summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd
AgeCommit message (Collapse)Author
2016-05-04drm/amdgpu: merge VM manager and VM context ID structureChristian König
No need to have two of them any more. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04drm/amdgpu: Set PFLIP_SUBMITTED for crtc after address updateAndrey Grodzovsky
Also add some pflip debug prints. This change allows us to wait on pflip status until the new surface address is actually submitted to the register. This reverts ed3020e923240829dcdfd3343f6e91dc02c63775 drm/amdgpu: Move MMIO flip out of spinlocked region The original change assumed DAL will aquire locks inside DAL implemetion of page_flip callback which eventaully didn't happen. This moves the flip before status update which makes sense for the non-DAL code pathes as well. Signed-off-by: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-04drm/amd/powerplay: add deep sleep divider id into DPM table on TongaEric Huang
Add a proper implementation for setting the deep sleep divider. Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: optionally enable GART debugfs fileChristian König
Keeping the pages array around can use a lot of system memory when you want a large GART. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: remove GART page addr arrayChristian König
Not needed any more. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: use BO pages instead of GART arrayChristian König
Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: change parameter passing in the VM codeChristian König
Make it more flexible by passing src and page addresses directly instead of the structures they contain. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: drop the GTT power of two limitChristian König
As far as I can see that isn't neccessary any more. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amd/powerplay: use min_clock_in_sr for deep sleep feature.Rex Zhu
This comes from the display handling code. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amd/amdgpu: Enable clockgating in UVD6 for StoneyTom St Denis
This patch enables clockgating for the UVD6 block in Stoney. Signed-off-by: Tom St Denis <tom.stdenis@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amd/amdgpu: Enable clockgating for UVD5 on TongaTom St Denis
This patch enables clock gating for the UVD5 block with Tonga. Signed-off-by: Tom St Denis <tom.stdenis@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amd/amdgpu: Add SW clock gating support to UVD 5 and 6Tom St Denis
This patch adds support for software clock gating to UVD 5 and UVD 6 blocks with a preliminary commented out hardware gating routine. Currently hardware gating does not work so it's not activated. Signed-off-by: Tom St Denis <tom.stdenis@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: refine code for code style.Rex Zhu
White space fix. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: No need to stop hw init although vce's state was not true.Rex Zhu
This is not a fatal error. v2: add comment why ignore the error here. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: fix issue that can't set vce clock gate.Rex Zhu
Need to soft reset VCE as part of the clockgating sequence. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: use ref to keep job aliveMonk Liu
this is to fix fatal page fault error that occured if: job is signaled/released after its timeout work is already put to the global queue (in this case the cancel_delayed_work will return false), which will lead to NX-protection error page fault during job_timeout_func. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: rework TDR in scheduler (v2)Monk Liu
Add two callbacks to scheduler to maintain jobs, and invoked for job timeout calculations. Now TDR measures time gap from job is processed by hw. v2: fix typo Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: get rid of incorrect TDRMonk Liu
original time out detect routine is incorrect, cuz it measures the gap from job scheduled, but we should only measure the gap from processed by hw. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: put job to list before doneMonk Liu
the mirror_list will be used for later time out detect feature. This is needed to properly detect a GPU timeout with the scheduler. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: delay job free to when it's finished (v2)Monk Liu
for those jobs submitted through scheduler, do not free it immediately after scheduled, instead free it in global workqueue by its sched fence signaling callback function. v2: call uf's bo_undef after job_run() call job's sync free after job_run() no static inline __amdgpu_job_free() anymore, just use kfree(job) to replace it. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Reviewed-by: Chunming Zhou <david1.zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu/dce11: fix vertical bars appear on monitorVitaly Prosyak
Fixed mc stop and resume hardware programming sequence. Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: use sched_job_init to initialize sched_jobMonk Liu
Consolidate job initialization in one place rather than duplicating it in multiple places. Signed-off-by: Monk Liu <Monk.Liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: patch cond exec for SDMAMonk Liu
More ground work for conditional execution on SDMA necessary for preemption. Signed-off-by: Monk Liu <monk.liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: support cond execMonk Liu
This adds the groundwork for conditional execution on SDMA which is necessary for preemption. Signed-off-by: Monk Liu <monk.liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: improve vmid assigment V2Chunming Zhou
V2: the signaled items on the LRU maintain their order Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: mark amdgpu_allowed_register_entry tables as 'const'Nils Wallménius
Signed-off-by: Nils Wallménius <nils.wallmenius@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: do not store bios_header_start in amdgpu_deviceNils Wallménius
It is only used locally in amdgpu_get_bios Signed-off-by: Nils Wallménius <nils.wallmenius@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-05-02drm/amdgpu: delete unused struct member suspend from amdgpu_deviceNils Wallménius
Signed-off-by: Nils Wallménius <nils.wallmenius@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-22Merge tag 'v4.6-rc3' into drm-nextDave Airlie
Backmerge 4.6-rc3 for i915. Linux 4.6-rc3
2016-04-07Merge branch 'drm-fixes-4.6' of git://people.freedesktop.org/~agd5f/linux ↵Dave Airlie
into drm-fixes Lots of misc bug fixes for radeon and amdgpu and one for ttm. - fix vram info fetching on Fiji and unposted boards - additional vblank fixes from the conversion to drm_vblank_on/off - UVD dGPU suspend and resume fixes - lots of powerplay fixes - fix a fence leak in the pageflip code - ttm fix for platforms where CPU is 32 bit, but physical addresses are >32bits * 'drm-fixes-4.6' of git://people.freedesktop.org/~agd5f/linux: (21 commits) drm/amdgpu: total vram size also reduces pin size drm/amd/powerplay: add uvd/vce dpm enabling flag default. drm/amd/powerplay: fix issue that resume back, dpm can't work on FIJI. drm/amdgpu: save and restore the firwmware cache part when suspend resume drm/amdgpu: save and restore UVD context with suspend and resume drm/ttm: use phys_addr_t for ttm_bus_placement drm/radeon: Only call drm_vblank_on/off between drm_vblank_init/cleanup drm/amdgpu: fence wait old rcu slot drm/amdgpu: fix leaking fence in the pageflip code drm/amdgpu: print vram type rather than just DDR drm/amdgpu/gmc: use proper register for vram type on Fiji drm/amdgpu/gmc: move vram type fetching into sw_init drm/amdgpu: Set vblank_disable_allowed = true drm/radeon: Set vblank_disable_allowed = true drm/amd/powerplay: Need to change boot to performance state in resume. drm/amd/powerplay: add new Fiji function for not setting same ps. drm/amdgpu: check dpm state before pm system fs initialized. drm/amd/powerplay: notify amdgpu whether dpm is enabled or not. drm/amdgpu: Not support disable dpm in powerplay. drm/amdgpu: add an cgs interface to notify amdgpu the dpm state. ...
2016-04-06Merge tag 'topic/drm-misc-2016-04-01' of ↵Dave Airlie
git://anongit.freedesktop.org/drm-intel into drm-next * tag 'topic/drm-misc-2016-04-01' of git://anongit.freedesktop.org/drm-intel: drm: Add new DCS commands in the enum list drm: Make uapi headers C89 pendantic compliant drm/atomic: export drm_atomic_helper_wait_for_fences() drm: Untangle __KERNEL__ guards drm: Move DRM_MODE_OBJECT_* to uapi headers drm: align #include directives with libdrm in uapi headers drm: Make drm.h uapi header safe for C++ vgacon: dummy implementation for vgacon_text_force drm/sysfs: Nuke TV/DVI property files drm/ttm: Remove TTM_HAS_AGP drm: bridge/dw-hdmi: Remove pre_enable/post_disable dummy funcs Revert "drm: Don't pass negative delta to ktime_sub_ns()" drm/atmel: Fixup drm_connector_/unplug/unregister/_all drm: Rename drm_connector_unplug_all() to drm_connector_unregister_all() drm: bridge: Make (pre/post) enable/disable callbacks optional
2016-04-05drm/amdgpu: total vram size also reduces pin sizeChunming Zhou
Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com>
2016-04-04drm/amd/powerplay: add uvd/vce dpm enabling flag default.Rex Zhu
These should be set by default otherwise the UVD/VCE performance won't be optimal. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-04drm/amd/powerplay: fix issue that resume back, dpm can't work on FIJI.Rex Zhu
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-04drm/amdgpu: save and restore the firwmware cache part when suspend resumeLeo Liu
Signed-off-by: Leo Liu <leo.liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-04drm/amdgpu: save and restore UVD context with suspend and resumeLeo Liu
and revert fix following it accordingly Revert "drm/amdgpu: stop trying to suspend UVD sessions v2" Revert "drm/amdgpu: fix the UVD suspend sequence order" Signed-off-by: Leo Liu <leo.liu@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-04Merge branch 'PAGE_CACHE_SIZE-removal'Linus Torvalds
Merge PAGE_CACHE_SIZE removal patches from Kirill Shutemov: "PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. Let's stop pretending that pages in page cache are special. They are not. The first patch with most changes has been done with coccinelle. The second is manual fixups on top. The third patch removes macros definition" [ I was planning to apply this just before rc2, but then I spaced out, so here it is right _after_ rc2 instead. As Kirill suggested as a possibility, I could have decided to only merge the first two patches, and leave the old interfaces for compatibility, but I'd rather get it all done and any out-of-tree modules and patches can trivially do the converstion while still also working with older kernels, so there is little reason to try to maintain the redundant legacy model. - Linus ] * PAGE_CACHE_SIZE-removal: mm: drop PAGE_CACHE_* and page_cache_{get,release} definition mm, fs: remove remaining PAGE_CACHE_* and page_cache_{get,release} usage mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
2016-04-04mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macrosKirill A. Shutemov
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01drm/amdgpu: fence wait old rcu slotChunming Zhou
since the rcu slot was initialized to be num_hw_submission, if command submission doesn't use scheduler, this limitation will be invalid like uvd test. Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu: fix leaking fence in the pageflip codeChristian König
This fixes a memory leak when we can't register the callback on a fence. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu: print vram type rather than just DDRAlex Deucher
We have the info, so use it rather than reporting just DDR. Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu/gmc: use proper register for vram type on FijiAlex Deucher
The offset changed on Fiji. Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2016-04-01drm/amdgpu/gmc: move vram type fetching into sw_initAlex Deucher
early_init gets called before atom asic init so on non-posted cards, the vram type is not initialized. Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org
2016-04-01drm/amdgpu: Set vblank_disable_allowed = trueMichel Dänzer
Without this, since the conversion from drm_vblank_pre/post_modeset to drm_vblank_on/off, the vblank interrupt could never be disabled after userspace triggered enabling it. Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amd/powerplay: Need to change boot to performance state in resume.Rex Zhu
Fixes slow performance on resume. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amd/powerplay: add new Fiji function for not setting same ps.Rex Zhu
Add comparison function used by powerplay to determine which power state to select. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu: check dpm state before pm system fs initialized.Rex Zhu
Make sure powerplay initialized properly before enabling debugfs pm files. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed- by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amd/powerplay: notify amdgpu whether dpm is enabled or not.Rex Zhu
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu: Not support disable dpm in powerplay.Rex Zhu
We don't support the dpm parameter in powerplay. Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2016-04-01drm/amdgpu: add an cgs interface to notify amdgpu the dpm state.Rex Zhu
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3-70-g09d2 (git 2.48.1) at 2025-07-31 13:41:47 +0000