Commit Graph

457282 Commits

Author SHA1 Message Date
Christian König 512d8afc5d drm/radeon: add radeon_bo_ref function
To be consistent with radeon_bo_unref, needed in the following patch.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:53 -04:00
Christian König e7e31600d3 drm/radeon: remove taking mclk_lock from radeon_bo_unref
It's causing lockdep warnings and why should
we access the memory that is freed up?

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:53 -04:00
Christian König dfc230f9af drm/radeon: adjust default radeon_vm_block_size v2
v2: rebase on vm_size scale change. Adjust vm_size default to 8,
    Better handle the default and smaller values.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:52 -04:00
Christian König f77c4f0223 drm/radeon: try to enable VM flushing once more
Let's try to fix bugs related to this instead of just disabling it.

Signed-off-by: Christian König <christian.koenig@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:51 -04:00
Alex Deucher 0aea5e4aa2 drm/radeon: use an intervall tree to manage the VMA v2
Scales much better than scanning the address range linearly.

v2: store pfn instead of address

Signed-off-by: Christian König <christian.koenig@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:51 -04:00
Christian König c265f24d5c drm/radeon: remove radeon_bo_clear_va
Won't work anyway, instead WARN_ON if the VA list isn't
empty when we free the BO.

Signed-off-by: Christian König <christian.koenig@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:50 -04:00
Christian König e31ad969bb drm/radeon: invalidate moved BOs in the VM (v2)
Don't wait for the BO to be used again, just
update the PT on the next VM use.

v2: remove stray semicolon.

Signed-off-by: Christian König <christian.koenig@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:49 -04:00
Alex Deucher c08abf1190 drm/radeon: re-enable dpm by default on BTC
This patch depends on:
e07929810f
(drm/radeon/dpm: fix typo in vddci setup for eg/btc)

bugs:
https://bugs.freedesktop.org/show_bug.cgi?id=73053
https://bugzilla.kernel.org/show_bug.cgi?id=68571

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
2014-08-05 08:53:49 -04:00
Alex Deucher 8f500af4ef drm/radeon: re-enable dpm by default on cayman
This patch depends on:
b0880e87c1
(drm/radeon/dpm: fix vddci setup typo on cayman)

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=69723

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
2014-08-05 08:53:48 -04:00
Michel Dänzer 404a6a512f drm/radeon: Only flush HDP cache from idle ioctl if BO is in VRAM
The HDP cache only applies to CPU access to VRAM.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:47 -04:00
Alex Deucher 6b57f20cb5 drm/radeon/dpm: handle voltage info fetching on hawaii
Some hawaii cards use a different method to fetch the
voltage info from the vbios.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=74250

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
2014-08-05 08:53:47 -04:00
Alex Deucher e9f274b2a1 drm/radeon/atom: add new voltage fetch function for hawaii
Some hawaii boards use a different method for fetching the
voltage information from the vbios.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
2014-08-05 08:53:46 -04:00
Michel Dänzer 72a9987edc drm/radeon: Always flush the HDP cache before submitting a CS to the GPU
This ensures the GPU sees all previous CPU writes to VRAM, which makes it
safe:

* For userspace to stream data from CPU to GPU via VRAM instead of GTT
* For IBs to be stored in VRAM instead of GTT
* For ring buffers to be stored in VRAM instead of GTT, if the HPD flush
  is performed via MMIO

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:45 -04:00
Michel Dänzer 124764f174 drm/radeon: s/ioctl_wait_idle/mmio_hpd_flush/
And clean up the function comment a little.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:44 -04:00
Christian König f28be81011 drm/radeon: separate ring and IB handling
Both on their own are complex enough.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:44 -04:00
Alex Deucher 3157c58972 drm/radeon: fix reversed logic in evergreen_mc_resume
Need to unblank the display when resuming the MC.  No
functional change as this code path is not currently
hit.  We always disable the displays entirely rather
than just blanking them.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:43 -04:00
Christian König f1d2a26b50 drm/radeon: set VM base addr using the PFP v2
Seems to make VM flushes more stable on SI and CIK.

v2: only use the PFP on the GFX ring on CIK

Signed-off-by: Christian König <christian.koenig@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:42 -04:00
Michel Dänzer f069dc1dbc drm/radeon/cik: Read back SDMA WPTR register after writing it
For symmetry with other *_set_wptr hooks.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:42 -04:00
Michel Dänzer 810b73d190 drm/radeon: Use write-combined CPU mappings of IBs on >= CIK
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:41 -04:00
Michel Dänzer 1490434f0d drm/radeon: Use write-combined CPU mappings of ring buffers with PCIe
PCI GART doesn't support unsnooped access. AGP GART already uses
write-combined CPU mappings.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:40 -04:00
Alex Deucher 5dc355325b drm/radeon: load the lm63 driver for an lm64 thermal chip.
Looks like the lm63 driver supports the lm64 as well.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
2014-08-05 08:53:40 -04:00
Alex Deucher ebdea82d2e drm/radeon: make radeon_connector_encoder_is_hbr2 static
It's only used in radeon_connectors.c.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:39 -04:00
Christian König a3a9226d29 drm/radeon: add trace_radeon_vm_flush
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:38 -04:00
Christian König 33fa9fe3dd drm/radeon: fix R600_PTE_GART handling
That didn't worked correctly any more and opened up a security problem.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:37 -04:00
Christian König ed5cb43f8f drm/radeon: remove discardable flag from radeon_gem_object_create
Unused and unimplemented. Also fix specifying the
kernel flag incorrectly at one occasion.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:37 -04:00
Alex Deucher 4807c5a8a0 drm/radeon: add a PX quirk list
Some PX laptops seems to have problems turning the dGPU on/off.
Add a quirk list to disable runpm by default on those systems.
Also convert the current PX d3 delay handling to a quirk.

bug:
https://bugzilla.kernel.org/show_bug.cgi?id=51381
https://bugzilla.kernel.org/show_bug.cgi?id=74551

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:36 -04:00
Alex Deucher 391bfec33c drm/radeon: remove visible vram size limit on bo allocation (v4)
Now that fallback to gtt is fixed for cpu access, we can
remove this limit.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=78717

v2: use new gart_pin_size to accurately track available gtt.
v3: fix comment
v4: clarify comment

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2014-08-05 08:53:35 -04:00
Alex Deucher 2c6316cbcd drm/radeon: use vram/gart pinned size in radeon_do_test_moves
Gives more accurate count and prevents failures when we can't
allocate memory for the tests.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2014-08-05 08:53:35 -04:00
Alex Deucher ccbe0060aa drm/radeon: use vram/gart pinned size in radeon_gem_info_ioctl
Gives a more accurate limit than the previous code.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2014-08-05 08:53:34 -04:00
Alex Deucher 71ecc97eaa drm/radeon: track pinned memory (v2)
So we know how large an allocation we can allow.

v2: incorporate Michel's comments

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
2014-08-05 08:53:33 -04:00
Michel Dänzer 02376d8282 drm/radeon: Allow write-combined CPU mappings of BOs in GTT (v2)
v2: fix rebase onto drm-fixes

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:33 -04:00
Michel Dänzer 77497f2735 drm/radeon: Pass GART page flags to radeon_gart_set_page() explicitly
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:32 -04:00
Michel Dänzer a3eb06dbca drm/radeon: Remove radeon_gart_restore()
Doesn't seem necessary, the GART table memory should be persistent.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:31 -04:00
Michel Dänzer 380670aebf drm/radeon: Demote 'BO allocation size too large' message to debug only
These clutter up dmesg during piglit runs. Userspace generally deals
gracefully with this failure.

Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:30 -04:00
Alex Deucher 377bd8a98d drm/radeon: use a fetch function to get the edid
We keep a cached version of the edid in radeon_connector which
we use for determining connectedness and when to enable certain
features like hdmi audio, etc.  When the user uses the firmware
interface to override the driver with some other edid the driver's
copy is never updated.  The fetch function will check if there
is a user supplied edid and update the driver's copy if there
is.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=80691

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:30 -04:00
Alex Deucher 72a5c970a5 drm/radeon: restructure edid fetching
Split radeon_ddc_get_modes() and move it into
radeon_connectors.c since that is the only place
that uses it.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:29 -04:00
Stefan Brüns 88fd4789cd drm/radeon/audio: break out of loops once we match connector
No need to continue with the loops once we've matched
the appropriate connector.
See commit 8a992ee145

Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:28 -04:00
Stefan Brüns c748990b7b drm/radeon: Use correct value for unknown audio/video latency
Valid values are 1 to 251 for 0 to 500 ms latency, 0 for unknown
and 255 for audio/video unsupported by sink, according to HDMI 1.3 spec.
Also matches Radeon HDA verb 0xf7b documentation.

Signed-off-by: Stefan Brüns <stefan.bruens@rwth-aachen.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:27 -04:00
Lauri Kasanen 59bc1d89d6 drm/radeon: Inline r100_mm_rreg, -wreg, v3
This was originally un-inlined by Andi Kleen in 2011 citing size concerns.
Indeed, a first attempt at inlining it grew radeon.ko by 7%.

However, 2% of cpu is spent in this function. Simply inlining it gave 1% more fps
in Urban Terror.

v2: We know the minimum MMIO size. Adding it to the if allows the compiler to
optimize the branch out, improving both performance and size.

The v2 patch decreases radeon.ko size by 2%. I didn't re-benchmark, but common sense
says perf is now more than 1% better.

v3: Also change _wreg, make the threshold a define.

Inlining _wreg increased the size a bit compared to v2, so now radeon.ko
is only 1% smaller.

Signed-off-by: Lauri Kasanen <cand@gmx.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:27 -04:00
Alex Deucher 3e22920fbd drm/radeon: consolidate vga and dvi get_modes functions (v2)
They are identical.

v2: rebase

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:26 -04:00
Alex Deucher da9976206c drm/radeon: enable display scaling on all connectors (v2)
This enables the display scaler on all connectors for r5xx
and newer asics.  Previously we only enabled the scaler for
fixed mode displays (eDP or LVDS) since they have to use the
scaler to support non-native modes.  Most other displays
are multi-sync or have a built in scaler to support non-native
modes.  The default scaling mode for non-fixed displays is
none which will use the scaler in the monitor.  Note that
we do not populate any fake modes like we do for fixed
displays so it will only use the modes in the edid.  For
other modes, you'll need to populate them manually.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=80868

v2: properly handle scaling with no modes defined

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:25 -04:00
Fabian Frederick 9c2448780f drm/radeon: remove null test before kfree
Fix checkpatch warning:
WARNING: kfree(NULL) is safe this check is probably not required

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:24 -04:00
Alex Deucher f2c6b0f452 drm/radeon/cik: Add support for new ucode format (v5)
This adds CIK support for the new ucode format.

v2: add size validation, integrate debug info
v3: add support for MEC2 on KV
v4: fix typos
v4: update to latest format

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:24 -04:00
Alex Deucher 629bd33c75 drm/radeon/si: Add support for new ucode format (v3)
This adds SI support for the new ucode format.

v2: add size validation, integrate debug info
v3: update to latest version

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:23 -04:00
Alex Deucher de5158226d drm/radeon: add new firmware header definitions (v3)
These are needed to properly handle more frequently
updated firmware.

v2: add new firmware helper functions as well.
v3: update to latest format

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:22 -04:00
Alex Deucher 6960948d48 drm/radeon: disable gfx cgcg on cik
This needs some tweaking to be stable with newer
ucode versions.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:22 -04:00
Alex Deucher 636e258265 drm/radeon/dpm: add support for SVI2 voltage for SI
Some newer boards use SVI2 for voltage control rather
than GPIO.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2014-08-05 08:53:21 -04:00
Daniel Vetter 168c02ec06 drm: Fix race when checking for fb in the generic kms obj lookup
In my review of

commit 98f75de40e
Author: Rob Clark <robdclark@gmail.com>
Date:   Fri May 30 11:37:03 2014 -0400

    drm: add object property typ

I asked for a check to make sure that we never leak an fb from the
generic mode object lookup since those have completely different
lifetime rules. Rob added it, but outside of the idr mutex, which
means that our dereference of obj->type can already chase free'd
memory.

Somehow I didn't spot this, so fix this asap.

v2: Simplify the conditionals as suggested by Chris.

Cc: Rob Clark <robdclark@gmail.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-08-05 11:06:52 +10:00
Chris Wilson dff01de1c3 drm: Unlink dead file_priv from list of active files first
In order to prevent external observers walking the list of open DRM
files from seeing an invalid drm_file_private in the process of being
torndown, the first operation we need to take is to unlink the
drm_file_private from that list.

	general protection fault: 0000 [#1] PREEMPT SMP
	Modules linked in: i915 i2c_algo_bit drm_kms_helper drm lpc_ich mfd_core nls_iso8859_1 i2c_hid video hid_generic usbhid hid e1000e ahci ptp libahci pps_core
	CPU: 3 PID: 8220 Comm: cat Not tainted 3.16.0-rc6+ #4
	Hardware name: Intel Corporation Shark Bay Client platform/WhiteTip Mountain 1, BIOS HSWLPTU1.86C.0119.R00.1303230105 03/23/2013
	task: ffff8800219642c0 ti: ffff880047024000 task.ti: ffff880047024000
	RIP: 0010:[<ffffffffa0137c70>]  [<ffffffffa0137c70>] per_file_stats+0x110/0x160 [i915]
	RSP: 0018:ffff880047027d48  EFLAGS: 00010246
	RAX: 6b6b6b6b6b6b6b6b RBX: ffff880047027e30 RCX: 0000000000000000
	RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff88003a05cd00
	RBP: ffff880047027d58 R08: 0000000000000001 R09: 0000000000000000
	R10: ffff8800219642c0 R11: 0000000000000000 R12: ffff88003a05cd00
	R13: 0000000000000000 R14: ffff88003a05cd00 R15: ffff880047027d88
	FS:  00007f5f73a13740(0000) GS:ffff88014e380000(0000) knlGS:0000000000000000
	CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
	CR2: 00000000023ff038 CR3: 0000000021a4b000 CR4: 00000000001407e0
	Stack:
	 0000000000000001 000000000000ffff ffff880047027dc8 ffffffff813438e4
	 ffff880047027e30 ffffffffa0137b60 ffff880021a8af58 ffff880021a8f1a0
	 ffff8800a2061fb0 ffff8800a2062048 ffff8800a2061fb0 ffff8800a1e23478
	Call Trace:
	 [<ffffffff813438e4>] idr_for_each+0xf4/0x180
	 [<ffffffffa0137b60>] ? i915_gem_stolen_list_info+0x1f0/0x1f0 [i915]
	 [<ffffffffa013a17a>] i915_gem_object_info+0x5ca/0x6a0 [i915]
	 [<ffffffff81193ec5>] seq_read+0xf5/0x3a0
	 [<ffffffff8116d950>] vfs_read+0x90/0x150
	 [<ffffffff8116e509>] SyS_read+0x49/0xb0
	 [<ffffffff815d8622>] tracesys+0xd0/0xd5
	Code: 01 00 00 49 39 84 24 08 01 00 00 74 55 49 8b 84 24 b8 00 00 00 48 01 43 18 31 c0 5b 41 5c 5d c3 0f 1f 00 49 8b 44 24 08 4c 89 e7 <48> 8b 70 28 48 81 c6 48 80 00 00 e8 80 14 01 00 84 c0 74 bc 49
	RIP  [<ffffffffa0137c70>] per_file_stats+0x110/0x160 [i915]
	RSP <ffff880047027d48>

Reported-by: "Ursulin, Tvrtko" <tvrtko.ursulin@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81712
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Ursulin, Tvrtko" <tvrtko.ursulin@intel.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-08-05 11:05:49 +10:00
Tetsuo Handa a91576d791 drm/ttm: Pass GFP flags in order to avoid deadlock.
Commit 7dc19d5a "drivers: convert shrinkers to new count/scan API" added
deadlock warnings that ttm_page_pool_free() and ttm_dma_page_pool_free()
are currently doing GFP_KERNEL allocation.

But these functions did not get updated to receive gfp_t argument.
This patch explicitly passes sc->gfp_mask or GFP_KERNEL to these functions,
and removes the deadlock warning.

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: stable <stable@kernel.org> [2.6.35+]
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-08-05 10:54:19 +10:00