- 19 Mar, 2019 13 commits
-
-
YueHaibing authored
Fixes gcc '-Wunused-but-set-variable' warning: drivers/gpu/drm/ttm/ttm_execbuf_util.c: In function 'ttm_eu_fence_buffer_objects': drivers/gpu/drm/ttm/ttm_execbuf_util.c:191:24: warning: variable 'bdev' set but not used [-Wunused-but-set-variable] It's not used any more and can be removed. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Yong Zhao authored
This is recommended by HW designers. Previously when it was set to 1, the PDE walk error in VM fault will be treated as PERMISSION_OR_INVALID_PAGE_FAULT rather than usually expected OTHER_FAULT. As a result, the retry control in VM_CONTEXT*_CNTL will change accordingly. The above behavior is kind of abnormal. Furthermore, the PDE_FAULT_CLASSIFICATION == 1 feature was targeted for very old ASICs and it never made it way to production. Therefore, we should set it to 0. Signed-off-by:
Yong Zhao <Yong.Zhao@amd.com> Acked-by:
Christian König <christian.koenig@amd.com> Acked-by:
-
Yong Zhao authored
All the gmc_*_set_pde_pte functions are the same across different ASICs, so we can eliminate the set_pde_pte function pointer and instead use a generic function. Signed-off-by:
-
Felix Kuehling authored
Don't account for them in other zones such as dma32. The kernel page allocator has its own heuristics to avoid exhausting special zones for regular kernel allocations. Signed-off-by:
Felix Kuehling <Felix.Kuehling@amd.com> Acked-by:
Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by:
-
Thomas Zimmermann authored
GEM defines DRM_FILE_PAGE_OFFSET_{START,SIZE} constants for the mmap-able range of addresses. TTM can use them as well. Signed-off-by:Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
Thomas Zimmermann authored
A BO's address has to be at least the minimum offset. Sharing this test in ttm_bo_mmap() removes code from drivers. A full buffer-address validation is still done within drm_vma_offset_lockup_locked(). Signed-off-by:
-
Thomas Zimmermann authored
The parameter file_page_offset is a constant shared by all drivers. Just replace it with the constant itself. Signed-off-by:
-
Thomas Zimmermann authored
Most TTM drivers define the constant DRM_FILE_PAGE_OFFSET of the same value. The only exception is vboxvideo, which is being converted to the new offset by this patch. Unifying the constants in a single place simplifies the driver code. Signed-off-by:
-
Thomas Zimmermann authored
The vboxvideo driver mmaps BOs at 0x10000000 or higher. Changing the offset to 0x100000000 aligns the driver with all other DRM drivers. Signed-off-by:
-
Philip Yang authored
Only select HMM_MIRROR will get kernel config dependency warnings if CONFIG_HMM is missing in the config. Add depends on HMM will solve the issue. Add conditional compilation to fix compilation errors if HMM_MIRROR is not enabled as HMM config is not enabled. Remove unused function amdgpu_ttm_tt_mark_user_pages. Signed-off-by:
Philip Yang <Philip.Yang@amd.com> Reviewed-by:
-
Philip Yang authored
Use HMM helper function hmm_vma_fault() to get physical pages backing userptr and start CPU page table update track of those pages. Then use hmm_vma_range_done() to check if those pages are updated before amdgpu_cs_submit for gfx or before user queues are resumed for kfd. If userptr pages are updated, for gfx, amdgpu_cs_ioctl will restart from scratch, for kfd, restore worker is rescheduled to retry. HMM simplify the CPU page table concurrent update check, so remove guptasklock, mmu_invalidations, last_set_pages fields from amdgpu_ttm_tt struct. HMM does not pin the page (increase page ref count), so remove related operations like release_pages(), put_page(), mark_page_dirty(). Signed-off-by:
-
Philip Yang authored
There is circular lock between gfx and kfd path with HMM change: lock(dqm) -> bo::reserve -> amdgpu_mn_lock To avoid this, move init/unint_mqd() out of lock(dqm), to remove nested locking between mmap_sem and bo::reserve. The locking order is: bo::reserve -> amdgpu_mn_lock(p->mn) Signed-off-by:
-
Philip Yang authored
Replace our MMU notifier with hmm_mirror_ops.sync_cpu_device_pagetables callback. Enable CONFIG_HMM and CONFIG_HMM_MIRROR as a dependency in DRM_AMDGPU_USERPTR Kconfig. It supports both KFD userptr and gfx userptr paths. Signed-off-by:
-
- 05 Mar, 2019 22 commits
-
-
Alex Deucher authored
Use BACO for reset of the platform supports it. Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by:
-
Alex Deucher authored
Several of the BACO functions are common across smu9-based asics. Split the common code out. Reviewed-by:
-
Alex Deucher authored
This implements BACO (Bus Active, Chip Off) support for vega12. Reviewed-by:
-
Mark McGarrity authored
Signed-off-by:
Mark McGarrity <mark.mcgarrity@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
Leo Li <sunpeng.li@amd.com> Signed-off-by:
-
Jun Lei authored
Add a dc_bounding_box_overrides struct to define bb overrides. It is loaded in during DC init. Signed-off-by:
Jun Lei <Jun.Lei@amd.com> Reviewed-by:
-
David Francis authored
[Why] dc_commit_updates_for_stream is called twice per stream: once with the flip data and once will all other data. This causes problems when these DC calls have different numbers of planes For example, a commit with a pageflip on plane A and a non-pageflip change on plane B will first call into DC with just plane A, causing plane B to be disabled. Then it will call into DC with both planes, re-enabling plane B [How] Merge flip and full into a single bundle Apart from the single DC call, the logic should not be changed by this patch Signed-off-by:
David Francis <David.Francis@amd.com> Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
-
Josip Pavic authored
[Why] eDP v1.4 allows panels to report link rates other than RBR/HBR/HBR2, that may be more optimal for the panel's timing. Power can be saved by using a link rate closer to the required bandwidth of the panel's timing. [How] Scan the table of reported link rates from the panel, and select the minimum link rate that satisfies the bandwidth requirements of the panel's timing. Include a flag to make the feature optional. Signed-off-by:
Josip Pavic <Josip.Pavic@amd.com> Reviewed-by:
Harry Wentland <Harry.Wentland@amd.com> Acked-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Jun Lei authored
Will be used to signify if P-state change is supported. Signed-off-by:
Eric Yang <eric.yang2@amd.com> Acked-by:
-
David Francis authored
[Why] The wait_for_vblank boolean in commit_tail was passed by reference into each stream commit, and if that commit was an asynchronous flip, it would disable vblank waits on all subsequent flips. This made the behaviour depend on crtc order in a non-intuitive way, although since the asynchronous pageflip flag is only used by the legacy IOCTLs at the moment it is never an issue [How] Find wait_for_vblank before doing any stream commits Signed-off-by:
-
Wesley Chalmers authored
[WHY] When doing split pipe, if one pipe is pending on flip, the entire plane's status should be flip pending, otherwise corruption can occur when OS writes to a surface prematurely. [HOW] Clear the flip pending bit before checking pipes, then OR the flip pending bits from all pipes together to create the flip pending status of the entire plane. Signed-off-by:
Wesley Chalmers <Wesley.Chalmers@amd.com> Reviewed-by:
Eryk Brol <Eryk.Brol@amd.com> Acked-by:
-
Anthony Koo authored
[Why] AUX arbitration occurs between SW and FW components. When AUX acquire fails, it causes engine->ddc to be NULL, which leads to an exception when we try to release the AUX engine. [How] When AUX engine acquire fails, it should return from the function without trying to continue the operation. The upper level will determine if it wants to retry. i.e. dce_aux_transfer_with_retries will be used and retry. Signed-off-by:
Aric Cyr <Aric.Cyr@amd.com> Acked-by:
-
Nicholas Kazlauskas authored
[Why] Seamless boot tries to reuse planes that were enabled for the first commit applied. In the case where Raven is booting with two monitors connected and the first commit contains two streams the screen corruption would occur because the second stream was trying to re-use a tg and plane that weren't previously enabled. The state on the first commit looks something like the following: TG0: enabled=1 TG1: enabled=0 TG2: enabled=0 TG3: enabled=0 New state: pipe=0, stream=0, plane=0, new_tg=0 New state: pipe=1, stream=1, plane=1, new_tg=1 New state: pipe=2, stream=NULL, plane=NULL, new_tg=NULL New state: pipe=3, stream=NULL, plane=NULL, new_tg=NULL Only one plane/tg is setup before we enter accelerated mode so we really want to disabling everything but that first plane. [How] Check if the stream is not NULL and if the tg is enabled before deciding whether to skip the plane disable. Also ensure we're also disabling on the current state's pipe_ctx so we don't overwrite the fields in the new pending state. Signed-off-by:
-
Dmytro Laktyushkin authored
This change changes dc add plane logic to allow plane-less resource reservation (pipe split). If a free pipe_ctx (no plane_state attached) is the head pipe, and is found with a bottom pipe attached, assign the plane to add on the bottom pipe. In addition, prepend dcn10 to dcn10-specific reset_back_end_for_pipe and reset_hw_ctx_wrap Signed-off-by:
Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com> Reviewed-by:
Charlene Liu <Charlene.Liu@amd.com> Acked-by:
-
Charlene Liu authored
Driver need to poll the SURFACE_INUSE register to determine when to start the new task and write data to the checked surface. Implement the wait functions, and add the necessary hubbub registers. Signed-off-by:
-
Eryk Brol authored
[Why] For later use by the DC VM implementation Signed-off-by:
-
Anthony Koo authored
[Why] Seamless boot code not working on all ASICs because of some underflow issues caused by some uninitialized HW state. [How] Keep some logical and power gating init code in hw_init. Move some per pipe init code to enable accelerated mode Signed-off-by:
-
Fatemeh Darbehani authored
[Why] DM has impelemented new pp_smu interface. 'Else' is not longer needed. Signed-off-by:
Fatemeh Darbehani <fatemeh.darbehani@amd.com> Reviewed-by:
Yongqiang Sun <yongqiang.sun@amd.com> Signed-off-by:
-
Eric Bernstein authored
This enum definition is shared, so move it to a shared location. Signed-off-by:
Eric Bernstein <eric.bernstein@amd.com> Reviewed-by:
-
mmcgarri authored
Signed-off-by:
-
Jun Lei authored
[Why] Make dml and integration with pplib clearer. [How] Change the way the dml formula is initialized to make its values more clear. Restructure DC interface with pplib into rv_funcs. Cap clocks received from pplib. Signed-off-by:
-
Charlene Liu authored
Added a "disable_tri_buf" DC debug option. When set to 1 feature will be off. Signed-off-by:
-
Mario Kleiner authored
In VRR mode, keep track of the vblank count of the last completed pageflip in amdgpu_crtc->last_flip_vblank, as recorded in the pageflip completion handler after each completed flip. Use that count to prevent mmio programming a new pageflip within the same vblank in which the last pageflip completed, iow. to throttle pageflips to at most one flip per video frame, while at the same time allowing to request a flip not only before start of vblank, but also anywhere within vblank. The old logic did the same, and made sense for regular fixed refresh rate flipping, but in vrr mode it prevents requesting a flip anywhere inside the possibly huge vblank, thereby reducing framerate in vrr mode instead of improving it, by delaying a slightly delayed flip requests up to a maximum vblank duration + 1 scanout duration. This would limit VRR usefulness to only help applications with a very high GPU demand, which can submit the flip request before start of vblank, but then have to wait long for fences to complete. With this method a flip can be both requested and - after fences have completed - executed, ie. it doesn't matter if the request (amdgpu_dm_do_flip()) gets delayed until deep into the extended vblank due to cpu execution delays. This also allows clients which want to regulate framerate within the vrr range a much more fine-grained control of flip timing, a feature that might be useful for video playback, and is very useful for neuroscience/vision research applications. In regular non-VRR mode, retain the old flip submission behavior. This to keep flip scheduling for fullscreen X11/GLX OpenGL clients intact, if they use the GLX_OML_sync_control extensions glXSwapBufferMscOML(, ..., target_msc,...) function with a specific target_msc target vblank count. glXSwapBuffersMscOML() or DRI3/Present PresentPixmap() will not flip at the proper target_msc for a non-zero target_msc if VRR mode is active with this patch. They'd often flip one frame too early. However, this limitation should not matter much in VRR mode, as scheduling based on vblank counts is pretty futile/unusable under variable refresh duration anyway, so no real extra harm is done. According to some testing already done with this patch by Nicholas on top of my tests, IGT tests didn't report any problems. If fixes stuttering and flickering when flipping at rates below the minimum vrr refresh rate. Fixes: bb47de73 ("drm/amdgpu: Set FreeSync state using drm VRR properties") Signed-off-by:
Mario Kleiner <mario.kleiner.de@gmail.com> Cc: <stable@vger.kernel.org> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Michel Dänzer <michel@daenzer.net> Tested-by:
Bruno Filipe <bmilreu@gmail.com> Reviewed-by:
Dave Airlie <airlied@redhat.com>
-
- 28 Feb, 2019 5 commits
-
-
Christian König authored
Clear the VM PDs/PTs only after initializing all the structures. Signed-off-by:
-
Nathan Chancellor authored
Clang warns when an expression that equals zero is used as a null pointer constant (in lieu of NULL): drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:4435:3: warning: expression which evaluates to zero treated as a null pointer constant of type 'const enum color_transfer_func *' [-Wnon-literal-null-conversion] TRANSFER_FUNC_UNKNOWN, ^~~~~~~~~~~~~~~~~~~~~ 1 warning generated. This warning is caused by commit bb47de73 ("drm/amdgpu: Set FreeSync state using drm VRR properties") and it could be solved by using NULL instead of TRANSFER_FUNC_UNKNOWN or casting TRANSFER_FUNC_UNKNOWN as a pointer. However, after looking into it, there doesn't appear to be a good reason to pass app_tf by reference as it is never mutated along the way. This is the only code path in which app_tf is used: mod_freesync_build_vrr_infopacket -> build_vrr_infopacket_v2 -> build_vrr_infopacket_fs2_data Neither mod_freesync_build_vrr_infopacket or build_vrr_infopacket_v2 modify app_tf's value and build_vrr_infopacket_fs2_data expects just the value so we can avoid dereferencing anything by just passing in app_tf's value to mod_freesync_build_vrr_infopacket and build_vrr_infopacket_v2. There is no functional change because build_vrr_infopacket_fs2_data doesn't do anything if TRANSFER_FUNC_UNKNOWN is passed to it, the same as not calling build_vrr_infopacket_fs2_data at all like before this change when NULL was used for app_tf. Reviewed-by:Nathan Chancellor <natechancellor@gmail.com> Signed-off-by:
-
Candice Li authored
This reverts commit 2172b89e. Signed-off-by:
Candice Li <candice.li@amd.com> Reviewed-by:
-
Evan Quan authored
Instead of the hard-coded ones from VBIOS. Signed-off-by:
Evan Quan <evan.quan@amd.com> Acked-by:
-
Evan Quan authored
Set the soft/hard max settings as max possible to not violate the OD settings. Signed-off-by:
-
