aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2016-07-29drm/amdgpu: enable UVD context buffer for older HWChristian König2-2/+27
Supported starting on certain FW versions. Signed-off-by: Christian König <[email protected]> Reviewed-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: fix default UVD context sizeChristian König1-1/+1
Context buffers should be denied by default, not allowed. Signed-off-by: Christian König <[email protected]> Reviewed-by: Leo Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: fix incorrect type of info_idHuang Rui1-5/+5
Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: make amdgpu_cgs_call_acpi_method as staticHuang Rui1-1/+1
Signed-off-by: Huang Rui <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: comment out unused defaults_staturn_pro static const structure ↵Slava Grigorev1-0/+2
to fix the build Signed-off-by: Slava Grigorev <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: enable UVD VM only on polarisChristian König1-1/+1
Stoney support it, but doesn't has unlimited session support. Signed-off-by: Christian König <[email protected]> Reviewed-by: Leo Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: increase timeout of IB testChunming Zhou1-1/+1
we should give enough time to IB test. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: add destroy session when generate VCE destroy msg.Rex Zhu1-7/+8
Signed-off-by: David Mao <[email protected]> Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd: fix deadlock of job_list_lock V2Chunming Zhou1-3/+6
run_job involves mutex, which could sleep. V2: use list_for_each_entry_safe, since the job might complete while we dropped the lock. Signed-off-by: Chunming Zhou <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd: reset hw count when reset jobChunming Zhou1-0/+3
Means the hw ring is empty after gpu reset. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: free handles after fini the contextLeo Liu1-4/+3
This will make sure all the submissions from different contexts gets finished, and then we close the session and free up the handles. This will fix the issue that session clean-up is not get done properly, when with the command `kill -9' Signed-off-by: Leo Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/ttm: partial revert "cleanup ttm_tt_(unbind|destroy)" v3Christian König3-7/+22
We still need to unbind explicitly during a move. This partial reverts commit ff20caa0bcbfef9f7686f8d1868a3b990921afd6. v2: remove unnecessary check and unused variable v3: fix typo in commit message Signed-off-by: Christian König <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: add a fence timeout for the IB tests v2Christian König11-57/+93
10ms should be enough for now. v2: fix some typos in CIK code Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: move UVD IB test into common code v2Christian König5-125/+40
Since we now raise the clocks from begin_use() we don't need a separate function for each hw generation any more. v2: remove unintentional lowering of the UVD clocks, fix typos for CIK hw. Signed-off-by: Christian König <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: use begin/end_use for VCE power/clock gatingChristian König5-13/+31
This fixes turning power and clock on when it is actually needed. Signed-off-by: Christian König <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: use begin/end_use for UVD power/clock gatingChristian König5-8/+17
This fixes turning power and clock on when it is actually needed. Signed-off-by: Christian König <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: add begin/end_use ring callbacksChristian König2-0/+13
For manual UVD/VCE power and clock gating. Signed-off-by: Christian König <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: remove fence_lockChristian König2-2/+0
Was never used as far as I can see. Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: fix indentation in struct amdgpu_ringChristian König1-2/+2
Signed-off-by: Christian König <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: fix issue can't enable vce dpm.Rex Zhu3-28/+22
Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Eric Huang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: add bypass mode for vce3.0Rex Zhu1-0/+15
Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Eric Huang <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29Revert "drm/amd/powerplay: workaround issue that when uvd dpm disabled,"Rex Zhu1-4/+0
This reverts commit <2ded8c7f04825bc5cde2624f6aa83f1ff62672c0> As we enabled bypass mode for uvd on polaris10 when clockgating. so no need to set uvd clock manually. Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Christian König<[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: fix typo error when set clock gate state.Rex Zhu2-2/+2
Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Christian König<[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: correct coding stylejimqu1-2/+2
Signed-off-by: JimQu <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: increment driver minorChristian König1-1/+2
We need to check on Polaris if UVD session context is allowed or not. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: implement UVD VM mode for Stoney v2Christian König2-2/+82
Starting with Stoney we support running UVD in VM mode as well. v2: rebased, only enable on Polaris for now. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: Delete unused functions in ppevvmath.hNils Wallménius1-58/+1
Signed-off-by: Nils Wallménius <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: Mark functions of ppevvmath.h staticNils Wallménius1-64/+64
This introduces some warnings due to unused functions, that are deleted in the following commit. Signed-off-by: Nils Wallménius <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/radeon: Remove deprecated create_singlethread_workqueueBhaktipriya Shridhar1-1/+1
alloc_workqueue replaces deprecated create_singlethread_workqueue(). Each hardware CRTC has a single flip work queue. When a radeon_flip_work_func item is queued, it needs to be executed ASAP because even a slight delay may cause the flip to be delayed by one refresh cycle. Hence, a dedicated workqueue with WQ_HIGHPRI set, has been used here since a delay can cause the outcome to miss the refresh cycle. Since there are only a fixed number of work items, explicit concurrency limit is unnecessary here. Reviewed-by: Christian König <[email protected]> Acked-by: Tejun Heo <[email protected]> Signed-off-by: Bhaktipriya Shridhar <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: Delete an unnecessary variable initialisation in ↵SF Markus Elfring1-1/+1
phm_dispatch_table() The variable "result" will be set to an appropriate value a bit later. Thus omit the explicit initialisation at the beginning. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: Change assignment for a buffer variable in ↵Markus Elfring1-1/+3
phm_dispatch_table() v2 The variable "temp_storage" was eventually reassigned with a pointer. Thus omit the explicit initialisation at the beginning. v2: agd: fix coding style Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: Change assignment for a variable in amdgpu_cgs_acpi_eval_object()Markus Elfring1-1/+2
Indicate successful function execution only at the end. Thus omit initialisation for the variable "result" at the beginning. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: Delete an unnecessary variable initialisation in ↵Markus Elfring1-1/+1
amdgpu_cgs_acpi_eval_object() The variable "argument" will be set to an appropriate value a bit later. Thus omit the explicit initialisation at the beginning. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: Delete a variable in amdgpu_cgs_acpi_eval_object()Markus Elfring1-2/+0
The local variable "func_no" was assigned a value at two places. But it was not read within this function. Thus delete it. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: One function call less in amdgpu_cgs_acpi_eval_object() after ↵Markus Elfring1-9/+9
error detection The kfree() function was called in one case by the amdgpu_cgs_acpi_eval_object() function during error handling even if the passed variable "obj" contained a null pointer. * Adjust jump targets according to the Linux coding style convention. * Delete unnecessary initialisations for the variables "obj" and "params" then. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: Delete an unnecessary check before ↵Markus Elfring1-3/+1
drm_gem_object_unreference_unlocked() The drm_gem_object_unreference_unlocked() function tests whether its argument is NULL and then returns immediately. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Reviewed-by: Christian König <[email protected]> Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29GPU-DRM-Radeon: Delete an unnecessary check before ↵Markus Elfring1-3/+1
drm_gem_object_unreference_unlocked() The drm_gem_object_unreference_unlocked() function tests whether its argument is NULL and then returns immediately. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Remove redundant itermediate return valEdward O'Callaghan5-14/+7
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Use canonical boolean form in various predicatesEdward O'Callaghan5-48/+53
V.2: Fixup by hand to remove a few instances of redundant '()' left over. Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Remove redundant casts on kzalloc() callsEdward O'Callaghan3-15/+10
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Remove redundant NULL check before kfree()Edward O'Callaghan2-4/+2
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Use canonical form in branch predicatesEdward O'Callaghan6-12/+12
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Use 'true/false' for bool typed variablesEdward O'Callaghan5-25/+25
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drivers/amdgpu: Remove spurious semicolonsEdward O'Callaghan4-4/+4
Found-by: Coccinelle Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Tom St Denis <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: enable DiDt feature for polaris10/11.Rex Zhu3-5/+554
DIDT is a power saving feature which helps limit power consumption in order to hit a target power allocation. v1: delete temp file added accidentally. Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amd/powerplay: populate SMC ACPI minimum voltage using VBIOS boot SCLK ↵Rex Zhu1-3/+2
and MCLK Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu/powerplay: partial revert of endian fixesAlex Deucher1-2/+2
This fixes a warning on big endian. Bitfields need to be handled properly. Cc: Arnd Bergmann <[email protected]> Acked-by: Arnd Bergmann <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/radeon: drop confusing message about backlight controlAlex Deucher1-7/+0
The message is harmless and confusing. On PX systems, there is one ATIF method, but potentially multiple GPUs leading to an error on the GPU with no backlight control. Bug: https://bugzilla.kernel.org/show_bug.cgi?id=115011 Signed-off-by: Alex Deucher <[email protected]> Cc: Hans de Goede <[email protected]>
2016-07-29drm/radeon/ci add comment to document intentionally unreachable codeNicholas Mc Guire1-1/+4
commit d967be9b80a5 ("drm/radeon/ci: disable needless sclk changes") introduces an unreachable if(C != C) conditional code section flagged by coccinelle script bad_conditional.cocci: Add a comment to make it clear that this is intentional. Fixes: d967be9b80a5 ("drm/radeon/ci: disable needless sclk changes") Signed-off-by: Nicholas Mc Guire <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-07-29drm/amdgpu: Disable RPM helpers while reprobing connectors on resumeLyude1-0/+16
Just about all of amdgpu's connector probing functions try to acquire runtime PM refs. If we try to do this in the context of amdgpu_resume_kms by calling drm_helper_hpd_irq_event(), we end up deadlocking the system. Since we're guaranteed to be holding the spinlock for RPM in amdgpu_resume_kms, and we already know the GPU is in working order, we need to prevent the RPM helpers from trying to run during the initial connector reprobe on resume. There's a couple of solutions I've explored for fixing this, but this one by far seems to be the simplest and most reliable (plus I'm pretty sure that's what disable_depth is there for anyway). Reproduction recipe: - Get any laptop dual GPUs using PRIME - Make sure runtime PM is enabled for amdgpu - Boot the machine - If the machine managed to boot without hanging, switch out of X to another VT. This should definitely cause X to hang infinitely. Changes since v1: - add appropriate #ifdef checks for CONFIG_PM. This is not very useful, but it appears some kernel test suites test compiling amdgpu with CONFIG_PM disabled, which results in this patch breaking the builds if we don't include this #ifdef Cc: [email protected] Cc: Alex Deucher <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Lyude <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.