* [PATCH v2 2/2] Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory"
2023-09-08 14:55 [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM Hamza Mahfooz
@ 2023-09-08 14:55 ` Hamza Mahfooz
2023-09-08 15:05 ` Harry Wentland
2023-09-08 15:06 ` [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM Alex Deucher
2023-09-11 8:38 ` Christian König
2 siblings, 1 reply; 5+ messages in thread
From: Hamza Mahfooz @ 2023-09-08 14:55 UTC (permalink / raw)
To: amd-gfx
Cc: Hamza Mahfooz, stable, Alex Deucher, Christian König,
Pan, Xinhui, David Airlie, Daniel Vetter, Harry Wentland, Leo Li,
Rodrigo Siqueira, Lijo Lazar, Hawking Zhang, Le Ma, Candice Li,
Lang Yu, Maarten Lankhorst, Victor Zhao, Mario Limonciello,
Srinivasan Shanmugam, Shashank Sharma, Qingqing Zhuo,
Aurabindo Pillai, Hersen Wu, Stylon Wang, Wayne Lin, Alan Liu,
Luben Tuikov, dri-devel, linux-kernel
This reverts commit 5b7a256c982636ebc4f16b708b40ff56d33c8a86.
Since, we now have an actual fix for this issue, we can get rid of this
workaround as it can cause pin failures if enough VRAM isn't carved out
by the BIOS.
Cc: stable@vger.kernel.org # 6.1+
Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
---
v2: new to the series
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 -
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 26 -------------------
.../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++--
3 files changed, 3 insertions(+), 29 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 83a9607a87b8..3a86d11d1605 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1316,7 +1316,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
void amdgpu_device_pci_config_reset(struct amdgpu_device *adev);
int amdgpu_device_pci_reset(struct amdgpu_device *adev);
bool amdgpu_device_need_post(struct amdgpu_device *adev);
-bool amdgpu_sg_display_supported(struct amdgpu_device *adev);
bool amdgpu_device_pcie_dynamic_switching_supported(void);
bool amdgpu_device_should_use_aspm(struct amdgpu_device *adev);
bool amdgpu_device_aspm_support_quirk(void);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 5f32e8d4f3d3..3d540b0cf0e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1358,32 +1358,6 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev)
return true;
}
-/*
- * On APUs with >= 64GB white flickering has been observed w/ SG enabled.
- * Disable S/G on such systems until we have a proper fix.
- * https://gitlab.freedesktop.org/drm/amd/-/issues/2354
- * https://gitlab.freedesktop.org/drm/amd/-/issues/2735
- */
-bool amdgpu_sg_display_supported(struct amdgpu_device *adev)
-{
- switch (amdgpu_sg_display) {
- case -1:
- break;
- case 0:
- return false;
- case 1:
- return true;
- default:
- return false;
- }
- if ((totalram_pages() << (PAGE_SHIFT - 10)) +
- (adev->gmc.real_vram_size / 1024) >= 64000000) {
- DRM_WARN("Disabling S/G due to >=64GB RAM\n");
- return false;
- }
- return true;
-}
-
/*
* Intel hosts such as Raptor Lake and Sapphire Rapids don't support dynamic
* speed switching. Until we have confirmation from Intel that a specific host
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 5f14cd9391ca..740a6fcafe4c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1654,8 +1654,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
}
break;
}
- if (init_data.flags.gpu_vm_support)
- init_data.flags.gpu_vm_support = amdgpu_sg_display_supported(adev);
+ if (init_data.flags.gpu_vm_support &&
+ (amdgpu_sg_display == 0))
+ init_data.flags.gpu_vm_support = false;
if (init_data.flags.gpu_vm_support)
adev->mode_info.gpu_vm_support = true;
--
2.41.0
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [PATCH v2 2/2] Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory"
2023-09-08 14:55 ` [PATCH v2 2/2] Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory" Hamza Mahfooz
@ 2023-09-08 15:05 ` Harry Wentland
0 siblings, 0 replies; 5+ messages in thread
From: Harry Wentland @ 2023-09-08 15:05 UTC (permalink / raw)
To: Hamza Mahfooz, amd-gfx
Cc: stable, Alex Deucher, Christian König, Pan, Xinhui,
David Airlie, Daniel Vetter, Leo Li, Rodrigo Siqueira, Lijo Lazar,
Hawking Zhang, Le Ma, Candice Li, Lang Yu, Maarten Lankhorst,
Victor Zhao, Mario Limonciello, Srinivasan Shanmugam,
Shashank Sharma, Qingqing Zhuo, Aurabindo Pillai, Hersen Wu,
Stylon Wang, Wayne Lin, Alan Liu, Luben Tuikov, dri-devel,
linux-kernel
Series is
Acked-by: Harry Wentland <harry.wentland@amd.com>
Harry
On 2023-09-08 10:55, Hamza Mahfooz wrote:
> This reverts commit 5b7a256c982636ebc4f16b708b40ff56d33c8a86.
>
> Since, we now have an actual fix for this issue, we can get rid of this
> workaround as it can cause pin failures if enough VRAM isn't carved out
> by the BIOS.
>
> Cc: stable@vger.kernel.org # 6.1+
> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
> ---
> v2: new to the series
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 -
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 26 -------------------
> .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++--
> 3 files changed, 3 insertions(+), 29 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> index 83a9607a87b8..3a86d11d1605 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> @@ -1316,7 +1316,6 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> void amdgpu_device_pci_config_reset(struct amdgpu_device *adev);
> int amdgpu_device_pci_reset(struct amdgpu_device *adev);
> bool amdgpu_device_need_post(struct amdgpu_device *adev);
> -bool amdgpu_sg_display_supported(struct amdgpu_device *adev);
> bool amdgpu_device_pcie_dynamic_switching_supported(void);
> bool amdgpu_device_should_use_aspm(struct amdgpu_device *adev);
> bool amdgpu_device_aspm_support_quirk(void);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 5f32e8d4f3d3..3d540b0cf0e1 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -1358,32 +1358,6 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev)
> return true;
> }
>
> -/*
> - * On APUs with >= 64GB white flickering has been observed w/ SG enabled.
> - * Disable S/G on such systems until we have a proper fix.
> - * https://gitlab.freedesktop.org/drm/amd/-/issues/2354
> - * https://gitlab.freedesktop.org/drm/amd/-/issues/2735
> - */
> -bool amdgpu_sg_display_supported(struct amdgpu_device *adev)
> -{
> - switch (amdgpu_sg_display) {
> - case -1:
> - break;
> - case 0:
> - return false;
> - case 1:
> - return true;
> - default:
> - return false;
> - }
> - if ((totalram_pages() << (PAGE_SHIFT - 10)) +
> - (adev->gmc.real_vram_size / 1024) >= 64000000) {
> - DRM_WARN("Disabling S/G due to >=64GB RAM\n");
> - return false;
> - }
> - return true;
> -}
> -
> /*
> * Intel hosts such as Raptor Lake and Sapphire Rapids don't support dynamic
> * speed switching. Until we have confirmation from Intel that a specific host
> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> index 5f14cd9391ca..740a6fcafe4c 100644
> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> @@ -1654,8 +1654,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
> }
> break;
> }
> - if (init_data.flags.gpu_vm_support)
> - init_data.flags.gpu_vm_support = amdgpu_sg_display_supported(adev);
> + if (init_data.flags.gpu_vm_support &&
> + (amdgpu_sg_display == 0))
> + init_data.flags.gpu_vm_support = false;
>
> if (init_data.flags.gpu_vm_support)
> adev->mode_info.gpu_vm_support = true;
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM
2023-09-08 14:55 [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM Hamza Mahfooz
2023-09-08 14:55 ` [PATCH v2 2/2] Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory" Hamza Mahfooz
@ 2023-09-08 15:06 ` Alex Deucher
2023-09-11 8:38 ` Christian König
2 siblings, 0 replies; 5+ messages in thread
From: Alex Deucher @ 2023-09-08 15:06 UTC (permalink / raw)
To: Hamza Mahfooz
Cc: amd-gfx, Alan Liu, Lijo Lazar, dri-devel, Mario Limonciello,
David Airlie, Shashank Sharma, Rodrigo Siqueira, Aurabindo Pillai,
Qingqing Zhuo, Harry Wentland, Stylon Wang, Victor Zhao,
Srinivasan Shanmugam, Leo Li, Maarten Lankhorst, Le Ma, Hersen Wu,
Yifan Zhang, Felix Kuehling, Pan, Xinhui, linux-kernel, stable,
Hawking Zhang, Daniel Vetter, Wayne Lin, Alex Deucher, Candice Li,
Lang Yu, Christian König, Luben Tuikov
Series is:
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
On Fri, Sep 8, 2023 at 10:56 AM Hamza Mahfooz <hamza.mahfooz@amd.com> wrote:
>
> From: Yifan Zhang <yifan1.zhang@amd.com>
>
> Dropping bit 31:4 of page table base is wrong, it makes page table
> base points to wrong address if phys addr is beyond 64GB; dropping
> page_table_start/end bit 31:4 is unnecessary since dcn20_vmid_setup
> will do that. Also, while we are at it, cleanup the assignments using
> upper_32_bits()/lower_32_bits() and AMDGPU_GPU_PAGE_SHIFT.
>
> Cc: stable@vger.kernel.org
> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2354
> Fixes: 81d0bcf99009 ("drm/amdgpu: make display pinning more flexible (v2)")
> Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
> ---
> v2: use upper_32_bits()/lower_32_bits() and AMDGPU_GPU_PAGE_SHIFT
> ---
> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> index 1bb1a394f55f..5f14cd9391ca 100644
> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> @@ -1283,11 +1283,15 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_
>
> pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
>
> - page_table_start.high_part = (u32)(adev->gmc.gart_start >> 44) & 0xF;
> - page_table_start.low_part = (u32)(adev->gmc.gart_start >> 12);
> - page_table_end.high_part = (u32)(adev->gmc.gart_end >> 44) & 0xF;
> - page_table_end.low_part = (u32)(adev->gmc.gart_end >> 12);
> - page_table_base.high_part = upper_32_bits(pt_base) & 0xF;
> + page_table_start.high_part = upper_32_bits(adev->gmc.gart_start >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_start.low_part = lower_32_bits(adev->gmc.gart_start >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_end.high_part = upper_32_bits(adev->gmc.gart_end >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_end.low_part = lower_32_bits(adev->gmc.gart_end >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_base.high_part = upper_32_bits(pt_base);
> page_table_base.low_part = lower_32_bits(pt_base);
>
> pa_config->system_aperture.start_addr = (uint64_t)logical_addr_low << 18;
> --
> 2.41.0
>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM
2023-09-08 14:55 [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM Hamza Mahfooz
2023-09-08 14:55 ` [PATCH v2 2/2] Revert "drm/amd: Disable S/G for APUs when 64GB or more host memory" Hamza Mahfooz
2023-09-08 15:06 ` [PATCH v2 1/2] drm/amd/display: fix the white screen issue when >= 64GB DRAM Alex Deucher
@ 2023-09-11 8:38 ` Christian König
2 siblings, 0 replies; 5+ messages in thread
From: Christian König @ 2023-09-11 8:38 UTC (permalink / raw)
To: Hamza Mahfooz, amd-gfx
Cc: Alan Liu, Lijo Lazar, dri-devel, Mario Limonciello, David Airlie,
Shashank Sharma, Rodrigo Siqueira, Aurabindo Pillai,
Qingqing Zhuo, Harry Wentland, Stylon Wang, Victor Zhao,
Srinivasan Shanmugam, Leo Li, Maarten Lankhorst, Le Ma, Hersen Wu,
Yifan Zhang, Felix Kuehling, Pan, Xinhui, linux-kernel, stable,
Hawking Zhang, Daniel Vetter, Wayne Lin, Alex Deucher, Candice Li,
Lang Yu, Christian König, Luben Tuikov
Am 08.09.23 um 16:55 schrieb Hamza Mahfooz:
> From: Yifan Zhang <yifan1.zhang@amd.com>
>
> Dropping bit 31:4 of page table base is wrong, it makes page table
> base points to wrong address if phys addr is beyond 64GB; dropping
> page_table_start/end bit 31:4 is unnecessary since dcn20_vmid_setup
> will do that. Also, while we are at it, cleanup the assignments using
> upper_32_bits()/lower_32_bits() and AMDGPU_GPU_PAGE_SHIFT.
>
> Cc: stable@vger.kernel.org
> Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2354
> Fixes: 81d0bcf99009 ("drm/amdgpu: make display pinning more flexible (v2)")
> Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com> for the series
as well.
> ---
> v2: use upper_32_bits()/lower_32_bits() and AMDGPU_GPU_PAGE_SHIFT
> ---
> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 14 +++++++++-----
> 1 file changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> index 1bb1a394f55f..5f14cd9391ca 100644
> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
> @@ -1283,11 +1283,15 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_
>
> pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
>
> - page_table_start.high_part = (u32)(adev->gmc.gart_start >> 44) & 0xF;
> - page_table_start.low_part = (u32)(adev->gmc.gart_start >> 12);
> - page_table_end.high_part = (u32)(adev->gmc.gart_end >> 44) & 0xF;
> - page_table_end.low_part = (u32)(adev->gmc.gart_end >> 12);
> - page_table_base.high_part = upper_32_bits(pt_base) & 0xF;
> + page_table_start.high_part = upper_32_bits(adev->gmc.gart_start >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_start.low_part = lower_32_bits(adev->gmc.gart_start >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_end.high_part = upper_32_bits(adev->gmc.gart_end >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_end.low_part = lower_32_bits(adev->gmc.gart_end >>
> + AMDGPU_GPU_PAGE_SHIFT);
> + page_table_base.high_part = upper_32_bits(pt_base);
> page_table_base.low_part = lower_32_bits(pt_base);
>
> pa_config->system_aperture.start_addr = (uint64_t)logical_addr_low << 18;
^ permalink raw reply [flat|nested] 5+ messages in thread