Re: [PATCH 2/7] drm/amdgpu: use scheduler load balancing for SDMA CS

On 08/02/2018 06:09 PM, Christian König wrote: Am 02.08.2018 um 07:50 schrieb Zhang, Jerry (Junwei): On 08/01/2018 07:31 PM, Christian König wrote: Start to use the scheduler load balancing for userspace SDMA command submissions. In this case, each SDMA could load all SDMA(instances) rqs, an

Re: [PATCH libdrm 1/6] amdgpu: always add all BOs to lockup table

On 08/02/2018 10:55 PM, Michel Dänzer wrote: On 2018-08-02 04:04 PM, Christian König wrote: This way we can always find a BO structure by its handle. Signed-off-by: Christian König Typo in the shortlog: should be "lookup" instead of "lockup". Also, this patch should really be after patch 4.

Re: [PATCH libdrm 3/6] amdgpu: add handle table implementation

On 08/02/2018 10:04 PM, Christian König wrote: The kernel handles are dense and the kernel always tries to use the lowest free id. Use this to implement a more efficient handle table by using a resizeable array instead of a hash. Signed-off-by: Christian König --- amdgpu/Makefile.sources | 4

Re: [PATCH] drm/scheduler: fix setting the priorty for entities

Hello Christian, AMD guys, this one _together_ with these series [PATCH 1/7] drm/amdgpu: use new scheduler load balancing for VMs https://lists.freedesktop.org/archives/amd-gfx/2018-August/024802.html on top of amd-staging-drm-next 53d5f1e4a6d9 freeze whole system (Intel Xeon X3470, RX580) duri

Re: [PATCH] drm/amd/display: Only require EDID read for HDMI and DVI

On 08/02/2018 03:38 PM, Harry Wentland wrote: [Why] VGA sometimes has trouble retrieving the EDID on very long cables, KVM switches, or old displays. [How] Only require EDID read for HDMI and DVI and exempt other types (DP, VGA). We currently don't support VGA but if anyone adds support in the f

Re: [PATCH] drm/amd/display: Only require EDID read for HDMI and DVI

On Thu, Aug 2, 2018 at 3:38 PM, Harry Wentland wrote: > [Why] > VGA sometimes has trouble retrieving the EDID on very long cables, KVM > switches, or old displays. > > [How] > Only require EDID read for HDMI and DVI and exempt other types (DP, > VGA). We currently don't support VGA but if anyone a

[PATCH] drm/amd/display: Only require EDID read for HDMI and DVI

[Why] VGA sometimes has trouble retrieving the EDID on very long cables, KVM switches, or old displays. [How] Only require EDID read for HDMI and DVI and exempt other types (DP, VGA). We currently don't support VGA but if anyone adds support in the future this might get overlooked. Signed-off-by:

Re: [PATCH v2] drm/amd/display: Report non-DP display as disconnected without EDID

On 2018-08-02 04:06 AM, Michel Dänzer wrote: > On 2018-08-01 09:51 PM, Harry Wentland wrote: >> [Why] >> Some boards seem to have a problem where HPD is high on HDMI even though >> no display is connected. We don't want to report these as connected. DP >> spec still requires us to report DP displ

Re: [PATCH xf86-video-ati 1/2] glamor: Set RADEON_CREATE_PIXMAP_DRI2 for DRI3 pixmaps

On Thu, Aug 2, 2018 at 1:05 PM, Michel Dänzer wrote: > From: Michel Dänzer > > Not doing this resulted in falling back to software for DRI3 client > presentation operations with ShadowPrimary. > > (Ported from amdgpu commit 2989d40ef74d9966e8e8df2ef7727b2cc48d4960) > > Signed-off-by: Michel Dänze

Re: [PATCH xf86-video-amdgpu] glamor: Use glamor_egl_create_textured_pixmap_from_gbm_bo when possible

On Thu, Aug 2, 2018 at 12:47 PM, Michel Dänzer wrote: > From: Michel Dänzer > > Inspired by the modesetting driver. > > (Ported from radeon commit db28d35ce9fd07a2a4703f3df0633d4c8291ff9b) > > Signed-off-by: Michel Dänzer Reviewed-by: Alex Deucher > --- > src/amdgpu_glamor.c | 32 +++

[PATCH xf86-video-ati 1/2] glamor: Set RADEON_CREATE_PIXMAP_DRI2 for DRI3 pixmaps

From: Michel Dänzer Not doing this resulted in falling back to software for DRI3 client presentation operations with ShadowPrimary. (Ported from amdgpu commit 2989d40ef74d9966e8e8df2ef7727b2cc48d4960) Signed-off-by: Michel Dänzer --- src/radeon_dri3.c | 1 + 1 file changed, 1 insertion(+) di

[PATCH xf86-video-ati 2/2] Store FB for each CRTC in drmmode_flipdata_rec

From: Michel Dänzer We were only storing the FB provided by the client, but on CRTCs with TearFree enabled, we use a separate FB. This could cause drmmode_flip_handler to fail to clear drmmode_crtc->flip_pending, which could result in a hang when waiting for the pending flip to complete. We were

[PATCH xf86-video-amdgpu] glamor: Use glamor_egl_create_textured_pixmap_from_gbm_bo when possible

From: Michel Dänzer Inspired by the modesetting driver. (Ported from radeon commit db28d35ce9fd07a2a4703f3df0633d4c8291ff9b) Signed-off-by: Michel Dänzer --- src/amdgpu_glamor.c | 32 ++-- 1 file changed, 18 insertions(+), 14 deletions(-) diff --git a/src/amdgpu_g

Re: [PATCH libdrm 3/6] amdgpu: add handle table implementation

Applies to patch #1 and patch #3, should lockup not be lookup? On 08/02/2018 09:04 AM, Christian König wrote: > The kernel handles are dense and the kernel always tries to use the > lowest free id. Use this to implement a more efficient handle table > by using a resizeable array instead of a hash

Re: [PATCH libdrm 3/6] amdgpu: add handle table implementation

On 2018-08-02 04:04 PM, Christian König wrote: > The kernel handles are dense and the kernel always tries to use the > lowest free id. Use this to implement a more efficient handle table > by using a resizeable array instead of a hash. > > Signed-off-by: Christian König > > [...] > > +drm_priv

Re: [PATCH libdrm 1/6] amdgpu: always add all BOs to lockup table

On 2018-08-02 04:04 PM, Christian König wrote: > This way we can always find a BO structure by its handle. > > Signed-off-by: Christian König Typo in the shortlog: should be "lookup" instead of "lockup". Also, this patch should really be after patch 4. > @@ -240,14 +237,11 @@ int amdgpu_bo_ex

[PATCH libdrm 6/6] amdgpu: remove the hash table implementation

Not used any more. Signed-off-by: Christian König --- amdgpu/Makefile.sources | 4 - amdgpu/util_hash.c | 383 --- amdgpu/util_hash.h | 103 - amdgpu/util_hash_table.c | 270 - amdgpu/util_hash

[PATCH libdrm 5/6] amdgpu: use handle table for flink names

Instead of the hash use the handle table. Signed-off-by: Christian König --- amdgpu/amdgpu_bo.c | 26 +- amdgpu/amdgpu_device.c | 14 -- amdgpu/amdgpu_internal.h | 2 +- 3 files changed, 14 insertions(+), 28 deletions(-) diff --git a/amdgpu/amdgpu_bo

[PATCH libdrm 4/6] amdgpu: use handle table for KMS handles

Instead of the hash use the handle table. Signed-off-by: Christian König --- amdgpu/amdgpu_bo.c | 19 ++- amdgpu/amdgpu_device.c | 2 -- amdgpu/amdgpu_internal.h | 3 ++- 3 files changed, 12 insertions(+), 12 deletions(-) diff --git a/amdgpu/amdgpu_bo.c b/amdgpu/amdgpu

[PATCH libdrm 1/6] amdgpu: always add all BOs to lockup table

This way we can always find a BO structure by its handle. Signed-off-by: Christian König --- amdgpu/amdgpu_bo.c | 16 +--- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/amdgpu/amdgpu_bo.c b/amdgpu/amdgpu_bo.c index d29be244..39228620 100644 --- a/amdgpu/amdgpu_bo.c +

[PATCH libdrm 2/6] amdgpu: stop using the hash table for fd_tab

We have so few devices that just walking a linked list is probably faster. Signed-off-by: Christian König --- amdgpu/amdgpu_device.c | 49 amdgpu/amdgpu_internal.h | 1 + 2 files changed, 17 insertions(+), 33 deletions(-) diff --git a/amdgpu/a

[PATCH libdrm 3/6] amdgpu: add handle table implementation

The kernel handles are dense and the kernel always tries to use the lowest free id. Use this to implement a more efficient handle table by using a resizeable array instead of a hash. Signed-off-by: Christian König --- amdgpu/Makefile.sources | 4 +++- amdgpu/handle_table.c | 59 ++

Re: [PATCH 3/3] drm/amd/pp: Add gfx power status check support on Rv/Vega12

If you use them enough for debugging that you think it's better to have them in tree, feel free to commit them. Maybe just add a note that they are just for debugging and add a note about the race condition. Alex On Thu, Aug 2, 2018 at 6:22 AM, Zhu, Rex wrote: > Currently, there is no use cas

Re: [PATCH 1/6] drm/amdgpu: move gfx definitions into amdgpu_gfx header

Am 02.08.2018 um 13:08 schrieb Huang Rui: Demangle amdgpu.h Signed-off-by: Huang Rui Acked-by: Christian König for the whole series. --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 276 + drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 34 +++- drivers/gpu/drm/amd/amdg

Re: [PATCH 5/6] drm/amdgpu: move psp macro into amdgpu_ucode into amdgpu_psp header

On Thu, Aug 02, 2018 at 07:08:14PM +0800, Huang Rui wrote: > Demangle amdgpu.h. Sorry, there is a typo in the subject. It should be "drm/amdgpu: move psp macro into amdgpu_ucode into amdgpu_psp header". Thanks, Ray > > Signed-off-by: Huang Rui > --- > drivers/gpu/drm/amd/amdgpu/amdgpu.h |

[PATCH 6/6] drm/amdgpu: move gem definitions into amdgpu_gem header

Demangle amdgpu.h. Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 62 + drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c| 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c| 1 + drivers/gpu/drm/amd/amdgpu/a

[PATCH 5/6] drm/amdgpu: move psp macro into amdgpu_ucode into amdgpu_psp header

Demangle amdgpu.h. Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h | 29 + 2 files changed, 17 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/

[PATCH 4/6] drm/amdgpu: move firmware definitions into amdgpu_ucode header

Demangle amdgpu.h. Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 27 --- drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h | 24 2 files changed, 24 insertions(+), 27 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.

[PATCH 3/6] drm/amdgpu: move sdma definitions into amdgpu_sdma header

Demangle amdgpu.h. Furthermore, SDMA is used for moving and clearing the data buffer, so the header also need be included in ttm. Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/Makefile | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu.h | 87 +- drivers/gpu/

[PATCH 2/6] drm/amdgpu: move ih definitions into amdgpu_ih header

Demangle amdgpu.h Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/amdgpu.h| 14 -- drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h | 15 +++ 2 files changed, 15 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdg

[PATCH 1/6] drm/amdgpu: move gfx definitions into amdgpu_gfx header

Demangle amdgpu.h Signed-off-by: Huang Rui --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 276 + drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 34 +++- drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 343 3 files changed, 336 insertions(+), 317 dele

Re: [PATCH 3/3] drm/amd/pp: Add gfx power status check support on Rv/Vega12

Currently, there is no use case except for debug. So I will keep the patches in my local. Thanks. Best Regards Rex From: Alex Deucher Sent: Wednesday, August 1, 2018 11:13 PM To: Zhu, Rex Cc: Deucher, Alexander; Quan, Evan; amd-gfx@lists.freedesktop.org Su

Re: [PATCH 2/7] drm/amdgpu: use scheduler load balancing for SDMA CS

Am 02.08.2018 um 07:50 schrieb Zhang, Jerry (Junwei): On 08/01/2018 07:31 PM, Christian König wrote: Start to use the scheduler load balancing for userspace SDMA command submissions. In this case, each SDMA could load all SDMA(instances) rqs, and UMD will not specify a ring id. If so, we may

Re: [PATCH v2] drm/amd/display: Report non-DP display as disconnected without EDID

On 2018-08-01 09:51 PM, Harry Wentland wrote: > [Why] > Some boards seem to have a problem where HPD is high on HDMI even though > no display is connected. We don't want to report these as connected. DP > spec still requires us to report DP displays as connected when HPD is > high but we can't read

Re: [PATCH] drm/scheduler: fix setting the priorty for entities

On Wed, Aug 1, 2018 at 7:58 PM Christian König < ckoenig.leichtzumer...@gmail.com> wrote: > Since we now deal with multiple rq we need to update all of them, not > just the current one. > > Signed-off-by: Christian König > Acked-by: Nayan Deshmukh > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.