linux-stable.git/drivers/gpu/drm/amd, branch linux-3.19.y

drm/amdkfd: Initialize only amdkfd's assigned pipelines

2015-04-19T08:10:15+00:00

commit 1365aa6266fad0669487240af3f098593796172c upstream.

This patch fixes a bug in the initialization of the pipelines. The
init_pipelines() function was called with a constant value of 0 in the
first_pipe argument. This is an error because amdkfd doesn't handle pipe 0.

The correct way is to pass the value that get_first_pipe() returns as the
argument for first_pipe.

This bug appeared in 3.19 (first version with amdkfd) and it causes around 15%
drop in CPU performance of Kaveri (A10-7850).

v2: Don't set get_first_pipe() as inline because it calls BUG_ON()

Signed-off-by: Oded Gabbay 
Tested-by: Michel Dänzer 
Signed-off-by: Greg Kroah-Hartman

drm/amdkfd: Don't create BUG due to incorrect user parameter

2015-02-02T07:45:24+00:00

This patch changes a BUG_ON() statement to pr_debug, in case the user tries to
update a non-existing queue.

Signed-off-by: Oded Gabbay 
Reviewed-by: Ben Goz 
Reviewed-by: Jammy Zhou 
Reviewed-by: Alex Deucher

drm/amdkfd: max num of queues can't be 0

2015-02-02T07:45:24+00:00

Signed-off-by: Oded Gabbay 
Reviewed-by: Jammy Zhou 
Reviewed-by: Alex Deucher

drm/amdkfd: Fix bug in accounting of queues

2015-02-02T07:45:24+00:00

Signed-off-by: Oded Gabbay 
Reviewed-by: Jammy Zhou 
Reviewed-by: Alex Deucher

drm/amdkfd: Fix bug in call to init_pipelines()

2015-01-22T10:50:37+00:00

This patch fixes a bug where the first_pipe index passed into init_pipelines()
was a #define instead of the value that is passed into amdkfd by radeon

Signed-off-by: Oded Gabbay 
Reviewed-by: Alex Deucher 
Reviewed-by: Jammy Zhou

drm/amdkfd: Fix bug in pipelines initialization

2015-01-22T09:03:42+00:00

This patch fixes a bug when calling to init_pipeline() interface.
The index that was passed to that function didn't take into account the
first_pipe value, which represents the first pipe index that is under amdkfd's
responsibility.

Signed-off-by: Oded Gabbay 
Reviewed-by: Alex Deucher 
Reviewed-by: Jammy Zhou

drm/amdkfd: Allow user to limit only queues per device

2015-01-18T11:18:01+00:00

This patch replaces the two current amdkfd module parameters with a new one.

The current parameters that are being replaced are:

- Maximum number of HSA processes
- Maximum number of queues per process

The new parameter that replaces them is called "Maximum queues per device"

This replacement achieves two goals:

- Allows the user to have as many HSA processes as it wants (until
  a maximum of 512 HSA processes in Kaveri).

- Removes the limitation the user had on maximum number of queues per HSA
  process. E.g. the user can now have processes which only have one queue and
  other processes which have hundreds of queues, while before the user
  couldn't have more than 128 queues per process (as default).

The default value of the new parameter is 4096 (32 * 128, which were the
defaults of the old parameters). There is almost no additional GART memory
required for the default case. As a reminder, this amount of queues requires a
little bit below 4MB of GART memory.

v2:
In addition, This patch defines a new counter for queues accounting in the DQM
structure. This is done because the current counter only counts active queues
which allows the user to create more queues than the
max_num_of_queues_per_device module parameter allows.

However, we need the current counter for the runlist packet build process, so
the solution is to have a dedicated counter for this accounting.

Signed-off-by: Oded Gabbay 
Reviewed-by: Ben Goz

drm/amdkfd: PQM handle queue creation fault

2015-01-15T15:14:47+00:00

If the first queue created was failed on DQM then PQM should
unregister the process from DQM.

Signed-off-by: Ben Goz 
Signed-off-by: Oded Gabbay

drm/amdkfd: Fix sparse warning (different address space)

2015-01-08T14:46:16+00:00

Signed-off-by: Oded Gabbay

drm/amdkfd: Drop interrupt SW ring buffer

2015-01-08T04:27:15+00:00

The work queue couldn't reliably prevent the SW ring buffer from
overflowing, so dmesg was spammed by

 kfd kfd: Interrupt ring overflow, dropping interrupt.

messages when running e.g. the Atlantis Substance demo from
https://wiki.unrealengine.com/Linux_Demos on Kaveri.

Since the SW ring buffer doesn't actually do anything at this point, just
remove it for now. When actual interrupt processing code is added to
amdkfd, it should try to do things immediately and only defer to work
queues when necessary.

Signed-off-by: Michel Dänzer 
Reviewed-by: Christian König 
Signed-off-by: Oded Gabbay