Am 09.09.2016 um 08:30 schrieb Flora Cui:
Change-Id: Ia2cb8a8f62d0933b7937991bec4eace560745f97
Signed-off-by: Flora Cui
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
b/drivers/gp
yes. please don't do this, I need them.
On Fri, Sep 09, 2016 at 02:41:16PM +0800, zhoucm1 wrote:
>
>
> On 2016年09月08日 21:41, Christian König wrote:
> >From: Christian König
> >
> >Either never used or not used in quite a while.
> No, I remember Flora's Direct GMA is using them like GDS use PRIV
NAK, that is clearly incorrect. The gpu_offset are MC addresses while
the vram_base_offset is a bud address.
Most of the time they are both zero, except for APUs which would break
with this change.
Regards,
Christian.
Am 09.09.2016 um 08:30 schrieb Flora Cui:
Change-Id: I8112e9d85866104559e
what about commit e6eb9e1cdb740f778fbbacc5f62edd0ce97c2d52?
should the start addr get updated either?
On Fri, Sep 09, 2016 at 09:18:47AM +0200, Christian König wrote:
> NAK, that is clearly incorrect. The gpu_offset are MC addresses while the
> vram_base_offset is a bud address.
>
> Most of the t
No that patch is correct. See we are dealing with different address
spaces here.
The TTM code works with MC addresses. E.g. the internal address space as
seen by everything which runs on system VM (copy operations, ring
buffers, UVD, VCE etc...).
The MC in turn then works with bus addresses
In this case please just add them back in your tree. That should be a
two liner.
For upstream it certainly doesn't make sense to keep them.
Regards,
Christian.
Am 09.09.2016 um 09:01 schrieb Flora Cui:
yes. please don't do this, I need them.
On Fri, Sep 09, 2016 at 02:41:16PM +0800, zhoucm1
Use error label to decrease global item unref and make codes more
readable.
Signed-off-by: Huang Rui
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 22 +-
1 file changed, 13 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/gpu/drm/
Am 09.09.2016 um 11:50 schrieb Huang Rui:
Use error label to decrease global item unref and make codes more
readable.
Signed-off-by: Huang Rui
Nice cleanup, please keep going with that. Probably more than once case
where we have things like this.
Patch is Reviewed-by: Christian König .
Re
On 09/09/2016 07:58 PM, Christian König wrote:
> Am 09.09.2016 um 11:50 schrieb Huang Rui:
>> Use error label to decrease global item unref and make codes more
>> readable.
>>
>> Signed-off-by: Huang Rui
>
> Nice cleanup, please keep going with that. Probably more than once case
> where we have
Hi Chris,
Removing the flag will make ttm_mem_type_from_place skip counting the
corresponding placement and thus have impact on mem region create and bo
movement. There is no guarantee that amdgpu would never introduce new memory
domain in future. In such case, I'd like to vote for keeping thes
Hi Hawking,
Removing the flag will make ttm_mem_type_from_place skip counting the
corresponding placement and thus have impact on mem region create and bo
movement.
And that is exactly the reason why I want to remove the unused flags.
There is no guarantee that amdgpu would never introduce
From: Christian König
We need to validate the offset to make sure that we don't write after the BO.
Additional to that a page should be enough and can make address space
handling much easier.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 6 ++
1 file changed,
Hi Chris,
Removing the flag will make ttm_mem_type_from_place skip counting the
corresponding placement and thus have impact on mem region create and bo
movement. There is no guarantee that amdgpu would never introduce new memory
domain in future. In such case, I'd like to vote for keeping thes
> -Original Message-
> From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
> Of Christian König
> Sent: Friday, September 09, 2016 7:24 AM
> To: Zhang, Hawking; Koenig, Christian; Cui, Flora; Zhou, David(ChunMing);
> amd-gfx@lists.freedesktop.org
> Cc: dri-de...@lists.fre
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/powerplay/hwmgr/iceland_hwmgr.c | 6 +++---
drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c | 10 +-
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/iceland_hwmgr.c
b/drivers/gpu/drm/am
Change-Id: I8d3a8cbcac10d65d0d7b724dfb6eb6fbaedcfda9
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 14 ++
1 file changed, 14 insertions(+)
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
b/drivers/gpu/drm/amd/powerplay/inc/h
Rex Zhu (9):
drm/amd/powerplay: mark symbols static where possible on tonga.
drm/amd/powerplay: add feature flags in hwmgr to enable/disable
special features.
drm/amd/powerplay: add module parameter for mask pp feature
drm/amd/powerplay: initialize platform caps in hwmgr_init.
drm/am
Reviewed-by: Alex Deucher
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 17 +
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 3 +++
2 files changed, 20 insertions(+)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
b/drivers/gpu/drm/amd/powerp
Change-Id: I9a03c3aac811259e32edbf7c20ef3eeca69d0252
Reviewed-by: Alex Deucher
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 7 +--
drivers/gpu/drm/amd/powerplay/hwmgr/iceland_hwmgr.c | 8
drivers/gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr
Change-Id: I832464f3de9cbacd2e0f6b6a22d2d7d9980be418
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c| 2 +-
drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c | 2 +-
drivers/gpu/drm/amd/powerplay/hwmgr/iceland_hwmgr.c | 2 +-
driv
Change-Id: I2ea5b1dc70883e8656fe0b1fd7a2b9a9c4ddcb27
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/inc/smumgr.h| 69 ++
drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c | 101 +-
2 files changed, 169 insertions(+), 1
implement smu7_hwmgr smu7_smumgr to manager asics with smu version 7.
so can delete duplicated code.
There are subtle differences in firmware image between different asics
implement asic_name_smc.c under smumgr to deal with fw related jobs.
Rex Zhu (20):
drm/amd/powerplay: add common interfac
Change-Id: Ia3dae5b7ef0889acd9479f83d5f3b8a91786f4bf
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/Makefile | 2 --
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 35 ++--
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h| 1 +
3 f
Change-Id: I22767dc03fcec497e5cd76e82459916abf130105
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/smumgr/Makefile |2 +-
.../gpu/drm/amd/powerplay/smumgr/polaris10_smc.c | 2284
.../gpu/drm/amd/powerplay/smumgr/polaris10_smc.h
Reviewed-by: Alex Deucher
Signed-off-by: Rex Zhu
---
drivers/gpu/drm/amd/powerplay/hwmgr/Makefile | 3 +--
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 25 +++--
2 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefile
b
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/Makefile | 5 ++---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 30 +---
2 files changed, 29 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefi
Change-Id: Id0017fcdd179cc726349eacc2ee56be96279b526
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/inc/smumgr.h | 1 +
drivers/gpu/drm/amd/powerplay/smumgr/Makefile | 2 +-
drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c | 571
Change-Id: Iee3c7ab9c1318ec7206bf5a5ae59d8c48b8e823b
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
Conflicts:
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
---
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/gpu/drm/amd/powerp
Change-Id: I8144e52141f7e6a49a18be3523540a6edd898ec0
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
.../gpu/drm/amd/powerplay/hwmgr/iceland_hwmgr.c| 50 +-
.../drm/amd/powerplay/hwmgr/iceland_powertune.c| 6 +-
.../gpu/drm/amd/powerplay/hwmgr/iceland_thermal.c | 2 +-
.../gpu
Change-Id: Ibad281cd7fd660abe612894df82be284c9b865a9
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
b/drivers/gpu/drm/amd/powerplay
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
.../gpu/drm/amd/powerplay/inc/polaris10_pwrvirus.h | 3 +-
.../gpu/drm/amd/powerplay/smumgr/polaris10_smc.c | 81 +--
.../drm/amd/powerplay/smumgr/polaris10_smumgr.c| 681 ++---
.../drm/amd/powerplay/smumgr/polaris10_s
Change-Id: Ie63ee99aa515e37943aee726103260819c9e2913
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/smumgr/fiji_smc.c| 69 +--
drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.c | 605 +
drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr
Change-Id: I5cbd25597b02a01e7de28347659139a5e0a30098
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c| 2 +-
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 704 +++--
drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.h
Change-Id: Ic5b83029c4e326ef4e02ee341ba8b1109d17328b
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
.../drm/amd/powerplay/hwmgr/iceland_powertune.h| 14 -
drivers/gpu/drm/amd/powerplay/smumgr/Makefile |5 +-
drivers/gpu/drm/amd/powerplay/smumgr/iceland_smc.c | 2577 +
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/Makefile | 4 +---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 36 +---
2 files changed, 28 insertions(+), 12 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/Makefi
Change-Id: I0fad3f4f96a57b41a5c47293e412fca40c741738
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 4
2 files changed, 5 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
b/drivers
Change-Id: Ia64aee14b196bef708aded00cc257c7eafcfbcb9
Reviewed-by: Alex Deucher
Signed-off-by: Rex Zhu
---
.../gpu/drm/amd/powerplay/hwmgr/polaris10_hwmgr.h| 5 +
.../drm/amd/powerplay/hwmgr/polaris10_powertune.h| 13 -
.../gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.h
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c | 57 +
drivers/gpu/drm/amd/powerplay/inc/hwmgr.h | 8 +++-
2 files changed, 63 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c
Reviewed-by: Alex Deucher
Signed-off-by: Rex Zhu
---
.../gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c | 34 ---
drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c| 109 -
.../gpu/drm/amd/powerplay/inc/hardwaremanager.h| 1 -
drivers/gpu/drm/amd/powerplay/inc/hwm
Change-Id: I1bab73135461c289264b14972fbf15eb424734ee
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/smumgr/Makefile |2 +-
drivers/gpu/drm/amd/powerplay/smumgr/fiji_smc.c| 2371
drivers/gpu/drm/amd/powerplay/smumgr/fiji_smc.h
Change-Id: If9ca14557f99f3e13e17086e8c0a90c53f232631
Signed-off-by: Rex Zhu
Reviewed-by: Alex Deucher
---
drivers/gpu/drm/amd/powerplay/smumgr/tonga_smc.c | 74 +--
.../gpu/drm/amd/powerplay/smumgr/tonga_smumgr.c| 657 +
.../gpu/drm/amd/powerplay/smumgr/tonga_smumgr.h
Am 09.09.2016 um 15:41 schrieb Deucher, Alexander:
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf
Of Christian König
Sent: Friday, September 09, 2016 7:24 AM
To: Zhang, Hawking; Koenig, Christian; Cui, Flora; Zhou, David(ChunMing);
amd-gfx@lists.
Patch #1-#7 and #9 are Acked-by: Christian König .
Patch #8: That looks fishy, usually we don't want to override static
functions tables.
So what is the specific use case here?
Regards,
Christian.
Am 09.09.2016 um 15:37 schrieb Rex Zhu:
Rex Zhu (9):
drm/amd/powerplay: mark symbols static
Am 09.09.2016 um 15:54 schrieb Emil Velikov:
On 9 September 2016 at 12:24, Christian König wrote:
Hi Hawking,
Removing the flag will make ttm_mem_type_from_place skip counting the
corresponding placement and thus have impact on mem region create and bo
movement.
And that is exactly the reaso
From: Christian König
An invalid offset to be used by custom VRAM/GGT manager when we
don't have an offset for an BO.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 2 ++
2 files changed, 3 insertions(+)
diff -
From: Christian König
Only allocate address space when we really need it.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/Makefile | 3 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 4 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c | 238 +++
From: Christian König
Doesn't make much sense to clear the GTT when it was never filled.
Signed-off-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/gpu/drm/amd/amdgpu/amdgp
Hi Christian,
I used smu7_hwmgr for all the asics of smu version 7 , CI and Vi. Please see
powerplay code refactoring patches.
But some functions in the table were different between those asics. so we
needed to overload those functions for some asics.
for example:
from tonga. we used p
Hi Rex,
exactly as I feared, please don't do this.
I for example have a system with multiple PEG slots to be able to test
different hardware generations at the same time.
This is rather common setup for testing and will break horrible and
quite unpredictable when the code suddenly overrides
Some comments inline.
Would it make sense to free the GTT address space in amdgpu_ttm_unbind?
Basically the counterpart of amdgpu_gtt_mgr_alloc, call it
amdgpu_gtt_mgr_free, that frees the address space but keeps the node
allocated and sets its node->start address back to AMDGPU_BO_INVALID_OFFSET?
Hi Christian,
you are right, I missed mult-gpu case. i will refine related patches.
Thanks.
Best Regards
Rex
From: Christian König
Sent: Saturday, September 10, 2016 1:51:30 AM
To: Zhu, Rex; amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 0/9] misc patch
The following series applies on top of either master or the drm-next-4.9
branch. Apart from the couple of misc fixes here we port out the
'get_local_mem_info()' helper out of the ROCm kernel patch mountain
and hook that up so to be mainlined.
Please review,
Edward O'Callaghan (7):
[PATCH 1/7]
Dereference the one time and unpack the lower and upper 32bit
portions with the proper kernel helper macros.
Signed-off-by: Edward O'Callaghan
Reviewed-by: Oded Gabbay
---
drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drive
Signed-off-by: Edward O'Callaghan
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c| 32 ---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h| 3 ++-
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c | 2 +-
d
pqm_uninit() will be called in kfd_process_notifier_release(), which
is when the process exits. Calling it in kfd_unbind_process_from_device()
is duplicate and in fact incorrect, because pqm should not be
uninitalized as long as the process still exists.
Original author: Yong Zhao
Hand ported for
Ensure we return a NULL on the fail branch so that the call
site may BUG_ON() it.
Signed-off-by: Edward O'Callaghan
---
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
b/drivers/gpu/drm/am
When the loop predicating timeout parameter passed happens to
not be a multiple of 20 the unsigned integer will overflow and
the loop will become unbounded.
Signed-off-by: Edward O'Callaghan
---
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 17 +
drivers/gpu/drm/amd/amdgpu/
The kfd_lookup_process_by_pasid() is just for that purpose,
so use it instead of repeating the code.
Signed-off-by: Edward O'Callaghan
---
drivers/gpu/drm/amd/amdkfd/kfd_process.c | 58 +---
1 file changed, 23 insertions(+), 35 deletions(-)
diff --git a/drivers/gpu/d
Signed-off-by: Edward O'Callaghan
Reviewed-by: Oded Gabbay
---
drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index 9beae87..291c69d 100644
--- a/d
59 matches
Mail list logo