Am 21.07.2017 um 04:52 schrieb Xiangliang.Yu:
> From: Xiangliang Yu <Xiangliang.Yu at amd.com>
>
> Change ioremap mode from noncache to write combine and it will
> reduce the read vbios time from 188ms to 8ms.
Nice improvement, but I think you need to improve the commit message a bit.
Changing the ioremap mode from noncache to write combine should actually
not improve the read performance (reads are uncached in both modes).
I think this patch makes a difference because VRAM is usually marked
write combined and so ioremap needs to change that for no good reason
and that in turn takes quite some time.
I think you should write something like "Avoid changing the mapping mode
of VRAM while reading the vbios from it, this reduce the time from....".
>
> Signed-off-by: Xiangliang Yu <Xiangliang.Yu at amd.com>
> Signed-off-by: Xiangliang.Yu <Xiangliang.Yu at amd.com>
You added the signed-off-by line twice, please fix this.
With that fixed and the commit message improved a bit the patch is
Reviewed-by: Christian König <christian.koenig at amd.com>.
Regards,
Christian.
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
> index ea3a250..c21adf6 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
> @@ -104,7 +104,7 @@ static bool igp_read_bios_from_vram(struct amdgpu_device *adev)
>
> adev->bios = NULL;
> vram_base = pci_resource_start(adev->pdev, 0);
> - bios = ioremap(vram_base, size);
> + bios = ioremap_wc(vram_base, size);
> if (!bios) {
> return false;
> }
