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: Alex Deucher <alexander.deucher@amd.com>

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: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Check to make sure we aren't touching a non-existent
display controller and simplify the code.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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 <alexander.deucher@amd.com>

Port of radeon commit:
8991668aReviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

It's only needed in dce8 crtc_cursor_set2.

Port of radeon commit:
2e007e61Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Port of radeon commit:
8991668aReviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

It's only needed in dce11 crtc_cursor_set2.

Port of radeon commit:
2e007e61Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Port of radeon commit:
8991668aReviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

It's only needed in dce10 crtc_cursor_set2.

Port of radeon commit:
2e007e61Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Better precision than the regular mul opcode.

v2: handle big endian properly.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Just skips a data section.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Basically a nop.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

These are buggy on some asics and not really used anymore
now that the GPU schedular is enabled.

Change-Id: I67182b409d64de308392a15d1a0a15018071dc0b
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Change-Id: Idce64f63e8422324996fc5d583d0bc9a5ac60d0c
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

need to decrease visible vram usage by default.
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

Calculate the driver state in sw_init and program the
registers in hw init.
Acked-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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: Alex Deucher <alexander.deucher@amd.com>

After this patch the register check loop does the same thing as before,
except that now gcc does better job optimizing it: it now sees that
end_reg was already checked against PACKET3_SET_CONTEXT_REG_END and can
optimize REG_SAFE_BM_SIZE comparison out of evergreen_is_safe_reg()
as (PACKET3_SET_CONTEXT_REG_END >> 7) < REG_SAFE_BM_SIZE.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

evergreen_cs_check_reg() is a large function and gcc doesn't want to
inline it. It has a quick check for reg_safe_bm[] to see if register
needs special handling, which often results in early exit. However
because the function is large, it has a long prologue/epilogue to
save/restore all the callee-save registers which according to perf is
taking significant amount of time. To avoid this, we can reuse
evergreen_is_safe_reg() to do the early check directly in register loop.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

To avoid having to distinguish between CAYMAN or older on every register
check, place a pointer in evergreen_cs_track and use it unconditionally.
Also make use of the fact that both reg_safe_bm[] arrays are of the same
length to remove another CAYMAN check.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

- initialize backlight from VBT as fallback (Jani)
- hpd A support from Ville
- various atomic polish all over (mostly from Maarten)
- first parts of virtualize gpu guest support on bdw from
  Zhiyuan Lv
- GuC fixes from Alex
- polish for the chv clocks code (Ville)
- various things all over, as usual

* tag 'drm-intel-next-2015-09-11' of git://anongit.freedesktop.org/drm-intel: (145 commits)
  drm/i915: Update DRIVER_DATE to 20150911
  drm/i915: Remove one very outdated comment
  drm/i915: Use crtc->state for duplication.
  drm/i915: Do not handle a null plane state.
  drm/i915: Remove legacy plane updates for cursor and sprite planes.
  drm/i915: Use atomic state when changing cursor visibility.
  drm/i915: Use the atomic state in intel_update_primary_planes.
  drm/i915: Use the plane state in intel_crtc_info.
  drm/i915: Use atomic plane state in the primary plane update.
  drm/i915: add attached connector to hdmi container
  drm/i915: don't hard code vlv backlight frequency if unset
  drm/i915: initialize backlight max from VBT
  drm/i915: use pch backlight override on hsw too
  drm/i915/bxt: Clean up bxt_init_clock_gating
  drm/i915: Fix cmdparser STORE/LOAD command descriptors
  drm/i915: Dump pfit state as hex
  drm/i915: access the PP_ON_DELAYS/PP_OFF_DELAYS regs only pre GEN5
  drm/i915: access the PP_CONTROL reg only pre GEN5
  drm/i915: Refactor common ringbuffer allocation code
  drm/i915: use the yesno helper for logging
  ...

Another attempt at drm-misc for 4.4 ...
- better atomic helpers for runtime pm drivers
- atomic fbdev
- dp aux i2c STATUS_UPDATE handling (for short i2c replies from the sink)
- bunch of constify patches
- inital kerneldoc for vga switcheroo
- some vblank code cleanups from Ville and Thierry
- various polish all over

* tag 'topic/drm-misc-2015-09-25' of git://anongit.freedesktop.org/drm-intel: (57 commits)
  drm/irq: Add drm_crtc_vblank_count_and_time()
  drm/irq: Rename drm_crtc -> crtc
  drm: drm_atomic_crtc_get_property should be static
  drm/gma500: Remove DP_LINK_STATUS_SIZE redefinition
  vga_switcheroo: Set active attribute to false for audio clients
  drm/core: Preserve the fb id on close.
  drm/core: Preserve the framebuffer after removing it.
  drm: Use vblank timestamps to guesstimate how many vblanks were missed
  drm: store_vblank() is never called with NULL timestamp
  drm: Clean up drm_calc_vbltimestamp_from_scanoutpos() vbl_status
  drm: Limit the number of .get_vblank_counter() retries
  drm: Pass flags to drm_update_vblank_count()
  drm/i915: Fix vblank count variable types
  drm: Kill pixeldur_ns
  drm: Stop using linedur_ns and pixeldur_ns for vblank timestamps
  drm: Move timestamping constants into drm_vblank_crtc
  drm/fbdev: Update legacy plane->fb refcounting for atomic restore
  drm: fix kernel-doc warnings in drm_crtc.h
  vga_switcheroo: Sort headers alphabetically
  drm: Spell vga_switcheroo consistently
  ...

Pull x86 fixes from Thomas Gleixner:
 "Two bugfixes from Andy addressing at least some of the subtle NMI
  related wreckage which has been reported by Sasha Levin"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/nmi/64: Fix a paravirt stack-clobbering bug in the NMI code
  x86/paravirt: Replace the paravirt nop with a bona fide empty function

Pull irq fix from Thomass Gleixner:
 "A bugfix for the atmel aic5 irq chip driver which caches the wrong
  data and thereby breaking resume"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/atmel-aic5: Use per chip mask caches in mask/unmask()