Re: [PATCH v2] tests/amdgpu: add deadlock test for sdma

Hi Alex, could you help to submit this patch upstream? 在 3/6/2019 4:08 PM, Christian König 写道: > Am 06.03.19 um 07:37 schrieb Cui, Flora: >> deadlock test for sdma will cause gpu recoverty. >> disable the test for now until GPU reset recovery could survive at least >> 1000 times test. >> >> v2: a

[PATCH] drm/amdgpu: error handling for xgmi and ras

This is a quick workaround. A more complete error handling around psp_hw_start is actually needed. Change-Id: I398efd652584e022debf237950207199a4ea78fc Signed-off-by: Evan Quan --- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 9 + 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/dr

Re: [PATCH 2/3] drm/amdgpu: trivial typo fix

Am 15.03.19 um 07:01 schrieb Evan Quan: "error" was not correctly spelled. Change-Id: Idc3acb6172469c1d63227cb5f0df75f912c10871 Signed-off-by: Evan Quan Reviewed-by: Christian König for this one. --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++-- 1 file changed, 2 insertions(+), 2

Re: [PATCH] drm/amdgpu: error handling for xgmi and ras

Am 15.03.19 um 09:33 schrieb Evan Quan: This is a quick workaround. A more complete error handling around psp_hw_start is actually needed. Change-Id: I398efd652584e022debf237950207199a4ea78fc Signed-off-by: Evan Quan --- drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 9 + 1 file changed,

[PATCH libdrm 2/4] tests/amdgpu: add memcpy dispatch test

add memcpy dispatch test for gfx9 Change-Id: Ic9b280db2ced21b15e228c1c4c217836ab3f44a8 Signed-off-by: Flora Cui Tested-by: Rui Teng Reviewed-by: Hawking Zhang --- tests/amdgpu/basic_tests.c | 163 - 1 file changed, 161 insertions(+), 2 deletions(-)

[PATCH libdrm 3/4] tests/amdgpu: add memset draw test

add memset draw test for gfx9 Change-Id: I3a1f2460ec0917e29ce1a891200eeaa1a0565fda Signed-off-by: Flora Cui Tested-by: Rui Teng --- tests/amdgpu/amdgpu_test.c | 5 + tests/amdgpu/basic_tests.c | 554 + 2 files changed, 559 insertions(+) diff --git

[PATCH libdrm 1/4] tests/amdgpu: add memset dispatch test

add memset dispatch test for gfx9 v2: disable dispatch test for other ASICs Change-Id: I55a410a4ea0a6d19d1aca1f172e1dfcbbd1112fa Signed-off-by: Flora Cui Tested-by: Rui Teng Reviewed-by: Hawking Zhang --- tests/amdgpu/amdgpu_test.c | 5 + tests/amdgpu/basic_tests.c | 273

[PATCH libdrm 4/4] tests/amdgpu: add memcpy draw test

add memcpy draw test for gfx9 Change-Id: Ib80e55b1ab7aa556c4b5adfdd39aedf7d58ba628 Signed-off-by: Flora Cui Tested-by: Rui Teng --- tests/amdgpu/basic_tests.c | 264 +++-- 1 file changed, 253 insertions(+), 11 deletions(-) diff --git a/tests/amdgpu/basic

RE: [PATCH] drm/amdgpu: error handling for xgmi and ras

Hw_init -> psp_hw_init -> psp_hw_start (failed) -> ras_pre_fini -> hw_fini -> psp_hw_fini On psp_hw_start failure, it destroys the cmd/fence/fw buffers. But the ras/xgmi deinit was delayed to psp_hw_fini. In ras_pre_fini, it calls the ras APIs to disable previously enabled ras features. And th

[PATCH 1/9] dma-buf: add new dma_fence_chain container v5

From: Christian König Lockless container implementation similar to a dma_fence_array, but with only two elements per node and automatic garbage collection. v2: properly document dma_fence_chain_for_each, add dma_fence_chain_find_seqno, drop prev reference during garbage collection if it's no

[PATCH 2/9] drm/syncobj: add new drm_syncobj_add_point interface v4

From: Christian König Use the dma_fence_chain object to create a timeline of fence objects instead of just replacing the existing fence. v2: rebase and cleanup v3: fix garbage collection parameters v4: add unorder point check, print a warn calltrace Signed-off-by: Christian König --- drivers/

[PATCH 3/9] drm/syncobj: add support for timeline point wait v8

points array is one-to-one match with syncobjs array. v2: add seperate ioctl for timeline point wait, otherwise break uapi. v3: userspace can specify two kinds waits:: a. Wait for time point to be completed. b. and wait for time point to become available v4: rebase v5: add comment for xxx_WAIT_AVAI

[PATCH 7/9] drm/syncobj: add transition iotcls between binary and timeline v2

we need to import/export timeline point. v2: unify to one transfer ioctl Signed-off-by: Chunming Zhou --- drivers/gpu/drm/drm_internal.h | 2 + drivers/gpu/drm/drm_ioctl.c| 2 + drivers/gpu/drm/drm_syncobj.c | 74 ++ include/uapi/drm/drm.h | 10 +++

[PATCH 9/9] drm/amdgpu: update version for timeline syncobj support in amdgpu

Signed-off-by: Chunming Zhou --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 8a0732088640..4d8db87048d3 100644 --- a/drivers/gpu/drm/amd/amdgp

[PATCH 4/9] drm/syncobj: add timeline payload query ioctl v6

user mode can query timeline payload. v2: check return value of copy_to_user v3: handle querying entry by entry v4: rebase on new chain container, simplify interface v5: query last signaled timeline point, not last point. v6: add unorder point check Signed-off-by: Chunming Zhou Cc: Daniel Rakos

[PATCH 8/9] drm/syncobj: add timeline signal ioctl for syncobj v3

v2: individually allocate chain array, since chain node is free independently. v3: all existing points must be already signaled before cpu perform signal operation, so add check condition for that. Signed-off-by: Chunming Zhou --- drivers/gpu/drm/drm_internal.h | 2 + drivers/gpu/drm/drm_

[PATCH 6/9] drm/amdgpu: add timeline support in amdgpu CS v3

syncobj wait/signal operation is appending in command submission. v2: separate to two kinds in/out_deps functions v3: fix checking for timeline syncobj Signed-off-by: Chunming Zhou Cc: Daniel Rakos Cc: Jason Ekstrand Cc: Bas Nieuwenhuizen Cc: Dave Airlie Cc: Christian König Cc: Chris Wilson

[PATCH 5/9] drm/syncobj: use the timeline point in drm_syncobj_find_fence v3

From: Christian König Implement finding the right timeline point in drm_syncobj_find_fence. v2: return -EINVAL when the point is not submitted yet. v3: fix reference counting bug, add flags handling as well v4: add timeout for find fence Signed-off-by: Christian König --- drivers/gpu/drm/drm_

[PATCH] drm/amd/display: Add debugfs entry for amdgpu_dm_visual_confirm

[Why] DC provides a few visual confirmation debug options that can be dynamically changed at runtime to help debug surface programming issues but we don't have any way to access it from userspace. [How] Add the amdgpu_dm_visual_confirm debugfs entry. It accepts a string containing the DC visual co

Re: [PATCH] drm/amd/display: Add debugfs entry for amdgpu_dm_visual_confirm

Am 15.03.19 um 13:42 schrieb Nicholas Kazlauskas: [Why] DC provides a few visual confirmation debug options that can be dynamically changed at runtime to help debug surface programming issues but we don't have any way to access it from userspace. [How] Add the amdgpu_dm_visual_confirm debugfs en

Re: [PATCH] drm/amd/display: Add debugfs entry for amdgpu_dm_visual_confirm

On 3/15/19 8:47 AM, Christian König wrote: > Am 15.03.19 um 13:42 schrieb Nicholas Kazlauskas: >> [Why] >> DC provides a few visual confirmation debug options that can be >> dynamically changed at runtime to help debug surface programming issues >> but we don't have any way to access it from usersp

[PATCH 02/26] drm/amd/display: change generic_reg_wait to void.

From: Yongqiang Sun we were only checking the return value in one place, thus changing generic_reg_wait from int to void and reading the register instead of getting it from generic_reg_wait, when we need the return value. Change-Id: Ib14a77e497960f39d9860ca8832355bdea1d9909 Signed-off-by: Yongqi

[PATCH 00/26] DC Patches 15 Mar 2019

Summary of changes *ODM related fixes and programming changes *DP audio regression fix *Multi-thread related fixes Aidan Wood (1): drm/amd/display: Fix multi-thread writing to 1 state Aric Cyr (3): drm/amd/display: Add a hysteresis to BTR frame multiplier drm/amd/display: 3.2.22 drm/amd/d

[PATCH 04/26] drm/amd/display: Move dm_read_reg_func to dc_helper.

From: Yongqiang Sun Move out of header to dc_helper.c, in preparation for future implementations. Change-Id: I004584f11db4a136e704c4c80f830a20ff05809d Signed-off-by: Yongqiang Sun Reviewed-by: Tony Cheng Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/dc_helper.c | 18 ++

[PATCH 06/26] drm/amd/display: fix DP 422 VID_M half the rate issue.

From: Charlene Liu [Description] when programming VID_TIMING, we were using the original VESA timing for DP_VIDM/N. for YCbCr420 or compressed YCbCr422, using half rate as YCbCr444. Change-Id: I8a60e252383c11f81789f190fd3477cb3083554b Signed-off-by: Charlene Liu Reviewed-by: Nikola Cornij Ac

[PATCH 03/26] drm/amd/display: Use drm helper for resetting plane state

From: Nicholas Kazlauskas [Why] To help prevent plane state not being set to the correct default value if any new properties are added in the future. [How] Use the drm helper - which seems to be the common solution among other DRM drivers. Change-Id: I951a36da92ead790a3b6888d59995e349fc98b3b Si

[PATCH 11/26] drm/amd/display: Programming correct VRR_EN bit in VTEM structure

From: Hugo Hu [Why] In HDMI plugfest, MTK report our EMP with VRR_EN bit = 0. VRR_EN bit is EMP-MD0-bit 0. Currently driver set 1 to bit 3. [How] Programming correct VRR_EN bit in EMP-MD0-bit0. Change-Id: I41d4f5b24896ca34f50fdb3d1d121678faa09034 Signed-off-by: Hugo Hu Reviewed-by: Reza Amini

[PATCH 08/26] drm/amd/display: Fix DP audio regression

From: Wesley Chalmers Audio was unmuted for HDMI only, need to do so for DP as well. Signed-off-by: Wesley Chalmers Reviewed-by: Chris Park Reviewed-by: Eric Bernstein Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c | 5 ++--- 1 file changed, 2 in

[PATCH 14/26] drm/amd/display: Use dc_is_hdmi_signal() instead of ENUM

From: Eric Bernstein Change-Id: I1a903bcc614aadbe63725713fddf1cb0574b3a33 Signed-off-by: Eric Bernstein Reviewed-by: Nevenko Stupar Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/core/dc_link.c | 4 ++-- drivers/gpu/drm/amd/display/dc/inc/core_types.h | 1 - 2 files changed,

[PATCH 01/26] drm/amd/display: Reset alpha state for planes to the correct values

From: Nicholas Kazlauskas [Why] The plane_reset callback is subclassed but hasn't been updated since the drm helper got updated to include resetting alpha related state (state->alpha and state->pixel_blend_mode). The overlay planes exposed by amdgpu_dm were therefore being rendered as invisible b

[PATCH 15/26] drm/amd/display: Fix multi-thread writing to 1 state

From: Aidan Wood [Why] Multiple threads were writing back to one global VBA in DC resulting in multiple threads overwriting eachother's data [How] Add an instance of DML (which contains VBA) to each context and change all calls that used dc->dml to use context->dml. Created a seperate copy const

[PATCH 26/26] drm/amd/display: Add debugfs dpcd interface

From: David Francis [Why] We need arbitrary read/write over DP AUX DPCD for debugging [How] Three debugfs entries Set the target address by writing to "aux_dpcd_address" (The first four bytes written are used) Set the transaction size in bytes by writing to "aux_dpcd_size" (The first four byte

[PATCH 18/26] drm/amd/display: Fix VTEM InfoPacket programming

From: Reza Amini Refactor setting bit fields. Correcting the offset of MD0. Initializing the InfoPacket header fields. Defining the field offsets and masks. Change-Id: Ifc05e2cb861102113bc6e70e5780321e5996dd3b Signed-off-by: Reza Amini Reviewed-by: Anthony Koo Acked-by: Bhawanpreet Lakha ---

[PATCH 19/26] drm/amd/display: 3.2.23

From: Aric Cyr Change-Id: I77d304828085ec379f0ebddcf5d9d143287312fe Signed-off-by: Aric Cyr Reviewed-by: Aric Cyr Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/dc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/driv

[PATCH 20/26] drm/amd/display: fix releasing planes when exiting odm

From: Dmytro Laktyushkin Releasing planes should not release the 2nd odm pipe right away, this change leaves us with 2 pipes with null planes and same stream when planes are released during odm. Change-Id: I9183a086d2240510b04fd6c76a9ce425128bb72c Signed-off-by: Dmytro Laktyushkin Reviewed-by:

[PATCH 24/26] drm/amd/display: Expose generic SDP message access interface

From: "Leo (Hanghong) Ma" [Why] We need to add DP SDP message test debugfs to make sdp message test more convenient and efficient. [How] Add a DM accessible SDP interface for custom data. Change-Id: I8ad12f9ffb8dd2b7d01d12f572bbc7eb0c6c2537 Signed-off-by: Leo (Hanghong) Ma Reviewed-by: Harry W

[PATCH 21/26] drm/amd/display: fix odm combine pipe reset

From: Dmytro Laktyushkin We fail to reset the second odm combine pipe. This change fixes odm pointer management. Change-Id: Ie4084f696636636a5965c9d75bb6291eb0626fa9 Signed-off-by: Dmytro Laktyushkin Reviewed-by: Tony Cheng Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/core/

[PATCH 22/26] drm/amd/display: add missing opp programming for odm

From: Dmytro Laktyushkin A number of places opp programming was missing for odm second pipe. This change fixes the oversight. Change-Id: I344c0480c76cf56a0b78a3547a5e5770d52fb547 Signed-off-by: Dmytro Laktyushkin Reviewed-by: Tony Cheng Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/dis

[PATCH 25/26] drm/amd/display: Generic SDP message access in amdgpu

From: "Leo (Hanghong) Ma" [Why] We need to add DP SDP message test debugfs to make sdp message test more convenient and efficient. [How] Add sdp_message debugfs entry in amdgpu. Change-Id: I3bb349566bb255c1515470694b1aff602a12cc77 Signed-off-by: Leo (Hanghong) Ma Reviewed-by: Sun peng Li Acke

[PATCH 16/26] drm/amd/display: add global master update lock interfaces

From: Wenjing Liu Add these interfacese for future use setup_global_lock() lock_global() pipe_control_lock_global() Change-Id: I0b21add5fb5fcb0cb866534a159dccfc34283149 Signed-off-by: Wenjing Liu Reviewed-by: Tony Cheng Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/inc/hw/ti

[PATCH 09/26] drm/amd/display: extend EDID support to 1kb

From: Jun Lei [why] There exist displays with EDIDs > 512 bytes, existing code will cause us to ignore all extension blocks. Change-Id: I0458baf7f8cc41f04405e25f6efe8456c0b05a57 Signed-off-by: Jun Lei Reviewed-by: Wenjing Liu Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/dc_

[PATCH 05/26] drm/amd/display: Fix setting DP_VID_N_MUL

From: Eric Bernstein [Why] Need to set VID_N_MUL for 4:2:0 cases [How] Move setting to enc1_stream_encoder_dp_unblank and ensure it is also set for non-4:2:0 cases. Change-Id: If17faba806244d6800ec9ec64990684675f82d0f Signed-off-by: Eric Bernstein --- .../drm/amd/display/dc/dce110/dce110_hw_s

[PATCH 17/26] drm/amd/display: Link train only when link is DP and backend is enabled

From: Samson Tam [Why] In certain cases we do link training when we don't have a backend. [How] In dc_link_set_preferred_link_settings(), store preferred link settings first and then verify that the link is DP and the link stream's backend is enabled. If either is false, then we will not do any

[PATCH 12/26] drm/amd/display: Update ABM crtc state on non-modeset

From: David Francis [Why] Somewhere in the atomic check reshuffle ABM got lost. ABM is a crtc property (copied from a connector property). It can change without a modeset, just like underscan. [How] In the skip_modeset branch of atomic check crtc updates, copy over the abm property. Change-Id:

[PATCH 13/26] drm/amd/display: cache additional dpcd caps for HDR capability check

From: Harmanprit Tatla [Why] Currently we are missing a few checks to see if HDR10 is allowed. In particular we never check for the extended colorimetry bit (whether its present or set to 1). Further we don't read in the dpcd block in DC that would provide these bits. [How] - Added in DC code to

[PATCH 23/26] drm/amd/display: fix odm pipe management

From: Dmytro Laktyushkin There are issues removing surfaces/streams when odm is active. This is a step to fix that Change-Id: Ib2aaa55ca4b504e1d8d75247e84422f1ac9b Signed-off-by: Dmytro Laktyushkin Reviewed-by: Tony Cheng Acked-by: Bhawanpreet Lakha --- .../gpu/drm/amd/display/dc/core/dc

[PATCH 10/26] drm/amd/display: 3.2.22

From: Aric Cyr Change-Id: Iab825a9085c97f9584db201e1b52c48909b57d81 Signed-off-by: Aric Cyr Reviewed-by: Aric Cyr Acked-by: Bhawanpreet Lakha --- drivers/gpu/drm/amd/display/dc/dc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/driv

[PATCH 07/26] drm/amd/display: Add a hysteresis to BTR frame multiplier

From: Aric Cyr [Why] Flickering is observed on some displays when the number of inserted BTR frames changes frequently. [How] Add in a margin of drift to prevent the inserted number of frames from jumping around too frequently. Change-Id: I592ce229f93faa74b5baa2e43a89c9e5a6174103 Signed-off-by:

Irc and Other Things

Greetings All, I was trying to ask a question on IRC but after registering it just complains and I have no idea why, if you want my nick to check it out it is was haweh. Anyhow I'm a student with interest in helping out with your mesa stack. I've already got a few very minor patches in gcc and

Re: Bug#924540: [PATCH xf86-video-ati] Use radeon_finish in drmmode_crtc_scanout_update

On Fri, Mar 15, 2019 at 4:25 AM Michel Dänzer wrote: > > From: Michel Dänzer > > radeon_glamor_finish only works if we're using glamor, otherwise it'll > crash. > > Fixes: ce7db51020d3 "Cancel pending scanout update in > drmmode_crtc_scanout_update" > Bug: https://bugs.debian.org/924540 > Signed

Re: Irc and Other Things

Hi Nick, I can see your username on the radeon IRC channel, so that much is working. When I try to chat it just hands up at "offering DCC CHAT to haweh" though. What kind of error message are you getting ? Thanks, John From: amd-gfx on behalf of nick Sen

Re: Irc and Other Things

On Fri, Mar 15, 2019 at 2:36 PM nick wrote: > > Greetings All, > > I was trying to ask a question on IRC but after registering it just complains > and I have no idea why, > if you want my nick to check it out it is was haweh. Anyhow I'm a student > with interest in helping out with > your mesa s

Re: Slow memory access when using OpenCL without X11

Hi Lauri, Thanks for your persistence. Seeing that this is reproducible on several boards with up-to-date BIOS is really helpful and gives me some confidence that it's more than a weird vendor or board-specific corner case and that we should be able to reproduce it. Yong is going to start looki

Re: [PATCH 5/9] drm/syncobj: use the timeline point in drm_syncobj_find_fence v3

Hi Chunming, I love your patch! Perhaps something to improve: [auto build test WARNING on linus/master] [also build test WARNING on next-20190306] [cannot apply to v5.0] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url: https://github.co