Am 19.10.2017 um 19:40 schrieb Harish Kasiviswanathan:
A single KFD eviction fence is attached to all the BOs of a process
including BOs imported. This fence ensures that all BOs belonging to
that process stays resident when the process queues are active.
Don't add this eviction fence to any syn
Am 20.10.2017 um 05:33 schrieb Monk Liu:
cleanups, now only operate on the given ring
Change-Id: I42ee081696ac348660d38be68807f34090ffcce2
Signed-off-by: Monk Liu
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 6 +++---
drivers/gpu/drm/amd/amdgpu/amdgpu_f
the variable ref_clock was assigned same
value twice in same function.
Change-Id: Id26ec6c8265ba53e0cb84511c825c9a446cb18f9
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
b/drivers/g
refresh_rate was not initialized when program
display gap.
this patch can fix vce ring test failed
when do S3 on Polaris10.
Change-Id: Ie7bb363a44df0ea5dca3c03b8d884ac42a5db0a7
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 6 +++---
1 file changed, 3 insertions(+)
Change-Id: Ic581d8f7acc4ba7b545a611c071815a5c53f8bb5
Signed-off-by: Evan Quan
---
drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c | 28 ++
1 file changed, 6 insertions(+), 22 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/rv_hwmgr.c
b/drivers/gpu/drm/amd/po
v2: update the ring_buffer_end address
v3: update the commit log
Change-Id: If3b79428b32ffab57b4e75f9c20c2b2d7e600223
Signed-off-by: Evan Quan
---
drivers/gpu/drm/amd/amdgpu/psp_v10_0.c | 16 ++--
drivers/gpu/drm/amd/amdgpu/psp_v3_1.c | 16 ++--
2 files changed, 28 ins
Am 20.10.2017 um 05:33 schrieb Monk Liu:
since gpu_scheduler source domain cannot access amdgpu variable
so need create the hang_limit membewr for sched, and it can
refer it for the upcoming GPU RESET patches
Change-Id: I977ae2717e55a8b87c59e58a288bffc3b458b653
Signed-off-by: Monk Liu
---
dri
Am 20.10.2017 um 05:33 schrieb Monk Liu:
merge the setting guilty on context into this function
to avoid implement extra routine.
Change-Id: I7a0063464fdc85d5ac9080046380e745565ff540
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c| 4 ++--
drivers/gpu/drm/amd/sche
Am 20.10.2017 um 05:33 schrieb Monk Liu:
for user context there will be a guilty pointer in entity
that points to the guilty member of its context, thus we
cant track if a given entity is not from kernel ctx and
if it is already marked guilty.
Change-Id: Ie0a30830d89f52a6c4a514e67206140698a46367
Am 20.10.2017 um 05:33 schrieb Monk Liu:
Change-Id: I44019f6475b1eaaba55633cf5f8bb84284f19a2c
Signed-off-by: Monk Liu
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
From: Christian König
Filter the placement mask before using it. In theory it could be that we
have other flags set here as well.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd
From: Christian König
We always use the BO mem now.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 4 ++--
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c| 6 +++---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h| 2 +
From: Christian König
No point in doing this.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index
From: Christian König
The GTT manager handles the GART address space anyway, so it is
completely pointless to keep the same information around twice.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 3 --
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 ++---
From: Christian König
use amdgpu_gtt_mgr_is_allocated() instead.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c| 15 +--
drivers/gpu/drm/amd/amdgp
On 20/10/17 11:20 AM, Christian König wrote:
> From: Christian König
>
> Filter the placement mask before using it. In theory it could be that we
> have other flags set here as well.
>
> Signed-off-by: Christian König
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 ++-
> 1 file changed, 2
> Don't set it directly, make it a parameter of amd_sched_init().
Okay
the indent is somehow always looks incorrect in email, but it already aligned
actually
-Original Message-
From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
Sent: 2017年10月20日 17:03
To: Liu, Monk ; amd-
> You need to walk the list of entities (with lock held) and check if
entity->fence_context == job->s_fence->scheduled.context and only if you
found the right one set the guilty flag there.
Yeah you remind me, I did this in the old gpu reset patch, will add them back !
BR Monk
-Original Mes
After starting VNC server or running CTS test, kernel will hang and
can see below call trace:
[961816] INFO: task khugepaged:42 blocked for more than 120 seconds.
[968581] Tainted: G OE 4.13.0 #1
[973495] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
this
Am 20.10.2017 um 11:45 schrieb Xiangliang.Yu:
After starting VNC server or running CTS test, kernel will hang and
can see below call trace:
[961816] INFO: task khugepaged:42 blocked for more than 120 seconds.
[968581] Tainted: G OE 4.13.0 #1
[973495] "echo 0 > /proc/sys/kernel/
This reverts commit 7ae4acd21e9e264afb079e23d43bcf2238c7dbea.
fix s3 hang issue.
Change-Id: Iefb61cf8072b83146d867961b22f13ad83ab0106
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 38 ---
1 file changed, 38 deletions(-)
diff --git a/drivers/
Bug: amdgpu_job_free_cb was accessing s_job->s_entity when the allocated
amdgpu_ctx (and the entity inside it) were already deallocated from
amdgpu_cs_parser_fini.
Fix: Save job's priority on it's creation instead of accessing it from
s_entity later on.
Signed-off-by: Andrey Grodzovsky
---
driv
It is intended to sabstitute the bounded fifo we are currently
using.
Signed-off-by: Andrey Grodzovsky
---
drivers/gpu/drm/amd/scheduler/spsc_queue.h | 120 +
1 file changed, 120 insertions(+)
create mode 100644 drivers/gpu/drm/amd/scheduler/spsc_queue.h
diff --git
Switch from kfifo to SPSC queue.
Bug:
Kfifo is limited at size, during GPU reset it would fill up to limit
and the pushing thread (producer) would wait for the scheduler worker to
consume the items in the fifo while holding reservation lock
on a BO. The gpu reset thread on the other hand blocks th
On Fri, Oct 20, 2017 at 3:21 AM, Rex Zhu wrote:
> the variable ref_clock was assigned same
> value twice in same function.
>
> Change-Id: Id26ec6c8265ba53e0cb84511c825c9a446cb18f9
> Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c | 2 --
> 1 fi
On Fri, Oct 20, 2017 at 3:21 AM, Rex Zhu wrote:
> refresh_rate was not initialized when program
> display gap.
> this patch can fix vce ring test failed
> when do S3 on Polaris10.
>
> Change-Id: Ie7bb363a44df0ea5dca3c03b8d884ac42a5db0a7
> Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
> ---
On Fri, Oct 20, 2017 at 6:52 AM, Rex Zhu wrote:
> This reverts commit 7ae4acd21e9e264afb079e23d43bcf2238c7dbea.
>
> fix s3 hang issue.
>
> Change-Id: Iefb61cf8072b83146d867961b22f13ad83ab0106
> Signed-off-by: Rex Zhu
I think the DAL guys may have a proper fix for this, but either way:
Acked-by:
Still looking for Reviewed-by for patches 2, 3, 22-24.
Harry
On 2017-10-19 01:46 PM, Harry Wentland wrote:
> * Fixed hotplug issue (mainly on HDMI, but affected DP as well)
> * Use new DRM state iterators
> * Couple of Dave's cleanups rerolled on new DML
> * Bunch of Raven work
>
> Anthony K
> -Original Message-
> From: Evan Quan [mailto:evan.q...@amd.com]
> Sent: Friday, October 20, 2017 4:32 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Zhu, Rex; Deucher, Alexander; Quan, Evan
> Subject: [PATCH 1/2] drm/amd/powerplay: retrieve the real-time coreClock
> values
There should be
> -Original Message-
> From: Evan Quan [mailto:evan.q...@amd.com]
> Sent: Friday, October 20, 2017 4:38 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Zhu, Rex; Deucher, Alexander; Quan, Evan
> Subject: [PATCH 2/2] drm/amdgpu/psp: prevent page fault by checking
> write_frame address(v3)
Nee
Add a new context creation function that allows specifying the context
priority.
A high priority context has the potential of starving lower priority
contexts. The current kernel driver implementation allows only apps
that hold CAP_SYS_NICE or DRM_MASTER to acquire a priority above
AMDGPU_CTX_PRIO
Generated using make headers_install from:
airlied/drm-next 282dc83 Merge tag 'drm-intel-next-2017-10-12' ...
Signed-off-by: Andres Rodriguez
---
include/drm/amdgpu_drm.h | 31 ++-
1 file changed, 30 insertions(+), 1 deletion(-)
diff --git a/include/drm/amdgpu_drm.h
Reviewed-by: Andrey Grodzovsky
Also just a reminder to submit a follow up to patch 15 to fix memory leak.
Thanks,
Andrey
On 2017-10-20 10:45 AM, Harry Wentland wrote:
Still looking for Reviewed-by for patches 2, 3, 22-24.
Harry
On 2017-10-19 01:46 PM, Harry Wentland wrote:
* Fixed hot
From: Michel Dänzer
We were leaking it.
(Ported from amdgpu commit cfccf4c4e7e5c73fe4040fabeb1b43283cf29b33)
Signed-off-by: Michel Dänzer
---
src/radeon_kms.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/radeon_kms.c b/src/radeon_kms.c
index df438768c..f79af3415 100644
--- a/src/ra
From: Michel Dänzer
We were leaking it.
(Inspired by amdgpu commit 9d84934309e4ccd9a43c73d958b8ff10ef2fc990)
Signed-off-by: Michel Dänzer
---
src/radeon_kms.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/radeon_kms.c b/src/radeon_kms.c
index 9ef51693a..df438768c 1
From: Michel Dänzer
We were not doing so in all cases, leaking memory allocated by the
latter.
(Cherry picked from amdgpu commit f6b39bcd45cb06976ba8a3600df77fc471c63995)
Signed-off-by: Michel Dänzer
---
src/drmmode_display.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --g
From: Michel Dänzer
We were leaking the memory allocated by TimerSet.
(Ported from amdgpu commit 84aad09f18fed6b52b0c073f0bbd675a6de07807)
Signed-off-by: Michel Dänzer
---
src/radeon_kms.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/radeon_kms.c b/src/radeon_kms.c
index f79af3415.
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Michel Dänzer
> Sent: Friday, October 20, 2017 11:28 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH xf86-video-ati 4/4] Always call drmModeFreeProperty after
> drmModeGetProperty
>
From: Keith Packard
Outputs may have NULL mode_output (connector) pointers if the
connector disappears while the server is running. Skip these when
resetting outputs with BAD link status.
(Ported from xserver commit 37f4e7651a2fd51efa613a08a1e705553be33e76)
Signed-off-by: Michel Dänzer
---
sr
From: Keith Packard
Outputs may have NULL mode_output (connector) pointers if the
connector disappears while the server is running. Skip these when
resetting outputs with BAD link status.
(Ported from xserver commit 37f4e7651a2fd51efa613a08a1e705553be33e76)
Signed-off-by: Michel Dänzer
---
sr
-Original Message-
From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
Sent: Friday, October 20, 2017 3:13 AM
To: Kasiviswanathan, Harish ;
amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/amdgpu: Sync KFD fence only for move/evict
Am 19.10.2017 um 19:40 schrieb Haris
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Michel Dänzer
> Sent: Friday, October 20, 2017 11:50 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH xf86-video-amdgpu] modesetting: Skip no-longer-present
> connectors when resetting
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Michel Dänzer
> Sent: Friday, October 20, 2017 11:50 AM
> To: amd-gfx@lists.freedesktop.org
> Subject: [PATCH xf86-video-ati] modesetting: Skip no-longer-present
> connectors when resetting BA
On 2017-10-20 09:32 AM, Andrey Grodzovsky wrote:
Bug: amdgpu_job_free_cb was accessing s_job->s_entity when the allocated
amdgpu_ctx (and the entity inside it) were already deallocated from
amdgpu_cs_parser_fini.
Fix: Save job's priority on it's creation instead of accessing it from
s_entity l
On 2017-10-20 11:59 AM, Andres Rodriguez wrote:
On 2017-10-20 09:32 AM, Andrey Grodzovsky wrote:
Bug: amdgpu_job_free_cb was accessing s_job->s_entity when the allocated
amdgpu_ctx (and the entity inside it) were already deallocated from
amdgpu_cs_parser_fini.
Fix: Save job's priority on it
On 2017-10-20 12:19 PM, Andrey Grodzovsky wrote:
On 2017-10-20 11:59 AM, Andres Rodriguez wrote:
On 2017-10-20 09:32 AM, Andrey Grodzovsky wrote:
Bug: amdgpu_job_free_cb was accessing s_job->s_entity when the allocated
amdgpu_ctx (and the entity inside it) were already deallocated from
am
On 2017-10-20 12:26 PM, Andres Rodriguez wrote:
On 2017-10-20 12:19 PM, Andrey Grodzovsky wrote:
On 2017-10-20 11:59 AM, Andres Rodriguez wrote:
On 2017-10-20 09:32 AM, Andrey Grodzovsky wrote:
Bug: amdgpu_job_free_cb was accessing s_job->s_entity when the
allocated
amdgpu_ctx (and th
A single KFD eviction fence is attached to all the BOs of a process
including BOs imported. This fence ensures that all BOs belonging to
that process stays resident when the process queues are active.
Don't add this eviction fence to any sync object unless it is a move or
evict job. These jobs are
From: "Leo (Sunpeng) Li"
Do not create dc_state within display_resume, since it's being
constructed within atomic check.
Signed-off-by: Leo (Sunpeng) Li
---
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4
1 file changed, 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/display/amdg
Ping.
On 2017-10-19 03:14 PM, Eric Huang wrote:
Eric Huang (2):
drm/amd/powerplay: add one smc message for Vega10
drm/amd/powerplay: fix performance drop on Vega10
drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c | 6 ++
drivers/gpu/drm/amd/powerplay/inc/vega10_ppsmc.h | 3 ++-
Am 20.10.2017 um 18:51 schrieb Andrey Grodzovsky:
On 2017-10-20 12:26 PM, Andres Rodriguez wrote:
On 2017-10-20 12:19 PM, Andrey Grodzovsky wrote:
On 2017-10-20 11:59 AM, Andres Rodriguez wrote:
On 2017-10-20 09:32 AM, Andrey Grodzovsky wrote:
Bug: amdgpu_job_free_cb was accessing s_jo
On 2017-10-20 01:59 PM, sunpeng...@amd.com wrote:
> From: "Leo (Sunpeng) Li"
>
> Do not create dc_state within display_resume, since it's being
> constructed within atomic check.
>
> Signed-off-by: Leo (Sunpeng) Li
Reviewed-by: Harry Wentland
Harry
> ---
> drivers/gpu/drm/amd/display/amdgp
Am 20.10.2017 um 19:18 schrieb Harish Kasiviswanathan:
A single KFD eviction fence is attached to all the BOs of a process
including BOs imported. This fence ensures that all BOs belonging to
that process stays resident when the process queues are active.
Don't add this eviction fence to any syn
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Eric Huang
> Sent: Friday, October 20, 2017 2:10 PM
> To: amd-gfx@lists.freedesktop.org
> Subject: Re: [PATCH 0/2] Some changes to fix performance drop on Vega10
>
> Ping.
Series is:
Acked-b
-Original Message-
From: Christian König [mailto:ckoenig.leichtzumer...@gmail.com]
Sent: Friday, October 20, 2017 2:38 PM
To: Kasiviswanathan, Harish ;
amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH v2] drm/amdgpu: Sync KFD fence only for move/evict
Am 20.10.2017 um 19:18 schrieb Ha
On 2017-10-20 02:24 PM, Christian König wrote:
Am 20.10.2017 um 18:51 schrieb Andrey Grodzovsky:
On 2017-10-20 12:26 PM, Andres Rodriguez wrote:
On 2017-10-20 12:19 PM, Andrey Grodzovsky wrote:
On 2017-10-20 11:59 AM, Andres Rodriguez wrote:
On 2017-10-20 09:32 AM, Andrey Grodzovsky
On 2017-10-20 04:45 PM, Jerry (Fangzhi) Zuo wrote:
> Programming sequence to frontend and backend has been switched.
> In such case, program_scaler() is getting called when programming
> frontend, and should be removed from backend programming routine.
>
> Signed-off-by: Jerry (Fangzhi) Zuo
Revi
A single KFD eviction fence is attached to all the BOs of a process
including BOs imported. This fence ensures that all BOs belonging to
that process stays resident when the process queues are active.
Don't add this eviction fence to any sync object unless it is a move or
evict job. These jobs are
This patch series improves interrupt handling latency, signal event
processing overhead and replaces some custom data structures with
standard kernel data structures (idr, kfifo, waitqueue).
It also increases the capacity of the number of signals that can be
processed from 256 to 4096. This breaks
The kfd_process doesn't own a reference to the mm_struct, so it can
disappear without warning even while the kfd_process still exists.
In fact, the delayed kfd_process teardown is triggered by an MMU
notifier when the mm_struct is destroyed. Permanently holding a
reference to the mm_struct would pr
From: Besar Wicaksono
This enables SDMA signalling with event interrupt.
Signed-off-by: Besar Wicaksono
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c | 3 +++
drivers/gpu/drm/amd/amdkfd/cik_int.h | 3 ++-
2 files changed, 5 insertions(+), 1 del
From: Andres Rodriguez
We don't need to wait for all work to complete in the IH exit function.
We only need to make sure the interrupt_work has finished executing to
guarantee that ih_kfifo is no longer in use.
Signed-off-by: Andres Rodriguez
---
drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c | 4
From: Oded Gabbay
Signed-off-by: Oded Gabbay
Reviewed-by: Ben Goz
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 2 +-
include/uapi/linux/kfd_ioctl.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_eve
From: Andres Rodriguez
Replace our implementation of a lockless ring buffer with the standard
linux kernel kfifo.
We shouldn't maintain our own version of a standard data structure.
Signed-off-by: Andres Rodriguez
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c |
Cleaned up the code while resolving some potential bugs and
inconsistencies in the process.
Clean-ups:
* Remove enum kfd_event_wait_result, which duplicates
KFD_IOC_EVENT_RESULT definitions
* alloc_event_waiters can be called without holding p->event_mutex
* Return an error code from copy_signal
This always identical with the index of the event_waiter in the array.
No need to store it in the waiter record.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 9 +++--
1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_e
From: Sean Keely
If kfd_wait_on_events can return immediately, we don't need to populate
the wait list and don't need to enter the sleep-loop.
Signed-off-by: Sean Keely
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 43 ++---
1 file cha
When an event with pending waiters is destroyed, those waiters may
end up sleeping forever unless they are notified and woken up.
Implement the notification by clearing the waiter->event pointer,
which becomes invalid anyway, when the event is freed, and waking
up the waiting tasks.
Waiters on an
Signal slots are identical to event IDs.
Replace the used_slot_bitmap and events hash table with an IDR to
allocate and lookup event IDs and signal slots more efficiently.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 230 ++--
drivers/g
From: Sean Keely
Signed-off-by: Sean Keely
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
b/drivers/gpu/drm/amd/amdkfd/kfd_events.c
index 64cc42
The first event page is always big enough to handle all events.
Handling of multiple events pages is not supported by user mode, and
not necessary.
Signed-off-by: Yong Zhao
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 197 +++-
drivers/
From: Andres Rodriguez
In systems under heavy load the IH work may experience significant
scheduling delays.
Under load + system workqueue:
Max Latency: 7.023695 ms
Avg Latency: 0.263994 ms
Under load + high priority workqueue:
Max Latency: 1.162568 ms
Avg Latency: 0.163213 ms
Signed-off-by: Kent Russell
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_events.c | 53 +++--
drivers/gpu/drm/amd/amdkfd/kfd_events.h | 3 +-
2 files changed, 19 insertions(+), 37 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c
This speeds up signal lookup when the IH ring entry includes a
valid context ID or partial context ID. Only if the context ID is
found to be invalid, fall back to an exhaustive search of all
signaled events.
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c | 7
From: Andres Rodriguez
A larger buffer will let us accommodate applications with a large amount
of semi-simultaneous event signals.
Signed-off-by: Andres Rodriguez
Signed-off-by: Felix Kuehling
---
drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
Programming sequence to frontend and backend has been switched.
In such case, program_scaler() is getting called when programming
frontend, and should be removed from backend programming routine.
Signed-off-by: Jerry (Fangzhi) Zuo
---
drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c |
76 matches
Mail list logo