We need to re-init the display hardware when going out of suspend. This
includes:
- Hooking the PCH to the reset logic
- Restoring CDCDLK
- Enabling the DDB power
Among those, only the CDCDLK one is a bit tricky. There's some
complexity in that:
- DPLL0 (which is the source for CDCLK) has two VCOs, each with a set
of supported frequencies. As eDP also uses DPLL0 for its link rate,
once DPLL0 is on, we restrict the possible eDP link rates the chosen
VCO.
- CDCLK also limits the bandwidth available to push pixels.
So, as a first step, this commit restore what the BIOS set, until I can
do more testing.
In case that's of interest for the reviewer, I've unit tested the
function that derives the decimal frequency field:
#include <stdio.h>
#include <stdint.h>
#include <assert.h>
#define ARRAY_SIZE(x) (sizeof(x) / sizeof(*(x)))
static const struct dpll_freq {
unsigned int freq;
unsigned int decimal;
} freqs[] = {
{ .freq = 308570, .decimal = 0b01001100111},
{ .freq = 337500, .decimal = 0b01010100001},
{ .freq = 432000, .decimal = 0b01101011110},
{ .freq = 450000, .decimal = 0b01110000010},
{ .freq = 540000, .decimal = 0b10000110110},
{ .freq = 617140, .decimal = 0b10011010000},
{ .freq = 675000, .decimal = 0b10101000100},
};
static void intbits(unsigned int v)
{
int i;
for(i = 10; i >= 0; i--)
putchar('0' + ((v >> i) & 1));
}
static unsigned int freq_decimal(unsigned int freq /* in kHz */)
{
return (freq - 1000) / 500;
}
static void test_freq(const struct dpll_freq *entry)
{
unsigned int decimal = freq_decimal(entry->freq);
printf("freq: %d, expected: ", entry->freq);
intbits(entry->decimal);
printf(", got: ");
intbits(decimal);
putchar('\n');
assert(decimal == entry->decimal);
}
int main(int argc, char **argv)
{
int i;
for (i = 0; i < ARRAY_SIZE(freqs); i++)
test_freq(&freqs[i]);
return 0;
}
v2:
- Rebase on top of -nightly
- Use (freq - 1000) / 500 for the decimal frequency (Ville)
- Fix setting the enable bit of HSW_NDE_RSTWRN_OPT (Ville)
- Rename skl_display_{resume,suspend} to skl_{init,uninit}_cdclk to
be consistent with the BXT code (Ville)
- Store boot CDCLK in ddi_pll_init (Ville)
- Merge dev_priv's skl_boot_cdclk into cdclk_freq
- Use LCPLL_PLL_LOCK instead of (1 << 30) (Ville)
- Replace various '0' by SKL_DPLL0 to be a bit more explicit that
we're programming DPLL0
- Busy poll the PCU before doing the frequency change. It takes about
3/4 cycles, each separated by 10us, to get the ACK from the CPU
(Ville)
v3:
- Restore dev_priv->skl_boot_cdclk, leaving unification with
dev_priv->cdclk_freq for a later patch (Daniel, Ville)
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
If the client stalls on a congested request, chosen to be 20ms old to
match throttling, allow the client a free RPS boost.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: s/rq/req/]
[danvet: s/0/NULL/ reported by 0-day build]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
If we have clients stalled waiting for requests, ignore the GPU if it
signals that it should downclock due to low load. This helps prevent
the automatic timeout from causing extremely long running batches from
taking even longer.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Now that we have internal clients, rather than faking a whole
drm_i915_file_private just for tracking RPS boosts, create a new struct
intel_rps_client and pass it along when waiting.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: s/rq/req/]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since we will often pageflip to an active surface, we will often have to
wait for the surface to be written before issuing the flip. Also we are
likely to wait on that surface in plenty of time before the vblank.
Since we have a mechanism for boosting when a flip misses the expected
vblank, curtain the number of times we RPS boost when simply waiting for
mmioflip.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: s/rq/req/]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Ring switches can occur many times per frame, and are often out of
control, causing frequent RPS boosting for no practical benefit. Treat
the sw semaphore synchronisation as a separate client and only allow it
to boost once per busy/idle cycle.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: s/rq/req/]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
This trims a little overhead from the common case of not needing to
synchronize between rings.
v2: execlists is special and likes to duplicate code.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Currently, we only track the last request globally across all engines.
This prevents us from issuing concurrent read requests on e.g. the RCS
and BCS engines (or more likely the render and media engines). Without
semaphores, we incur costly stalls as we synchronise between rings -
greatly impacting the current performance of Broadwell versus Haswell in
certain workloads (like video decode). With the introduction of
reference counted requests, it is much easier to track the last request
per ring, as well as the last global write request so that we can
optimise inter-engine read read requests (as well as better optimise
certain CPU waits).
v2: Fix inverted readonly condition for nonblocking waits.
v3: Handle non-continguous engine array after waits
v4: Rebase, tidy, rewrite ring list debugging
v5: Use obj->active as a bitfield, it looks cool
v6: Micro-optimise, mostly involving moving code around
v7: Fix retire-requests-upto for execlists (and multiple rq->ringbuf)
v8: Rebase
v9: Refactor i915_gem_object_sync() to allow the compiler to better
optimise it.
Benchmark: igt/gem_read_read_speed
hsw:gt3e (with semaphores):
Before: Time to read-read 1024k: 275.794µs
After: Time to read-read 1024k: 123.260µs
hsw:gt3e (w/o semaphores):
Before: Time to read-read 1024k: 230.433µs
After: Time to read-read 1024k: 124.593µs
bdw-u (w/o semaphores): Before After
Time to read-read 1x1: 26.274µs 10.350µs
Time to read-read 128x128: 40.097µs 21.366µs
Time to read-read 256x256: 77.087µs 42.608µs
Time to read-read 512x512: 281.999µs 181.155µs
Time to read-read 1024x1024: 1196.141µs 1118.223µs
Time to read-read 2048x2048: 5639.072µs 5225.837µs
Time to read-read 4096x4096: 22401.662µs 21137.067µs
Time to read-read 8192x8192: 89617.735µs 85637.681µs
Testcase: igt/gem_concurrent_blit (read-read and friends)
Cc: Lionel Landwerlin <lionel.g.landwerlin@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> [v8]
[danvet: s/\<rq\>/req/g]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The merged seqno->request conversion from John called request
variables req, but some (not all) of Chris' recent patches changed
those to just rq. We've had a lenghty (and inconclusive) discussion on
irc which is the more meaningful name with maybe at most a slight bias
towards req.
Given that the "don't change names without good reason to avoid
conflicts" rule applies, so lets go back to a req everywhere for
consistency. I'll sed any patches for which this will cause conflicts
before applying.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Harrison <John.C.Harrison@Intel.com>
[danvet: s/origina/merged/ as pointed out by Chris - the first
mass-conversion patch was from Chris, the merged one from John.]
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
v2:
- set the override disable flag too on stepping F0 (mika)
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
On B0 and C0 steppings the workaround enable bit would be overriden by
default, so the overriding must be disabled.
The WA was added in
commit 83a24979c4
Author: Nick Hoath <nicholas.hoath@intel.com>
Date: Fri Apr 10 13:12:26 2015 +0100
drm/i915/bxt: Add WaForceContextSaveRestoreNonCoherent
Spotted-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Our driver compiles clean (nowadays thanks to 0day) but for me, at least,
it would be beneficial if the compiler threw an error rather than a
warning when it found a piece of suspect code. (I use this to
compile-check patch series and want to break on the first compiler error
in order to fix the patch.)
v2: Kick off a new "Debugging" submenu for i915.ko
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Jani Nikula <jani.nikula@intel.com>
[danvet: Add "DRM i915" to the menu name as requested by Chris.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
The macros we use there are the magic ones that can take either dev or
dev_priv. We'd like to move as much as possible towards dev_priv though.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Currently bxt_resume_prepare() is only used in the runtime-resume path.
Add it to the full S3/S4 path as well.
v2: Rebase on top of the vlv_resume_prepare() shuffling around
Cc: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com> (v1)
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Since DRM_ROTATE is counter clockwise (which is compliant with Xrandr),
and HW rotation is clockwise, swapping 90/270 to work as expected from
userspace.
v2: Rebased
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Explain why a few fields of the new pipe_config have their values
preserved, while the others are zeroed.
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Also make the WA comment consistent with the rest, where the stepping
info is not shown.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
ARGB8888 is used for cursors on all platforms so we need to allow it
everywhere.
ABGR8888 is currently only honoured:
- on VLV/CHV in sprite planes
- on SKL+ for primary and sprite planes
so only allow it for those platforms.
Note that we only support ARGB8888/ABGR8888 on the primary plane for
SKL/BXT because we have in line of sight the pipe bottom color on those
platforms and because the primary plane programming on VLV/CHV doesn't
anything different for those formats today.
v2: Fix the logic to forbid the creation ABGR2101010 fbs (Ville)
v3: Still allow the creation of ARGB8888 fbs now that cursor planes use
real fb objects (found by PRTS).
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Making lane stagger calculation common for HDMI and DP
v2: Imre's comments addressed
- Remove lane stagger from bxt_clk_div and make it a local variable in
ddi_pll_select
Signed-off-by: Vandana Kannan <vandana.kannan@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
BUN 1: prop_coeff, int_coeff, tdctargetcnt programming updated and tied to
VCO frequencies. Program i_lockthresh in PORT_PLL_9.
VCO calculated based on the formula:
Desired Output = Port bit rate in MHz (DisplayPort HBR2 is 5400 MHz)
Fast Clock = Desired Output / 2
VCO = Fast Clock * P1 * P2
Prop_coeff, int_coeff, and tdctargetcnt modified according to above
calculation.
BUN 2: Port PLLs require additional programming at certain frequencies -
DCO amplitude in PORT_PLL_10
Review comments from Siva which were addressed in the initial version of the
patch.
- Change PORT_PLL_LOCK_THRESHOLD to PORT_PLL_LOCK_THRESHOLD_MASK
- Calculate for HDMI
- Correct values for vco = 5.4
- return in case of invalid vco range
v2: Imre's review comments addressed
- change dcoampovr_en to dcoampovr_en_h
- change PORT_PLL_DCO_AMP_OVR_EN to PORT_PLL_DCO_AMP_OVR_EN_H
- Correct lane stagger value for 324MHz
- Make coef common for HDMI and DP
- remove superfluous comments
v3: Imre's comments addressed
- Remove Prop_coeff, int_coeff, tdctargetcnt, dcoampovr_en, gain_ctl,
dcoampovr_en_h from bxt_clk_div and make them local variables.
Signed-off-by: Vandana Kannan <vandana.kannan@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com> [v1]
Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Be in line with other features that we have.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We just have have VLV and CHV sprites programming the hardware
differently for the ABGR2101010 so keep them working.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
That define makes it hard to figure out what is the actual list of
formats at a glance. Expand it then.
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Mika encountered one pathological scenario under X where acquiring all
the mm locks (required to insert a mmu notifier) was very slow, so slow
that by the time we tried to lock the struct_mutex with the usual call
to i915_mutex_lock_interruptible(), X's signal timer had fired causing
us to restart the ioctl (and so looped indefinitely).
While I suspect this is the result of another bug (something leaking mm
perhaps?) we can forgo the error checking and interuptible nature of the
lock here so we only have to pay the expense once and get on with it.
This does expose the userptr creation routine to a driver livelock
though by not being interruptible.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
[danvet: Init ret to avoid issues reported by PRTS.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
When the modeset code is reached with a CRTC that only needs a flip, the
code that assigns PLLs is skipped. But since there is still a state swap
for that CRTC, the current PLL assignment needs to be preserved. I
missed the ddi_pll_sel field in the following commit, which causes
warnings in DDI platforms.
commit 4978cc93d9
Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Date: Tue Apr 21 17:13:21 2015 +0300
drm/i915: Preserve shared DPLL information in new pipe_config
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90410
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
In the following commit, the place where the contents of dpll_hw_state
in crtc_state where zeroed was changed. Prior to that commit, it
happened when the new state was allocated, but now that happens just
before the call the .crtc_compute_clock() hook. The DP code for SKL,
however, sets up the (private) PLL in the encoder compute config
function that has already run by the time that memset() is reached,
causing the previous value to be lost.
This patch fixes the issue by moving the memset() down the call chain,
so that it is only called if the values in dpll_hw_state are going to be
updated.
commit 4978cc93d9
Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Date: Tue Apr 21 17:13:21 2015 +0300
drm/i915: Preserve shared DPLL information in new pipe_config
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90462
Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Reported-and-tested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Add one explicit discard of __iomem address space qualifier in
validate_vbt(), and respect it otherwise. This adds clarity in the code,
and reduces the sparse warnings from the module to just one.
Quoting Daniel, "The vbt really is plain old memory. Except that it's
reserved in the e820 table as something special and hence treated as io
range by the kernel. But it is memory, hence casting away the __iomem is
imo the right approach."
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Just so it is grouped logically in line with other data and makes a
rather verbose output a bit shorter.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Chandra Konduru <chandra.konduru@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Chandra Konduru <chandra.konduru@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We never pass a non-NULL vbt to validate_vbt, and we can safely expect
the callers to not change.
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Ville noticed in another patch we we didn't need them at all, so remove
them. It's worth saying that it makes no difference to code generated as
gcc is clever enough to optimize it out.
v2: Remove 'break' after 'return' in switches (Ville)
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Currently DSI PLL N1 is hardcoded off. Make it possible to use it
later. This should have no functional changes for now.
v2: s/ffz(~(n))/ffs(n) - 1/ (Ville)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Added docbook info regarding context save and restore (CSR)
firmware support added from gen9 onwards to drive newly added
DMC (Display microcontroller) in display engine.
v1: Initial version as RFC.
v2: Used "DOC:" tag for csr description based on review comment from Daniel.
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-by: A.Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reading from disconnected ports will spit out timeout error
on the dmesg. Skip the attempted read if the port is not
connected and avoid confusing users/testcases about
expected timeouts.
This new dpcd debugfs entry was introduced by commit aa7471d228
("drm/i915: add i915 specific connector debugfs file for DPCD")
v2 by Jani: move the check at the top, out of the loop.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90060
Tested-by: yex.tian@intel.com
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Bob Paauwe <bob.j.paauwe@intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
There is an extra semi-colon on the if statement so the debug output
always says "Failed to write EDID checksum" even when it didn't fail.
Fixes: 559be30cb7 ('drm/i915: Implement the intel_dp_autotest_edid function for DP EDID complaince tests')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We now prefix our functions/enums/data with the first platform it has
been introduced. Do that for the primary plane formats.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
[danvet: s/gen2/i8xx/ and s/gen4/i965/ ...]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We advertize C8 in the primary plane formats didn't have the
corresponding code to set PLANE_CTL accordingly.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Let's be consistent with the others skl_plane_ctl_*() functions and use
a MISSING_CASE(). Not only that, but it's a rude to BUG() the whole
machine here.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
No reason to not follow the 80 chars rule, renaming the local variable
makes it easy.
Cc: Chandra Konduru <chandra.konduru@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
We usually use a new line before those kind of return statements. Also
the various skl_plane_ctl*() functions weren't consistent.
Cc: Chandra Konduru <chandra.konduru@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Improve readability. No functional changes.
v2: use more rational types (Ville)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau