Chris Wilson
a740f5c5f6
drm/i915/selftests: Skip energy consumption tests if not controlling freq
...
If we can not manipulate the frequency with RPS, then comparing min/max
power consumption is pointless / misleading. We will leave the warning
about not being able to control the frequency selection via RPS to other
tests so as not to confuse this more specialised check.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200420172739.11620-2-chris@chris-wilson.co.uk
2020-04-20 20:08:06 +01:00
Chris Wilson
4ba74e53ad
drm/i915/selftests: Verify frequency scaling with RPS
...
One of the core tenents of reclocking the GPU is that its throughput
scales with the clock frequency. We can observe this by incrementing a
loop counter on the GPU, and compare the different execution rates at
the notional RPS frequencies.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200420172739.11620-1-chris@chris-wilson.co.uk
2020-04-20 20:08:06 +01:00
Ville Syrjälä
f0617ff0b8
drm/i915: Push MST link retraining to the hotplug work
...
We shouldn't try to do link retraining from the short hpd handler.
We can't take any modeset locks there so this is racy as hell.
Push the whole thing into the hotplug work like we do with SST.
We'll just have to adjust the SST retraining code to deal with
the MST encoders and multiple pipes.
TODO: I have a feeling we should just rip this all out and
do a full modeset instead. Stuff like port sync and the tgl+
MST master transcoder stuff maybe doesn't work well if we
try to retrain without following the proper modeset sequence.
So far haven't done any actual tests to confirm that though.
Cc: Lyude Paul <lyude@redhat.com >
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417152734.464-2-ville.syrjala@linux.intel.com
Reviewed-by: Lyude Paul <lyude@redhat.com >
2020-04-20 21:21:11 +03:00
Ville Syrjälä
3c0ec2c2d5
drm/i915: Flatten intel_dp_check_mst_status() a bit
...
Make intel_dp_check_mst_status() somewhat legible by humans.
Note that the return value of drm_dp_mst_hpd_irq() is always
either 0 or -ENOMEM, and we never did anything with the latter
so we can just ignore the whole thing.
We can also get rid of the direct drm_dp_mst_topology_mgr_set_mst(false)
call since returning -EINVAL causes the caller to do the very same call
for us.
Cc: Lyude Paul <lyude@redhat.com >
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417152734.464-1-ville.syrjala@linux.intel.com
Reviewed-by: Lyude Paul <lyude@redhat.com >
2020-04-20 21:21:11 +03:00
Ville Syrjälä
eed22a46b9
drm/i915: Pass encoder all the way to intel_ddi_transcoder_func_reg_val_get()
...
Pass the encoder all the way down to
intel_ddi_transcoder_func_reg_val_get(). Allows us eliminate the
intel_ddi_get_crtc_encoder() eyesore.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417134720.16654-4-ville.syrjala@linux.intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Acked-by: Jani Nikula <jani.nikula@intel.com >
2020-04-20 21:21:10 +03:00
Ville Syrjälä
7c2fedd760
drm/i915: Push TRANS_DDI_FUNC_CTL into the encoder->enable() hook
...
Push the TRANS_DDI_FUNC_CTL into the encoder enable hook. The disable
is already there, and as a followup will enable us to pass the encoder
all the way down.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417134720.16654-3-ville.syrjala@linux.intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Acked-by: Jani Nikula <jani.nikula@intel.com >
2020-04-20 21:21:10 +03:00
Ville Syrjälä
c38730987b
drm/i915: Move the TRANS_DDI_FUNC_CTL enable to a later point
...
No reason that I can see why we should enable TRANS_DDI_FUNC_CTL
before we set up the watermarks of configure the mbus stuff.
In fact reordering these seems to match the bspec sequence better,
and crucially will allow us to push the TRANS_DDI_FUNC_CTL enable
into the encoder enable hook as a followup.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417134720.16654-2-ville.syrjala@linux.intel.com
Acked-by: Chris Wilson <chris@chris-wilson.co.uk >
Acked-by: Jani Nikula <jani.nikula@intel.com >
2020-04-20 21:21:10 +03:00
Ville Syrjälä
02a715c371
drm/i915: Pass encoder to intel_ddi_enable_pipe_clock()
...
Since intel_ddi_enable_pipe_clock() was pushed down into the
encoder hooks we can pass on the encoder instead of having
to use intel_ddi_get_crtc_encoder().
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417134720.16654-1-ville.syrjala@linux.intel.com
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Acked-by: Jani Nikula <jani.nikula@intel.com >
2020-04-20 21:20:58 +03:00
Lyude Paul
d082119f42
drm/i915/dpcd_bl: Unbreak enable_dpcd_backlight modparam
...
Looks like I accidentally made it so you couldn't force DPCD backlight
support on, whoops. Fix that.
Signed-off-by: Lyude Paul <lyude@redhat.com >
Fixes: 17f5d57915ajax@redhat.com >
Cc: Jani Nikula <jani.nikula@linux.intel.com >
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com >
Cc: "Ville Syrjälä" <ville.syrjala@linux.intel.com >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200413214407.1851002-1-lyude@redhat.com
(cherry picked from commit d7fb38ae36rodrigo.vivi@intel.com >
2020-04-20 10:12:58 -07:00
Jani Nikula
a05b289917
drm/i915: fix Sphinx build duplicate label warning
...
Fix the warning caused by enabling the autosectionlabel extension in the
kernel Sphinx build:
Documentation/gpu/i915.rst:610: WARNING: duplicate label
gpu/i915:layout, other instance in Documentation/gpu/i915.rst
The autosectionlabel extension adds labels to each section title for
cross-referencing, but forbids identical section titles in a
document. With kernel-doc, this includes sections titles in the included
kernel-doc comments.
In the warning message, Sphinx is unable to reference the labels in
their true locations in the kernel-doc comments in source. In this case,
there's "Layout" sections in both gt/intel_workarounds.c and
i915_reg.h. Rename the section in the latter to "File Layout".
Fixes: 58ad30cf91chris@chris-wilson.co.uk >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417130109.12791-1-jani.nikula@intel.com
(cherry picked from commit 27be41de45rodrigo.vivi@intel.com >
2020-04-20 10:12:53 -07:00
José Roberto de Souza
1239902875
drm/i915/display: Load DP_TP_CTL/STATUS offset before use it
...
Right now dp.regs.dp_tp_ctl/status are only set during the encoder
pre_enable() hook, what is causing all reads and writes to those
registers to go to offset 0x0 before pre_enable() is executed.
So if i915 takes the BIOS state and don't do a modeset any following
link retraing will fail.
In the case that i915 needs to do a modeset, the DDI disable sequence
will write to a wrong register not disabling DP 'Transport Enable' in
DP_TP_CTL, making a HDMI modeset in the same port/transcoder to
not light up the monitor.
So here for GENs older than 12, that have those registers fixed at
port offset range it is loading at encoder/port init while for GEN12
it will keep setting it at encoder pre_enable() and during HW state
readout.
Fixes: 4444df6e20matthew.d.roper@intel.com >
Cc: Lucas De Marchi <lucas.demarchi@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Reviewed-by: Matt Roper <matthew.d.roper@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414230442.262092-1-jose.souza@intel.com
(cherry picked from commit edcb9028d6rodrigo.vivi@intel.com >
2020-04-20 10:12:46 -07:00
Matt Roper
335f62e760
drm/i915/tgl: TBT AUX should use TC power well ops
...
As on ICL, we want to use the Type-C aux handlers for the TBT aux wells
to ensure the DP_AUX_CH_CTL_TBT_IO flag is set properly.
Fixes: 656409bbafjose.souza@intel.com >
Cc: Imre Deak <imre.deak@intel.com >
Signed-off-by: Matt Roper <matthew.d.roper@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415233435.3064257-1-matthew.d.roper@intel.com
Reviewed-by: José Roberto de Souza <jose.souza@intel.com >
(cherry picked from commit 3cbdb97564rodrigo.vivi@intel.com >
2020-04-20 10:12:43 -07:00
Oliver Barta
ced633266b
drm/i915: HDCP: fix Ri prime check done during link check
...
The check was always succeeding even in case of a mismatch due to the
HDCP_STATUS_ENC bit being set. Make sure both bits are actually set.
Signed-off-by: Oliver Barta <oliver.barta@aptiv.com >
Fixes: 2320175febramalingam.c@intel.com >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200328104100.12162-1-oliver.barta@aptiv.com
(cherry picked from commit 3ffaf56e91rodrigo.vivi@intel.com >
2020-04-20 10:12:40 -07:00
Chris Wilson
e1eb075c50
drm/i915/gt: Update PMINTRMSK holding fw
...
If we use a non-forcewaked write to PMINTRMSK, it does not take effect
until much later, if at all, causing a loss of RPS interrupts and no GPU
reclocking, leaving the GPU running at the wrong frequency for long
periods of time.
Reported-by: Francisco Jerez <currojerez@riseup.net >
Suggested-by: Francisco Jerez <currojerez@riseup.net >
Fixes: 35cc7f32c2chris@chris-wilson.co.uk >
Cc: Francisco Jerez <currojerez@riseup.net >
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com >
Cc: Andi Shyti <andi.shyti@intel.com >
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com >
Reviewed-by: Andi Shyti <andi.shyti@intel.com >
Reviewed-by: Francisco Jerez <currojerez@riseup.net >
Cc: <stable@vger.kernel.org > # v5.6+
Link: https://patchwork.freedesktop.org/patch/msgid/20200415170318.16771-2-chris@chris-wilson.co.uk
(cherry picked from commit a080bd994crodrigo.vivi@intel.com >
2020-04-20 10:12:36 -07:00
Matt Roper
81fdd7bfeb
drm/i915/tgl: Add Wa_14010477008:tgl
...
Media decompression support should not be advertised on any display
planes for steppings A0-C0.
Bspec: 53273
Fixes: 2dfbf9d287matthew.s.atwood@intel.com >
Signed-off-by: Matt Roper <matthew.d.roper@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414211118.2787489-3-matthew.d.roper@intel.com
Reviewed-by: José Roberto de Souza <jose.souza@intel.com >
(cherry picked from commit dbff5a8db9rodrigo.vivi@intel.com >
2020-04-20 10:12:32 -07:00
Peter Jones
6b7fc6a3e6
Make the "Reducing compressed framebufer size" message be DRM_INFO_ONCE()
...
This was sort of annoying me:
random:~$ dmesg | tail -1
[523884.039227] [drm] Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS.
random:~$ dmesg | grep -c "Reducing the compressed"
47
This patch makes it DRM_INFO_ONCE() just like the similar message
farther down in that function is pr_info_once().
Cc: stable@vger.kernel.org
Signed-off-by: Peter Jones <pjones@redhat.com >
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com >
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1745
Link: https://patchwork.freedesktop.org/patch/msgid/20180706190424.29194-1-pjones@redhat.com
[vsyrjala: Rebase due to per-device logging]
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
2020-04-20 19:14:59 +03:00
Chris Wilson
f153f6395a
drm/i915/gt: Move the late flush_submission in retire to the end
...
Avoid flushing the submission queue (of others) under the client's
timeline lock, but instead move it to the end so that we may catch more.
References: https://gitlab.freedesktop.org/drm/intel/-/issues/1066
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Matthew Auld <matthew.auld@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200420125356.26614-2-chris@chris-wilson.co.uk
2020-04-20 16:56:23 +01:00
Chris Wilson
a95f3ac21d
drm/i915/gem: Remove object_is_locked assertion from unpin_from_display_plane
...
Since moving the obj->vma.list to a spin_lock, and the vm->bound_list to
its vm->mutex, along with tracking shrinkable status under its own
spinlock, we no long require the object to be locked by the caller.
This is fortunate as it appears we can be called with the lock along an
error path in flipping:
<4> [139.942851] WARN_ON(debug_locks && !lock_is_held(&(&((obj)->base.resv)->lock.base)->dep_map))
<4> [139.943242] WARNING: CPU: 0 PID: 1203 at drivers/gpu/drm/i915/gem/i915_gem_domain.c:405 i915_gem_object_unpin_from_display_plane+0x70/0x130 [i915]
<4> [139.943263] Modules linked in: snd_hda_intel i915 vgem snd_hda_codec_realtek snd_hda_codec_generic coretemp snd_intel_dspcfg snd_hda_codec snd_hwdep snd_hda_core r8169 lpc_ich snd_pcm realtek prime_numbers [last unloaded: i915]
<4> [139.943347] CPU: 0 PID: 1203 Comm: kms_flip Tainted: G U 5.6.0-gd0fda5c2cf3f1-drmtip_474+ #1
<4> [139.943363] Hardware name: /D510MO, BIOS MOPNV10J.86A.0311.2010.0802.2346 08/02/2010
<4> [139.943589] RIP: 0010:i915_gem_object_unpin_from_display_plane+0x70/0x130 [i915]
<4> [139.943589] Code: 85 28 01 00 00 be ff ff ff ff 48 8d 78 60 e8 d7 9b f0 e2 85 c0 75 b9 48 c7 c6 50 b9 38 c0 48 c7 c7 e9 48 3c c0 e8 20 d4 e9 e2 <0f> 0b eb a2 48 c7 c1 08 bb 38 c0 ba 0a 01 00 00 48 c7 c6 88 a3 35
<4> [139.943589] RSP: 0018:ffffb774c0603b48 EFLAGS: 00010282
<4> [139.943589] RAX: 0000000000000000 RBX: ffff9a142fa36e80 RCX: 0000000000000006
<4> [139.943589] RDX: 000000000000160d RSI: ffff9a142c1a88f8 RDI: ffffffffa434a64d
<4> [139.943589] RBP: ffff9a1410a513c0 R08: ffff9a142c1a88f8 R09: 0000000000000000
<4> [139.943589] R10: 0000000000000000 R11: 0000000000000000 R12: ffff9a1436ee94b8
<4> [139.943589] R13: 0000000000000001 R14: 00000000ffffffff R15: ffff9a1410960000
<4> [139.943589] FS: 00007fc73a744e40(0000) GS:ffff9a143da00000(0000) knlGS:0000000000000000
<4> [139.943589] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [139.943589] CR2: 00007fc73997e098 CR3: 000000002f5fe000 CR4: 00000000000006f0
<4> [139.943589] Call Trace:
<4> [139.943589] intel_pin_and_fence_fb_obj+0x1c9/0x1f0 [i915]
<4> [139.943589] intel_plane_pin_fb+0x3f/0xd0 [i915]
<4> [139.943589] intel_prepare_plane_fb+0x13b/0x5c0 [i915]
<4> [139.943589] drm_atomic_helper_prepare_planes+0x85/0x110
<4> [139.943589] intel_atomic_commit+0xda/0x390 [i915]
<4> [139.943589] drm_atomic_helper_page_flip+0x9c/0xd0
<4> [139.943589] ? drm_event_reserve_init+0x46/0x60
<4> [139.943589] drm_mode_page_flip_ioctl+0x587/0x5d0
This completes the symmetry lost in commit 8b1c78e06ehttps://gitlab.freedesktop.org/drm/intel/-/issues/1743
Fixes: 8b1c78e06echris@chris-wilson.co.uk >
Cc: Matthew Auld <matthew.auld@intel.com >
Cc: Andi Shyti <andi.shyti@intel.com >
Cc: <stable@vger.kernel.org > # v5.6+
Reviewed-by: Matthew Auld <matthew.auld@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200420125356.26614-1-chris@chris-wilson.co.uk
2020-04-20 16:23:24 +01:00
Thomas Zimmermann
f0adbc382b
drm/ast: Allocate initial CRTC state of the correct size
...
The ast driver inherits from DRM's CRTC state, but still uses the atomic
helper for struct drm_crtc_funcs.reset, drm_atomic_helper_crtc_reset().
The helper only allocates enough memory for the core CRTC state. That
results in an out-ouf-bounds access when duplicating the initial CRTC
state. Simplified backtrace shown below:
[ 21.469321] ==================================================================
[ 21.469434] BUG: KASAN: slab-out-of-bounds in ast_crtc_atomic_duplicate_state+0x84/0x100 [ast]
[ 21.469445] Read of size 8 at addr ffff888036c1c5f8 by task systemd-udevd/382
[ 21.469451]
[ 21.469464] CPU: 2 PID: 382 Comm: systemd-udevd Tainted: G E 5.5.0-rc6-1-default+ #214
[ 21.469473] Hardware name: Sun Microsystems SUN FIRE X2270 M2/SUN FIRE X2270 M2, BIOS 2.05 07/01/2010
[ 21.469480] Call Trace:
[ 21.469501] dump_stack+0xb8/0x110
[ 21.469528] print_address_description.constprop.0+0x1b/0x1e0
[ 21.469557] ? ast_crtc_atomic_duplicate_state+0x84/0x100 [ast]
[ 21.469581] ? ast_crtc_atomic_duplicate_state+0x84/0x100 [ast]
[ 21.469597] __kasan_report.cold+0x1a/0x35
[ 21.469640] ? ast_crtc_atomic_duplicate_state+0x84/0x100 [ast]
[ 21.469665] kasan_report+0xe/0x20
[ 21.469693] ast_crtc_atomic_duplicate_state+0x84/0x100 [ast]
[ 21.469733] drm_atomic_get_crtc_state+0xbf/0x1c0
[ 21.469768] __drm_atomic_helper_set_config+0x81/0x5a0
[ 21.469803] ? drm_atomic_plane_check+0x690/0x690
[ 21.469843] ? drm_client_rotation+0xae/0x240
[ 21.469876] drm_client_modeset_commit_atomic+0x230/0x390
[ 21.469888] ? __mutex_lock+0x8f0/0xbe0
[ 21.469929] ? drm_client_firmware_config.isra.0+0xa60/0xa60
[ 21.469948] ? drm_client_modeset_commit_force+0x28/0x230
[ 21.470031] ? memset+0x20/0x40
[ 21.470078] drm_client_modeset_commit_force+0x90/0x230
[ 21.470110] drm_fb_helper_restore_fbdev_mode_unlocked+0x5f/0xc0
[ 21.470132] drm_fb_helper_set_par+0x59/0x70
[ 21.470155] fbcon_init+0x61d/0xad0
[ 21.470185] ? drm_fb_helper_restore_fbdev_mode_unlocked+0xc0/0xc0
[ 21.470232] visual_init+0x187/0x240
[ 21.470266] do_bind_con_driver+0x2e3/0x460
[ 21.470321] do_take_over_console+0x20a/0x290
[ 21.470371] do_fbcon_takeover+0x85/0x100
[ 21.470402] register_framebuffer+0x2fd/0x490
[ 21.470425] ? kzalloc.constprop.0+0x10/0x10
[ 21.470503] __drm_fb_helper_initial_config_and_unlock+0xf2/0x140
[ 21.470533] drm_fbdev_client_hotplug+0x162/0x250
[ 21.470563] drm_fbdev_generic_setup+0xd2/0x155
[ 21.470602] ast_driver_load+0x688/0x850 [ast]
<...>
[ 21.472625] ==================================================================
Allocating enough memory for struct ast_crtc_state in a custom ast CRTC
reset handler fixes the problem.
v2:
* implement according to drm_atomic_helper_crtc_reset()
* update state with __drm_atomic_helper_crtc_reset()
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de >
Fixes: 83be6a3cebdaniel.vetter@ffwll.ch >
Cc: Gerd Hoffmann <kraxel@redhat.com >
Cc: Dave Airlie <airlied@redhat.com >
Cc: Daniel Vetter <daniel.vetter@ffwll.ch >
Cc: Alex Deucher <alexander.deucher@amd.com >
Cc: "Noralf Trønnes" <noralf@tronnes.org >
Cc: Sam Ravnborg <sam@ravnborg.org >
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200130094012.32140-1-tzimmermann@suse.de
2020-04-20 15:48:23 +02:00
Lyude Paul
d7fb38ae36
drm/i915/dpcd_bl: Unbreak enable_dpcd_backlight modparam
...
Looks like I accidentally made it so you couldn't force DPCD backlight
support on, whoops. Fix that.
Signed-off-by: Lyude Paul <lyude@redhat.com >
Fixes: 17f5d57915ajax@redhat.com >
Cc: Jani Nikula <jani.nikula@linux.intel.com >
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com >
Cc: "Ville Syrjälä" <ville.syrjala@linux.intel.com >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200413214407.1851002-1-lyude@redhat.com
2020-04-20 11:18:26 +03:00
Tomi Valkeinen
9da67433f6
drm/tidss: fix crash related to accessing freed memory
...
tidss uses devm_kzalloc to allocate DRM plane, encoder and crtc objects.
This is not correct as the lifetime of those objects should be longer
than the underlying device's.
When unloading tidss module, the devm_kzalloc'ed objects have already
been freed when tidss_release() is called, and the driver will accesses
freed memory possibly causing a crash, a kernel WARN, or other undefined
behavior, and also KASAN will give a bug.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415092006.26675-1-tomi.valkeinen@ti.com
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch >
2020-04-20 10:07:35 +03:00
Jani Nikula
b4ed131dbf
drm/i915/audio: error log non-zero audio power refcount after unbind
...
We have some module unload/reload tests hitting an issue with i915
unbinding the component interface before the audio driver has properly
put the power. Log an error about it for ease of debugging. (Normally
this leads to a wakeref debug splat on the power well.)
Cc: Kai Vehmanen <kai.vehmanen@linux.intel.com >
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417065132.23048-1-jani.nikula@intel.com
2020-04-20 09:41:10 +03:00
Jani Nikula
27be41de45
drm/i915: fix Sphinx build duplicate label warning
...
Fix the warning caused by enabling the autosectionlabel extension in the
kernel Sphinx build:
Documentation/gpu/i915.rst:610: WARNING: duplicate label
gpu/i915:layout, other instance in Documentation/gpu/i915.rst
The autosectionlabel extension adds labels to each section title for
cross-referencing, but forbids identical section titles in a
document. With kernel-doc, this includes sections titles in the included
kernel-doc comments.
In the warning message, Sphinx is unable to reference the labels in
their true locations in the kernel-doc comments in source. In this case,
there's "Layout" sections in both gt/intel_workarounds.c and
i915_reg.h. Rename the section in the latter to "File Layout".
Fixes: 58ad30cf91chris@chris-wilson.co.uk >
Signed-off-by: Jani Nikula <jani.nikula@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417130109.12791-1-jani.nikula@intel.com
2020-04-20 09:07:08 +03:00
Michael J. Ruhl
31a02eb70b
drm/i915: Refactor setting dma info to a common helper
...
DMA_MASK bit values are different for different generations.
This will become more difficult to manage over time with the open
coded usage of different versions of the device.
Fix by:
disallow setting of dma mask in AGP path (< GEN(5) for i915,
add dma_mask_size to the device info configuration,
updating open code call sequence to the latest interface,
refactoring into a common function for setting the dma segment
and mask info
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com >
cc: Brian Welty <brian.welty@intel.com >
cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com >
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417195107.68732-1-michael.j.ruhl@intel.com
2020-04-18 07:49:11 +01:00
Colin Ian King
7479f3c90a
drm/i915: remove redundant assignment to variable test_result
...
The variable test_result is being initialized with a value that is
never read and it is being updated later with a new value. The
initialization is redundant and can be removed.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com >
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417160829.112776-1-colin.king@canonical.com
2020-04-18 07:47:22 +01:00
Anshuman Gupta
65bb9dd0ec
drm/i915: Add ICL PG3 PW ID for EHL
...
Gen11 onwards PG3 contains functions for pipe B,
external displays, and VGA. Add missing ICL_DISP_PW_3
for ehl_power_wells.
Cc: Animesh Manna <animesh.manna@intel.com >
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1737
Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com >
Acked-by: Chris Wilson <chris@chris-wilson.co.uk >
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417172835.15461-1-anshuman.gupta@intel.com
2020-04-18 07:44:56 +01:00
Mikita Lipski
7bfc1fec1a
drm/dp_mst: Zero assigned PBN when releasing VCPI slots
...
Zero Port's PBN together with VCPI slots when releasing
allocated VCPI slots. That way when disabling the connector
it will not cause issues in drm_dp_mst_atomic_check verifying
branch bw limit.
Signed-off-by: Mikita Lipski <mikita.lipski@amd.com >
Signed-off-by: Lyude Paul <lyude@redhat.com >
Fixes: cd82d82cbcstable@vger.kernel.org > # v5.6+
Link: https://patchwork.freedesktop.org/patch/msgid/20200407160717.27976-1-mikita.lipski@amd.com
2020-04-17 18:25:28 -04:00
José Roberto de Souza
edcb9028d6
drm/i915/display: Load DP_TP_CTL/STATUS offset before use it
...
Right now dp.regs.dp_tp_ctl/status are only set during the encoder
pre_enable() hook, what is causing all reads and writes to those
registers to go to offset 0x0 before pre_enable() is executed.
So if i915 takes the BIOS state and don't do a modeset any following
link retraing will fail.
In the case that i915 needs to do a modeset, the DDI disable sequence
will write to a wrong register not disabling DP 'Transport Enable' in
DP_TP_CTL, making a HDMI modeset in the same port/transcoder to
not light up the monitor.
So here for GENs older than 12, that have those registers fixed at
port offset range it is loading at encoder/port init while for GEN12
it will keep setting it at encoder pre_enable() and during HW state
readout.
Fixes: 4444df6e20matthew.d.roper@intel.com >
Cc: Lucas De Marchi <lucas.demarchi@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Reviewed-by: Matt Roper <matthew.d.roper@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414230442.262092-1-jose.souza@intel.com
2020-04-17 15:08:53 -07:00
José Roberto de Souza
0f8925090a
drm/i915/tc: Do not warn when aux power well of static TC ports timeout
...
This is a expected timeout of static TC ports not conneceted, so
not throwing warnings that would taint CI.
v3:
- moved checks to tc_phy_aux_timeout_expected()
v4:
- moved and add comments to tc_phy_aux_timeout_expected()
v5:
- only checking tc_legacy_port for TC ports
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-8-jose.souza@intel.com
2020-04-17 15:01:34 -07:00
José Roberto de Souza
3ed347d1a7
drm/i915/tc: Catch TC users accessing FIA registers without enable aux
...
As described in "drm/i915/tc/icl: Implement TC cold sequences" users
of TC functions should held aux power well during access to avoid
read garbage due HW in TC cold state.
v3:
- renamed is_tc_cold_blocked() to assert_tc_cold_blocked()
- restored the removed 0xffffffff checks
Reviewed-by: Imre Deak <imre.deak@intel.com >
Tested-by: You-Sheng Yang <vicamo.yang@canonical.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-7-jose.souza@intel.com
2020-04-17 15:01:34 -07:00
José Roberto de Souza
3c02934b24
drm/i915/tc/tgl: Implement TC cold sequences
...
TC ports can enter in TCCOLD to save power and is required to request
to PCODE to exit this state before use or read to TC registers.
For TGL there is a new MBOX command to do that with a parameter to ask
PCODE to exit and block TCCOLD entry or unblock TCCOLD entry.
So adding a new power domain to reuse the refcount and only allow
TC cold when all TC ports are not in use.
v2:
- fixed missing case in intel_display_power_domain_str()
- moved tgl_tc_cold_request to intel_display_power.c
- renamed TGL_TC_COLD_OFF to TGL_TC_COLD_OFF_POWER_DOMAINS
- added all TC and TBT aux power domains to
TGL_TC_COLD_OFF_POWER_DOMAINS
v3:
- added one msec sleep when PCODE returns -EAGAIN
- added timeout of 5msec to not loop forever if
sandybridge_pcode_write_timeout() keeps returning -EAGAIN
v4:
- Made failure to block or unblock TC cold a error
- removed 5msec timeout, instead giving PCODE 1msec by up 3 times to
recover from the internal error
v5:
- only sleeping 1msec when ret is -EAGAIN
BSpec: 49294
Cc: Imre Deak <imre.deak@intel.com >
Cc: Cooper Chiou <cooper.chiou@intel.com >
Cc: Kai-Heng Feng <kai.heng.feng@canonical.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-6-jose.souza@intel.com
2020-04-17 15:01:34 -07:00
José Roberto de Souza
7ce40a6715
drm/i915/tc: Skip ref held check for TC legacy aux power wells
...
As part of ICL TC cold exit sequences we need to request aux power
well before lock the access to TC ports, so skiping the
intel_tc_port_ref_held() check for TC legacy ports.
Reviewed-by: Imre Deak <imre.deak@intel.com >
Tested-by: You-Sheng Yang <vicamo.yang@canonical.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-5-jose.souza@intel.com
2020-04-17 15:01:34 -07:00
José Roberto de Souza
feb7e0ef5f
drm/i915/tc/icl: Implement TC cold sequences
...
This is required for legacy/static TC ports as IOM is not aware of
the connection and will not trigger the TC cold exit.
Just request PCODE to exit TCCOLD is not enough as it could enter
again before driver makes use of the port, to prevent it BSpec states
that aux powerwell should be held.
So here embedding the TC cold exit sequence into ICL aux enable,
it will enable aux and then request TC cold to exit.
The TC cold block(exit and aux hold) and unblock was added to some
exported TC functions for the others and to access PHY registers,
callers should enable and keep aux powerwell enabled during access.
Also adding TC cold check and warnig in tc_port_load_fia_params() as
at this point of the driver initialization we can't request power
wells, if we get this warning we will need to figure out how to handle
it.
v2:
- moved ICL TC cold exit function to intel_display_power
- using dig_port->tc_legacy_port to only execute sequences for legacy
ports, hopefully VBTs will have this right
- fixed check to call _hsw_power_well_continue_enable()
- calling _hsw_power_well_continue_enable() unconditionally in
icl_tc_phy_aux_power_well_enable(), if needed we will surpress timeout
warnings of TC legacy ports
- only blocking TC cold around fia access
v3:
- added timeout of 5msec to not loop forever if
sandybridge_pcode_write_timeout() keeps returning -EAGAIN
returning -EAGAIN in in icl_tc_cold_exit()
- removed leftover tc_cold_wakeref
- added one msec sleep when PCODE returns -EAGAIN
v4:
- removed 5msec timeout, instead giving 1msec to whoever is using
PCODE to finish it up to 3 times
- added a comment about turn TC cold exit failure as a error in future
BSpec: 21750
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1296
Cc: Imre Deak <imre.deak@intel.com >
Cc: Cooper Chiou <cooper.chiou@intel.com >
Cc: Kai-Heng Feng <kai.heng.feng@canonical.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-4-jose.souza@intel.com
2020-04-17 15:00:50 -07:00
José Roberto de Souza
f8bb28e63a
drm/i915/display: Split hsw_power_well_enable() into two
...
This is a preparation for ICL TC cold exit sequences.
v2:
- renamed new functions to hsw_power_well_enable_prepare()/complete()
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Tested-by: You-Sheng Yang <vicamo.yang@canonical.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-3-jose.souza@intel.com
2020-04-17 14:55:30 -07:00
José Roberto de Souza
dba6b0b4ea
drm/i915/display: Add intel_legacy_aux_to_power_domain()
...
This is a similar function to intel_aux_power_domain() but it do not
care about TBT ports, this will be needed by ICL TC sequences.
v2:
- renamed to intel_legacy_aux_to_power_domain()
Cc: Imre Deak <imre.deak@intel.com >
Cc: Cooper Chiou <cooper.chiou@intel.com >
Cc: Kai-Heng Feng <kai.heng.feng@canonical.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Tested-by: You-Sheng Yang <vicamo.yang@canonical.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-2-jose.souza@intel.com
2020-04-17 14:55:29 -07:00
José Roberto de Souza
34a3f0b273
drm/i915/display: Move out code to return the digital_port of the aux ch
...
Moving the code to return the digital port of the aux channel also
removing the intel_phy_is_tc() to make it generic.
digital_port will be needed in icl_tc_phy_aux_power_well_enable()
so adding it as a parameter to icl_tc_port_assert_ref_held().
While at at removing the duplicated call to icl_tc_phy_aux_ch() in
icl_tc_port_assert_ref_held().
v2:
- fixed build when DRM_I915_DEBUG_RUNTIME_PM is not set
- moved to before hsw_wait_for_power_well_enable() as it will be
needed by hsw_wait_for_power_well_enable() in a future patch
v4:
- fixed action of if (!dig_port), continue instead of return
Cc: You-Sheng Yang <vicamo@gmail.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Tested-by: You-Sheng Yang <vicamo.yang@canonical.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200414194956.164323-1-jose.souza@intel.com
2020-04-17 14:55:29 -07:00
Alex Deucher
7daec99fdc
drm/amdgpu/display: give aux i2c buses more meaningful names
...
Mirror what we do for i2c display buses.
Reviewed-by: Harry Wentland <harry.wentland@amd.com >
Signed-off-by: Alex Deucher <alexander.deucher@amd.com >
2020-04-17 17:31:38 -04:00
Alex Deucher
00a8037e20
drm/amdgpu/display: fix aux registration (v2)
...
We were registering the aux device in the MST late_register
rather than the regular one.
v2: handle eDP as well
Fixes: 405a1f9090https://gitlab.freedesktop.org/drm/amd/-/issues/1100
Signed-off-by: Alex Deucher <alexander.deucher@amd.com >
Reviewed-by: Harry Wentland <harry.wentland@amd.com >
2020-04-17 17:31:37 -04:00
Sandeep Raghuraman
bbc25dadc7
drm/amdgpu: Correctly initialize thermal controller for GPUs with Powerplay table v0 (e.g Hawaii)
...
Initialize thermal controller fields in the PowerPlay table for Hawaii
GPUs, so that fan speeds are reported.
Signed-off-by: Sandeep Raghuraman <sandy.8925@gmail.com >
Signed-off-by: Alex Deucher <alexander.deucher@amd.com >
2020-04-17 17:31:37 -04:00
Prike Liang
45a5e63954
drm/amd/powerplay: fix resume failed as smu table initialize early exit
...
When the amdgpu in the suspend/resume loop need notify the dpm disabled,
otherwise the smu table will be uninitialize and result in resume failed.
Signed-off-by: Prike Liang <Prike.Liang@amd.com >
Tested-by: Mengbing Wang <Mengbing.Wang@amd.com >
Reviewed-by: Alex Deucher <alexander.deucher@amd.com >
Reviewed-by: Huang Rui <ray.huang@amd.com >
Signed-off-by: Alex Deucher <alexander.deucher@amd.com >
2020-04-17 17:31:36 -04:00
José Roberto de Souza
c0ff9e5e69
drm/i915: Add missing deinitialization cases of load failure
...
The intel_display_power_put_async() used in TC cold sequences made
easy to hit the missing deinitialization of driver in case of load
failure as seen in the stack trace bellow.
intel_modeset_driver_remove_noirq() had to be removed from
i915_driver_modeset_remove_noirq() as those are different
initialialition steps with IRQ and GEM initialization in between then.
[drm:__intel_engine_init_ctx_wa [i915]] Initialized 3 context workarounds on rcs'0
[drm:__i915_inject_probe_error [i915]] Injecting failure -19 at checkpoint 36 [__uc_init:294]
[drm:i915_hdcp_component_unbind [i915]] I915 HDCP comp unbind
[drm:edp_panel_vdd_off_sync [i915]] Turning [ENCODER:275:DDI A] VDD off
[drm:edp_panel_vdd_off_sync [i915]] PP_STATUS: 0x00000000 PP_CONTROL: 0x00000060
[drm:intel_power_well_disable [i915]] disabling AUX A
general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b6b6b: 0000 [#1 ] PREEMPT SMP NOPTI
CPU: 3 PID: 1142 Comm: kworker/u16:20 Tainted: G U 5.6.0-CI-Patchwork_17226+ #1
Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWI1.R00.2457.A16.1912270059 12/27/2019
Workqueue: events_unbound intel_display_power_put_async_work [i915]
RIP: 0010:__intel_display_power_put_domain+0xa5/0x180 [i915]
Code: 48 85 c0 78 54 44 89 e1 41 bd 01 00 00 00 49 c7 c4 80 44 41 a0 49 d3 e5 eb 0d 48 83 eb 10 48 3b 9d 08 ad 00 00 78 32 48 8b 03 <4c> 85 68 10 74 ea 8b 53 08 85 d2 74 2d 83 ea 01 85 d2 89 53 08 75
RSP: 0018:ffffc9000061fdb0 EFLAGS: 00010206
RAX: 6b6b6b6b6b6b6b6b RBX: ffff8884948f5df0 RCX: 000000000000003d
RDX: 0000000080000001 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffff888479be0000 R08: ffff88849a180920 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffffffa0414480
R13: 2000000000000000 R14: ffff888479beb320 R15: 2000000000000000
FS: 0000000000000000(0000) GS:ffff88849ff80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005634fa8ed670 CR3: 0000000005610004 CR4: 0000000000760ee0
PKRU: 55555554
Call Trace:
release_async_put_domains+0x9b/0x110 [i915]
intel_display_power_put_async_work+0x91/0xf0 [i915]
process_one_work+0x260/0x600
? worker_thread+0xc9/0x380
worker_thread+0x37/0x380
? process_one_work+0x600/0x600
kthread+0x119/0x130
? kthread_park+0x80/0x80
ret_from_fork+0x24/0x50
Modules linked in: i915(+) vgem snd_hda_codec_hdmi mei_hdcp x86_pkg_temp_thermal coretemp crct10dif_pclmul crc32_pclmul cdc_ether usbnet mii snd_intel_dspcfg ghash_clmulni_intel snd_hda_codec snd_hwdep snd_hda_core e1000e ptp mei_me snd_pcm pps_core mei intel_lpss_pci prime_numbers [last unloaded: i915]
---[ end trace b402d1b4060f8b97 ]---
BUG: sleeping function called from invalid context at kernel/sched/completion.c:99
in_atomic(): 0, irqs_disabled(): 0, non_block: 0, pid: 1142, name: kworker/u16:20
INFO: lockdep is turned off.
Preemption disabled at:
[<0000000000000000>] 0x0
CPU: 3 PID: 1142 Comm: kworker/u16:20 Tainted: G UD 5.6.0-CI-Patchwork_17226+ #1
Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWI1.R00.2457.A16.1912270059 12/27/2019
Workqueue: events_unbound intel_display_power_put_async_work [i915]
Call Trace:
dump_stack+0x71/0x9b
___might_sleep+0x178/0x260
wait_for_completion+0x37/0x1a0
virt_efi_query_variable_info+0x161/0x1b0
efi_query_variable_store+0xb3/0x1a0
? efivar_entry_set_safe+0x19c/0x220
efivar_entry_set_safe+0x19c/0x220
? efi_pstore_write+0x10b/0x150
? efi_pstore_write+0xa0/0x150
efi_pstore_write+0x10b/0x150
pstore_dump+0x123/0x340
kmsg_dump+0x87/0x1b0
oops_end+0x3e/0x90
do_general_protection+0x1c3/0x2f0
general_protection+0x2d/0x40
RIP: 0010:__intel_display_power_put_domain+0xa5/0x180 [i915]
Code: 48 85 c0 78 54 44 89 e1 41 bd 01 00 00 00 49 c7 c4 80 44 41 a0 49 d3 e5 eb 0d 48 83 eb 10 48 3b 9d 08 ad 00 00 78 32 48 8b 03 <4c> 85 68 10 74 ea 8b 53 08 85 d2 74 2d 83 ea 01 85 d2 89 53 08 75
RSP: 0018:ffffc9000061fdb0 EFLAGS: 00010206
RAX: 6b6b6b6b6b6b6b6b RBX: ffff8884948f5df0 RCX: 000000000000003d
RDX: 0000000080000001 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffff888479be0000 R08: ffff88849a180920 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffffffa0414480
R13: 2000000000000000 R14: ffff888479beb320 R15: 2000000000000000
release_async_put_domains+0x9b/0x110 [i915]
intel_display_power_put_async_work+0x91/0xf0 [i915]
process_one_work+0x260/0x600
? worker_thread+0xc9/0x380
worker_thread+0x37/0x380
? process_one_work+0x600/0x600
kthread+0x119/0x130
? kthread_park+0x80/0x80
ret_from_fork+0x24/0x50
------------[ cut here ]------------
WARNING: CPU: 3 PID: 1142 at kernel/rcu/tree_plugin.h:293 rcu_note_context_switch+0x87/0x650
Modules linked in: i915(+) vgem snd_hda_codec_hdmi mei_hdcp x86_pkg_temp_thermal coretemp crct10dif_pclmul crc32_pclmul cdc_ether usbnet mii snd_intel_dspcfg ghash_clmulni_intel snd_hda_codec snd_hwdep snd_hda_core e1000e ptp mei_me snd_pcm pps_core mei intel_lpss_pci prime_numbers [last unloaded: i915]
v2:
- fixed handling in case of failure in drm_vblank_init()
- moved i915_gem_driver_remove() call to before
i915_driver_modeset_remove_noirq() this match initialization order too
v3:
- reverting call swap between i915_reset_error_state() and i915_gem_driver_remove()
call order
- improved label naming in i915_driver_modeset_probe_noirq()
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1647
Cc: Imre Deak <imre.deak@intel.com >
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com >
Cc: Jani Nikula <jani.nikula@intel.com >
Reviewed-by: Imre Deak <imre.deak@intel.com >
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Link: https://patchwork.freedesktop.org/patch/msgid/20200416185841.125686-1-jose.souza@intel.com
2020-04-17 22:30:19 +01:00
Radhakrishna Sripada
c4310defd8
drm/i915/icl: Update forcewake firmware ranges
...
Some workarounds are not sticking across suspend resume cycles. The
forcewake ranges table has been updated and would reflect the hardware
appropriately.
Closes: https://gitlab.freedesktop.org/drm/intel/issues/1222
v2: Simplify the table and use 0 for some unused ranges(Matt)
Cc: Matt Roper <matthew.d.roper@intel.com >
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com >
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com >
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Link: https://patchwork.freedesktop.org/patch/msgid/20200416164610.15422-1-radhakrishna.sripada@intel.com
2020-04-17 19:35:43 +01:00
Chris Wilson
c43dd6b414
drm/i915/selftests: Check power consumption at min/max frequencies
...
A basic premise of RPS is that at lower frequencies, not only do we run
slower, but we save power compared to higher frequencies. For example,
when idle, we set the minimum frequency just in case there is some
residual current. Since the power curve should be a physical
relationship, if we find no power saving it's likely that we've broken
our frequency handling, so test!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com >
Cc: Andi Shyti <andi.shyti@intel.com >
Reviewed-by: Andi Shyti <andi.shyti@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417152018.13079-2-chris@chris-wilson.co.uk
2020-04-17 18:48:51 +01:00
Chris Wilson
d4e3d455a1
drm/i915/selftests: Move gpu energy measurement into its own little lib
...
Move the handy utility to measure the GPU energy consumption using RAPL
msr into a common lib so that it can be reused easily.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Andi Shyti <andi.shyti@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417152018.13079-1-chris@chris-wilson.co.uk
2020-04-17 18:48:51 +01:00
Stanislav Lisovskiy
680e1af713
drm/i915: Add pre/post plane updates for SAGV
...
Lets have a unified way to handle SAGV changes,
espoecially considering the upcoming Gen12 changes.
Current "standard" way of doing this in commit_tail
is pre/post plane updates, when everything which
has to be forbidden and not supported in new config
has to be restricted before update and relaxed after
plane update.
v2: - Removed unneeded returns(Ville)
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com >
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415143911.10244-5-stanislav.lisovskiy@intel.com
2020-04-17 20:41:00 +03:00
Stanislav Lisovskiy
a389c49fac
drm/i915: Prepare to extract gen specific functions from intel_can_enable_sagv
...
Addressing one of the comments, recommending to extract platform
specific code from intel_can_enable_sagv as a preparation, before
we are going to add support for tgl+.
v2: - Removed whitespace
v3: - Removed premature debug and new cycle introduction(Ville)
- Added missing no active pipes check(Ville)
v4: - Fixed stupid mistake with plane_state caused by stupid macro change
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com >
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415145740.28241-1-stanislav.lisovskiy@intel.com
2020-04-17 20:41:00 +03:00
Stanislav Lisovskiy
442e7ee834
drm/i915: Add intel_atomic_get_bw_*_state helpers
...
Add correspondent helpers to be able to get old/new bandwidth
global state object.
v2: - Fixed typo in function call
v3: - Changed new functions naming to use convention proposed
by Jani Nikula, i.e intel_bw_* in intel_bw.c file.
v4: - Change function naming back to intel_atomic* pattern,
was decided to rename in a separate patch series.
v5: - Fix function naming to match existing practices(Ville)
v6: - Removed spurious whitespace
v7: - Removed bw_state NULL checks(Ville)
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com >
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415143911.10244-3-stanislav.lisovskiy@intel.com
2020-04-17 20:41:00 +03:00
Linus Walleij
ac2caae61e
drm/tegra: Clean up GPIO includes
...
The Tegra DRM drivers includes the legacy GPIO headers
<linux/gpio.h> and <linux/of_gpio.h> but what it really
uses is <linux/gpio/consumer.h> since only gpio_desc
structs are ever referenced.
Include the right header on the top level tegra/drm.h
file and drop all the surplus includes.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org >
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch >
Link: https://patchwork.freedesktop.org/patch/msgid/20200415122427.111769-1-linus.walleij@linaro.org
2020-04-17 17:16:22 +02:00
Chris Wilson
a50717dbf4
drm/i915/selftests: Take the engine wakeref around __rps_up_interrupt
...
Since we are touching the device to read the registers, we are required
to ensure the device is awake at the time. Currently, we believe
ourselves to be inside the active request [thus an active engine
wakeref], but since that may be retired in the background, we can
spontaneously lose the wakeref and the ability to probe the HW.
<4> [379.686703] RPM wakelock ref not held during HW access
<4> [379.686805] WARNING: CPU: 7 PID: 4869 at ./drivers/gpu/drm/i915/intel_runtime_pm.h:115 gen12_fwtable_read32+0x233/0x300 [i915]
<4> [379.686808] Modules linked in: i915(+) vgem snd_hda_codec_hdmi mei_hdcp x86_pkg_temp_thermal coretemp crct10dif_pclmul crc32_pclmul ax88179_178a usbnet mii ghash_clmulni_intel snd_intel_dspcfg snd_hda_codec snd_hwdep snd_hda_core snd_pcm e1000e mei_me ptp mei pps_core intel_lpss_pci prime_numbers [last unloaded: i915]
<4> [379.686827] CPU: 7 PID: 4869 Comm: i915_selftest Tainted: G U 5.7.0-rc1-CI-CI_DRM_8313+ #1
<4> [379.686830] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWI1.R00.2457.A13.1912190237 12/19/2019
<4> [379.686883] RIP: 0010:gen12_fwtable_read32+0x233/0x300 [i915]
<4> [379.686887] Code: d8 ea e0 0f 0b e9 19 fe ff ff 80 3d ad 12 2d 00 00 0f 85 17 fe ff ff 48 c7 c7 b0 32 3e a0 c6 05 99 12 2d 00 01 e8 2d d8 ea e0 <0f> 0b e9 fd fd ff ff 8b 05 c4 75 56 e2 85 c0 0f 85 84 00 00 00 48
<4> [379.686889] RSP: 0018:ffffc90000727970 EFLAGS: 00010286
<4> [379.686892] RAX: 0000000000000000 RBX: ffff88848cc20ee8 RCX: 0000000000000001
<4> [379.686894] RDX: 0000000080000001 RSI: ffff88843b1f0900 RDI: 00000000ffffffff
<4> [379.686896] RBP: 0000000000000000 R08: ffff88843b1f0900 R09: 0000000000000000
<4> [379.686898] R10: 0000000000000000 R11: 0000000000000000 R12: 000000000000a058
<4> [379.686900] R13: 0000000000000001 R14: ffff88848cc2bf30 R15: 00000000ffffffea
<4> [379.686902] FS: 00007f7d63f5e300(0000) GS:ffff8884a0180000(0000) knlGS:0000000000000000
<4> [379.686904] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<4> [379.686907] CR2: 000055e5c30f4988 CR3: 000000042e190002 CR4: 0000000000760ee0
<4> [379.686910] PKRU: 55555554
<4> [379.686911] Call Trace:
<4> [379.686986] live_rps_interrupt+0xb14/0xc10 [i915]
<4> [379.687051] ? intel_rps_unpark+0xb0/0xb0 [i915]
<4> [379.687057] ? __trace_bprintk+0x57/0x80
<4> [379.687143] __i915_subtests+0xb8/0x210 [i915]
<4> [379.687222] ? __i915_live_teardown+0x50/0x50 [i915]
<4> [379.687291] ? __intel_gt_live_setup+0x30/0x30 [i915]
<4> [379.687361] __run_selftests+0x112/0x170 [i915]
<4> [379.687431] i915_live_selftests+0x2c/0x60 [i915]
<4> [379.687491] i915_pci_probe+0x93/0x1b0 [i915]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Andi Shyti <andi.shyti@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417093928.17822-2-chris@chris-wilson.co.uk
2020-04-17 14:47:30 +01:00
Chris Wilson
9d7e560f43
drm/i915/selftests: Delay spinner before waiting for an interrupt
...
It seems that although (perhaps because of the memory stall?) the
spinner has signaled that it has started, it still takes some time to
spin up to 100% utilisation of the HW. Since the test depends on the
full utilisation of the HW to trigger the RPS interrupt, wait a little
bit and flush the interrupt status to be sure that the event we see if
from the spinner.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk >
Reviewed-by: Andi Shyti <andi.shyti@intel.com >
Link: https://patchwork.freedesktop.org/patch/msgid/20200417093928.17822-1-chris@chris-wilson.co.uk
2020-04-17 14:47:15 +01:00
