- 15 Oct, 2015 3 commits
-
-
git://people.freedesktop.org/~agd5f/linuxDave Airlie authored
This is the first radeon and amdgpu pull for drm-next. Highlights include: - Efficiency improvements to the CS checker for pre-SI asics - Cursor fixes ported from radeon to amdgpu - Enable GPU scheduler by default - Add a bunch of GPUVM debugging options - Add support for some new atombios opcodes - Misc cleanups and fixes * 'drm-next-4.4' of git://people.freedesktop.org/~agd5f/linux: (42 commits) drm/amdgpu: fix lockup when clean pending fences drm/amdgpu: add timer to fence to detect scheduler lockup drm/amdgpu: add VM CS mapping trace point drm/amdgpu: add option to clear VM page tables after every submit drm/amdgpu: add option to stop on VM fault drm/amdgpu: only print meaningful VM faults drm/amdgpu: also trace already allocated VMIDs drm/amdgpu: Drop unnecessary #include <linux/vga_switcheroo.h> drm/radeon: Drop unnecessary #include <linux/vga_switcheroo.h> drm/amdgpu: clean up pageflip interrupt handling drm/amdgpu: rework sdma structures drm/amdgpu: unpin cursor BOs on suspend and pin them again on resume drm/amdgpu/dce8: Fold set_cursor() into show_cursor() drm/amdgpu/dce8: Clean up reference counting and pinning of the cursor BOs drm/amdgpu/dce8: Move hotspot handling out of set_cursor drm/amdgpu/dce8: Re-show the cursor after a modeset (v2) drm/amdgpu/dce8: Use cursor_set2 hook for enabling / disabling the HW cursor drm/amdgpu/dce11: Fold set_cursor() into show_cursor() drm/amdgpu/dce11: Clean up reference counting and pinning of the cursor BOs drm/amdgpu/dce11: Move hotspot handling out of set_cursor ...
-
git://ftp.arm.linux.org.uk/~rmk/linux-armDave Airlie authored
* removes the now unused DRM slave encoder support, which all users have migrated away from, allowing us to simplify the code. * ensure all pending interrupts are processed together, rather than needing the handler to be re-entered each time. * use more HDMI helpers to setup the info frames. * fix EDID read handling by ensuring that we always wait the specified time before attempting to read the EDID, no matter where the EDID read request came from. * 'drm-tda998x-devel' of git://ftp.arm.linux.org.uk/~rmk/linux-arm: drm/i2c: tda998x: clean up after struct tda998x_priv2 removal drm/i2c: tda998x: kill struct tda998x_priv2 drm/i2c: tda998x: move connector into struct tda998x_priv drm/i2c: tda998x: remove encoder pointer drm/i2c: tda998x: remove DRM slave encoder support drm/i2c: tda998x: use more HDMI helpers drm/i2c: tda998x: handle all outstanding interrupts drm/i2c: tda998x: convert to u8/u16/u32 types drm/i2c: tda998x: re-implement "Fix EDID read timeout on HDMI connect" drm/i2c: tda998x: report whether we actually handled the IRQ drm/i2c: tda998x: remove useless NULL checks
-
git://ftp.arm.linux.org.uk/~rmk/linux-armDave Airlie authored
* remove support for the non-component support from the Armada DRM driver, switching it to component-only mode. * create a "armada plane" to allow the primary and overlay planes to share some code. * increase efficiency by using inherently atomic operations, rather than spinlocking to achieve atomicity. Eg, if we want to exchange a value, using xchg(). * increase PM savings by stopping the external pixel clock when we're in DPMS mode. * 'drm-armada-devel' of git://ftp.arm.linux.org.uk/~rmk/linux-arm: drm/armada: move frame wait wakeup into plane work drm/armada: convert overlay plane vbl worker to a armada plane worker drm/armada: move CRTC flip work to primary plane work drm/armada: move frame wait into armada_frame drm/armada: move the locking for armada_drm_vbl_event_remove() drm/armada: move the update of dplane->ctrl0 out of spinlock drm/armada: move write to dma_ctrl0 to armada_drm_crtc_plane_disable() drm/armada: provide a common helper to disable a plane drm/armada: allocate primary plane ourselves drm/armada: add primary plane creation drm/armada: introduce generic armada_plane struct drm/armada: update armada overlay to use drm_universal_plane_init() drm/armada: use xchg() to atomically update dplane->old_fb drm/armada: factor out retirement of old fb drm/armada: rename overlay identifiers drm/armada: redo locking and atomics for armada_drm_crtc_complete_frame_work() drm/armada: disable CRTC clock during DPMS drm/armada: use drm_plane_force_disable() to disable the overlay plane drm/armada: move vbl code into armada_crtc drm/armada: remove non-component support
-
- 14 Oct, 2015 31 commits
-
-
Junwei Zhang authored
The first lockup fence will lock the fence list of scheduler. Then cancel the delayed workqueues for all clean pending fences without waiting the workqueues to finish. Change-Id: I9bec826de1aa49d587b0662f3fb4a95333979429 Signed-off-by:
Junwei Zhang <Jerry.Zhang@amd.com> Reviewed-by:
Christian König <christian.koenig@amd.com>
-
Junwei Zhang authored
Change-Id: I67e987db0efdca28faa80b332b75571192130d33 Signed-off-by:
David Zhou <david1.zhou@amd.com> Reviewed-by:
-
Christian König authored
Output all VM mappings a command submission uses. Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Christian König authored
This makes it much easier to find when userspace misses to send some buffers. Signed-off-by:
-
Christian König authored
Signed-off-by:
-
Christian König authored
Port of radeon commit 9b7d786b. Signed-off-by:
-
Christian König authored
Signed-off-by:
-
Lukas Wunner authored
This was added to two radeon files even though they don't use any vga_switcheroo symbols, the amdgpu fork inherited them: Added to amdgpu_acpi.c by commit d7a2952f ("drm/radeon: Add support for the ATIF ACPI method to the radeon driver"). Added to amdgpu_bios.c by commit 6a9ee8af ("vga_switcheroo: initial implementation (v15)"). Signed-off-by:
Lukas Wunner <lukas@wunner.de> Signed-off-by:
-
Lukas Wunner authored
This was added to three files even though they don't use any vga_switcheroo symbols: Added to radeon_acpi.c by commit d7a2952f ("drm/radeon: Add support for the ATIF ACPI method to the radeon driver"). Added to radeon_asic.c by commit 0a10c851 ("drm/radeon: create radeon_asic.c"). Added to radeon_bios.c by commit 6a9ee8af ("vga_switcheroo: initial implementation (v15)"). Signed-off-by:
-
Alex Deucher authored
Check to make sure we aren't touching a non-existent display controller and simplify the code. Reviewed-by:
-
Alex Deucher authored
Rework the sdma structures in the driver to consolidate all of the sdma info into a single structure and allow for asics that may have different numbers of sdma instances. Reviewed-by:
-
Alex Deucher authored
Everything is evicted from VRAM before suspend, so we need to make sure all BOs are unpinned and re-pinned after resume. Fixes broken mouse cursor after resume introduced by commit b9729b17. Port of radeon commit: f3cbb17bReviewed-by:
Michel Dänzer <michel.daenzer@amd.com> Signed-off-by:
-
Alex Deucher authored
Port of radeon commit: 8991668aReviewed-by:
-
Alex Deucher authored
Take a GEM reference for and pin the new cursor BO, unpin and drop the GEM reference for the old cursor BO in dce8 crtc_cursor_set2, and use amdgpu_crtc->cursor_addr in dce8 set_cursor. This fixes dce8 cursor_reset accidentally incrementing the cursor BO pin count, and cleans up the code a little. Port of radeon commit: cd404af0Reviewed-by:
-
Alex Deucher authored
It's only needed in dce8 crtc_cursor_set2. Port of radeon commit: 2e007e61Reviewed-by:
-
Alex Deucher authored
Setting a mode seems to clear the cursor registers, so we need to re-program them to make sure the cursor is visible. Port of radeon commit: 6d3759fa v2: change radeon reference in error message Reviewed-by:
-
Alex Deucher authored
The cursor_set2 hook provides the cursor hotspot position within the cursor image. When the hotspot position changes, we can adjust the cursor position such that the hotspot doesn't move on the screen. This prevents the cursor from appearing to intermittently jump around on the screen when the position of the hotspot within the cursor image changes. Port of radeon commits: 78b1a601 3feba08dReviewed-by:
-
Alex Deucher authored
Port of radeon commit: 8991668aReviewed-by:
-
Alex Deucher authored
Take a GEM reference for and pin the new cursor BO, unpin and drop the GEM reference for the old cursor BO in dce11 crtc_cursor_set2, and use amdgpu_crtc->cursor_addr in dce11 set_cursor. This fixes dce11 cursor_reset accidentally incrementing the cursor BO pin count, and cleans up the code a little. Port of radeon commit: cd404af0Reviewed-by:
-
Alex Deucher authored
It's only needed in dce11 crtc_cursor_set2. Port of radeon commit: 2e007e61Reviewed-by:
-
Alex Deucher authored
Setting a mode seems to clear the cursor registers, so we need to re-program them to make sure the cursor is visible. Port of radeon commit: 6d3759fa v2: change radeon reference in error output Reviewed-by:
-
Alex Deucher authored
The cursor_set2 hook provides the cursor hotspot position within the cursor image. When the hotspot position changes, we can adjust the cursor position such that the hotspot doesn't move on the screen. This prevents the cursor from appearing to intermittently jump around on the screen when the position of the hotspot within the cursor image changes. Port of radeon commits: 78b1a601 3feba08dReviewed-by:
-
Alex Deucher authored
Port of radeon commit: 8991668aReviewed-by:
-
Alex Deucher authored
Take a GEM reference for and pin the new cursor BO, unpin and drop the GEM reference for the old cursor BO in dce10 crtc_cursor_set2, and use amdgpu_crtc->cursor_addr in dce10 set_cursor. This fixes dce10 cursor_reset accidentally incrementing the cursor BO pin count, and cleans up the code a little. Port of radeon commit: cd404af0Reviewed-by:
-
Alex Deucher authored
It's only needed in dce10 crtc_cursor_set2. Port of radeon commit: 2e007e61Reviewed-by:
-
Alex Deucher authored
Setting a mode seems to clear the cursor registers, so we need to re-program them to make sure the cursor is visible. Port of radeon commit: 6d3759fa v2: change radeon reference in error message Reviewed-by:
-
Alex Deucher authored
The cursor_set2 hook provides the cursor hotspot position within the cursor image. When the hotspot position changes, we can adjust the cursor position such that the hotspot doesn't move on the screen. This prevents the cursor from appearing to intermittently jump around on the screen when the position of the hotspot within the cursor image changes. Port of radeon commits: 78b1a601 3feba08dReviewed-by:
-
Alex Deucher authored
Better precision than the regular div opcode. v2: drop 64 bit divide v3: fix op handling. This actually is a 64 bit divide. Reviewed-by:
-
Alex Deucher authored
Better precision than the regular mul opcode. v2: handle big endian properly. Reviewed-by:
-
Alex Deucher authored
Just skips a data section. Reviewed-by:
-
Alex Deucher authored
Basically a nop. Reviewed-by:
-
- 08 Oct, 2015 5 commits
-
-
Alex Deucher authored
These are buggy on some asics and not really used anymore now that the GPU schedular is enabled. Change-Id: I67182b409d64de308392a15d1a0a15018071dc0b Signed-off-by: -
Chunming Zhou authored
Change-Id: Idce64f63e8422324996fc5d583d0bc9a5ac60d0c Signed-off-by:
Jammy Zhou <Jammy.Zhou@amd.com> Reviewed-by:
-
Chunming Zhou authored
need to decrease visible vram usage by default. Signed-off-by:
monk.liu <monk.liu@amd.com> Reviewed-by:
-
Chunming Zhou authored
Signed-off-by:
-
Alex Deucher authored
Calculate the driver state in sw_init and program the registers in hw init. Acked-by:
Leo Liu <leo.liu@amd.com> Reviewed-by:
-
- 02 Oct, 2015 1 commit
-
-
Grazvydas Ignotas authored
There doesn't seem to be any need to have 'ib' volatile, the code is not even consistent with it and some places already miss it. As it is now it's just making gcc produce worse code. If there are special requirements for that memory, then proper primitives like memory barriers or accessor functions should be used, but it doesn't look like that is needed here. While at it, change the type to match the one in radeon_ib structure. Reviewed-by:
Dave Airlie <airlied@redhat.com> Signed-off-by:
Grazvydas Ignotas <notasas@gmail.com> Signed-off-by:
-
