奎

Subject: [PATCH V2] amdgpu: optimization-- reduce noneed mutex_lock area

Maybe we could reduce the mutex_lock(&mem->lock)`s protected code area,
and noneed to protect pr_debug.

Signed-off-by: Bernard Zhao <[email protected]>

Changes since V1:
*commit message improve

Link for V1:
*https://lore.kernel.org/patchwork/patch/1226588/
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 327317c54f7c..3c3769e57174 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1285,17 +1285,18 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
struct bo_vm_reservation_context ctx;
struct ttm_validate_buffer *bo_list_entry;
int ret;
+ unsigned int mapped_to_gpu_memory;

mutex_lock(&mem->lock);
+ mapped_to_gpu_memory = mem->mapped_to_gpu_memory;
+ mutex_unlock(&mem->lock);

- if (mem->mapped_to_gpu_memory > 0) {
+ if (mapped_to_gpu_memory > 0) {
pr_debug("BO VA 0x%llx size 0x%lx is still mapped.\n",
mem->va, bo_size);
- mutex_unlock(&mem->lock);
return -EBUSY;
}

- mutex_unlock(&mem->lock);
/* lock is not needed after this, since mem is unused and will
* be freed anyway
*/
--
2.26.2

2020-04-21 04:22:10

by Felix Kuehling

Subject: Re: [PATCH V2] amdgpu: optimization-- reduce noneed mutex_lock area

Hi Bernard,

Thank you for the patch. Please see some comments inline.

Am 2020-04-20 um 10:25 p.m. schrieb Bernard Zhao:
> Maybe we could reduce the mutex_lock(&mem->lock)`s protected code area,
> and noneed to protect pr_debug.

Typo: noneed -> no need


>
> Signed-off-by: Bernard Zhao <[email protected]>
>
> Changes since V1:
> *commit message improve
>
> Link for V1:
> *https://lore.kernel.org/patchwork/patch/1226588/
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> index 327317c54f7c..3c3769e57174 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> @@ -1285,17 +1285,18 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
> struct bo_vm_reservation_context ctx;
> struct ttm_validate_buffer *bo_list_entry;
> int ret;
> + unsigned int mapped_to_gpu_memory;
>
> mutex_lock(&mem->lock);
> + mapped_to_gpu_memory = mem->mapped_to_gpu_memory;
> + mutex_unlock(&mem->lock);
>
> - if (mem->mapped_to_gpu_memory > 0) {
> + if (mapped_to_gpu_memory > 0) {
> pr_debug("BO VA 0x%llx size 0x%lx is still mapped.\n",
> mem->va, bo_size);
> - mutex_unlock(&mem->lock);
> return -EBUSY;
> }
>
> - mutex_unlock(&mem->lock);
> /* lock is not needed after this, since mem is unused and will
> * be freed anyway
> */

Please move this comment along with the mutex_unlock.

Regards,
  Felix