drm/amdgpu/sriov: Need to set in_gpu_reset flag to back after gpu reset

Ping... > -Original Message- > From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of > Emily Deng > Sent: Wednesday, April 25, 2018 2:38 PM > To: amd-gfx@lists.freedesktop.org > Cc: Deng, Emily > Subject: [PATCH 1/2] drm/amdgpu/sriov: Need to set

RE: [PATCH] drm/amdgpu: For sriov reset, move IB test into exclusive mode

Ping... > -Original Message- > From: Emily Deng [mailto:emily.d...@amd.com] > Sent: Wednesday, April 25, 2018 11:26 AM > To: amd-gfx@lists.freedesktop.org > Cc: Deng, Emily > Subject: [PATCH] drm/amdgpu: For sriov reset, move IB test into exclusive > mode > > When

[PATCH 14/32] drm/amdgpu: add VEGAM GFX golden settings

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 39 +++ 1 file changed, 39

[PATCH 15/32] drm/amdgpu: initialize VEGAM GFX

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 21 +++-- 1 file changed, 15 insertions(+), 6

[PATCH 04/32] drm/amdgpu: specify VEGAM ucode SMU load method

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c | 1 + 1 file changed, 1 insertion(+) diff --git

[PATCH 01/32] drm/amdgpu: add VEGAM ASIC type

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 + include/drm/amd_asic_type.h| 1 + 2 files

[PATCH 11/32] drm/amdgpu: add VEGAM SDMA firmware support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 11 --- 1 file changed, 8 insertions(+), 3 deletions(-)

[PATCH 07/32] drm/amdgpu: add VEGAM dc support check

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff

[PATCH 06/32] drm/amdgpu/virtual_dce: add VEGAM support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/dce_virtual.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git

Re: [PATCH] drm/scheduler: don't update last scheduled fence in TDR

NAK on it. First of all, without this patch, Does it cause any issue? second, entity->last_scheduled present the last submiting job. Your this change will break this meaning and don't work, e.g. 1. mirror list has jobA and jobB, assuming they are belonged to same entity, then the

[PATCH 23/32] drm/amd/display: Implement VEGAM device IDs in DM

From: "Jerry (Fangzhi) Zuo" Add CHIP_VEGAM Signed-off-by: Jerry (Fangzhi) Zuo Reviewed-by: Harry Wentland Reviewed-by: Alex Deucher Signed-off-by: Harry Wentland Signed-off-by:

[PATCH 22/32] drm/amd/display: Implement VEGAM device IDs in DC

From: "Jerry (Fangzhi) Zuo" Implement device IDs for VEGAM Signed-off-by: Jerry (Fangzhi) Zuo Reviewed-by: Harry Wentland Acked-by: Alex Deucher Signed-off-by: Harry Wentland

[PATCH 30/32] drm/amd/powerplay: add smumgr support for VEGAM (v2)

From: Eric Huang The smumgr handles communication between the driver and the SMU for power management. v2: fix typo (Alex) Signed-off-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher

[PATCH 20/32] drm/amdgpu: add VEGAM support to vi

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vi.c | 31 +-- 1 file changed, 29 insertions(+), 2

[PATCH 24/32] drm/amdgpu: Add VEGAM support to the legacy DCE 11 module

DC is preferred. Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/dce_v11_0.c | 13 ++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index

[PATCH 26/32] drm/amd/powerplay: add smu75 header files

From: Eric Huang Signed-off-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/powerplay/inc/smu75.h | 760 ++

[PATCH 18/32] drm/amdgpu: add VEGAM VCE firmware support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff

[PATCH 32/32] drm/powerplay: Add powertune table for VEGAM

From: Eric Huang Add the powertune table for VEGAM. Signed-off-by: Eric Huang Signed-off-by: Alex Deucher --- .../gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c | 189 + 1 file changed, 189

[PATCH 09/32] drm/amdgpu: add VEGAM GMC golden settings

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 1 + 1 file changed, 1 insertion(+) diff --git

[PATCH 25/32] drm/amd/display: Use HBR2 if eDP monitor it doesn't advertise link rate

From: Harry Wentland Some eDP displays use the extra link rate table to advertise link rate support. If they do that they don't need to provide link rate through the usual registers. Since we don't currently have support for the extra link rate table default to HBR2 for

[PATCH 16/32] drm/amdgpu: add VEGAM UVD firmware support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c | 5 + 1 file changed, 5 insertions(+) diff --git

[PATCH 12/32] drm/amdgpu: add VEGAM SDMA golden settings

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c | 1 + 1 file changed, 1 insertion(+) diff --git

[PATCH 19/32] drm/amdgpu: add VEGAM to VCE harvest config

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/vce_v3_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git

[PATCH 21/32] drm/amdgpu: add VEGAM pci ids

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 3 +++ 1 file changed, 3 insertions(+) diff --git

[PATCH 02/32] drm/amdgpu: bypass GPU info firmware load for VEGAM

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff

[PATCH 03/32] drm/amdgpu: set VEGAM to ASIC family and ip blocks

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff

[PATCH 05/32] drm/amdgpu: add VEGAM SMU firmware support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 3 +++ drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c | 1 + 2

[PATCH 29/32] drm/amd/powerplay: update process pptables

From: Eric Huang Add functionality to fetch gpio table from vbios. Signed-off-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher ---

[PATCH 10/32] drm/amdgpu: initialize VEGAM GMC (v2)

From: Leo Liu v2: use proper register rather than hardcoding. Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 6 -- 1 file

[PATCH 27/32] drm/amd: add a new struct in atombios.h

From: Eric Huang Signed-off-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/include/atombios.h | 7 +++ 1 file changed, 7 insertions(+)

[PATCH 08/32] drm/amdgpu: skip VEGAM MC firmware load

From: Leo Liu Directly loaded by VBIOS Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 1 + 1 file changed, 1 insertion(+) diff

[PATCH 13/32] drm/amdgpu: add VEGAM GFX firmware support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 42 ++- 1 file changed, 26

[PATCH 17/32] drm/amdgpu: add VEGAM UVD encode support

From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

Re: [PATCH 1/2] drm/amdgpu: abstract bo_base init function

On 04/24/2018 03:35 PM, Chunming Zhou wrote: Change-Id: Ia3e57dbacff05f32b6c02e29aeadabd36f08028e Signed-off-by: Chunming Zhou Reviewed-by: Junwei Zhang some trivial comments. --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 75

RE: [PATCH] drm/scheduler: don't update last scheduled fence in TDR

Reviewed-by : Monk Liu -Original Message- From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of Pixel Ding Sent: 2018年4月25日 16:40 To: amd-gfx@lists.freedesktop.org Cc: Ding, Pixel Subject: [PATCH] drm/scheduler: don't update

Re: [PATCH 2/2] drm/amdgpu: link shadow bo as well

On 04/24/2018 03:35 PM, Chunming Zhou wrote: Shadow BO is located on GTT and its parent (PT and PD) BO could located on VRAM. In some case, the BO on GTT could be evicted but the parent did not. This may cause the shadow BO not be put in the evict list and could not be invalidate correctly.

RE: [PATCH] drm/amdgpu/sriov: reject kms open if TDR not finished or failed

Hi Michel, > This seems like a bad idea, as it would cause basically any userspace which > wants to use the GPU to fail to start during a GPU reset. [Emily] Yes, this is what the change want to do, when driver is doing gpu recover or hardware is doing reset, it doesn't want to be

Re: [PATCH] drm/scheduler: don't update last scheduled fence in TDR

Hi Monk, Please review it. Thanks. — Sincerely Yours, Pixel On 2018/4/25, 4:39 PM, "Pixel Ding" wrote: The current sequence in scheduler thread is: 1. update last sched fence 2. job begin (adding to mirror list) 3. job finish (remove from mirror list)

Re: [PATCH 01/32] drm/amdgpu: add VEGAM ASIC type

On 04/26/2018 10:18 AM, Alex Deucher wrote: From: Leo Liu Signed-off-by: Leo Liu Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 +

[PATCH] drm/amdgpu: For sriov reset, move IB test into exclusive mode

When put the IB test out of exclusive mode, and do sriov reset, the IB test will randomly fail. As out of exclusive mode it uses kiq to do read and write registers, but as it has world switch, the kiq read and write time will be random, sometimes it will beyond the MAX_KIQ_REG_WAIT and then the

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 7:48 AM, Christoph Hellwig wrote: > On Tue, Apr 24, 2018 at 09:32:20PM +0200, Daniel Vetter wrote: >> Out of curiosity, how much virtual flushing stuff is there still out >> there? At least in drm we've pretty much ignore this, and seem to be >> getting

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 8:13 AM, Daniel Vetter wrote: > On Wed, Apr 25, 2018 at 7:48 AM, Christoph Hellwig wrote: >> On Tue, Apr 24, 2018 at 09:32:20PM +0200, Daniel Vetter wrote: >>> Out of curiosity, how much virtual flushing stuff is there still out >>>

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 8:43 AM, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 08:23:15AM +0200, Daniel Vetter wrote: >> For more fun: >> >> https://www.spinics.net/lists/dri-devel/msg173630.html >> >> Yeah, sometimes we want to disable the iommu because the on-gpu >>

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 08:23:15AM +0200, Daniel Vetter wrote: > For more fun: > > https://www.spinics.net/lists/dri-devel/msg173630.html > > Yeah, sometimes we want to disable the iommu because the on-gpu > pagetables are faster ... I am not on this list, but remote NAK from here. This needs

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 09:02:17AM +0200, Daniel Vetter wrote: > Can we please not nack everything right away? Doesn't really motivate > me to show you all the various things we're doing in gpu to make the > dma layer work for us. That kind of noodling around in lower levels to > get them to do

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 02:24:36AM -0400, Alex Deucher wrote: > > It has a non-coherent transaction mode (which the chipset can opt to > > not implement and still flush), to make sure the AGP horror show > > doesn't happen again and GPU folks are happy with PCIe. That's at > > least my

Re: [PATCH] drm/amdgpu: fix amdgpu_atpx_get_client_id()'s return type

On Tue, Apr 24, 2018 at 9:15 AM, Luc Van Oostenryck wrote: > The method struct vga_switcheroo_handler::get_client_id() is defined > as returning an 'enum vga_switcheroo_client_id' but the implementation > in this driver, amdgpu_atpx_get_client_id(), returns an 'int'.

[PATCH] drm/scheduler: don't update last scheduled fence in TDR

The current sequence in scheduler thread is: 1. update last sched fence 2. job begin (adding to mirror list) 3. job finish (remove from mirror list) 4. back to 1 Since we update last sched prior to joining mirror list, the jobs in mirror list already pass the last sched fence. TDR just run the

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 1:48 AM, Christoph Hellwig wrote: > On Tue, Apr 24, 2018 at 09:32:20PM +0200, Daniel Vetter wrote: >> Out of curiosity, how much virtual flushing stuff is there still out >> there? At least in drm we've pretty much ignore this, and seem to be >> getting

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On Tue, Apr 24, 2018 at 05:37:08PM -0400, Andrey Grodzovsky wrote: > > > On 04/24/2018 05:21 PM, Eric W. Biederman wrote: > > Andrey Grodzovsky writes: > > > > > On 04/24/2018 03:44 PM, Daniel Vetter wrote: > > > > On Tue, Apr 24, 2018 at 05:46:52PM +0200, Michel

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 2:13 AM, Daniel Vetter wrote: > On Wed, Apr 25, 2018 at 7:48 AM, Christoph Hellwig wrote: >> On Tue, Apr 24, 2018 at 09:32:20PM +0200, Daniel Vetter wrote: >>> Out of curiosity, how much virtual flushing stuff is there still out >>>

[PATCH] drm/amdgpu/sriov: reject kms open if TDR not finished or failed

Signed-off-by: Monk Liu --- drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 6d55cae..adeca71 100644 ---

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 12:09:05AM -0700, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 09:02:17AM +0200, Daniel Vetter wrote: > > Can we please not nack everything right away? Doesn't really motivate > > me to show you all the various things we're doing in gpu to make the > > dma layer work

Re: [PATCH] drm/amdgpu/sriov: reject kms open if TDR not finished or failed

On 2018-04-25 08:39 AM, Emily Deng wrote: > Signed-off-by: Monk Liu > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c > b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c > index

[PATCH 1/2] drm/amdgpu/sriov: Need to set in_gpu_reset flag to back after gpu reset

After host os reset gpu reset, need to set flag in_gpu_reset to zero. Signed-off-by: Emily Deng --- drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c

Re: [PATCH] drm/amdkfd: Integer overflows in ioctl

On Tue, Apr 24, 2018 at 02:58:10PM -0400, Felix Kuehling wrote: > Reviewed-by: Felix Kuehling > > We could probably add a sanity check for n_devices to avoid user mode > causing excessive memory allocations in the kernel. There is no good > reason for this to be bigger

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 01:54:39AM -0700, Christoph Hellwig wrote: > [discussion about this patch, which should have been cced to the iommu > and linux-arm-kernel lists, but wasn't: > https://www.spinics.net/lists/dri-devel/msg173630.html] > > On Wed, Apr 25, 2018 at 09:41:51AM +0200, Thierry

Re: [PATCH] drm/radeon: Change the default to PCI on PowerPC

On 2018-04-24 09:55 PM, Mathieu Malaterre wrote: > AGP mode is unstable on PowerPC. Symptoms are generally of the form: > > [ 1228.795711] radeon :00:10.0: ring 0 stalled for more than 10240msec > > Reviewed-by: Christian König > Signed-off-by: Mathieu Malaterre

Re: [PATCH xf86-video-amdgpu 03/19] Use ODEV_ATTRIB_PATH where possible for the device node.

On 2018-04-24 08:13 PM, Emil Velikov wrote: > On 10 April 2018 at 09:27, Michel Dänzer wrote: >> On 2018-04-04 04:29 PM, Emil Velikov wrote: >>> From: Emil Velikov >>> >>> Signed-off-by: Emil Velikov >>> --- >>>

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On Wed, Apr 25, 2018 at 3:22 PM, Oleg Nesterov wrote: > On 04/24, Daniel Vetter wrote: >> >> wait_event_killabel doesn't check for fatal_signal_pending before calling >> schedule, so definitely has a nice race there. > > This is fine. See the signal_pending_state() check in

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/25/2018 03:14 AM, Daniel Vetter wrote: On Tue, Apr 24, 2018 at 05:37:08PM -0400, Andrey Grodzovsky wrote: On 04/24/2018 05:21 PM, Eric W. Biederman wrote: Andrey Grodzovsky writes: On 04/24/2018 03:44 PM, Daniel Vetter wrote: On Tue, Apr 24, 2018 at

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/24/2018 05:40 PM, Daniel Vetter wrote: On Tue, Apr 24, 2018 at 05:02:40PM -0400, Andrey Grodzovsky wrote: On 04/24/2018 03:44 PM, Daniel Vetter wrote: On Tue, Apr 24, 2018 at 05:46:52PM +0200, Michel Dänzer wrote: Adding the dri-devel list, since this is driver independent code. On

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

Andrey Grodzovsky writes: > On 04/25/2018 11:29 AM, Eric W. Biederman wrote: > >> Another issue is changing wait_event_killable to wait_event_timeout where I >> need >> to understand >> what TO value is acceptable for all the drivers using the scheduler, or >> maybe

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/25, Andrey Grodzovsky wrote: > > here (drm_sched_entity_fini) is also a bad idea, but we still want to be > able to exit immediately > and not wait for GPU jobs completion when the reason for reaching this code > is because of KILL > signal to the user process who opened the device file.

noveau vs arm dma ops

[discussion about this patch, which should have been cced to the iommu and linux-arm-kernel lists, but wasn't: https://www.spinics.net/lists/dri-devel/msg173630.html] On Wed, Apr 25, 2018 at 09:41:51AM +0200, Thierry Reding wrote: > > API from the iommu/dma-mapping code. Drivers have no

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/24, Eric W. Biederman wrote: > > Let me respectfully suggest that the wait_event_killable on that code > path is wrong. I tend to agree even if I don't know this code. But if it can be called from f_op->release() then any usage of "current" or signals looks suspicious. Simply because

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 09:30:39AM +0200, Daniel Vetter wrote: > On Wed, Apr 25, 2018 at 12:09:05AM -0700, Christoph Hellwig wrote: > > On Wed, Apr 25, 2018 at 09:02:17AM +0200, Daniel Vetter wrote: > > > Can we please not nack everything right away? Doesn't really motivate > > > me to show you

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 09:56:43AM +0200, Thierry Reding wrote: > And to add to the confusion, none of this seems to be an issue on 64-bit > ARM where the generic DMA/IOMMU code from drivers/iommu/dma-iommu.c is > used. In the long term I want everyone to use that code. Help welcome!

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/25, Daniel Vetter wrote: > > On Wed, Apr 25, 2018 at 3:22 PM, Oleg Nesterov wrote: > > On 04/24, Daniel Vetter wrote: > >> > >> wait_event_killabel doesn't check for fatal_signal_pending before calling > >> schedule, so definitely has a nice race there. > > > > This is

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

Andrey Grodzovsky writes: > On 04/25/2018 03:14 AM, Daniel Vetter wrote: >> On Tue, Apr 24, 2018 at 05:37:08PM -0400, Andrey Grodzovsky wrote: >>> >>> On 04/24/2018 05:21 PM, Eric W. Biederman wrote: Andrey Grodzovsky writes: >

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/24, Daniel Vetter wrote: > > wait_event_killabel doesn't check for fatal_signal_pending before calling > schedule, so definitely has a nice race there. This is fine. See the signal_pending_state() check in __schedule(). And this doesn't differ from wait_event_interruptible(), it too

[pull] amdgpu drm-fixes-4.17

Hi Dave, A few fixes for 4.17: - Fix a hang on CZ boards with EDC enabled - Fix hangs related to DP MST handling - Fix a deadlock in irq handling in DC The following changes since commit 221bda4b5f1abfd74159d7bf3703affa62468030: Merge branch 'drm-next-4.17' of

Re: [PATCH 3/3] drm/amdgpu: Switch to interrupted wait to recover from ring hang.

On 04/24/2018 12:30 PM, Eric W. Biederman wrote: "Panariti, David" writes: Andrey Grodzovsky writes: Kind of dma_fence_wait_killable, except that we don't have such API (maybe worth adding ?) Depends on how many places it would be

Re: [PATCH] drm/amdkfd: Integer overflows in ioctl

On 2018-04-25 05:21 AM, Dan Carpenter wrote: > On Tue, Apr 24, 2018 at 02:58:10PM -0400, Felix Kuehling wrote: >> Reviewed-by: Felix Kuehling >> >> We could probably add a sanity check for n_devices to avoid user mode >> causing excessive memory allocations in the kernel.

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/25/2018 11:29 AM, Eric W. Biederman wrote: Another issue is changing wait_event_killable to wait_event_timeout where I need to understand what TO value is acceptable for all the drivers using the scheduler, or maybe it should come as a property of drm_sched_entity. It would not surprise

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Tue, Apr 24, 2018 at 11:43:35PM -0700, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 08:23:15AM +0200, Daniel Vetter wrote: > > For more fun: > > > > https://www.spinics.net/lists/dri-devel/msg173630.html > > > > Yeah, sometimes we want to disable the iommu because the on-gpu > >

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 12:09:05AM -0700, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 09:02:17AM +0200, Daniel Vetter wrote: > > Can we please not nack everything right away? Doesn't really motivate > > me to show you all the various things we're doing in gpu to make the > > dma layer work

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 12:04:29PM +0200, Daniel Vetter wrote: > > Coordinating the backport of a trivial helper in the arm tree is not > > the end of the world. Really, this cowboy attitude is a good reason > > why graphics folks have such a bad rep. You keep poking into random > > kernel

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 01:54:39AM -0700, Christoph Hellwig wrote: > [discussion about this patch, which should have been cced to the iommu > and linux-arm-kernel lists, but wasn't: > https://www.spinics.net/lists/dri-devel/msg173630.html] > > On Wed, Apr 25, 2018 at 09:41:51AM +0200, Thierry

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/24, Andrey Grodzovsky wrote: > > --- a/drivers/gpu/drm/scheduler/gpu_scheduler.c > +++ b/drivers/gpu/drm/scheduler/gpu_scheduler.c > @@ -227,9 +227,10 @@ void drm_sched_entity_do_release(struct > drm_gpu_scheduler *sched, > return; > /** >* The client will not

Re: [PATCH 1/3] signals: Allow generation of SIGKILL to exiting task.

On 04/24, Andrey Grodzovsky wrote: > > Currently calling wait_event_killable as part of exiting process > will stall forever since SIGKILL generation is suppresed by PF_EXITING. See my reply to 2/3, > In our partilaur case AMDGPU driver wants to flush all GPU jobs in > flight before shutting

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 2:41 AM, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 02:24:36AM -0400, Alex Deucher wrote: >> > It has a non-coherent transaction mode (which the chipset can opt to >> > not implement and still flush), to make sure the AGP horror show >> >

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

On 04/25/2018 01:17 PM, Oleg Nesterov wrote: On 04/25, Andrey Grodzovsky wrote: here (drm_sched_entity_fini) is also a bad idea, but we still want to be able to exit immediately and not wait for GPU jobs completion when the reason for reaching this code is because of KILL signal to the user

Re: Ryzen 1600X with mem_encrypt=on

On Wed, Apr 25, 2018 at 1:13 PM, Sami Farin wrote: > Is it supposed to work? Now if I use that option, monitor powers off > (no signal) when it's initializing amdgpu and I have to reset the computer. > Instead of crashing, could it not just give an error or something?

Re: [PATCH v2 2/3] drm/amdgpu: Allow dma_map_sg() coalescing

On 4/17/2018 5:13 PM, Sinan Kaya wrote: > Tested-by: Sinan Kaya > > using QDF2400 and XFX Vega64 GPU for the first two patches. > > ./builddir/tests/amdgpu/amdgpu_test -s 1 > > Suite: Basic Tests > Test: Userptr Test ...passed > > Userptr Test fails without this patch.

Re: [PATCH 3/3] drm/amdgpu: Switch to interrupted wait to recover from ring hang.

Andrey Grodzovsky writes: > On 04/24/2018 12:30 PM, Eric W. Biederman wrote: >> "Panariti, David" writes: >> >>> Andrey Grodzovsky writes: Kind of dma_fence_wait_killable, except that we don't have such API

Re: [Linaro-mm-sig] [PATCH 4/8] dma-buf: add peer2peer flag

On Wed, Apr 25, 2018 at 10:44 AM, Alex Deucher wrote: > On Wed, Apr 25, 2018 at 2:41 AM, Christoph Hellwig wrote: >> On Wed, Apr 25, 2018 at 02:24:36AM -0400, Alex Deucher wrote: >>> > It has a non-coherent transaction mode (which the chipset can opt to

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 5:33 PM, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 12:04:29PM +0200, Daniel Vetter wrote: >> > Coordinating the backport of a trivial helper in the arm tree is not >> > the end of the world. Really, this cowboy attitude is a good reason >> >

Re: [PATCH 1/5] dma-buf: use parameter structure for dma_buf_attach

Hi Christian, I love your patch! Yet something to improve: [auto build test ERROR on drm/drm-next] [also build test ERROR on v4.17-rc2 next-20180424] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] url:

Re: [PATCH 2/3] drm/scheduler: Don't call wait_event_killable for signaled process.

Andrey Grodzovsky writes: > On 04/25/2018 01:17 PM, Oleg Nesterov wrote: >> On 04/25, Andrey Grodzovsky wrote: >>> here (drm_sched_entity_fini) is also a bad idea, but we still want to be >>> able to exit immediately >>> and not wait for GPU jobs completion when the

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 08:33:12AM -0700, Christoph Hellwig wrote: > On Wed, Apr 25, 2018 at 12:04:29PM +0200, Daniel Vetter wrote: > > - dma api hides the cache flushing requirements from us. GPUs love > > non-snooped access, and worse give userspace control over that. We want > > a strict

Re: noveau vs arm dma ops

On Wed, Apr 25, 2018 at 11:35:13PM +0200, Daniel Vetter wrote: > On arm that doesn't work. The iommu api seems like a good fit, except > the dma-api tends to get in the way a bit (drm/msm apparently has > similar problems like tegra), and if you need contiguous memory > dma_alloc_coherent is the