[PATCH 1/2] drm/radeon: Don't use WC for VRAM if !RADEON_GEM_GTT_WC

Tiezhu Yang yangtiezhu at loongson.cn
Wed Sep 9 09:39:01 UTC 2020


On 09/09/2020 12:21 PM, Huacai Chen wrote:
> Though RADEON_GEM_GTT_WC is initially used for GTT, but this flag is
> bound to drm_arch_can_wc_memory(), and if arch doesn't support WC, then
> VRAM should not use WC.

+cc RADEON and AMDGPU DRM DRIVERS maintainer
Alex Deucher <alexander.deucher at amd.com>
Christian König <christian.koenig at amd.com>
amd-gfx at lists.freedesktop.org

Hi all,

In the current code, if CONFIG_CPU_LOONGSON64 is set, 
drm_arch_can_wc_memory()
returns false, and then bo->flags clears the flag RADEON_GEM_GTT_WC, so with
this patch, TTM_PL_FLAG_WC of VRAM is removed on the Loongson platform,
the writecombine issue for Loongson64 can be fixed [1].

I find this is done by commit 221004c66a58 ("drm: Loongson-3 doesn't fully
support wc memory"), but I want to know why drm_arch_can_wc_memory() returns
false for Loongson64, is there some historical reasons?

include/drm/drm_cache.h
static inline bool drm_arch_can_wc_memory(void)
{
#if defined(CONFIG_PPC) && !defined(CONFIG_NOT_COHERENT_CACHE)
     return false;
#elif defined(CONFIG_MIPS) && defined(CONFIG_CPU_LOONGSON64)
     return false;
#elif defined(CONFIG_ARM) || defined(CONFIG_ARM64)
...
}

drivers/gpu/drm/radeon/radeon_object.c
int radeon_bo_create()
{
...
     if (!drm_arch_can_wc_memory())
         bo->flags &= ~RADEON_GEM_GTT_WC;
...
}

[1] https://lore.kernel.org/patchwork/patch/1285542/
gpu/drm: Remove TTM_PL_FLAG_WC of VRAM to fix writecombine issue for 
Loongson64
[2] 
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=221004c66a58

Thanks,
Tiezhu

>
> Signed-off-by: Huacai Chen <chenhc at lemote.com>
> ---
>   drivers/gpu/drm/radeon/radeon_object.c | 14 ++++++++++----
>   1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
> index f3dee01..07b82d9 100644
> --- a/drivers/gpu/drm/radeon/radeon_object.c
> +++ b/drivers/gpu/drm/radeon/radeon_object.c
> @@ -117,10 +117,16 @@ void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain)
>   						     TTM_PL_FLAG_VRAM;
>   		}
>   
> -		rbo->placements[c].fpfn = 0;
> -		rbo->placements[c++].flags = TTM_PL_FLAG_WC |
> -					     TTM_PL_FLAG_UNCACHED |
> -					     TTM_PL_FLAG_VRAM;
> +		if (rbo->flags & RADEON_GEM_GTT_WC) {
> +			rbo->placements[c].fpfn = 0;
> +			rbo->placements[c++].flags = TTM_PL_FLAG_WC |
> +						     TTM_PL_FLAG_UNCACHED |
> +						     TTM_PL_FLAG_VRAM;
> +		} else {
> +			rbo->placements[c].fpfn = 0;
> +			rbo->placements[c++].flags = TTM_PL_FLAG_UNCACHED |
> +						     TTM_PL_FLAG_VRAM;
> +		}
>   	}
>   
>   	if (domain & RADEON_GEM_DOMAIN_GTT) {



More information about the amd-gfx mailing list