Re: [PATCH v4 1/2] drm/amdgpu/gmc: steal the appropriate amount of vram for fw hand-over

On 2018-04-12 06:25 AM, Andrey Grodzovsky wrote: > On 04/12/2018 12:16 AM, Alex Deucher wrote: >> On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky >> wrote: >>> From: Alex Deucher >>> >>> Steal 9 MB for vga emulation and fb if vga is enabled,

Re: RFC for a render API to support adaptive sync and VRR

On 2018-04-12 01:30 AM, Cyr, Aric wrote: >> From: Michel Dänzer [mailto:mic...@daenzer.net] >> Sent: Wednesday, April 11, 2018 05:50 >> On 2018-04-11 08:57 AM, Nicolai Hähnle wrote: >>> On 10.04.2018 23:45, Cyr, Aric wrote: How does it work fine today given that all kernel seems to know is

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

Am 12.04.2018 um 11:11 schrieb Lucas Stach: Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: Am 11.04.2018 um 19:11 schrieb Robin Murphy: For dma_map_sg(), DMA API implementations are free to merge consecutive segments into a single DMA mapping if conditions are suitable, thus

Re: [PATCH V2] drm/amdgpu: limit DMA size to PAGE_SIZE for scatter-gather buffers

On Wed, Apr 11, 2018 at 01:03:59PM +0100, Robin Murphy wrote: > On 10/04/18 21:59, Sinan Kaya wrote: >> Code is expecing to observe the same number of buffers returned from >> dma_map_sg() function compared to sg_alloc_table_from_pages(). This >> doesn't hold true universally especially for

Re: [PATCH v4 1/2] drm/amdgpu/gmc: steal the appropriate amount of vram for fw hand-over

On Thu, Apr 12, 2018 at 12:25:08PM +0800, Grodzovsky, Andrey wrote: > > > On 04/12/2018 12:16 AM, Alex Deucher wrote: > > On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky > > wrote: > >> From: Alex Deucher > >> > >> Steal 9 MB for vga

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: > Am 11.04.2018 um 19:11 schrieb Robin Murphy: > > For dma_map_sg(), DMA API implementations are free to merge consecutive > > segments into a single DMA mapping if conditions are suitable, thus the > > resulting DMA addresses may

Re: [PATCH V2] drm/amdgpu: limit DMA size to PAGE_SIZE for scatter-gather buffers

Am 12.04.2018 um 08:26 schrieb Christoph Hellwig: On Wed, Apr 11, 2018 at 01:03:59PM +0100, Robin Murphy wrote: On 10/04/18 21:59, Sinan Kaya wrote: Code is expecing to observe the same number of buffers returned from dma_map_sg() function compared to sg_alloc_table_from_pages(). This doesn't

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

Am Donnerstag, den 12.04.2018, 11:35 +0200 schrieb Christian König: > Am 12.04.2018 um 11:11 schrieb Lucas Stach: > > Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: > > > Am 11.04.2018 um 19:11 schrieb Robin Murphy: > > > > For dma_map_sg(), DMA API implementations are free to

[PATCH 09/11] drm/amdgpu: counter for every bo creation

Change-Id: I877b2d5c54b3e47b66f2d58454dcf1e5f5a68972 Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c

[PATCH 08/11] drm/ttm: add bo index

Change-Id: I4abf5cf0aaf946162dabd08fc1fd0406c2abf418 Signed-off-by: Chunming Zhou --- drivers/gpu/drm/ttm/ttm_bo.c| 1 + include/drm/ttm/ttm_bo_driver.h | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c

[PATCH 10/11] drm/ttm: some good fixes for per-vm-lru

Change-Id: Ib68bff91fd127162cf8c72516101546e1fe014df Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 1 - drivers/gpu/drm/ttm/ttm_bo.c | 39 -- 2 files changed, 26 insertions(+), 14 deletions(-) diff

[PATCH 11/11] drm/ttm: bulk move per vm bo

Change-Id: I5b6afbdd715e28e5266b5099ca9a34399d1fc3a1 Signed-off-by: Chunming Zhou --- drivers/gpu/drm/ttm/ttm_bo.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

Am 12.04.2018 um 11:49 schrieb Lucas Stach: Am Donnerstag, den 12.04.2018, 11:35 +0200 schrieb Christian König: Am 12.04.2018 um 11:11 schrieb Lucas Stach: Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: Am 11.04.2018 um 19:11 schrieb Robin Murphy: For dma_map_sg(), DMA API

Re: RFC for a render API to support adaptive sync and VRR

On 12.04.2018 01:30, Cyr, Aric wrote: At least with VDPAU, video players are already explicitly specifying the target presentation time, so no changes should be required at that level. Don't know about other video APIs. The X11 Present extension protocol is also prepared for specifying the

[PATCH 04/11] drm/amdgpu: pass process to tbo

Change-Id: I4de8146567b858ae07a8a27cadf71d13d490e8ac Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c| 7 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 4 +++- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git

[PATCH 03/11] drm/amdgpu: add kernel process

Change-Id: Ie43f3c73cc65526a449208f3ce927b1dfad5cf6b Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 1 + 2 files changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h

[PATCH] drm/gpu-sched: fix force APP kill hang(v3)

issue: there are VMC page fault occurred if force APP kill during 3dmark test, the cause is in entity_fini we manually signal all those jobs in entity's queue which confuse the sync/dep mechanism: 1)page fault occurred in sdma's clear job which operate on shadow buffer, and shadow buffer's Gart

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On 04/12/2018 12:32 AM, Alex Deucher wrote: On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky wrote: Reserved VRAM is used to avoid overriding pre OS FB. Once our display stack takes over we don't need the reserved VRAM anymore. v2: Remove comment, we know

[PATCH 05/11] drm/ttm: add per process lru

Change-Id: Id2333f69119222a7e9bdb0357bbed97cf08636da Signed-off-by: Chunming Zhou --- drivers/gpu/drm/ttm/ttm_bo.c| 59 ++--- include/drm/ttm/ttm_bo_driver.h | 3 ++- 2 files changed, 52 insertions(+), 10 deletions(-) diff --git

[PATCH 06/11] drm/amdgpu: pass ttm process to buffer object

Change-Id: Ifb0dc95db6a358cf7f76e2a99f94c58637ad6ee6 Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu.h| 3 +- drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c| 17 ++-

[PATCH 07/11] drm/ttm: use RB tree instead of link list

Change-Id: I0f533c6512f3b72fcf2fbf11d738f38d9e087f26 Signed-off-by: Chunming Zhou --- drivers/gpu/drm/ttm/ttm_bo.c| 39 +++ include/drm/ttm/ttm_bo_api.h| 3 ++- include/drm/ttm/ttm_bo_driver.h | 2 +- 3 files changed, 34

Re: [PATCH xf86-video-amdgpu 3/5] Keep CRTC properties consistent

On 2018-04-11 11:26 PM, Leo Li wrote: > On 2018-04-11 04:39 AM, Michel Dänzer wrote: >> >> Hmm. So either legacy or non-legacy clients won't work at all, or >> they'll step on each other's toes, clobbering the HW gamma LUT from >> each other. >> >> I'm afraid neither of those alternatives sound

[PATCH 01/11] drm/ttm: add ttm process struct

Change-Id: I34924a40392653e72f143c30ab312cbbf9fa Signed-off-by: Chunming Zhou --- drivers/gpu/drm/ttm/ttm_bo.c| 23 +++ include/drm/ttm/ttm_bo_api.h| 1 + include/drm/ttm/ttm_bo_driver.h | 10 ++ 3 files changed, 34 insertions(+)

[PATCH 00/11] ***[WIP] TTM per process lru ***

Since per-process-bo feature is introduced, old lru isn't working for it. old lru order is depending on BO list order, which will be updated by bo list after every command submission. But for per-process-bo, which aren't in bo list, so it have no chance to refresh its order in lru. Which also will

[PATCH 02/11] drm/amdgpu: use ttm process in amdgpu vm

Change-Id: I2cf802e641d8b2cdb2bf8bdf1957f3f4f27afaba Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c

[PATCH 2/2] Revert "drm/amd/display: disable CRTCs with NULL FB on their primary plane (V2)"

This seems to cause flickering and lock-ups for a wide range of users. Revert until we've found a proper fix for the flickering and lock-ups. This reverts commit 36cc549d59864b7161f0e23d710c1c4d1b9cf022. Cc: Shirish S Cc: Alex Deucher Cc:

[PATCH 1/2] Revert "drm/amd/display: fix dereferencing possible ERR_PTR()"

This reverts commit cd2d6c92a8e39d7e50a5af9fcc67d07e6a89e91d. Cc: Shirish S Cc: Alex Deucher Cc: sta...@vger.kernel.org Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 --- 1 file

Re: [PATCH V2] drm/amdgpu: limit DMA size to PAGE_SIZE for scatter-gather buffers

On 12/04/18 10:42, Christian König wrote: Am 12.04.2018 um 08:26 schrieb Christoph Hellwig: On Wed, Apr 11, 2018 at 01:03:59PM +0100, Robin Murphy wrote: On 10/04/18 21:59, Sinan Kaya wrote: Code is expecing to observe the same number of buffers returned from dma_map_sg() function compared to

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

On 12/04/18 11:33, Christian König wrote: Am 12.04.2018 um 11:49 schrieb Lucas Stach: Am Donnerstag, den 12.04.2018, 11:35 +0200 schrieb Christian König: Am 12.04.2018 um 11:11 schrieb Lucas Stach: Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: Am 11.04.2018 um 19:11

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On 2018-04-12 03:33 PM, Alex Deucher wrote: > On Thu, Apr 12, 2018 at 7:17 AM, Andrey Grodzovsky > wrote: >> On 04/12/2018 12:32 AM, Alex Deucher wrote: >>> >>> On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky >>> wrote: Reserved

Re: RFC for a render API to support adaptive sync and VRR

On 2018-04-12 01:39 PM, Nicolai Hähnle wrote: > On 12.04.2018 01:30, Cyr, Aric wrote: >>> At least with VDPAU, video players are already explicitly specifying the >>> target presentation time, so no changes should be required at that >>> level. Don't know about other video APIs. >>> >>> The X11

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On 04/12/2018 10:33 AM, Michel Dänzer wrote: On 2018-04-12 03:33 PM, Alex Deucher wrote: On Thu, Apr 12, 2018 at 7:17 AM, Andrey Grodzovsky wrote: On 04/12/2018 12:32 AM, Alex Deucher wrote: On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky

Re: [PATCH 1/2] drm/prime: Iterate SG DMA addresses separately

On 2018-04-12 06:33, Christian König wrote: Am 12.04.2018 um 11:49 schrieb Lucas Stach: Am Donnerstag, den 12.04.2018, 11:35 +0200 schrieb Christian König: Am 12.04.2018 um 11:11 schrieb Lucas Stach: Am Mittwoch, den 11.04.2018, 20:26 +0200 schrieb Christian König: Am 11.04.2018 um 19:11

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On Thu, Apr 12, 2018 at 7:17 AM, Andrey Grodzovsky wrote: > > > On 04/12/2018 12:32 AM, Alex Deucher wrote: >> >> On Thu, Apr 12, 2018 at 12:08 AM, Andrey Grodzovsky >> wrote: >>> >>> Reserved VRAM is used to avoid overriding pre OS FB. >>>

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Can you be more specific, Christian? Mesa has this, I don't think it needs anything else: Completely agree, that's what I suggested to implement. The point is this kernel change now needs to be reworked and adapted to what Mesa is doing. Regards, Christian. Am 12.04.2018 um 18:40 schrieb

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

> The point is this kernel change now needs to be reworked and adapted to what > Mesa is doing. Three options have been brought up for kernel, 1) Turn off immediate mode when flipping between VRAM/GTT. 2) Check the domain of the current fb and then adjust the new one before pinning it. 3) Use

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

1) Turn off immediate mode when flipping between VRAM/GTT. That must be implemented independently. See working around the hardware bug should be a different patch than implementing a placement policy. As per discussion, the 3rd one, which is the current patch, seems the best so far. And I

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Patch #1: Avoid the hardware bug! E.g. even when we avoid different placements it would be good to have a patch which turns off immediate flipping when switching from VRAM to GTT. That is as safety net and to document that we need to avoid this condition on the hardware. Patch #2: Go into

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Can you be more specific, Christian? Mesa has this, I don't think it needs anything else: https://cgit.freedesktop.org/mesa/mesa/commit/?id=7d2079908d9ef05ec3f35b7078833e57846cab5b Marek On Wed, Mar 28, 2018 at 3:46 AM, Christian König < ckoenig.leichtzumer...@gmail.com> wrote: > Am 28.03.2018

Re: [PATCH xf86-video-amdgpu 3/5] Keep CRTC properties consistent

On 2018-04-12 06:30 AM, Michel Dänzer wrote: On 2018-04-11 11:26 PM, Leo Li wrote: On 2018-04-11 04:39 AM, Michel Dänzer wrote: Hmm. So either legacy or non-legacy clients won't work at all, or they'll step on each other's toes, clobbering the HW gamma LUT from each other. I'm afraid

Re: RFC for a render API to support adaptive sync and VRR

On 2018-04-12 07:39 AM, Nicolai Hähnle wrote: > On 12.04.2018 01:30, Cyr, Aric wrote: >>> At least with VDPAU, video players are already explicitly specifying the >>> target presentation time, so no changes should be required at that >>> level. Don't know about other video APIs. >>> >>> The X11

Re: [PATCH 2/2] drm/amdgpu: Allow dma_map_sg() coalescing

On 11/04/18 19:28, Christian König wrote: Am 11.04.2018 um 19:11 schrieb Robin Murphy: Now that drm_prime_sg_to_page_addr_arrays() understands the case where dma_map_sg() has coalesced segments and returns 0 < count < nents, we can relax the check to only consider genuine failure. That

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Please clarify, Christian. How would you like it to be implemented? Sam On 2018-04-12 02:00 PM, Christian König wrote: >> 1) Turn off immediate mode when flipping between VRAM/GTT. > That must be implemented independently. > > See working around the hardware bug should be a different patch

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Hi Sam, yeah sorry for that. It's already rather late here and I got a bit taken away. But I still insist that you at least try as I advised, I'm pretty sure that this approach will work and cover all use cases discussed so far. Regards, Christian. Am 12.04.2018 um 22:42 schrieb Samuel

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On 04/12/2018 11:10 AM, Michel Dänzer wrote: On 2018-04-12 04:58 PM, Andrey Grodzovsky wrote: On 04/12/2018 10:33 AM, Michel Dänzer wrote: On 2018-04-12 03:33 PM, Alex Deucher wrote: On Thu, Apr 12, 2018 at 7:17 AM, Andrey Grodzovsky wrote: On 04/12/2018 12:32

Re: [PATCH 2/2] drm/amdgpu: Allow dma_map_sg() coalescing

Am 12.04.2018 um 19:53 schrieb Robin Murphy: On 11/04/18 19:28, Christian König wrote: Am 11.04.2018 um 19:11 schrieb Robin Murphy: Now that drm_prime_sg_to_page_addr_arrays() understands the case where dma_map_sg() has coalesced segments and returns 0 < count < nents, we can relax the check

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

The 4th proposal :) > In other words add something like the following: > > if (domain & AMDGPU_GEM_DOMAIN_GTT && bo->preferred_domains & > AMDGPU_GEM_DOMAIN_GTT) > domain = AMDGPU_GEM_DOMAIN_GTT; > > That should be everything we need here. This is basically against the idea of Marek's

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Am 12.04.2018 um 22:01 schrieb Samuel Li: The 4th proposal :) In other words add something like the following: if (domain & AMDGPU_GEM_DOMAIN_GTT && bo->preferred_domains & AMDGPU_GEM_DOMAIN_GTT) domain = AMDGPU_GEM_DOMAIN_GTT; That should be everything we need here. This is basically

Re: [PATCH v2 1/1] drm/amdgpu: Enable scatter gather display support

Hi Christian, If you have any good proposal, please describe here. Otherwise kindly avoid saying anything based on your emotion. Regards, Sam On 2018-04-12 04:13 PM, Christian König wrote: > Am 12.04.2018 um 22:01 schrieb Samuel Li: >> The 4th proposal :) >> >>> In other words add something

[pull] radeon and amdgpu drm-next-4.17

Hi Dave, More fixes for 4.17 and stable: - Add a PX quirk for radeon - Fix flickering and stability issues with DC on some platforms - Fix HDMI audio regression - Few other misc DC and base driver fixes The following changes since commit 871e899db19da3dbd17a5d263b555dc5b7d8fed5: Merge branch

Re: RFC for a render API to support adaptive sync and VRR

On Tue, Apr 10, 2018 at 12:37 AM, Michel Dänzer wrote: > On 2018-04-10 08:45 AM, Christian König wrote: >> Am 09.04.2018 um 23:45 schrieb Manasi Navare: >>> Thanks for initiating the discussion. Find my comments below: >>> On Mon, Apr 09, 2018 at 04:00:21PM -0400, Harry

[PATCH] drm/radeon: change function signature to pass full range

In function ‘radeon_process_i2c_ch’ a comparison of a u8 value against 255 is done. Since it is always false, change the signature of this function to use an `int` instead, which match the type used in caller: `radeon_atom_hw_i2c_xfer`. Fix the following warning triggered with W=1: CC [M]

RE: [PATCH 2/2] Revert "drm/amd/display: disable CRTCs with NULL FB on their primary plane (V2)"

Hi Harry, Alex, The solution given while reviewing my patch was that "DC should support enabling a CRTC without a framebuffer." Since the revert is a temporary workaround to address issue at hand and considering the bigger regression it will cause on ChromeOS(explained below), I would strongly

Re: [PATCH 10/21] drm/amd/display: Add back code to allow for rounding error

Den tis 10 apr. 2018 23:11Harry Wentland skrev: > From: Anthony Koo > > Signed-off-by: Anthony Koo > Reviewed-by: Aric Cyr > Acked-by: Harry Wentland > --- >

RE: [PATCH] drm/gpu-sched: fix force APP kill hang(v3)

Ping Best Wishes, Emily Deng > -Original Message- > From: Emily Deng [mailto:emily.d...@amd.com] > Sent: Thursday, April 12, 2018 6:22 PM > To: amd-gfx@lists.freedesktop.org > Cc: Deng, Emily ; Liu, Monk > Subject: [PATCH] drm/gpu-sched: fix

[PATCH] drm/amdgpu: defer test IBs on the rings at boot

amdgpu_ib_ring_tests() runs test IB's on rings at boot contributes to ~500 ms of amdgpu driver's boot time. This patch defers it and adds a check to report in amdgpu_info_ioctl() if it was scheduled or not. Signed-off-by: Shirish S --- drivers/gpu/drm/amd/amdgpu/amdgpu.h

Re: [PATCH] drm/gpu-sched: fix force APP kill hang(v3)

Hi Christian & Emily This v3 version looks pretty good to me, but still some parts need to improve: e.g. 1)entity->finished doesn't presenting what it really means, better rename it to entity->last_scheduled. 2)drm_sched_entity_fini_job_cb() better renamed to

Re: [PATCH 2/2] Revert "drm/amd/display: disable CRTCs with NULL FB on their primary plane (V2)"

On 2018-04-12 04:51 PM, Harry Wentland wrote: > This seems to cause flickering and lock-ups for a wide range of users. > Revert until we've found a proper fix for the flickering and lock-ups. > > This reverts commit 36cc549d59864b7161f0e23d710c1c4d1b9cf022. > > Cc: Shirish S

Re: [PATCH v4 2/2] drm/amdgpu: Free VGA stolen memory as soon as possible.

On 2018-04-12 04:58 PM, Andrey Grodzovsky wrote: > On 04/12/2018 10:33 AM, Michel Dänzer wrote: >> On 2018-04-12 03:33 PM, Alex Deucher wrote: >>> On Thu, Apr 12, 2018 at 7:17 AM, Andrey Grodzovsky >>> wrote: On 04/12/2018 12:32 AM, Alex Deucher wrote: > On