[Public]
Yeah, seems to be a mix. I don't have a strong opinion as long as it has MIT.
Alex
________________________________
From: Kuehling, Felix <felix.kuehl...@amd.com>
Sent: Tuesday, September 6, 2022 9:46 AM
To: Jingyu Wang <jingyuwang_...@163.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Koenig, Christian <christian.koe...@amd.com>; Pan,
Xinhui <xinhui....@amd.com>; airl...@linux.ie <airl...@linux.ie>;
dan...@ffwll.ch <dan...@ffwll.ch>
Cc: amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>;
dri-de...@lists.freedesktop.org <dri-de...@lists.freedesktop.org>;
linux-ker...@vger.kernel.org <linux-ker...@vger.kernel.org>
Subject: Re: [PATCH] drm/amdgpu: cleanup coding style in amdgpu_amdkfd_gpuvm.c
Am 2022-09-05 um 04:38 schrieb Jingyu Wang:
> Fix everything checkpatch.pl complained about in amdgpu_amdkfd_gpuvm.c
>
> Signed-off-by: Jingyu Wang <jingyuwang_...@163.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> index cbd593f7d553..eff596c60c89 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> @@ -1,3 +1,4 @@
> +// SPDX-License-Identifier: MIT
I'm not sure if this is correct. We've used "GPL-2.0 OR MIT" in KFD. In
amdgpu there is currently a mix of licenses. Alex, do you want to make a
call on a consistent one to use in amdgpu?
Other than that, this patch is
Reviewed-by: Felix Kuehling <felix.kuehl...@amd.com>
> /*
> * Copyright 2014-2018 Advanced Micro Devices, Inc.
> *
> @@ -1612,6 +1613,7 @@ size_t amdgpu_amdkfd_get_available_memory(struct
> amdgpu_device *adev)
> uint64_t reserved_for_pt =
> ESTIMATE_PT_SIZE(amdgpu_amdkfd_total_mem_size);
> size_t available;
> +
> spin_lock(&kfd_mem_limit.mem_limit_lock);
> available = adev->gmc.real_vram_size
> - adev->kfd.vram_used_aligned
> @@ -2216,7 +2218,7 @@ int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct
> amdgpu_device *adev,
> {
> if (atomic_read(&adev->gmc.vm_fault_info_updated) == 1) {
> *mem = *adev->gmc.vm_fault_info;
> - mb();
> + mb(); /* make sure read happened */
> atomic_set(&adev->gmc.vm_fault_info_updated, 0);
> }
> return 0;
>
> base-commit: e47eb90a0a9ae20b82635b9b99a8d0979b757ad8
