Re: ttm crash on init

On 2018-09-17 12:21 p.m., Tom St Denis wrote: (attached).  I'll try to bisect in a second.  Is anyone aware of this? Tom Bisection led to: a327772a5655ff4fb104c8aae6515faa461df466 is the first bad commit commit a327772a5655ff4fb104c8aae6515faa461df466 Author: Christian König Date: Fr

ttm crash on init

(attached). I'll try to bisect in a second. Is anyone aware of this? Tom [0.00] Linux version 4.19.0-rc1+ (root@raven) (gcc version 8.1.1 20180712 (Red Hat 8.1.1-5) (GCC)) #29 SMP Fri Sep 14 07:30:30 EDT 2018 [0.00] Command line: BOOT_IMAGE=/vmlinuz-4.19.0-rc1+ root=UUID=66163c8

Re: [PATCH 1/2] drm/ttm: set ttm_buffer_object pointer as null after it's freed

o since my workstation depends on our staging tree to work properly...). Tom Am 10.09.2018 um 14:59 schrieb Tom St Denis: Hi Christian, Are you adding new traces or turning on existing ones?  Would you like me to try them out in my setup? Tom On 2018-09-10 8:49 a.m., Christian König

Re: [PATCH 1/2] drm/ttm: set ttm_buffer_object pointer as null after it's freed

Hi Christian, Are you adding new traces or turning on existing ones? Would you like me to try them out in my setup? Tom On 2018-09-10 8:49 a.m., Christian König wrote: Am 10.09.2018 um 14:05 schrieb Huang Rui: On Mon, Sep 10, 2018 at 05:25:48PM +0800, Koenig, Christian wrote: Am 10.09.201

Re: [PATCH] drm/amdgpu: set bulk_moveable to false when a per VM is released

On 2018-09-08 5:12 a.m., Huang Rui wrote: On Wed, Sep 05, 2018 at 05:08:26PM +0200, Christian König wrote: Otherwise we might run into a use after free during bulk move. Signed-off-by: Christian König Is this patch able to fix the KASAN? [ 66.143009] ==

Re: two KASANs in TTM logic

th piglit. May I know the steps/configurations to repro it? Thanks, Ray -Original Message- From: amd-gfx On Behalf Of Tom St Denis Sent: Wednesday, September 5, 2018 9:27 PM To: Koenig, Christian ; Daenzer, Michel ; amd-gfx@lists.freedesktop.org; Deucher, Alexander Subject: Re: two

Re: two KASANs in TTM logic

Logs attached. Tom On 09/05/2018 08:02 AM, Christian König wrote: Still not the slightest idea what is causing this and the patch definitely fixes things a lot. Can you try to enable list debugging in your kernel? Thanks, Christian. Am 04.09.2018 um 19:18 schrieb Tom St Denis: Sure

Re: two KASANs in TTM logic

the patch definitely fixes things a lot. Can you try to enable list debugging in your kernel? Thanks, Christian. Am 04.09.2018 um 19:18 schrieb Tom St Denis: Sure: d2917f399e0b250f47d07da551a335843a24f835 is the first bad commit commit d2917f399e0b250f47d07da551a335843a24f835 Author: Christian K

Re: two KASANs in TTM logic

I can run xonotic-glx and piglit on my Carrizo without a KASAN. Tom On 09/04/2018 10:05 AM, Christian König wrote: The first one should already be fixed. Not sure where the second comes from. Can you narrow that down further? Christian. Am 04.09.2018 um 15:46 schrieb Tom St Denis: First is

Re: drm-next build broken

I use a defconfig + enable AMDGPU $ grep AMDGPU .config CONFIG_DRM_AMDGPU=m CONFIG_DRM_AMDGPU_SI=y CONFIG_DRM_AMDGPU_CIK=y # CONFIG_DRM_AMDGPU_USERPTR is not set # CONFIG_DRM_AMDGPU_GART_DEBUGFS is not set Tom On 08/28/2018 09:28 AM, Lin, Amber wrote: Hi Tom, May I ask which config I can use

drm-next build broken

Hi Amber, This commit: commit 7c3a4287bfc05a26baa9d32d060feffe387b9b84 (HEAD -> amd-staging-drm-next, origin/amd-staging-drm-next, origin/HEAD) Author: Amber Lin Date: Thu Aug 23 10:52:34 2018 -0400 drm/amdgpu: Move KFD parameters to amdgpu After merging KFD into amdgpu, move modu

Re: new KASAN running piglit

stripped Tom On 07/18/2018 12:44 PM, Michel Dänzer wrote: On 2018-07-18 06:17 PM, Michel Dänzer wrote: On 2018-07-18 06:05 PM, Tom St Denis wrote: Hi Christian, This patch: [root@raven linux]# git bisect bad 90f362bdf0d0d06a126a5fd35b084436dd8250ad is the first bad commit commit

Re: new KASAN running piglit

Here's the dmesg I should have attached ... Tom On 07/18/2018 12:05 PM, Tom St Denis wrote: Hi Christian, This patch: [root@raven linux]# git bisect bad 90f362bdf0d0d06a126a5fd35b084436dd8250ad is the first bad commit commit 90f362bdf0d0d06a126a5fd35b084436dd8250ad Author: Christian

new KASAN running piglit

Hi Christian, This patch: [root@raven linux]# git bisect bad 90f362bdf0d0d06a126a5fd35b084436dd8250ad is the first bad commit commit 90f362bdf0d0d06a126a5fd35b084436dd8250ad Author: Christian König Date: Mon Jul 16 14:58:48 2018 +0200 drm/amdgpu: change ring priority after pushing the jo

Re: dmesg warning during init

Finally was able to reproduce with your commit too (a79a94c676e5b06d43b7b236dbf22b70d2b11b39). It's definitely not 100% triggerable hence my bisect missing it. Tom On 07/16/2018 01:14 PM, Tom St Denis wrote: With the tip of drm-next I get this KASAN immediately. Tom On 07/16/2018 12:

Re: dmesg warning during init

With the tip of drm-next I get this KASAN immediately. Tom On 07/16/2018 12:40 PM, Michel Dänzer wrote: On 2018-07-16 06:36 PM, Tom St Denis wrote: Will in a bit.  I'm tracking down a KASAN oops that the tip of drm-next causes with unigine-heaven. Is it different from the one I report

Re: dmesg warning during init

Might be though it's definitely a heisenbug since I zeroed in on that commit easily but now I can't reproduce it with 10+ cycles of loading/killing unigine-heaven On 07/16/2018 12:54 PM, Tom St Denis wrote: On 07/16/2018 12:40 PM, Michel Dänzer wrote: On 2018-07-16 06:36 PM, To

Re: dmesg warning during init

On 07/16/2018 12:40 PM, Michel Dänzer wrote: On 2018-07-16 06:36 PM, Tom St Denis wrote: Will in a bit.  I'm tracking down a KASAN oops that the tip of drm-next causes with unigine-heaven. Is it different from the one I reported an hour ago? https://lists.freedesktop.org/archives/am

Re: dmesg warning during init

Will in a bit. I'm tracking down a KASAN oops that the tip of drm-next causes with unigine-heaven. (geez, on vacation for a week and my Raven is suffering hehehehehe) Tom On 07/16/2018 12:35 PM, Michel Dänzer wrote: On 2018-07-16 05:55 PM, Tom St Denis wrote: I have the same now an

Re: dmesg warning during init

I have the same now and get the attached dmesg. Tom On 07/16/2018 10:25 AM, Michel Dänzer wrote: On 2018-07-16 02:10 PM, Tom St Denis wrote: Hi all, Back from vacation, booting up the latest drm-next kernel and noticed this warning on boot. Attached. I like the comment in the commit

dmesg warning during init

Hi all, Back from vacation, booting up the latest drm-next kernel and noticed this warning on boot. Attached. I like the comment in the commit 5c200f56506a142451f17dbea7befa76dd74a942: "This shouldn't happen, but if it does, we'll get a backtrace of the caller, and update the pin_size values

Re: SOC15 DC support warning

rning On Tue, Jul 3, 2018 at 12:36 PM, Michel Dänzer wrote: On 2018-07-03 06:13 PM, Felix Kuehling wrote: On 2018-07-03 10:19 AM, Tom St Denis wrote: Hi all, This block #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) amdgpu_device_ip_block_add(a

Re: SOC15 DC support warning

Hi Felix, I can (v2) the patch I sent to the list. Ultimately a build warning alone I think is the wrong approach. Cheers, Tom On 07/03/2018 12:13 PM, Felix Kuehling wrote: On 2018-07-03 10:19 AM, Tom St Denis wrote: Hi all, This block #if defined(CONFIG_DRM_AMD_DC)     else if

[PATCH] drm/amd/amdgpu: Convert DC warning to runtime warning.

Use DRM_WARN instead of preprocessor warning so that pre-built kernels will warn if DC is disabled on SOC15 ASICs. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/soc15.c | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b

SOC15 DC support warning

Hi all, This block #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) amdgpu_device_ip_block_add(adev, &dm_ip_block); #else # warning "Enable CONFIG_DRM_AMD_DC for display support on SOC15." #endif in soc15_set_ip_blocks()

[PATCH umr] remove casts and replace with portable formatters in output

Signed-off-by: Tom St Denis --- src/lib/dump_ib.c | 18 +-- src/lib/read_vram.c | 210 +- src/lib/umr_llvm_disasm.c | 8 +- src/lib/umr_read_pm4_stream.c | 4 +- 4 files changed, 120 insertions(+), 120 deletions(-) diff --git a/src

Re: DCN/PP regression

On 06/26/2018 05:09 AM, Michel Dänzer wrote: On 2018-06-26 10:39 AM, Michel Dänzer wrote: On 2018-06-25 03:58 PM, Tom St Denis wrote: With a Tonga dGPU installed in a Raven1 system I see (attached) dmesg warning from init. Also, every warm reboot results in my Tonga failing to init (also

Re: DCN/PP regression

Note: The bug (fail to init tonga) doesn't occur on my Carrizo system so it definitely has something to do with the call to dcn_bw_update_from_pplib() on my Raven. Tom On 06/25/2018 09:58 AM, Tom St Denis wrote: With a Tonga dGPU installed in a Raven1 system I see (attached) dmesg wa

[PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs (v2)

This adds what should be a stable interface to read GPU load from userspace. (v2): Fix comments and name of file per recommendations. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 47 ++ 1 file changed, 47 insertions(+) diff --git a

Re: [PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs

On 06/20/2018 10:37 AM, Abramov, Slava wrote: I see some functions in amdgpu_pm.c have function level documentation, so that it would be good to have this for newly added functions. Sure I can add some comments/docs. Another comment is inline. From: amd-gfx on behalf of Tom St Denis  

Re: [PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs

On 06/20/2018 10:37 AM, Abramov, Slava wrote: I see some functions in amdgpu_pm.c have function level documentation, so that it would be good to have this for newly added functions. Sure I can add some comments/docs. Another comment is inline. From: amd-gfx on behalf of Tom St Denis

[PATCH] drm/amd/amdgpu: Add a GPU_LOAD entry to sysfs

This adds what should be a stable interface to read GPU load from userspace. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c | 41 ++ 1 file changed, 41 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm

[PATCH umr] Add UVD/VCN IB detection to pm4 stream decoder

Signed-off-by: Tom St Denis --- src/lib/umr_read_pm4_stream.c | 49 ++- 1 file changed, 48 insertions(+), 1 deletion(-) diff --git a/src/lib/umr_read_pm4_stream.c b/src/lib/umr_read_pm4_stream.c index b17f232a5381..9da804fc6793 100644 --- a/src/lib

[PATCH umr] Update disasm CPU names to match what mesa uses

Updates the CPU names for vega12/vegam/raven1. Signed-off-by: Tom St Denis --- src/lib/umr_llvm_disasm.c | 10 ++ 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/src/lib/umr_llvm_disasm.c b/src/lib/umr_llvm_disasm.c index 40e31774457f..f772db2eff29 100644 --- a/src/lib

[PATCH umr] PM4 packet streaming support (v2)

. (v2): Add ability to specify ring name and documentation Signed-off-by: Tom St Denis --- doc/sphinx/source/libpm4_stream.rst | 102 +++ doc/sphinx/source/libumr_api.rst| 1 + doc/umr.1 | 6 +- src/app/main.c | 12 +- src/app

[PATCH umr] PM4 packet streaming support

. Signed-off-by: Tom St Denis --- src/app/print_waves.c | 37 ++- src/lib/CMakeLists.txt| 1 + src/lib/umr_read_pm4_stream.c | 243 ++ src/umr.h | 19 4 files changed, 298 insertions(+), 2 deletions(-) create

[PATCH] drm/amd/amdgpu: Convert comments for amdgpu_ttm.c

We're moving to Sphinx style documentation. Conver the comments to that format. Also document some new content. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 254 +++- 1 file changed, 185 insertions(+), 69 deletions(-) diff --

Re: [PATCH] drm/amd/amdgpu: Code comments for the amdgpu_ttm.c driver. (v2)

On 05/18/2018 05:52 AM, Christian König wrote: Am 17.05.2018 um 17:34 schrieb Alex Deucher: On Tue, May 15, 2018 at 10:02 AM, Tom St Denis wrote: NFC just comments. (v2):  Updated based on feedback from Alex Deucher. Signed-off-by: Tom St Denis Reviewed-by: Alex Deucher Just one

[PATCH umr] Introduction of Vega20.

Signed-off-by: Tom St Denis --- scripts/soc15_asic.sh | 3 +- src/lib/asic/CMakeLists.txt | 1 + src/lib/asic/vega20.c | 51 src/lib/asic/vega20.i | 189 src/lib/discover_by_did.c | 6 ++ src/lib

[PATCH] drm/amd/amdgpu: Code comments for the amdgpu_ttm.c driver. (v2)

NFC just comments. (v2): Updated based on feedback from Alex Deucher. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 347 +++- 1 file changed, 340 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b

[PATCH umr] Re-factor ring reading out of app side of build

This will be used later on by the ring/ib decoder library API that is coming. Signed-off-by: Tom St Denis --- src/app/ring_read.c | 24 src/lib/CMakeLists.txt | 1 + src/lib/umr_read_ring_data.c | 53 src/umr.h

[PATCH umr] Update power sensors for --top

The kernel has dropped all but the average Wattage. Signed-off-by: Tom St Denis --- src/app/top.c | 9 - 1 file changed, 9 deletions(-) diff --git a/src/app/top.c b/src/app/top.c index a22f4dd6f7cc..7109e127d081 100644 --- a/src/app/top.c +++ b/src/app/top.c @@ -256,9 +256,6 @@ static

[PATCH] drm/amd/amdgpu: Add some documentation to the debugfs entries

Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 193 +++- 1 file changed, 189 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index c98e59721444

Re: vcn regression on raven1

ig, Christian; amd-gfx@lists.freedesktop.org Subject: Re: vcn regression on raven1 On 05/01/2018 09:34 PM, Tom St Denis wrote: Hi all, I've noticed that on the tip of drm-next vcn playback of video is broken (see dmesg below).  I've bisected it to this commit It may be fixed here as a co

[PATCH] drm/amd/amdgpu: vcn10 Add callback for emit_reg_write_reg_wait

The callback .emit_reg_write_reg_wait was missing for vcn decode which resulted in a kernel oops. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu

vcn regression on raven1

Hi all, I've noticed that on the tip of drm-next vcn playback of video is broken (see dmesg below). I've bisected it to this commit [root@raven linux]# git bisect good 701372349fd55b5396b335580e979ac4dde3dd02 is the first bad commit commit 701372349fd55b5396b335580e979ac4dde3dd02 Author: Alex

[PATCH umr] Introduction of VegaM

Signed-off-by: Tom St Denis --- src/lib/asic/CMakeLists.txt | 1 + src/lib/asic/vegam.c| 40 src/lib/discover_by_did.c | 2 ++ src/lib/discover_by_name.c | 1 + src/umr.h | 1 + 5 files changed, 45 insertions(+) create

[PATCH umr] Capture COMPUTE shaders in ring decoding as well.

Signed-off-by: Tom St Denis --- src/lib/ring_decode.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/lib/ring_decode.c b/src/lib/ring_decode.c index 42265e0a74c9..7e3218a46145 100644 --- a/src/lib/ring_decode.c +++ b/src/lib/ring_decode.c @@ -774,9 +774,9 @@ static

[PATCH umr] Add profiler using scanned wave data arrays.

Signed-off-by: Tom St Denis --- doc/sphinx/source/index.rst| 1 + doc/sphinx/source/profiler.rst | 36 doc/umr.1 | 4 ++ src/app/CMakeLists.txt | 1 + src/app/main.c | 15 - src/app/print_waves.c | 4 +- src/app

[PATCH umr] refactor out wave scanning to new function umr_scan_wave_data()

Signed-off-by: Tom St Denis --- doc/sphinx/source/libwave_status.rst | 28 +++ src/app/print_waves.c| 379 +-- src/lib/CMakeLists.txt | 1 + src/lib/scan_waves.c | 97 + src/umr.h

[PATCH umr] Allow specifying a zero size for --vm-disasm to have it size the shader automatically

Signed-off-by: Tom St Denis --- doc/umr.1 | 3 ++- src/app/main.c | 9 - 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/doc/umr.1 b/doc/umr.1 index 83706b887894..f1f5fec55946 100644 --- a/doc/umr.1 +++ b/doc/umr.1 @@ -115,7 +115,8 @@ Write 'size' bytes (

[PATCH] drm/amd/amdgpu: Add missing DCN CMx_TEST_DEBUG bitfields

Only the bitfields for CM0 were initially added. This adds the rest for CM1..CM3 that way umr can pick up the bitfields for all of the debug registers. Signed-off-by: Tom St Denis --- .../gpu/drm/amd/include/asic_reg/dcn/dcn_1_0_sh_mask.h| 15 +++ 1 file changed, 15 insertions

Re: [PATCH 3/3] drm/amd/pp: Remove struct pp_gpu_power

On 04/11/2018 01:23 PM, Alex Deucher wrote: On Wed, Apr 11, 2018 at 2:31 AM, Rex Zhu wrote: This struct can't be used for all asics. Currently smu only calculate average gpu power in real time. for vddc/vddci/max power, we can read them per 1ms through some registers. but 1. the type of re

Re: [PATCH] drm/amd/amdgpu: Fix amdgpu_debugfs_gpr_read debugfs entry

On 04/11/2018 08:17 AM, Christian König wrote: Am 11.04.2018 um 14:03 schrieb Tom St Denis: On 04/11/2018 07:58 AM, Christian König wrote:   -    if (size & 3 || *pos & 3) +    if (size & 3 || size > (4 * AMDGPU_DEBUGFS_MAX_SGPR_READ)) I think checking the position al

Re: [PATCH] drm/amd/amdgpu: Fix amdgpu_debugfs_gpr_read debugfs entry

be preceded by a seek to set the higher order bits anyways. Cheers, Tom Christian. Am 11.04.2018 um 13:55 schrieb Tom St Denis: Ping? On 04/09/2018 08:16 AM, Tom St Denis wrote: We don't need to check the alignment of the offset and there was potential a buffer overflow as well.

Re: [PATCH] drm/amd/amdgpu: Fix amdgpu_debugfs_gpr_read debugfs entry

Ping? On 04/09/2018 08:16 AM, Tom St Denis wrote: We don't need to check the alignment of the offset and there was potential a buffer overflow as well. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++-- 1 file changed, 6 insertions(+), 2 dele

[PATCH umr] Add 'disasm_early_term' option.

For UMDs that don't use the 0xBF9F shader terminator marker this option allows the disassembler to stop once it hits the first s_endpgm opcode. Signed-off-by: Tom St Denis --- doc/sphinx/source/basic.rst | 69 +++-- doc/umr.1

[PATCH umr] Fix VMID of chained IBs

We were using the VMID field literally when inside an IB it's inherited instead Signed-off-by: Tom St Denis --- src/lib/dump_ib.c | 8 src/lib/ring_decode.c | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/lib/dump_ib.c b/src/lib/dump_ib.c

Re: [PATCH] drm/amd/display: Don't spam debug messages

Tested-by: Tom St Denis Works for me! Thanks, Tom On 04/09/2018 02:06 PM, Harry Wentland wrote: Signed-off-by: Harry Wentland --- drivers/gpu/drm/amd/display/include/logger_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/include

spam in dmesg log from display

This commit commit 5fadc6fe773862f59f8a54480f06f097f0719c26 Author: Bhawanpreet Lakha Date: Thu Mar 15 13:01:46 2018 -0400 drm/amd/display: Add Dynamic debug prints Created Macros for DC_LOG_XXX to pr_debug() & DRM_DEBUG_KMS. Signed-off-by: Bhawanpreet Lakha Reviewed-by: Ha

[PATCH] drm/amd/amdgpu: Fix amdgpu_debugfs_gpr_read debugfs entry

We don't need to check the alignment of the offset and there was potential a buffer overflow as well. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/a

[PATCH umr] Add trap temp SGPRs to the read_gpr call

If trap_en and priv are set in the SQ_WAVE_STATUS then read 12 or 16 words from 0x6C (SQ_WAVE_TTMP0...) and display inline with rest of SGPRs. Signed-off-by: Tom St Denis --- src/app/print_waves.c | 33 ++--- src/lib/read_gpr.c| 16 +++- src/lib

Re: Add SQ_WAVE_TTMPx support for gfx9

SGPRs, and 0x6c is the operand encoding of TTMP0. I think umr should just use the SGPR read path with the correct index. Thanks, Nicolai On 06.04.2018 16:51, Tom St Denis wrote: Patches attached for both umr/kernel. Tested on my Raven1.  I'll circle ba

[PATCH umr] Add SQ_WAVE_TTMP registers to no_kernel code path

Signed-off-by: Tom St Denis --- src/lib/wave_status.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/src/lib/wave_status.c b/src/lib/wave_status.c index 3b537abefcc8..516e8d33a948 100644 --- a/src/lib/wave_status.c +++ b/src/lib/wave_status.c @@ -73,8 +73,10

Re: [PATCH umr] add SQ_WAVE_TTMP registers for gfx8 devices

Both the vi and ai read only 32 words (in umr). I'll fix those to 64 words before committing these. Also I'm rushing a bit. I didn't add no_kernel support but I'll do that in a later commit since I want to get the functionality out sooner than later. On 04/06/2018 12

[PATCH umr] add SQ_WAVE_TTMP registers for gfx8 devices

Signed-off-by: Tom St Denis --- src/lib/wave_status.c | 18 +++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/src/lib/wave_status.c b/src/lib/wave_status.c index 92ce624c07a9..e2081bcd2377 100644 --- a/src/lib/wave_status.c +++ b/src/lib/wave_status.c @@ -101,7

[PATCH] drm/amd/amdgpu: Add SQ_WAVE_TTMP registers for gfx8 debugfs

Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index b0e591eaa71a..8a53869b7937 100644 --- a/drivers/gpu/drm/amd

Add SQ_WAVE_TTMPx support for gfx9

Patches attached for both umr/kernel. Tested on my Raven1. I'll circle back to adding gfx8 after lunch. Tom >From e8171c0abbb2806d644065f5e212a7d8b35ff7f8 Mon Sep 17 00:00:00 2001 From: Tom St Denis Date: Fri, 6 Apr 2018 10:34:47 -0400 Subject: [PATCH] drm/amd/amdgpu: Add SQ_W

[PATCH] drm/amd/amdgpu: Merge the smc/pcie/didt debugfs read/write methods (v2)

Fold the read/write methods of the various register access debugfs entries into op functions. (v2): Merge pcie/didt/smc op function into one Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 175 +--- 1 file changed, 56 insertions(+), 119

Re: [PATCH] drm/amd/amdgpu: Merge the smc/pcie/didt debugfs read/write methods

Hi Christian, I was thinking of merging the didt/smc/pcie ops into one themselves. I'll see what I can come up with without spending too much time on it. Cheers, Tom On 04/04/2018 10:02 AM, Christian König wrote: Am 04.04.2018 um 15:35 schrieb Tom St Denis: Fold the read/write metho

[PATCH] drm/amd/amdgpu: Merge the smc/pcie/didt debugfs read/write methods

Fold the read/write methods of the various register access debugfs entries into op functions. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 140 +++- 1 file changed, 57 insertions(+), 83 deletions(-) diff --git a/drivers/gpu/drm/amd

[PATCH umr 1/2] enable follow_ib by default now

Signed-off-by: Tom St Denis --- doc/sphinx/source/basic.rst | 66 ++--- doc/sphinx/source/ring.rst | 17 doc/umr.1 | 4 +-- src/app/main.c | 7 ++--- 4 files changed, 39 insertions(+), 55 deletions(-) diff

[PATCH umr 2/2] Simplify the ability to read multiple registers at once

0x22010010 vcn10.mmUVD_SEMA_ADDR_LOW => 0x vcn10.mmUVD_SEMA_ADDR_HIGH => 0x vcn10.mmUVD_UDEC_DBW_UV_ADDR_CONFIG => 0x22010010 ...... Signed-off-by: Tom St Denis --- doc/sphinx/source/register_access.rst | 9 + doc/umr.1 | 7 ++- src/app/main.c

regression on raven (gmc9)

This commit [root@raven linux]# git bisect bad c704d7a15d292e6e222b7e990ed7e9f41617cdd8 is the first bad commit commit c704d7a15d292e6e222b7e990ed7e9f41617cdd8 Author: Alex Deucher Date: Tue Mar 27 17:10:56 2018 -0500 drm/amdgpu/gmc9: use amdgpu_ring_emit_reg_write_reg_wait in gpu tlb fl

[PATCH] Add srbm selection to both umr and kernel

tions were filtering down into the library. Cheers, Tom >From 236e1b84a6a8a3a91cf487a73fe4dd48aa245b3b Mon Sep 17 00:00:00 2001 From: Tom St Denis Date: Thu, 29 Mar 2018 08:41:28 -0400 Subject: [PATCH umr] Add support for SRBM selection (v3) Also clean up grbm debugfs addressing by hoistin

Re: amdgpu hangs on boot or shutdown on AMD Raven Ridge CPU (Engineer Sample)

I have a retail Ryzen 2400G in an ASUS B350-TUF motherboard and it works fine (well better than this) on the amd-staging-drm-next kernel. Tom On 03/22/2018 05:09 AM, Daniel Drake wrote: Hi Alex, On Tue, Feb 20, 2018 at 10:18 PM, Alex Deucher wrote: It seems that we are not alone seeing am

[PATCH umr] Fix PBA addressing on APUs with a non-zero FB_OFFSET

Tested with my Carrizo and Polaris10. Signed-off-by: Tom St Denis --- src/lib/read_vram.c | 17 + 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/src/lib/read_vram.c b/src/lib/read_vram.c index 86864c6076bd..de9b6497f419 100644 --- a/src/lib/read_vram.c +++ b/src

[PATCH] drm/amd/amdgpu: fix offset into page with amdgpu_iomem debugfs file

The offset inside the page wasn't included in the copy call meaning the start of the page was being read/written instead. Reported-by: Jay Cornwall Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --

[PATCH] drm/amd/display: Correct VGA handoff for DCN1 devices

Partial revert of 1b0ff66bc0bf1a0559255cb7b066a65d55491974 Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc

Re: vga handover for raven

to have fixed the issue. Thanks, Tom On 03/14/2018 10:08 AM, Harry Wentland wrote: On 2018-03-14 09:54 AM, Tom St Denis wrote: Hi, In the original version of the patch we had: @@ -358,7 +360,9 @@ struct dce_hwseq_registers {     HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, OTG0_),\    

Re: vga handover for raven

Hi Harry, Reverting that on top of the current amd-staging-drm-next seems to have fixed the issue. Thanks, Tom On 03/14/2018 10:08 AM, Harry Wentland wrote: On 2018-03-14 09:54 AM, Tom St Denis wrote: Hi, In the original version of the patch we had: @@ -358,7 +360,9 @@ struct

vga handover for raven

Hi, In the original version of the patch we had: @@ -358,7 +360,9 @@ struct dce_hwseq_registers { HWSEQ_PIXEL_RATE_MASK_SH_LIST(mask_sh, OTG0_),\ HWS_SF1(OTG0_, PHYPLL_PIXEL_RATE_CNTL, PHYPLL_PIXEL_RATE_SOURCE, mask_sh), \ HWS_SF(, DCHUBBUB_GLOBAL_TIMER_CNTL, DCHUBB

[PATCH] drm/amd/amdgpu: Add missing "DDR4" label

The commit d296278fd372003fc69588acfd0c0c5edbdf4874 added support for detecting DDR4 but omitted the label that is printed out in amdgpu_bo_init() resulting in a KASAN error. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 3 ++- 1 file changed, 2 insertions(+), 1

Re: Modesetting (amdgpudrmfb) artifacts on RAVEN APU

It's an old patch but does drm-fixes-4.16 have this patch commit e92b44fdee3ec0b679bacdb9c7e95e55699167f0 Refs: v4.14-rc3-1871-ge92b44fdee3e Author: Tony Cheng AuthorDate: Thu Oct 5 14:38:46 2017 -0400 Commit: Alex Deucher CommitDate: Sat Oct 21 16:52:21 2017 -0400 drm

Re: Modesetting (amdgpudrmfb) artifacts on RAVEN APU

tended on that branch? Harry Harry On 2018-03-07 07:55 AM, Tom St Denis wrote: It's in the next set of display patches for drm-next but I don't think it'll be in 4.16 at this point unless it gets cherry picked out. Alex would know for sure since he co-ordinates that with

Re: Modesetting (amdgpudrmfb) artifacts on RAVEN APU

It's in the next set of display patches for drm-next but I don't think it'll be in 4.16 at this point unless it gets cherry picked out. Alex would know for sure since he co-ordinates that with Dave. Tom On 03/06/2018 12:38 PM, KARBOWSKI Piotr wrote: On 2018-03-06 18:28, Tom St

Re: Modesetting (amdgpudrmfb) artifacts on RAVEN APU

I routinely rebase this patch on top of our amd-staging-drm-next tree at fdo. https://cgit.freedesktop.org/~agd5f/linux/log/?h=amd-staging-drm-next Tom On 03/06/2018 12:27 PM, KARBOWSKI Piotr wrote: Hi, On 2018-03-06 12:18, Tom St Denis wrote: I believe this is the same issue I had which

Re: Modesetting (amdgpudrmfb) artifacts on RAVEN APU

I believe this is the same issue I had which is a VGA handoff problem. Can you try this patch the display team sent me? Harry: Will this patch be promoted in the next cycle? Tom On 03/05/2018 11:40 AM, KARBOWSKI Piotr wrote: Hi list, I'd like to report a very odd screen artifacts while runn

[PATCH] drm/amd/amdgpu: Mask rptr as well in ring debugfs

The read/write pointers on sdma4 devices increment beyond the ring size and should be masked. Tested on my Ryzen 2400G. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu

[PATCH] drm/amd/amdgpu: Correct VRAM width for APUs with GMC9

DDR4 has a 64-bit width not 128-bits. It was reporting twice the width. Tested with my Ryzen 2400G. Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers

[PATCH] drm/amdgpu: replace iova debugfs file with iomem (v3)

This allows access to pages allocated through the driver with optional IOMMU mapping. v2: Fix number of bytes copied and add write method v3: drop check for kmap return Original-by: Christian König Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 102

Re: [PATCH v3 00/13] gpu: drm: amd: remove unused headers

This will break umr since we source the headers from the kernel. The kernel might not use them but the different IP blocks have deltas that umr is aware of. One might argue that we should then publish the headers somewhere else that is public but the kernel is our vehicle right now. Thought

Re: [PATCH] drm/amdgpu: replace iova debugfs file with iomem (v2)

On 12/02/18 12:16 PM, Christian König wrote: Am 12.02.2018 um 17:40 schrieb Tom St Denis: On 09/02/18 01:44 PM, Christian König wrote: Am 09.02.2018 um 19:19 schrieb Tom St Denis: On 09/02/18 01:17 PM, Christian König wrote: Am 09.02.2018 um 18:28 schrieb Tom St Denis: On 09/02/18 12:27 PM

Re: [PATCH] drm/amdgpu: replace iova debugfs file with iomem (v2)

On 09/02/18 01:44 PM, Christian König wrote: Am 09.02.2018 um 19:19 schrieb Tom St Denis: On 09/02/18 01:17 PM, Christian König wrote: Am 09.02.2018 um 18:28 schrieb Tom St Denis: On 09/02/18 12:27 PM, Tom St Denis wrote: From: Christian König Oops, I'll remove this from the c

[PATCH] drm/amdgpu: replace iova debugfs file with iomem (v3)

From: Christian König This allows access to pages allocated through the driver with optional IOMMU mapping. v2: Fix number of bytes copied and add write method v3: drop check for kmap return Original-by: Christian König Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c

Re: [PATCH] drm/amdgpu: replace iova debugfs file with iomem (v2)

On 09/02/18 01:17 PM, Christian König wrote: Am 09.02.2018 um 18:28 schrieb Tom St Denis: On 09/02/18 12:27 PM, Tom St Denis wrote: From: Christian König Oops, I'll remove this from the commit message before pushing :-) I did give you credit below though. The patch before this one

Re: [PATCH] drm/amdgpu: replace iova debugfs file with iomem (v2)

On 09/02/18 12:27 PM, Tom St Denis wrote: From: Christian König Oops, I'll remove this from the commit message before pushing :-) I did give you credit below though. Tom This allows access to pages allocated through the driver with optional IOMMU mapping. v2: Fix number of bytes c

[PATCH] drm/amdgpu: replace iova debugfs file with iomem (v2)

From: Christian König This allows access to pages allocated through the driver with optional IOMMU mapping. v2: Fix number of bytes copied and add write method Original-by: Christian König Signed-off-by: Tom St Denis --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 110

Re: [PATCH 5/5] drm/amdgpu: replace iova debugfs file with iomem

On 09/02/18 09:56 AM, Christian König wrote: Am 09.02.2018 um 15:51 schrieb Tom St Denis: On 09/02/18 09:12 AM, Christian König wrote: No, there is simply no need to initialize the system domain. What are the values of p->mapping and adev->mman.bdev.dev_mapping when they don't matc

Re: [PATCH 5/5] drm/amdgpu: replace iova debugfs file with iomem

On 09/02/18 09:12 AM, Christian König wrote: No, there is simply no need to initialize the system domain. What are the values of p->mapping and adev->mman.bdev.dev_mapping when they don't match? Maybe we are allocating memory before initializing adev->mman.bdev.dev_mapping. In my test setup I

<    1   2   3   4   5   6   7   8   9   10   >