linux_dsm_epyc7002/drivers/gpu/drm/i915/display/intel_psr.c

1681 lines
49 KiB
C
Raw Normal View History

/*
* Copyright © 2014 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*/
#include <drm/drm_atomic_helper.h>
#include "display/intel_dp.h"
#include "i915_drv.h"
#include "intel_atomic.h"
#include "intel_display_types.h"
#include "intel_psr.h"
#include "intel_sprite.h"
#include "intel_hdmi.h"
/**
* DOC: Panel Self Refresh (PSR/SRD)
*
* Since Haswell Display controller supports Panel Self-Refresh on display
* panels witch have a remote frame buffer (RFB) implemented according to PSR
* spec in eDP1.3. PSR feature allows the display to go to lower standby states
* when system is idle but display is on as it eliminates display refresh
* request to DDR memory completely as long as the frame buffer for that
* display is unchanged.
*
* Panel Self Refresh must be supported by both Hardware (source) and
* Panel (sink).
*
* PSR saves power by caching the framebuffer in the panel RFB, which allows us
* to power down the link and memory controller. For DSI panels the same idea
* is called "manual mode".
*
* The implementation uses the hardware-based PSR support which automatically
* enters/exits self-refresh mode. The hardware takes care of sending the
* required DP aux message and could even retrain the link (that part isn't
* enabled yet though). The hardware also keeps track of any frontbuffer
* changes to know when to exit self-refresh mode again. Unfortunately that
* part doesn't work too well, hence why the i915 PSR support uses the
* software frontbuffer tracking to make sure it doesn't miss a screen
* update. For this integration intel_psr_invalidate() and intel_psr_flush()
* get called by the frontbuffer tracking code. Note that because of locking
* issues the self-refresh re-enable code is done from a work queue, which
* must be correctly synchronized/cancelled when shutting down the pipe."
*
* DC3CO (DC3 clock off)
*
* On top of PSR2, GEN12 adds a intermediate power savings state that turns
* clock off automatically during PSR2 idle state.
* The smaller overhead of DC3co entry/exit vs. the overhead of PSR2 deep sleep
* entry/exit allows the HW to enter a low-power state even when page flipping
* periodically (for instance a 30fps video playback scenario).
*
* Every time a flips occurs PSR2 will get out of deep sleep state(if it was),
* so DC3CO is enabled and tgl_dc3co_disable_work is schedule to run after 6
* frames, if no other flip occurs and the function above is executed, DC3CO is
* disabled and PSR2 is configured to enter deep sleep, resetting again in case
* of another flip.
* Front buffer modifications do not trigger DC3CO activation on purpose as it
* would bring a lot of complexity and most of the moderns systems will only
* use page flips.
*/
static bool psr_global_enabled(struct drm_i915_private *i915)
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
{
switch (i915->psr.debug & I915_PSR_DEBUG_MODE_MASK) {
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
case I915_PSR_DEBUG_DEFAULT:
return i915_modparams.enable_psr;
case I915_PSR_DEBUG_DISABLE:
return false;
default:
return true;
}
}
static bool intel_psr2_enabled(struct drm_i915_private *dev_priv,
const struct intel_crtc_state *crtc_state)
{
/* Cannot enable DSC and PSR2 simultaneously */
drm_WARN_ON(&dev_priv->drm, crtc_state->dsc.compression_enable &&
crtc_state->has_psr2);
switch (dev_priv->psr.debug & I915_PSR_DEBUG_MODE_MASK) {
case I915_PSR_DEBUG_DISABLE:
case I915_PSR_DEBUG_FORCE_PSR1:
return false;
default:
return crtc_state->has_psr2;
}
}
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
static void psr_irq_control(struct drm_i915_private *dev_priv)
{
enum transcoder trans_shift;
u32 mask, val;
i915_reg_t imr_reg;
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
/*
* gen12+ has registers relative to transcoder and one per transcoder
* using the same bit definition: handle it as TRANSCODER_EDP to force
* 0 shift in bit definition
*/
if (INTEL_GEN(dev_priv) >= 12) {
trans_shift = 0;
imr_reg = TRANS_PSR_IMR(dev_priv->psr.transcoder);
} else {
trans_shift = dev_priv->psr.transcoder;
imr_reg = EDP_PSR_IMR;
}
mask = EDP_PSR_ERROR(trans_shift);
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
if (dev_priv->psr.debug & I915_PSR_DEBUG_IRQ)
mask |= EDP_PSR_POST_EXIT(trans_shift) |
EDP_PSR_PRE_ENTRY(trans_shift);
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
/* Warning: it is masking/setting reserved bits too */
val = intel_de_read(dev_priv, imr_reg);
val &= ~EDP_PSR_TRANS_MASK(trans_shift);
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
val |= ~mask;
intel_de_write(dev_priv, imr_reg, val);
}
static void psr_event_print(struct drm_i915_private *i915,
u32 val, bool psr2_enabled)
{
drm_dbg_kms(&i915->drm, "PSR exit events: 0x%x\n", val);
if (val & PSR_EVENT_PSR2_WD_TIMER_EXPIRE)
drm_dbg_kms(&i915->drm, "\tPSR2 watchdog timer expired\n");
if ((val & PSR_EVENT_PSR2_DISABLED) && psr2_enabled)
drm_dbg_kms(&i915->drm, "\tPSR2 disabled\n");
if (val & PSR_EVENT_SU_DIRTY_FIFO_UNDERRUN)
drm_dbg_kms(&i915->drm, "\tSU dirty FIFO underrun\n");
if (val & PSR_EVENT_SU_CRC_FIFO_UNDERRUN)
drm_dbg_kms(&i915->drm, "\tSU CRC FIFO underrun\n");
if (val & PSR_EVENT_GRAPHICS_RESET)
drm_dbg_kms(&i915->drm, "\tGraphics reset\n");
if (val & PSR_EVENT_PCH_INTERRUPT)
drm_dbg_kms(&i915->drm, "\tPCH interrupt\n");
if (val & PSR_EVENT_MEMORY_UP)
drm_dbg_kms(&i915->drm, "\tMemory up\n");
if (val & PSR_EVENT_FRONT_BUFFER_MODIFY)
drm_dbg_kms(&i915->drm, "\tFront buffer modification\n");
if (val & PSR_EVENT_WD_TIMER_EXPIRE)
drm_dbg_kms(&i915->drm, "\tPSR watchdog timer expired\n");
if (val & PSR_EVENT_PIPE_REGISTERS_UPDATE)
drm_dbg_kms(&i915->drm, "\tPIPE registers updated\n");
if (val & PSR_EVENT_REGISTER_UPDATE)
drm_dbg_kms(&i915->drm, "\tRegister updated\n");
if (val & PSR_EVENT_HDCP_ENABLE)
drm_dbg_kms(&i915->drm, "\tHDCP enabled\n");
if (val & PSR_EVENT_KVMR_SESSION_ENABLE)
drm_dbg_kms(&i915->drm, "\tKVMR session enabled\n");
if (val & PSR_EVENT_VBI_ENABLE)
drm_dbg_kms(&i915->drm, "\tVBI enabled\n");
if (val & PSR_EVENT_LPSP_MODE_EXIT)
drm_dbg_kms(&i915->drm, "\tLPSP mode exited\n");
if ((val & PSR_EVENT_PSR_DISABLE) && !psr2_enabled)
drm_dbg_kms(&i915->drm, "\tPSR disabled\n");
}
void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir)
{
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
enum transcoder cpu_transcoder = dev_priv->psr.transcoder;
enum transcoder trans_shift;
i915_reg_t imr_reg;
ktime_t time_ns = ktime_get();
if (INTEL_GEN(dev_priv) >= 12) {
trans_shift = 0;
imr_reg = TRANS_PSR_IMR(dev_priv->psr.transcoder);
} else {
trans_shift = dev_priv->psr.transcoder;
imr_reg = EDP_PSR_IMR;
}
if (psr_iir & EDP_PSR_PRE_ENTRY(trans_shift)) {
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
dev_priv->psr.last_entry_attempt = time_ns;
drm_dbg_kms(&dev_priv->drm,
"[transcoder %s] PSR entry attempt in 2 vblanks\n",
transcoder_name(cpu_transcoder));
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
}
if (psr_iir & EDP_PSR_POST_EXIT(trans_shift)) {
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
dev_priv->psr.last_exit = time_ns;
drm_dbg_kms(&dev_priv->drm,
"[transcoder %s] PSR exit completed\n",
transcoder_name(cpu_transcoder));
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
if (INTEL_GEN(dev_priv) >= 9) {
u32 val = intel_de_read(dev_priv,
PSR_EVENT(cpu_transcoder));
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
bool psr2_enabled = dev_priv->psr.psr2_enabled;
intel_de_write(dev_priv, PSR_EVENT(cpu_transcoder),
val);
psr_event_print(dev_priv, val, psr2_enabled);
}
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
}
if (psr_iir & EDP_PSR_ERROR(trans_shift)) {
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
u32 val;
drm_warn(&dev_priv->drm, "[transcoder %s] PSR aux error\n",
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
transcoder_name(cpu_transcoder));
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
dev_priv->psr.irq_aux_error = true;
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
/*
* If this interruption is not masked it will keep
* interrupting so fast that it prevents the scheduled
* work to run.
* Also after a PSR error, we don't want to arm PSR
* again so we don't care about unmask the interruption
* or unset irq_aux_error.
*/
val = intel_de_read(dev_priv, imr_reg);
val |= EDP_PSR_ERROR(trans_shift);
intel_de_write(dev_priv, imr_reg, val);
schedule_work(&dev_priv->psr.work);
}
}
static bool intel_dp_get_alpm_status(struct intel_dp *intel_dp)
{
u8 alpm_caps = 0;
if (drm_dp_dpcd_readb(&intel_dp->aux, DP_RECEIVER_ALPM_CAP,
&alpm_caps) != 1)
return false;
return alpm_caps & DP_ALPM_CAP;
}
static u8 intel_dp_get_sink_sync_latency(struct intel_dp *intel_dp)
{
struct drm_i915_private *i915 = dp_to_i915(intel_dp);
u8 val = 8; /* assume the worst if we can't read the value */
if (drm_dp_dpcd_readb(&intel_dp->aux,
DP_SYNCHRONIZATION_LATENCY_IN_SINK, &val) == 1)
val &= DP_MAX_RESYNC_FRAME_COUNT_MASK;
else
drm_dbg_kms(&i915->drm,
"Unable to get sink synchronization latency, assuming 8 frames\n");
return val;
}
static u16 intel_dp_get_su_x_granulartiy(struct intel_dp *intel_dp)
{
struct drm_i915_private *i915 = dp_to_i915(intel_dp);
u16 val;
ssize_t r;
/*
* Returning the default X granularity if granularity not required or
* if DPCD read fails
*/
if (!(intel_dp->psr_dpcd[1] & DP_PSR2_SU_GRANULARITY_REQUIRED))
return 4;
r = drm_dp_dpcd_read(&intel_dp->aux, DP_PSR2_SU_X_GRANULARITY, &val, 2);
if (r != 2)
drm_dbg_kms(&i915->drm,
"Unable to read DP_PSR2_SU_X_GRANULARITY\n");
/*
* Spec says that if the value read is 0 the default granularity should
* be used instead.
*/
if (r != 2 || val == 0)
val = 4;
return val;
}
void intel_psr_init_dpcd(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv =
to_i915(dp_to_dig_port(intel_dp)->base.base.dev);
if (dev_priv->psr.dp) {
drm_warn(&dev_priv->drm,
"More than one eDP panel found, PSR support should be extended\n");
return;
}
drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT, intel_dp->psr_dpcd,
sizeof(intel_dp->psr_dpcd));
if (!intel_dp->psr_dpcd[0])
return;
drm_dbg_kms(&dev_priv->drm, "eDP panel supports PSR version %x\n",
intel_dp->psr_dpcd[0]);
if (drm_dp_has_quirk(&intel_dp->desc, 0, DP_DPCD_QUIRK_NO_PSR)) {
drm_dbg_kms(&dev_priv->drm,
"PSR support not currently available for this panel\n");
return;
}
if (!(intel_dp->edp_dpcd[1] & DP_EDP_SET_POWER_CAP)) {
drm_dbg_kms(&dev_priv->drm,
"Panel lacks power state control, PSR cannot be enabled\n");
return;
}
dev_priv->psr.sink_support = true;
drm/i915/psr: Set idle frame count based on sink synchronization latency DPCD 2009h "Synchronization latency in sink" has bits that tell us the maximum number of frames sink can take to resynchronize to source timing when exiting PSR. More importantly, as per eDP 1.4b, this is the "Minimum number of frames following PSR exit that the Source device needs to wait for PSR entry." We currently use this value only to setup the number frames to wait before PSR2 selective update. But, based on the above description it makes more sense to use this to configure idle frames for both PSR1 and and PSR2. This will ensure we wait the required number of frames before activation whether it is PSR1 or PSR2. The minimum number of idle frames remains 6, while allowing sink synchronization latency and VBT to increase this value. This also solves the flip-flop between sink and source frames that I noticed on my Thinkpad X260 during PSR exit. This specific panel has a value of 8h, which according to the spec means the "Source device must wait for more than eight active frames after PSR exit before initiating PSR entry. (In this case, should be provided by the panel supplier.)" VBT however has a value of 0. Cc: Jani Nikula <jani.nikula@intel.com> Cc: Jose Roberto de Souza <jose.souza@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180525033047.7596-1-dhinakaran.pandiyan@intel.com
2018-05-25 10:30:47 +07:00
dev_priv->psr.sink_sync_latency =
intel_dp_get_sink_sync_latency(intel_dp);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
dev_priv->psr.dp = intel_dp;
if (INTEL_GEN(dev_priv) >= 9 &&
(intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_IS_SUPPORTED)) {
bool y_req = intel_dp->psr_dpcd[1] &
DP_PSR2_SU_Y_COORDINATE_REQUIRED;
bool alpm = intel_dp_get_alpm_status(intel_dp);
/*
* All panels that supports PSR version 03h (PSR2 +
* Y-coordinate) can handle Y-coordinates in VSC but we are
* only sure that it is going to be used when required by the
* panel. This way panel is capable to do selective update
* without a aux frame sync.
*
* To support PSR version 02h and PSR version 03h without
* Y-coordinate requirement panels we would need to enable
* GTC first.
*/
dev_priv->psr.sink_psr2_support = y_req && alpm;
drm_dbg_kms(&dev_priv->drm, "PSR2 %ssupported\n",
dev_priv->psr.sink_psr2_support ? "" : "not ");
if (dev_priv->psr.sink_psr2_support) {
dev_priv->psr.colorimetry_support =
intel_dp_get_colorimetry_status(intel_dp);
dev_priv->psr.su_x_granularity =
intel_dp_get_su_x_granulartiy(intel_dp);
}
}
}
static void hsw_psr_setup_aux(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 aux_clock_divider, aux_ctl;
int i;
static const u8 aux_msg[] = {
[0] = DP_AUX_NATIVE_WRITE << 4,
[1] = DP_SET_POWER >> 8,
[2] = DP_SET_POWER & 0xff,
[3] = 1 - 1,
[4] = DP_SET_POWER_D0,
};
u32 psr_aux_mask = EDP_PSR_AUX_CTL_TIME_OUT_MASK |
EDP_PSR_AUX_CTL_MESSAGE_SIZE_MASK |
EDP_PSR_AUX_CTL_PRECHARGE_2US_MASK |
EDP_PSR_AUX_CTL_BIT_CLOCK_2X_MASK;
BUILD_BUG_ON(sizeof(aux_msg) > 20);
for (i = 0; i < sizeof(aux_msg); i += 4)
intel_de_write(dev_priv,
EDP_PSR_AUX_DATA(dev_priv->psr.transcoder, i >> 2),
intel_dp_pack_aux(&aux_msg[i], sizeof(aux_msg) - i));
aux_clock_divider = intel_dp->get_aux_clock_divider(intel_dp, 0);
/* Start with bits set for DDI_AUX_CTL register */
aux_ctl = intel_dp->get_aux_send_ctl(intel_dp, sizeof(aux_msg),
aux_clock_divider);
/* Select only valid bits for SRD_AUX_CTL */
aux_ctl &= psr_aux_mask;
intel_de_write(dev_priv, EDP_PSR_AUX_CTL(dev_priv->psr.transcoder),
aux_ctl);
}
static void intel_psr_enable_sink(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u8 dpcd_val = DP_PSR_ENABLE;
/* Enable ALPM at sink for psr2 */
if (dev_priv->psr.psr2_enabled) {
drm_dp_dpcd_writeb(&intel_dp->aux, DP_RECEIVER_ALPM_CONFIG,
DP_ALPM_ENABLE |
DP_ALPM_LOCK_ERROR_IRQ_HPD_ENABLE);
dpcd_val |= DP_PSR_ENABLE_PSR2 | DP_PSR_IRQ_HPD_WITH_CRC_ERRORS;
} else {
if (dev_priv->psr.link_standby)
dpcd_val |= DP_PSR_MAIN_LINK_ACTIVE;
if (INTEL_GEN(dev_priv) >= 8)
dpcd_val |= DP_PSR_CRC_VERIFICATION;
}
drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, dpcd_val);
drm_dp_dpcd_writeb(&intel_dp->aux, DP_SET_POWER, DP_SET_POWER_D0);
}
static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 val = 0;
if (INTEL_GEN(dev_priv) >= 11)
val |= EDP_PSR_TP4_TIME_0US;
if (i915_modparams.psr_safest_params) {
val |= EDP_PSR_TP1_TIME_2500us;
val |= EDP_PSR_TP2_TP3_TIME_2500us;
goto check_tp3_sel;
}
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
if (dev_priv->vbt.psr.tp1_wakeup_time_us == 0)
val |= EDP_PSR_TP1_TIME_0us;
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
else if (dev_priv->vbt.psr.tp1_wakeup_time_us <= 100)
val |= EDP_PSR_TP1_TIME_100us;
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
else if (dev_priv->vbt.psr.tp1_wakeup_time_us <= 500)
val |= EDP_PSR_TP1_TIME_500us;
else
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR_TP1_TIME_2500us;
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us == 0)
val |= EDP_PSR_TP2_TP3_TIME_0us;
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 100)
val |= EDP_PSR_TP2_TP3_TIME_100us;
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
else if (dev_priv->vbt.psr.tp2_tp3_wakeup_time_us <= 500)
val |= EDP_PSR_TP2_TP3_TIME_500us;
else
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR_TP2_TP3_TIME_2500us;
check_tp3_sel:
if (intel_dp_source_supports_hbr2(intel_dp) &&
drm_dp_tps3_supported(intel_dp->dpcd))
val |= EDP_PSR_TP1_TP3_SEL;
else
val |= EDP_PSR_TP1_TP2_SEL;
return val;
}
static u8 psr_compute_idle_frames(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
int idle_frames;
/* Let's use 6 as the minimum to cover all known cases including the
* off-by-one issue that HW has in some cases.
*/
idle_frames = max(6, dev_priv->vbt.psr.idle_frames);
idle_frames = max(idle_frames, dev_priv->psr.sink_sync_latency + 1);
if (drm_WARN_ON(&dev_priv->drm, idle_frames > 0xf))
idle_frames = 0xf;
return idle_frames;
}
static void hsw_activate_psr1(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 max_sleep_time = 0x1f;
u32 val = EDP_PSR_ENABLE;
val |= psr_compute_idle_frames(intel_dp) << EDP_PSR_IDLE_FRAME_SHIFT;
val |= max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT;
if (IS_HASWELL(dev_priv))
val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
if (dev_priv->psr.link_standby)
val |= EDP_PSR_LINK_STANDBY;
val |= intel_psr1_get_tp_time(intel_dp);
if (INTEL_GEN(dev_priv) >= 8)
val |= EDP_PSR_CRC_ENABLE;
val |= (intel_de_read(dev_priv, EDP_PSR_CTL(dev_priv->psr.transcoder)) &
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK);
intel_de_write(dev_priv, EDP_PSR_CTL(dev_priv->psr.transcoder), val);
}
static u32 intel_psr2_get_tp_time(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 val = 0;
if (i915_modparams.psr_safest_params)
return EDP_PSR2_TP2_TIME_2500us;
if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us >= 0 &&
dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 50)
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR2_TP2_TIME_50us;
else if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 100)
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR2_TP2_TIME_100us;
else if (dev_priv->vbt.psr.psr2_tp2_tp3_wakeup_time_us <= 500)
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR2_TP2_TIME_500us;
else
drm/i915/psr: vbt change for psr For psr block #9, the vbt description has moved to options [0-3] for TP1,TP2,TP3 Wakeup time from decimal value without any change to vbt structure. Since spec does not mention from which VBT version this change was added to vbt.bsf file, we cannot depend on bdb->version check to change for all the platforms. There is RCR inplace for GOP team to provide the version number to make generic change. Since Kabylake with bdb version 209 is having this change, limiting this change to gen9_bc and version 209+ to unblock google. Tested on skl(bdb version 203,without options) and kabylake(bdb version 209,212) having new options. bspec 20131 v2: (Jani and Rodrigo) move the 165 version check to intel_bios.c v3: Jani Move the abstraction to intel_bios. v4: Jani Rename tp*_wakeup_time to have "us" suffix. For values outside range[0-3],default to max 2500us. Old decimal value was wake up time in multiples of 100us. v5: Jani and Rodrigo Handle option 2 in default condition. Print oustide range value. For negetive values default to 2500us. v6: Jani Handle default first and then fall through for case 2. v7: Rodrigo Apply this change for IS_GEN9_BC and vbt version > 209 v8: Puthik Add new function vbt_psr_to_us. v9: Jani Change to v7 version as it's more readable. DK add comment /*fall through*/ after case2. Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Puthikorn Voravootivat <puthik@chromium.org> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Maulik V Vaghela <maulik.v.vaghela@intel.com> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/1526981243-2745-1-git-send-email-vathsala.nagaraju@intel.com
2018-05-22 16:27:23 +07:00
val |= EDP_PSR2_TP2_TIME_2500us;
return val;
}
static void hsw_activate_psr2(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 val;
val = psr_compute_idle_frames(intel_dp) << EDP_PSR2_IDLE_FRAME_SHIFT;
val |= EDP_PSR2_ENABLE | EDP_SU_TRACK_ENABLE;
if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
val |= EDP_Y_COORDINATE_ENABLE;
val |= EDP_PSR2_FRAME_BEFORE_SU(dev_priv->psr.sink_sync_latency + 1);
val |= intel_psr2_get_tp_time(intel_dp);
/*
* PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is
* recommending keep this bit unset while PSR2 is enabled.
*/
intel_de_write(dev_priv, EDP_PSR_CTL(dev_priv->psr.transcoder), 0);
intel_de_write(dev_priv, EDP_PSR2_CTL(dev_priv->psr.transcoder), val);
}
static bool
transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder trans)
{
drm/i915: Do not read PSR2 register in transcoders without PSR2 This fix unclaimed access warnings: [ 245.525788] ------------[ cut here ]------------ [ 245.525884] Unclaimed read from register 0x62900 [ 245.526154] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526160] Modules linked in: i915 x86_pkg_temp_thermal ax88179_178a coretemp usbnet crct10dif_pclmul mii crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: i915] [ 245.526191] CPU: 0 PID: 1234 Comm: kms_fullmodeset Not tainted 5.1.0-rc6+ #915 [ 245.526197] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWR1.D00.2081.A10.1904182155 04/18/2019 [ 245.526273] RIP: 0010:__unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526281] Code: 74 05 5b 5d 41 5c c3 45 84 e4 48 c7 c0 76 97 21 a0 48 c7 c6 6c 97 21 a0 89 ea 48 0f 44 f0 48 c7 c7 7f 97 21 a0 e8 4f 1e fe e0 <0f> 0b 83 2d 6f d9 1c 00 01 5b 5d 41 5c c3 66 90 41 57 41 56 41 55 [ 245.526288] RSP: 0018:ffffc900006bf7d8 EFLAGS: 00010086 [ 245.526297] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 245.526304] RDX: 0000000000000007 RSI: 0000000000000000 RDI: 00000000ffffffff [ 245.526310] RBP: 0000000000061900 R08: 0000000000000000 R09: 0000000000000001 [ 245.526317] R10: 0000000000000006 R11: 0000000000000000 R12: 0000000000000001 [ 245.526324] R13: 0000000000000000 R14: ffff8882914f0d58 R15: 0000000000000206 [ 245.526332] FS: 00007fed2a3c39c0(0000) GS:ffff8882a8600000(0000) knlGS:0000000000000000 [ 245.526340] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 245.526347] CR2: 00007fed28dff000 CR3: 00000002a086c006 CR4: 0000000000760ef0 [ 245.526354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 245.526361] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 245.526367] PKRU: 55555554 [ 245.526373] Call Trace: [ 245.526454] gen11_fwtable_read32+0x219/0x250 [i915] [ 245.526576] intel_psr_activate+0x57/0x400 [i915] [ 245.526697] intel_psr_enable_locked+0x367/0x4b0 [i915] [ 245.526828] intel_psr_enable+0xa4/0xd0 [i915] [ 245.526946] intel_enable_ddi+0x127/0x2f0 [i915] [ 245.527075] intel_encoders_enable.isra.79+0x62/0x90 [i915] [ 245.527202] haswell_crtc_enable+0x2a2/0x850 [i915] [ 245.527337] intel_update_crtc+0x51/0x360 [i915] [ 245.527466] skl_update_crtcs+0x26c/0x300 [i915] [ 245.527603] intel_atomic_commit_tail+0x3e5/0x13c0 [i915] [ 245.527757] intel_atomic_commit+0x24d/0x2d0 [i915] [ 245.527782] drm_atomic_helper_set_config+0x7b/0x90 [ 245.527799] drm_mode_setcrtc+0x1b4/0x6f0 [ 245.527856] ? drm_mode_getcrtc+0x180/0x180 [ 245.527867] drm_ioctl_kernel+0xad/0xf0 [ 245.527886] drm_ioctl+0x2f4/0x3b0 [ 245.527902] ? drm_mode_getcrtc+0x180/0x180 [ 245.527935] ? rcu_read_lock_sched_held+0x6f/0x80 [ 245.527956] do_vfs_ioctl+0xa0/0x6d0 [ 245.527970] ? __task_pid_nr_ns+0xb6/0x200 [ 245.527991] ksys_ioctl+0x35/0x70 [ 245.528009] __x64_sys_ioctl+0x11/0x20 [ 245.528020] do_syscall_64+0x55/0x180 [ 245.528034] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 245.528042] RIP: 0033:0x7fed2cc7c3c7 [ 245.528050] Code: 00 00 90 48 8b 05 c9 3a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 99 3a 0d 00 f7 d8 64 89 01 48 [ 245.528057] RSP: 002b:00007ffe36944378 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 245.528067] RAX: ffffffffffffffda RBX: 00007ffe369443b0 RCX: 00007fed2cc7c3c7 [ 245.528074] RDX: 00007ffe369443b0 RSI: 00000000c06864a2 RDI: 0000000000000003 [ 245.528081] RBP: 00007ffe369443b0 R08: 0000000000000000 R09: 0000564c0173ae98 [ 245.528088] R10: 0000564c0173aeb8 R11: 0000000000000246 R12: 00000000c06864a2 [ 245.528095] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000 [ 245.528128] irq event stamp: 140866 [ 245.528138] hardirqs last enabled at (140865): [<ffffffff819a63dc>] _raw_spin_unlock_irqrestore+0x4c/0x60 [ 245.528148] hardirqs last disabled at (140866): [<ffffffff819a624d>] _raw_spin_lock_irqsave+0xd/0x50 [ 245.528158] softirqs last enabled at (140860): [<ffffffff81c0038c>] __do_softirq+0x38c/0x499 [ 245.528170] softirqs last disabled at (140853): [<ffffffff810b4a09>] irq_exit+0xa9/0xc0 [ 245.528247] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.528254] ---[ end trace 366069676e98a410 ]--- Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-7-lucas.demarchi@intel.com
2019-08-17 16:38:33 +07:00
if (INTEL_GEN(dev_priv) < 9)
return false;
else if (INTEL_GEN(dev_priv) >= 12)
return trans == TRANSCODER_A;
else
return trans == TRANSCODER_EDP;
}
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
static u32 intel_get_frame_time_us(const struct intel_crtc_state *cstate)
{
if (!cstate || !cstate->hw.active)
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
return 0;
return DIV_ROUND_UP(1000 * 1000,
drm_mode_vrefresh(&cstate->hw.adjusted_mode));
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
}
static void psr2_program_idle_frames(struct drm_i915_private *dev_priv,
u32 idle_frames)
{
u32 val;
idle_frames <<= EDP_PSR2_IDLE_FRAME_SHIFT;
val = intel_de_read(dev_priv, EDP_PSR2_CTL(dev_priv->psr.transcoder));
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
val &= ~EDP_PSR2_IDLE_FRAME_MASK;
val |= idle_frames;
intel_de_write(dev_priv, EDP_PSR2_CTL(dev_priv->psr.transcoder), val);
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
}
static void tgl_psr2_enable_dc3co(struct drm_i915_private *dev_priv)
{
psr2_program_idle_frames(dev_priv, 0);
intel_display_power_set_target_dc_state(dev_priv, DC_STATE_EN_DC3CO);
}
static void tgl_psr2_disable_dc3co(struct drm_i915_private *dev_priv)
{
struct intel_dp *intel_dp = dev_priv->psr.dp;
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
intel_display_power_set_target_dc_state(dev_priv, DC_STATE_EN_UPTO_DC6);
psr2_program_idle_frames(dev_priv, psr_compute_idle_frames(intel_dp));
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
}
static void tgl_dc3co_disable_work(struct work_struct *work)
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
{
struct drm_i915_private *dev_priv =
container_of(work, typeof(*dev_priv), psr.dc3co_work.work);
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
mutex_lock(&dev_priv->psr.lock);
/* If delayed work is pending, it is not idle */
if (delayed_work_pending(&dev_priv->psr.dc3co_work))
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
goto unlock;
tgl_psr2_disable_dc3co(dev_priv);
unlock:
mutex_unlock(&dev_priv->psr.lock);
}
static void tgl_disallow_dc3co_on_psr2_exit(struct drm_i915_private *dev_priv)
{
if (!dev_priv->psr.dc3co_enabled)
return;
cancel_delayed_work(&dev_priv->psr.dc3co_work);
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
/* Before PSR2 exit disallow dc3co*/
tgl_psr2_disable_dc3co(dev_priv);
}
static void
tgl_dc3co_exitline_compute_config(struct intel_dp *intel_dp,
struct intel_crtc_state *crtc_state)
{
const u32 crtc_vdisplay = crtc_state->uapi.adjusted_mode.crtc_vdisplay;
struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
u32 exit_scanlines;
if (!(dev_priv->csr.allowed_dc_mask & DC_STATE_EN_DC3CO))
return;
/* B.Specs:49196 DC3CO only works with pipeA and DDIA.*/
if (to_intel_crtc(crtc_state->uapi.crtc)->pipe != PIPE_A ||
dig_port->base.port != PORT_A)
return;
/*
* DC3CO Exit time 200us B.Spec 49196
* PSR2 transcoder Early Exit scanlines = ROUNDUP(200 / line time) + 1
*/
exit_scanlines =
intel_usecs_to_scanlines(&crtc_state->uapi.adjusted_mode, 200) + 1;
if (drm_WARN_ON(&dev_priv->drm, exit_scanlines > crtc_vdisplay))
return;
crtc_state->dc3co_exitline = crtc_vdisplay - exit_scanlines;
}
static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
struct intel_crtc_state *crtc_state)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
int crtc_hdisplay = crtc_state->hw.adjusted_mode.crtc_hdisplay;
int crtc_vdisplay = crtc_state->hw.adjusted_mode.crtc_vdisplay;
int psr_max_h = 0, psr_max_v = 0, max_bpp = 0;
if (!dev_priv->psr.sink_psr2_support)
return false;
if (!transcoder_has_psr2(dev_priv, crtc_state->cpu_transcoder)) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 not supported in transcoder %s\n",
transcoder_name(crtc_state->cpu_transcoder));
return false;
}
/*
* Some platforms lack PSR2 HW tracking and instead require manual
* tracking by software. In this case, the driver is required to track
* the areas that need updates and program hardware to send selective
* updates.
*
* So until the software tracking is implemented, PSR2 needs to be
* disabled for platforms without PSR2 HW tracking.
*/
if (!HAS_PSR_HW_TRACKING(dev_priv)) {
drm_dbg_kms(&dev_priv->drm,
"No PSR2 HW tracking in the platform\n");
return false;
}
/*
* DSC and PSR2 cannot be enabled simultaneously. If a requested
* resolution requires DSC to be enabled, priority is given to DSC
* over PSR2.
*/
if (crtc_state->dsc.compression_enable) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 cannot be enabled since DSC is enabled\n");
return false;
}
if (INTEL_GEN(dev_priv) >= 12) {
psr_max_h = 5120;
psr_max_v = 3200;
max_bpp = 30;
} else if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) {
psr_max_h = 4096;
psr_max_v = 2304;
max_bpp = 24;
} else if (IS_GEN(dev_priv, 9)) {
psr_max_h = 3640;
psr_max_v = 2304;
max_bpp = 24;
}
if (crtc_hdisplay > psr_max_h || crtc_vdisplay > psr_max_v) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 not enabled, resolution %dx%d > max supported %dx%d\n",
crtc_hdisplay, crtc_vdisplay,
psr_max_h, psr_max_v);
return false;
}
if (crtc_state->pipe_bpp > max_bpp) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 not enabled, pipe bpp %d > max supported %d\n",
crtc_state->pipe_bpp, max_bpp);
return false;
}
/*
* HW sends SU blocks of size four scan lines, which means the starting
* X coordinate and Y granularity requirements will always be met. We
* only need to validate the SU block width is a multiple of
* x granularity.
*/
if (crtc_hdisplay % dev_priv->psr.su_x_granularity) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 not enabled, hdisplay(%d) not multiple of %d\n",
crtc_hdisplay, dev_priv->psr.su_x_granularity);
return false;
}
if (crtc_state->crc_enabled) {
drm_dbg_kms(&dev_priv->drm,
"PSR2 not enabled because it would inhibit pipe CRC calculation\n");
return false;
}
tgl_dc3co_exitline_compute_config(intel_dp, crtc_state);
return true;
}
void intel_psr_compute_config(struct intel_dp *intel_dp,
struct intel_crtc_state *crtc_state)
{
struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
const struct drm_display_mode *adjusted_mode =
&crtc_state->hw.adjusted_mode;
int psr_setup_time;
if (!CAN_PSR(dev_priv))
return;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
if (intel_dp != dev_priv->psr.dp)
return;
if (!psr_global_enabled(dev_priv))
return;
/*
* HSW spec explicitly says PSR is tied to port A.
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
* BDW+ platforms have a instance of PSR registers per transcoder but
* for now it only supports one instance of PSR, so lets keep it
* hardcoded to PORT_A
*/
if (dig_port->base.port != PORT_A) {
drm_dbg_kms(&dev_priv->drm,
"PSR condition failed: Port not supported\n");
return;
}
if (dev_priv->psr.sink_not_reliable) {
drm_dbg_kms(&dev_priv->drm,
"PSR sink implementation is not reliable\n");
return;
}
if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) {
drm_dbg_kms(&dev_priv->drm,
"PSR condition failed: Interlaced mode enabled\n");
return;
}
psr_setup_time = drm_dp_psr_setup_time(intel_dp->psr_dpcd);
if (psr_setup_time < 0) {
drm_dbg_kms(&dev_priv->drm,
"PSR condition failed: Invalid PSR setup time (0x%02x)\n",
intel_dp->psr_dpcd[1]);
return;
}
if (intel_usecs_to_scanlines(adjusted_mode, psr_setup_time) >
adjusted_mode->crtc_vtotal - adjusted_mode->crtc_vdisplay - 1) {
drm_dbg_kms(&dev_priv->drm,
"PSR condition failed: PSR setup time (%d us) too long\n",
psr_setup_time);
return;
}
crtc_state->has_psr = true;
crtc_state->has_psr2 = intel_psr2_config_valid(intel_dp, crtc_state);
crtc_state->infoframes.enable |= intel_hdmi_infoframe_enable(DP_SDP_VSC);
}
static void intel_psr_activate(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
drm/i915: Do not read PSR2 register in transcoders without PSR2 This fix unclaimed access warnings: [ 245.525788] ------------[ cut here ]------------ [ 245.525884] Unclaimed read from register 0x62900 [ 245.526154] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526160] Modules linked in: i915 x86_pkg_temp_thermal ax88179_178a coretemp usbnet crct10dif_pclmul mii crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: i915] [ 245.526191] CPU: 0 PID: 1234 Comm: kms_fullmodeset Not tainted 5.1.0-rc6+ #915 [ 245.526197] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWR1.D00.2081.A10.1904182155 04/18/2019 [ 245.526273] RIP: 0010:__unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526281] Code: 74 05 5b 5d 41 5c c3 45 84 e4 48 c7 c0 76 97 21 a0 48 c7 c6 6c 97 21 a0 89 ea 48 0f 44 f0 48 c7 c7 7f 97 21 a0 e8 4f 1e fe e0 <0f> 0b 83 2d 6f d9 1c 00 01 5b 5d 41 5c c3 66 90 41 57 41 56 41 55 [ 245.526288] RSP: 0018:ffffc900006bf7d8 EFLAGS: 00010086 [ 245.526297] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 245.526304] RDX: 0000000000000007 RSI: 0000000000000000 RDI: 00000000ffffffff [ 245.526310] RBP: 0000000000061900 R08: 0000000000000000 R09: 0000000000000001 [ 245.526317] R10: 0000000000000006 R11: 0000000000000000 R12: 0000000000000001 [ 245.526324] R13: 0000000000000000 R14: ffff8882914f0d58 R15: 0000000000000206 [ 245.526332] FS: 00007fed2a3c39c0(0000) GS:ffff8882a8600000(0000) knlGS:0000000000000000 [ 245.526340] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 245.526347] CR2: 00007fed28dff000 CR3: 00000002a086c006 CR4: 0000000000760ef0 [ 245.526354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 245.526361] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 245.526367] PKRU: 55555554 [ 245.526373] Call Trace: [ 245.526454] gen11_fwtable_read32+0x219/0x250 [i915] [ 245.526576] intel_psr_activate+0x57/0x400 [i915] [ 245.526697] intel_psr_enable_locked+0x367/0x4b0 [i915] [ 245.526828] intel_psr_enable+0xa4/0xd0 [i915] [ 245.526946] intel_enable_ddi+0x127/0x2f0 [i915] [ 245.527075] intel_encoders_enable.isra.79+0x62/0x90 [i915] [ 245.527202] haswell_crtc_enable+0x2a2/0x850 [i915] [ 245.527337] intel_update_crtc+0x51/0x360 [i915] [ 245.527466] skl_update_crtcs+0x26c/0x300 [i915] [ 245.527603] intel_atomic_commit_tail+0x3e5/0x13c0 [i915] [ 245.527757] intel_atomic_commit+0x24d/0x2d0 [i915] [ 245.527782] drm_atomic_helper_set_config+0x7b/0x90 [ 245.527799] drm_mode_setcrtc+0x1b4/0x6f0 [ 245.527856] ? drm_mode_getcrtc+0x180/0x180 [ 245.527867] drm_ioctl_kernel+0xad/0xf0 [ 245.527886] drm_ioctl+0x2f4/0x3b0 [ 245.527902] ? drm_mode_getcrtc+0x180/0x180 [ 245.527935] ? rcu_read_lock_sched_held+0x6f/0x80 [ 245.527956] do_vfs_ioctl+0xa0/0x6d0 [ 245.527970] ? __task_pid_nr_ns+0xb6/0x200 [ 245.527991] ksys_ioctl+0x35/0x70 [ 245.528009] __x64_sys_ioctl+0x11/0x20 [ 245.528020] do_syscall_64+0x55/0x180 [ 245.528034] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 245.528042] RIP: 0033:0x7fed2cc7c3c7 [ 245.528050] Code: 00 00 90 48 8b 05 c9 3a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 99 3a 0d 00 f7 d8 64 89 01 48 [ 245.528057] RSP: 002b:00007ffe36944378 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 245.528067] RAX: ffffffffffffffda RBX: 00007ffe369443b0 RCX: 00007fed2cc7c3c7 [ 245.528074] RDX: 00007ffe369443b0 RSI: 00000000c06864a2 RDI: 0000000000000003 [ 245.528081] RBP: 00007ffe369443b0 R08: 0000000000000000 R09: 0000564c0173ae98 [ 245.528088] R10: 0000564c0173aeb8 R11: 0000000000000246 R12: 00000000c06864a2 [ 245.528095] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000 [ 245.528128] irq event stamp: 140866 [ 245.528138] hardirqs last enabled at (140865): [<ffffffff819a63dc>] _raw_spin_unlock_irqrestore+0x4c/0x60 [ 245.528148] hardirqs last disabled at (140866): [<ffffffff819a624d>] _raw_spin_lock_irqsave+0xd/0x50 [ 245.528158] softirqs last enabled at (140860): [<ffffffff81c0038c>] __do_softirq+0x38c/0x499 [ 245.528170] softirqs last disabled at (140853): [<ffffffff810b4a09>] irq_exit+0xa9/0xc0 [ 245.528247] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.528254] ---[ end trace 366069676e98a410 ]--- Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-7-lucas.demarchi@intel.com
2019-08-17 16:38:33 +07:00
if (transcoder_has_psr2(dev_priv, dev_priv->psr.transcoder))
drm_WARN_ON(&dev_priv->drm,
intel_de_read(dev_priv, EDP_PSR2_CTL(dev_priv->psr.transcoder)) & EDP_PSR2_ENABLE);
drm/i915: Do not read PSR2 register in transcoders without PSR2 This fix unclaimed access warnings: [ 245.525788] ------------[ cut here ]------------ [ 245.525884] Unclaimed read from register 0x62900 [ 245.526154] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526160] Modules linked in: i915 x86_pkg_temp_thermal ax88179_178a coretemp usbnet crct10dif_pclmul mii crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: i915] [ 245.526191] CPU: 0 PID: 1234 Comm: kms_fullmodeset Not tainted 5.1.0-rc6+ #915 [ 245.526197] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWR1.D00.2081.A10.1904182155 04/18/2019 [ 245.526273] RIP: 0010:__unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526281] Code: 74 05 5b 5d 41 5c c3 45 84 e4 48 c7 c0 76 97 21 a0 48 c7 c6 6c 97 21 a0 89 ea 48 0f 44 f0 48 c7 c7 7f 97 21 a0 e8 4f 1e fe e0 <0f> 0b 83 2d 6f d9 1c 00 01 5b 5d 41 5c c3 66 90 41 57 41 56 41 55 [ 245.526288] RSP: 0018:ffffc900006bf7d8 EFLAGS: 00010086 [ 245.526297] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 245.526304] RDX: 0000000000000007 RSI: 0000000000000000 RDI: 00000000ffffffff [ 245.526310] RBP: 0000000000061900 R08: 0000000000000000 R09: 0000000000000001 [ 245.526317] R10: 0000000000000006 R11: 0000000000000000 R12: 0000000000000001 [ 245.526324] R13: 0000000000000000 R14: ffff8882914f0d58 R15: 0000000000000206 [ 245.526332] FS: 00007fed2a3c39c0(0000) GS:ffff8882a8600000(0000) knlGS:0000000000000000 [ 245.526340] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 245.526347] CR2: 00007fed28dff000 CR3: 00000002a086c006 CR4: 0000000000760ef0 [ 245.526354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 245.526361] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 245.526367] PKRU: 55555554 [ 245.526373] Call Trace: [ 245.526454] gen11_fwtable_read32+0x219/0x250 [i915] [ 245.526576] intel_psr_activate+0x57/0x400 [i915] [ 245.526697] intel_psr_enable_locked+0x367/0x4b0 [i915] [ 245.526828] intel_psr_enable+0xa4/0xd0 [i915] [ 245.526946] intel_enable_ddi+0x127/0x2f0 [i915] [ 245.527075] intel_encoders_enable.isra.79+0x62/0x90 [i915] [ 245.527202] haswell_crtc_enable+0x2a2/0x850 [i915] [ 245.527337] intel_update_crtc+0x51/0x360 [i915] [ 245.527466] skl_update_crtcs+0x26c/0x300 [i915] [ 245.527603] intel_atomic_commit_tail+0x3e5/0x13c0 [i915] [ 245.527757] intel_atomic_commit+0x24d/0x2d0 [i915] [ 245.527782] drm_atomic_helper_set_config+0x7b/0x90 [ 245.527799] drm_mode_setcrtc+0x1b4/0x6f0 [ 245.527856] ? drm_mode_getcrtc+0x180/0x180 [ 245.527867] drm_ioctl_kernel+0xad/0xf0 [ 245.527886] drm_ioctl+0x2f4/0x3b0 [ 245.527902] ? drm_mode_getcrtc+0x180/0x180 [ 245.527935] ? rcu_read_lock_sched_held+0x6f/0x80 [ 245.527956] do_vfs_ioctl+0xa0/0x6d0 [ 245.527970] ? __task_pid_nr_ns+0xb6/0x200 [ 245.527991] ksys_ioctl+0x35/0x70 [ 245.528009] __x64_sys_ioctl+0x11/0x20 [ 245.528020] do_syscall_64+0x55/0x180 [ 245.528034] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 245.528042] RIP: 0033:0x7fed2cc7c3c7 [ 245.528050] Code: 00 00 90 48 8b 05 c9 3a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 99 3a 0d 00 f7 d8 64 89 01 48 [ 245.528057] RSP: 002b:00007ffe36944378 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 245.528067] RAX: ffffffffffffffda RBX: 00007ffe369443b0 RCX: 00007fed2cc7c3c7 [ 245.528074] RDX: 00007ffe369443b0 RSI: 00000000c06864a2 RDI: 0000000000000003 [ 245.528081] RBP: 00007ffe369443b0 R08: 0000000000000000 R09: 0000564c0173ae98 [ 245.528088] R10: 0000564c0173aeb8 R11: 0000000000000246 R12: 00000000c06864a2 [ 245.528095] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000 [ 245.528128] irq event stamp: 140866 [ 245.528138] hardirqs last enabled at (140865): [<ffffffff819a63dc>] _raw_spin_unlock_irqrestore+0x4c/0x60 [ 245.528148] hardirqs last disabled at (140866): [<ffffffff819a624d>] _raw_spin_lock_irqsave+0xd/0x50 [ 245.528158] softirqs last enabled at (140860): [<ffffffff81c0038c>] __do_softirq+0x38c/0x499 [ 245.528170] softirqs last disabled at (140853): [<ffffffff810b4a09>] irq_exit+0xa9/0xc0 [ 245.528247] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.528254] ---[ end trace 366069676e98a410 ]--- Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-7-lucas.demarchi@intel.com
2019-08-17 16:38:33 +07:00
drm_WARN_ON(&dev_priv->drm,
intel_de_read(dev_priv, EDP_PSR_CTL(dev_priv->psr.transcoder)) & EDP_PSR_ENABLE);
drm_WARN_ON(&dev_priv->drm, dev_priv->psr.active);
lockdep_assert_held(&dev_priv->psr.lock);
/* psr1 and psr2 are mutually exclusive.*/
if (dev_priv->psr.psr2_enabled)
hsw_activate_psr2(intel_dp);
else
hsw_activate_psr1(intel_dp);
dev_priv->psr.active = true;
}
static void intel_psr_enable_source(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
u32 mask;
/* Only HSW and BDW have PSR AUX registers that need to be setup. SKL+
* use hardcoded values PSR AUX transactions
*/
if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
hsw_psr_setup_aux(intel_dp);
if (dev_priv->psr.psr2_enabled && (IS_GEN(dev_priv, 9) &&
!IS_GEMINILAKE(dev_priv))) {
i915_reg_t reg = CHICKEN_TRANS(cpu_transcoder);
u32 chicken = intel_de_read(dev_priv, reg);
chicken |= PSR2_VSC_ENABLE_PROG_HEADER |
PSR2_ADD_VERTICAL_LINE_COUNT;
intel_de_write(dev_priv, reg, chicken);
}
/*
* Per Spec: Avoid continuous PSR exit by masking MEMUP and HPD also
* mask LPSP to avoid dependency on other drivers that might block
* runtime_pm besides preventing other hw tracking issues now we
* can rely on frontbuffer tracking.
*/
mask = EDP_PSR_DEBUG_MASK_MEMUP |
EDP_PSR_DEBUG_MASK_HPD |
EDP_PSR_DEBUG_MASK_LPSP |
EDP_PSR_DEBUG_MASK_MAX_SLEEP;
if (INTEL_GEN(dev_priv) < 11)
mask |= EDP_PSR_DEBUG_MASK_DISP_REG_WRITE;
intel_de_write(dev_priv, EDP_PSR_DEBUG(dev_priv->psr.transcoder),
mask);
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
psr_irq_control(dev_priv);
if (crtc_state->dc3co_exitline) {
u32 val;
/*
* TODO: if future platforms supports DC3CO in more than one
* transcoder, EXITLINE will need to be unset when disabling PSR
*/
val = intel_de_read(dev_priv, EXITLINE(cpu_transcoder));
val &= ~EXITLINE_MASK;
val |= crtc_state->dc3co_exitline << EXITLINE_SHIFT;
val |= EXITLINE_ENABLE;
intel_de_write(dev_priv, EXITLINE(cpu_transcoder), val);
}
}
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
static void intel_psr_enable_locked(struct drm_i915_private *dev_priv,
const struct intel_crtc_state *crtc_state,
const struct drm_connector_state *conn_state)
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
{
struct intel_dp *intel_dp = dev_priv->psr.dp;
struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
struct intel_encoder *encoder = &intel_dig_port->base;
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
u32 val;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
drm_WARN_ON(&dev_priv->drm, dev_priv->psr.enabled);
dev_priv->psr.psr2_enabled = intel_psr2_enabled(dev_priv, crtc_state);
dev_priv->psr.busy_frontbuffer_bits = 0;
dev_priv->psr.pipe = to_intel_crtc(crtc_state->uapi.crtc)->pipe;
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
dev_priv->psr.dc3co_enabled = !!crtc_state->dc3co_exitline;
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
dev_priv->psr.transcoder = crtc_state->cpu_transcoder;
/* DC5/DC6 requires at least 6 idle frames */
val = usecs_to_jiffies(intel_get_frame_time_us(crtc_state) * 6);
dev_priv->psr.dc3co_exit_delay = val;
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
/*
* If a PSR error happened and the driver is reloaded, the EDP_PSR_IIR
* will still keep the error set even after the reset done in the
* irq_preinstall and irq_uninstall hooks.
* And enabling in this situation cause the screen to freeze in the
* first time that PSR HW tries to activate so lets keep PSR disabled
* to avoid any rendering problems.
*/
if (INTEL_GEN(dev_priv) >= 12) {
val = intel_de_read(dev_priv,
TRANS_PSR_IIR(dev_priv->psr.transcoder));
val &= EDP_PSR_ERROR(0);
} else {
val = intel_de_read(dev_priv, EDP_PSR_IIR);
val &= EDP_PSR_ERROR(dev_priv->psr.transcoder);
}
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
if (val) {
dev_priv->psr.sink_not_reliable = true;
drm_dbg_kms(&dev_priv->drm,
"PSR interruption error set, not enabling PSR\n");
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
return;
}
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
drm_dbg_kms(&dev_priv->drm, "Enabling PSR%s\n",
dev_priv->psr.psr2_enabled ? "2" : "1");
intel_dp_compute_psr_vsc_sdp(intel_dp, crtc_state, conn_state,
&dev_priv->psr.vsc);
intel_write_dp_vsc_sdp(encoder, crtc_state, &dev_priv->psr.vsc);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
intel_psr_enable_sink(intel_dp);
intel_psr_enable_source(intel_dp, crtc_state);
dev_priv->psr.enabled = true;
intel_psr_activate(intel_dp);
}
/**
* intel_psr_enable - Enable PSR
* @intel_dp: Intel DP
* @crtc_state: new CRTC state
* @conn_state: new CONNECTOR state
*
* This function can only be called after the pipe is fully trained and enabled.
*/
void intel_psr_enable(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
const struct drm_connector_state *conn_state)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
drm/i915/psr: Force PSR probe only after full initialization Commit 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") was forcing the state compute too earlier causing errors because not everything was initialized, so here moving to the end of i915_driver_modeset_probe() when the display is all initialized. Also fixing the place where it disarm the force probe as during the atomic check phase errors could happen like the ones due locking and it would cause PSR to never be enabled if that happens. Leaving the disarm to the atomic commit phase, intel_psr_enable() or intel_psr_update() will be called even if the current state do not allow PSR to be enabled. v2: Check if intel_dp is null in intel_psr_force_mode_changed_set() v3: Check intel_dp before get dev_priv v4: - renamed intel_psr_force_mode_changed_set() to intel_psr_set_force_mode_changed() - removed the set parameter from intel_psr_set_force_mode_changed() - not calling intel_psr_set_force_mode_changed() from intel_psr_enable/update(), directly setting it after the same checks that intel_psr_set_force_mode_changed() does - moved intel_psr_set_force_mode_changed() arm call to i915_driver_modeset_probe() as it is a better for a PSR call, all the functions calls happening between the old and the new function call will cause issue Fixes: 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1151 Tested-by: Ross Zwisler <zwisler@google.com> Reported-by: Ross Zwisler <zwisler@google.com> Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200221212635.11614-1-jose.souza@intel.com
2020-02-22 04:26:35 +07:00
if (!CAN_PSR(dev_priv) || dev_priv->psr.dp != intel_dp)
return;
drm/i915/psr: Force PSR probe only after full initialization Commit 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") was forcing the state compute too earlier causing errors because not everything was initialized, so here moving to the end of i915_driver_modeset_probe() when the display is all initialized. Also fixing the place where it disarm the force probe as during the atomic check phase errors could happen like the ones due locking and it would cause PSR to never be enabled if that happens. Leaving the disarm to the atomic commit phase, intel_psr_enable() or intel_psr_update() will be called even if the current state do not allow PSR to be enabled. v2: Check if intel_dp is null in intel_psr_force_mode_changed_set() v3: Check intel_dp before get dev_priv v4: - renamed intel_psr_force_mode_changed_set() to intel_psr_set_force_mode_changed() - removed the set parameter from intel_psr_set_force_mode_changed() - not calling intel_psr_set_force_mode_changed() from intel_psr_enable/update(), directly setting it after the same checks that intel_psr_set_force_mode_changed() does - moved intel_psr_set_force_mode_changed() arm call to i915_driver_modeset_probe() as it is a better for a PSR call, all the functions calls happening between the old and the new function call will cause issue Fixes: 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1151 Tested-by: Ross Zwisler <zwisler@google.com> Reported-by: Ross Zwisler <zwisler@google.com> Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200221212635.11614-1-jose.souza@intel.com
2020-02-22 04:26:35 +07:00
dev_priv->psr.force_mode_changed = false;
if (!crtc_state->has_psr)
return;
drm_WARN_ON(&dev_priv->drm, dev_priv->drrs.dp);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
mutex_lock(&dev_priv->psr.lock);
if (!psr_global_enabled(dev_priv)) {
drm_dbg_kms(&dev_priv->drm, "PSR disabled by flag\n");
goto unlock;
}
intel_psr_enable_locked(dev_priv, crtc_state, conn_state);
unlock:
mutex_unlock(&dev_priv->psr.lock);
}
static void intel_psr_exit(struct drm_i915_private *dev_priv)
{
u32 val;
if (!dev_priv->psr.active) {
drm/i915: Do not read PSR2 register in transcoders without PSR2 This fix unclaimed access warnings: [ 245.525788] ------------[ cut here ]------------ [ 245.525884] Unclaimed read from register 0x62900 [ 245.526154] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526160] Modules linked in: i915 x86_pkg_temp_thermal ax88179_178a coretemp usbnet crct10dif_pclmul mii crc32_pclmul ghash_clmulni_intel e1000e [last unloaded: i915] [ 245.526191] CPU: 0 PID: 1234 Comm: kms_fullmodeset Not tainted 5.1.0-rc6+ #915 [ 245.526197] Hardware name: Intel Corporation Tiger Lake Client Platform/TigerLake U DDR4 SODIMM RVP, BIOS TGLSFWR1.D00.2081.A10.1904182155 04/18/2019 [ 245.526273] RIP: 0010:__unclaimed_reg_debug+0x40/0x50 [i915] [ 245.526281] Code: 74 05 5b 5d 41 5c c3 45 84 e4 48 c7 c0 76 97 21 a0 48 c7 c6 6c 97 21 a0 89 ea 48 0f 44 f0 48 c7 c7 7f 97 21 a0 e8 4f 1e fe e0 <0f> 0b 83 2d 6f d9 1c 00 01 5b 5d 41 5c c3 66 90 41 57 41 56 41 55 [ 245.526288] RSP: 0018:ffffc900006bf7d8 EFLAGS: 00010086 [ 245.526297] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 [ 245.526304] RDX: 0000000000000007 RSI: 0000000000000000 RDI: 00000000ffffffff [ 245.526310] RBP: 0000000000061900 R08: 0000000000000000 R09: 0000000000000001 [ 245.526317] R10: 0000000000000006 R11: 0000000000000000 R12: 0000000000000001 [ 245.526324] R13: 0000000000000000 R14: ffff8882914f0d58 R15: 0000000000000206 [ 245.526332] FS: 00007fed2a3c39c0(0000) GS:ffff8882a8600000(0000) knlGS:0000000000000000 [ 245.526340] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 245.526347] CR2: 00007fed28dff000 CR3: 00000002a086c006 CR4: 0000000000760ef0 [ 245.526354] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 245.526361] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 245.526367] PKRU: 55555554 [ 245.526373] Call Trace: [ 245.526454] gen11_fwtable_read32+0x219/0x250 [i915] [ 245.526576] intel_psr_activate+0x57/0x400 [i915] [ 245.526697] intel_psr_enable_locked+0x367/0x4b0 [i915] [ 245.526828] intel_psr_enable+0xa4/0xd0 [i915] [ 245.526946] intel_enable_ddi+0x127/0x2f0 [i915] [ 245.527075] intel_encoders_enable.isra.79+0x62/0x90 [i915] [ 245.527202] haswell_crtc_enable+0x2a2/0x850 [i915] [ 245.527337] intel_update_crtc+0x51/0x360 [i915] [ 245.527466] skl_update_crtcs+0x26c/0x300 [i915] [ 245.527603] intel_atomic_commit_tail+0x3e5/0x13c0 [i915] [ 245.527757] intel_atomic_commit+0x24d/0x2d0 [i915] [ 245.527782] drm_atomic_helper_set_config+0x7b/0x90 [ 245.527799] drm_mode_setcrtc+0x1b4/0x6f0 [ 245.527856] ? drm_mode_getcrtc+0x180/0x180 [ 245.527867] drm_ioctl_kernel+0xad/0xf0 [ 245.527886] drm_ioctl+0x2f4/0x3b0 [ 245.527902] ? drm_mode_getcrtc+0x180/0x180 [ 245.527935] ? rcu_read_lock_sched_held+0x6f/0x80 [ 245.527956] do_vfs_ioctl+0xa0/0x6d0 [ 245.527970] ? __task_pid_nr_ns+0xb6/0x200 [ 245.527991] ksys_ioctl+0x35/0x70 [ 245.528009] __x64_sys_ioctl+0x11/0x20 [ 245.528020] do_syscall_64+0x55/0x180 [ 245.528034] entry_SYSCALL_64_after_hwframe+0x49/0xbe [ 245.528042] RIP: 0033:0x7fed2cc7c3c7 [ 245.528050] Code: 00 00 90 48 8b 05 c9 3a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 99 3a 0d 00 f7 d8 64 89 01 48 [ 245.528057] RSP: 002b:00007ffe36944378 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 245.528067] RAX: ffffffffffffffda RBX: 00007ffe369443b0 RCX: 00007fed2cc7c3c7 [ 245.528074] RDX: 00007ffe369443b0 RSI: 00000000c06864a2 RDI: 0000000000000003 [ 245.528081] RBP: 00007ffe369443b0 R08: 0000000000000000 R09: 0000564c0173ae98 [ 245.528088] R10: 0000564c0173aeb8 R11: 0000000000000246 R12: 00000000c06864a2 [ 245.528095] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000 [ 245.528128] irq event stamp: 140866 [ 245.528138] hardirqs last enabled at (140865): [<ffffffff819a63dc>] _raw_spin_unlock_irqrestore+0x4c/0x60 [ 245.528148] hardirqs last disabled at (140866): [<ffffffff819a624d>] _raw_spin_lock_irqsave+0xd/0x50 [ 245.528158] softirqs last enabled at (140860): [<ffffffff81c0038c>] __do_softirq+0x38c/0x499 [ 245.528170] softirqs last disabled at (140853): [<ffffffff810b4a09>] irq_exit+0xa9/0xc0 [ 245.528247] WARNING: CPU: 0 PID: 1234 at drivers/gpu/drm/i915/intel_uncore.c:1100 __unclaimed_reg_debug+0x40/0x50 [i915] [ 245.528254] ---[ end trace 366069676e98a410 ]--- Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190823082055.5992-7-lucas.demarchi@intel.com
2019-08-17 16:38:33 +07:00
if (transcoder_has_psr2(dev_priv, dev_priv->psr.transcoder)) {
val = intel_de_read(dev_priv,
EDP_PSR2_CTL(dev_priv->psr.transcoder));
drm_WARN_ON(&dev_priv->drm, val & EDP_PSR2_ENABLE);
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
}
val = intel_de_read(dev_priv,
EDP_PSR_CTL(dev_priv->psr.transcoder));
drm_WARN_ON(&dev_priv->drm, val & EDP_PSR_ENABLE);
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
return;
}
if (dev_priv->psr.psr2_enabled) {
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
tgl_disallow_dc3co_on_psr2_exit(dev_priv);
val = intel_de_read(dev_priv,
EDP_PSR2_CTL(dev_priv->psr.transcoder));
drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR2_ENABLE));
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
val &= ~EDP_PSR2_ENABLE;
intel_de_write(dev_priv,
EDP_PSR2_CTL(dev_priv->psr.transcoder), val);
} else {
val = intel_de_read(dev_priv,
EDP_PSR_CTL(dev_priv->psr.transcoder));
drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR_ENABLE));
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
val &= ~EDP_PSR_ENABLE;
intel_de_write(dev_priv,
EDP_PSR_CTL(dev_priv->psr.transcoder), val);
}
dev_priv->psr.active = false;
}
static void intel_psr_disable_locked(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
i915_reg_t psr_status;
u32 psr_status_mask;
lockdep_assert_held(&dev_priv->psr.lock);
if (!dev_priv->psr.enabled)
return;
drm_dbg_kms(&dev_priv->drm, "Disabling PSR%s\n",
dev_priv->psr.psr2_enabled ? "2" : "1");
intel_psr_exit(dev_priv);
if (dev_priv->psr.psr2_enabled) {
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
psr_status = EDP_PSR2_STATUS(dev_priv->psr.transcoder);
psr_status_mask = EDP_PSR2_STATUS_STATE_MASK;
} else {
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
psr_status = EDP_PSR_STATUS(dev_priv->psr.transcoder);
psr_status_mask = EDP_PSR_STATUS_STATE_MASK;
}
/* Wait till PSR is idle */
if (intel_de_wait_for_clear(dev_priv, psr_status,
psr_status_mask, 2000))
drm_err(&dev_priv->drm, "Timed out waiting PSR idle state\n");
/* Disable PSR on Sink */
drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, 0);
if (dev_priv->psr.psr2_enabled)
drm_dp_dpcd_writeb(&intel_dp->aux, DP_RECEIVER_ALPM_CONFIG, 0);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
dev_priv->psr.enabled = false;
}
/**
* intel_psr_disable - Disable PSR
* @intel_dp: Intel DP
* @old_crtc_state: old CRTC state
*
* This function needs to be called before disabling pipe.
*/
void intel_psr_disable(struct intel_dp *intel_dp,
const struct intel_crtc_state *old_crtc_state)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
if (!old_crtc_state->has_psr)
return;
if (drm_WARN_ON(&dev_priv->drm, !CAN_PSR(dev_priv)))
return;
mutex_lock(&dev_priv->psr.lock);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
intel_psr_disable_locked(intel_dp);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
mutex_unlock(&dev_priv->psr.lock);
cancel_work_sync(&dev_priv->psr.work);
cancel_delayed_work_sync(&dev_priv->psr.dc3co_work);
}
static void psr_force_hw_tracking_exit(struct drm_i915_private *dev_priv)
{
if (INTEL_GEN(dev_priv) >= 9)
/*
* Display WA #0884: skl+
* This documented WA for bxt can be safely applied
* broadly so we can force HW tracking to exit PSR
* instead of disabling and re-enabling.
* Workaround tells us to write 0 to CUR_SURFLIVE_A,
* but it makes more sense write to the current active
* pipe.
*/
intel_de_write(dev_priv, CURSURFLIVE(dev_priv->psr.pipe), 0);
else
/*
* A write to CURSURFLIVE do not cause HW tracking to exit PSR
* on older gens so doing the manual exit instead.
*/
intel_psr_exit(dev_priv);
}
/**
* intel_psr_update - Update PSR state
* @intel_dp: Intel DP
* @crtc_state: new CRTC state
* @conn_state: new CONNECTOR state
*
* This functions will update PSR states, disabling, enabling or switching PSR
* version when executing fastsets. For full modeset, intel_psr_disable() and
* intel_psr_enable() should be called instead.
*/
void intel_psr_update(struct intel_dp *intel_dp,
const struct intel_crtc_state *crtc_state,
const struct drm_connector_state *conn_state)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
struct i915_psr *psr = &dev_priv->psr;
bool enable, psr2_enable;
if (!CAN_PSR(dev_priv) || READ_ONCE(psr->dp) != intel_dp)
return;
drm/i915/psr: Force PSR probe only after full initialization Commit 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") was forcing the state compute too earlier causing errors because not everything was initialized, so here moving to the end of i915_driver_modeset_probe() when the display is all initialized. Also fixing the place where it disarm the force probe as during the atomic check phase errors could happen like the ones due locking and it would cause PSR to never be enabled if that happens. Leaving the disarm to the atomic commit phase, intel_psr_enable() or intel_psr_update() will be called even if the current state do not allow PSR to be enabled. v2: Check if intel_dp is null in intel_psr_force_mode_changed_set() v3: Check intel_dp before get dev_priv v4: - renamed intel_psr_force_mode_changed_set() to intel_psr_set_force_mode_changed() - removed the set parameter from intel_psr_set_force_mode_changed() - not calling intel_psr_set_force_mode_changed() from intel_psr_enable/update(), directly setting it after the same checks that intel_psr_set_force_mode_changed() does - moved intel_psr_set_force_mode_changed() arm call to i915_driver_modeset_probe() as it is a better for a PSR call, all the functions calls happening between the old and the new function call will cause issue Fixes: 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1151 Tested-by: Ross Zwisler <zwisler@google.com> Reported-by: Ross Zwisler <zwisler@google.com> Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200221212635.11614-1-jose.souza@intel.com
2020-02-22 04:26:35 +07:00
dev_priv->psr.force_mode_changed = false;
mutex_lock(&dev_priv->psr.lock);
enable = crtc_state->has_psr && psr_global_enabled(dev_priv);
psr2_enable = intel_psr2_enabled(dev_priv, crtc_state);
if (enable == psr->enabled && psr2_enable == psr->psr2_enabled) {
/* Force a PSR exit when enabling CRC to avoid CRC timeouts */
if (crtc_state->crc_enabled && psr->enabled)
psr_force_hw_tracking_exit(dev_priv);
else if (INTEL_GEN(dev_priv) < 9 && psr->enabled) {
/*
* Activate PSR again after a force exit when enabling
* CRC in older gens
*/
if (!dev_priv->psr.active &&
!dev_priv->psr.busy_frontbuffer_bits)
schedule_work(&dev_priv->psr.work);
}
goto unlock;
}
if (psr->enabled)
intel_psr_disable_locked(intel_dp);
if (enable)
intel_psr_enable_locked(dev_priv, crtc_state, conn_state);
unlock:
mutex_unlock(&dev_priv->psr.lock);
}
/**
* intel_psr_wait_for_idle - wait for PSR1 to idle
* @new_crtc_state: new CRTC state
* @out_value: PSR status in case of failure
*
* This function is expected to be called from pipe_update_start() where it is
* not expected to race with PSR enable or disable.
*
* Returns: 0 on success or -ETIMEOUT if PSR status does not idle.
*/
int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state,
u32 *out_value)
2018-06-28 03:02:49 +07:00
{
struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc);
struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
2018-06-28 03:02:49 +07:00
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
if (!dev_priv->psr.enabled || !new_crtc_state->has_psr)
return 0;
/* FIXME: Update this for PSR2 if we need to wait for idle */
if (READ_ONCE(dev_priv->psr.psr2_enabled))
return 0;
2018-06-28 03:02:49 +07:00
/*
* From bspec: Panel Self Refresh (BDW+)
* Max. time for PSR to idle = Inverse of the refresh rate + 6 ms of
* exit training time + 1.5 ms of aux channel handshake. 50 ms is
* defensive enough to cover everything.
2018-06-28 03:02:49 +07:00
*/
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
return __intel_wait_for_register(&dev_priv->uncore,
EDP_PSR_STATUS(dev_priv->psr.transcoder),
EDP_PSR_STATUS_STATE_MASK,
EDP_PSR_STATUS_STATE_IDLE, 2, 50,
out_value);
2018-06-28 03:02:49 +07:00
}
static bool __psr_wait_for_idle_locked(struct drm_i915_private *dev_priv)
{
i915_reg_t reg;
u32 mask;
int err;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
if (!dev_priv->psr.enabled)
return false;
if (dev_priv->psr.psr2_enabled) {
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
reg = EDP_PSR2_STATUS(dev_priv->psr.transcoder);
mask = EDP_PSR2_STATUS_STATE_MASK;
} else {
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
reg = EDP_PSR_STATUS(dev_priv->psr.transcoder);
mask = EDP_PSR_STATUS_STATE_MASK;
}
mutex_unlock(&dev_priv->psr.lock);
err = intel_de_wait_for_clear(dev_priv, reg, mask, 50);
if (err)
drm_err(&dev_priv->drm,
"Timed out waiting for PSR Idle for re-enable\n");
/* After the unlocked wait, verify that PSR is still wanted! */
mutex_lock(&dev_priv->psr.lock);
return err == 0 && dev_priv->psr.enabled;
}
static int intel_psr_fastset_force(struct drm_i915_private *dev_priv)
{
struct drm_device *dev = &dev_priv->drm;
struct drm_modeset_acquire_ctx ctx;
struct drm_atomic_state *state;
struct intel_crtc *crtc;
int err;
state = drm_atomic_state_alloc(dev);
if (!state)
return -ENOMEM;
drm_modeset_acquire_init(&ctx, DRM_MODESET_ACQUIRE_INTERRUPTIBLE);
state->acquire_ctx = &ctx;
retry:
for_each_intel_crtc(dev, crtc) {
struct intel_crtc_state *crtc_state =
intel_atomic_get_crtc_state(state, crtc);
if (IS_ERR(crtc_state)) {
err = PTR_ERR(crtc_state);
goto error;
}
if (crtc_state->hw.active && crtc_state->has_psr) {
/* Mark mode as changed to trigger a pipe->update() */
crtc_state->uapi.mode_changed = true;
break;
}
}
err = drm_atomic_commit(state);
error:
if (err == -EDEADLK) {
drm_atomic_state_clear(state);
err = drm_modeset_backoff(&ctx);
if (!err)
goto retry;
}
drm_modeset_drop_locks(&ctx);
drm_modeset_acquire_fini(&ctx);
drm_atomic_state_put(state);
return err;
}
int intel_psr_debug_set(struct drm_i915_private *dev_priv, u64 val)
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
{
const u32 mode = val & I915_PSR_DEBUG_MODE_MASK;
u32 old_mode;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
int ret;
if (val & ~(I915_PSR_DEBUG_IRQ | I915_PSR_DEBUG_MODE_MASK) ||
mode > I915_PSR_DEBUG_FORCE_PSR1) {
drm_dbg_kms(&dev_priv->drm, "Invalid debug mask %llx\n", val);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
return -EINVAL;
}
ret = mutex_lock_interruptible(&dev_priv->psr.lock);
if (ret)
return ret;
old_mode = dev_priv->psr.debug & I915_PSR_DEBUG_MODE_MASK;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
dev_priv->psr.debug = val;
drm/i915/psr: Only handle interruptions of the transcoder in use It was enabling and checking PSR interruptions in every transcoder while it should keep the interruptions on the non-used transcoders masked. While doing this it gives us trouble on Tiger Lake if we are reading/writing to registers of disabled transcoders since from gen12 onwards the registers are relative to the transcoder. Instead of forcing them ON to access those registers, just avoid the accesses as they are not needed. v2 (Lucas): - Explain why we can't keep accessing all transcoders - Remove TODO about extending the irq handling to multiple instances: when/if implementing multiple instances it's pretty clear by the singleton psr that it needs to be extended - Fix intel_psr_debug_set() calling psr_irq_control() with psr.transcoder not set yet (from Imre). Now we only set the debug register right away if psr is already enabled. Otherwise we just record the value to be set when enabling the source. - Do not depend on the value of TRANSCODER_A. Just be relative to it (from Imre) - handle psr error last so we don't schedule the work before handling the other flags v3: - Adding a warning about setting reserverd bits on EDP_PSR_IMR Cc: Imre Deak <imre.deak@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Matt Roper <matthew.d.roper@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190904213419.27547-2-jose.souza@intel.com
2019-09-05 04:34:14 +07:00
/*
* Do it right away if it's already enabled, otherwise it will be done
* when enabling the source.
*/
if (dev_priv->psr.enabled)
psr_irq_control(dev_priv);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
mutex_unlock(&dev_priv->psr.lock);
if (old_mode != mode)
ret = intel_psr_fastset_force(dev_priv);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
return ret;
}
static void intel_psr_handle_irq(struct drm_i915_private *dev_priv)
{
struct i915_psr *psr = &dev_priv->psr;
intel_psr_disable_locked(psr->dp);
psr->sink_not_reliable = true;
/* let's make sure that sink is awaken */
drm_dp_dpcd_writeb(&psr->dp->aux, DP_SET_POWER, DP_SET_POWER_D0);
}
static void intel_psr_work(struct work_struct *work)
{
struct drm_i915_private *dev_priv =
container_of(work, typeof(*dev_priv), psr.work);
mutex_lock(&dev_priv->psr.lock);
if (!dev_priv->psr.enabled)
goto unlock;
if (READ_ONCE(dev_priv->psr.irq_aux_error))
intel_psr_handle_irq(dev_priv);
/*
* We have to make sure PSR is ready for re-enable
* otherwise it keeps disabled until next full enable/disable cycle.
* PSR might take some time to get fully disabled
* and be ready for re-enable.
*/
2018-06-28 03:02:49 +07:00
if (!__psr_wait_for_idle_locked(dev_priv))
goto unlock;
/*
* The delayed work can race with an invalidate hence we need to
* recheck. Since psr_flush first clears this and then reschedules we
* won't ever miss a flush when bailing out here.
*/
if (dev_priv->psr.busy_frontbuffer_bits || dev_priv->psr.active)
goto unlock;
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
intel_psr_activate(dev_priv->psr.dp);
unlock:
mutex_unlock(&dev_priv->psr.lock);
}
/**
* intel_psr_invalidate - Invalidade PSR
* @dev_priv: i915 device
* @frontbuffer_bits: frontbuffer plane tracking bits
drm/i915/psr: Use more PSR HW tracking. So far we are using frontbuffer tracking for everything and ignoring that PSR has a HW capable HW tracking for many modern usages of GPU on Core platforms and newer Atom ones. One reason for that is that we were trying to keep same infrastructure in place for VLV/CHV than the rest of platforms. But also because when this infrastructure was created the front-buffer-tracking origin wasn't that good and stable how it is today after Paulo reworked it to attend FBC cases. However this PSR implementation without HW tracking died on gen8LP. And newer platforms are starting to demand more HW tracking specially with PSR2 cases in mind. By disabling and re-enabling PSR totally every time we believe someone is going to change the front buffer content we don't allow PSR HW tracking to do this job and specially compromising the whole idea of PSR2 case where the HW tracking detect only the damaged area and do a partial screen update. So, from now on, on the platforms that has hw_tracking let's rely more on HW tracking. This also is the case in used by other drivers and more validated by SV teams. So I hope that this will lead us to less misterious bugs. v2: Only do this for platform that actually has hw tracking. v3 from DK Do this only for flips, small gradual changes are better. Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Jose Roberto de Souza <jose.souza@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307033420.3086-3-dhinakaran.pandiyan@intel.com
2018-03-07 10:34:20 +07:00
* @origin: which operation caused the invalidate
*
* Since the hardware frontbuffer tracking has gaps we need to integrate
* with the software frontbuffer tracking. This function gets called every
* time frontbuffer rendering starts and a buffer gets dirtied. PSR must be
* disabled if the frontbuffer mask contains a buffer relevant to PSR.
*
* Dirty frontbuffers relevant to PSR are tracked in busy_frontbuffer_bits."
*/
void intel_psr_invalidate(struct drm_i915_private *dev_priv,
drm/i915/psr: Use more PSR HW tracking. So far we are using frontbuffer tracking for everything and ignoring that PSR has a HW capable HW tracking for many modern usages of GPU on Core platforms and newer Atom ones. One reason for that is that we were trying to keep same infrastructure in place for VLV/CHV than the rest of platforms. But also because when this infrastructure was created the front-buffer-tracking origin wasn't that good and stable how it is today after Paulo reworked it to attend FBC cases. However this PSR implementation without HW tracking died on gen8LP. And newer platforms are starting to demand more HW tracking specially with PSR2 cases in mind. By disabling and re-enabling PSR totally every time we believe someone is going to change the front buffer content we don't allow PSR HW tracking to do this job and specially compromising the whole idea of PSR2 case where the HW tracking detect only the damaged area and do a partial screen update. So, from now on, on the platforms that has hw_tracking let's rely more on HW tracking. This also is the case in used by other drivers and more validated by SV teams. So I hope that this will lead us to less misterious bugs. v2: Only do this for platform that actually has hw tracking. v3 from DK Do this only for flips, small gradual changes are better. Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Jose Roberto de Souza <jose.souza@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307033420.3086-3-dhinakaran.pandiyan@intel.com
2018-03-07 10:34:20 +07:00
unsigned frontbuffer_bits, enum fb_op_origin origin)
{
if (!CAN_PSR(dev_priv))
return;
if (origin == ORIGIN_FLIP)
drm/i915/psr: Use more PSR HW tracking. So far we are using frontbuffer tracking for everything and ignoring that PSR has a HW capable HW tracking for many modern usages of GPU on Core platforms and newer Atom ones. One reason for that is that we were trying to keep same infrastructure in place for VLV/CHV than the rest of platforms. But also because when this infrastructure was created the front-buffer-tracking origin wasn't that good and stable how it is today after Paulo reworked it to attend FBC cases. However this PSR implementation without HW tracking died on gen8LP. And newer platforms are starting to demand more HW tracking specially with PSR2 cases in mind. By disabling and re-enabling PSR totally every time we believe someone is going to change the front buffer content we don't allow PSR HW tracking to do this job and specially compromising the whole idea of PSR2 case where the HW tracking detect only the damaged area and do a partial screen update. So, from now on, on the platforms that has hw_tracking let's rely more on HW tracking. This also is the case in used by other drivers and more validated by SV teams. So I hope that this will lead us to less misterious bugs. v2: Only do this for platform that actually has hw tracking. v3 from DK Do this only for flips, small gradual changes are better. Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Jose Roberto de Souza <jose.souza@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307033420.3086-3-dhinakaran.pandiyan@intel.com
2018-03-07 10:34:20 +07:00
return;
mutex_lock(&dev_priv->psr.lock);
if (!dev_priv->psr.enabled) {
mutex_unlock(&dev_priv->psr.lock);
return;
}
frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(dev_priv->psr.pipe);
dev_priv->psr.busy_frontbuffer_bits |= frontbuffer_bits;
if (frontbuffer_bits)
intel_psr_exit(dev_priv);
mutex_unlock(&dev_priv->psr.lock);
}
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
/*
* When we will be completely rely on PSR2 S/W tracking in future,
* intel_psr_flush() will invalidate and flush the PSR for ORIGIN_FLIP
* event also therefore tgl_dc3co_flush() require to be changed
* accordingly in future.
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
*/
static void
tgl_dc3co_flush(struct drm_i915_private *dev_priv,
unsigned int frontbuffer_bits, enum fb_op_origin origin)
{
mutex_lock(&dev_priv->psr.lock);
if (!dev_priv->psr.dc3co_enabled)
goto unlock;
if (!dev_priv->psr.psr2_enabled || !dev_priv->psr.active)
goto unlock;
/*
* At every frontbuffer flush flip event modified delay of delayed work,
* when delayed work schedules that means display has been idle.
*/
if (!(frontbuffer_bits &
INTEL_FRONTBUFFER_ALL_MASK(dev_priv->psr.pipe)))
goto unlock;
tgl_psr2_enable_dc3co(dev_priv);
mod_delayed_work(system_wq, &dev_priv->psr.dc3co_work,
dev_priv->psr.dc3co_exit_delay);
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
unlock:
mutex_unlock(&dev_priv->psr.lock);
}
/**
* intel_psr_flush - Flush PSR
* @dev_priv: i915 device
* @frontbuffer_bits: frontbuffer plane tracking bits
* @origin: which operation caused the flush
*
* Since the hardware frontbuffer tracking has gaps we need to integrate
* with the software frontbuffer tracking. This function gets called every
* time frontbuffer rendering has completed and flushed out to memory. PSR
* can be enabled again if no other frontbuffer relevant to PSR is dirty.
*
* Dirty frontbuffers relevant to PSR are tracked in busy_frontbuffer_bits.
*/
void intel_psr_flush(struct drm_i915_private *dev_priv,
unsigned frontbuffer_bits, enum fb_op_origin origin)
{
if (!CAN_PSR(dev_priv))
return;
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
if (origin == ORIGIN_FLIP) {
tgl_dc3co_flush(dev_priv, frontbuffer_bits, origin);
drm/i915/psr: Use more PSR HW tracking. So far we are using frontbuffer tracking for everything and ignoring that PSR has a HW capable HW tracking for many modern usages of GPU on Core platforms and newer Atom ones. One reason for that is that we were trying to keep same infrastructure in place for VLV/CHV than the rest of platforms. But also because when this infrastructure was created the front-buffer-tracking origin wasn't that good and stable how it is today after Paulo reworked it to attend FBC cases. However this PSR implementation without HW tracking died on gen8LP. And newer platforms are starting to demand more HW tracking specially with PSR2 cases in mind. By disabling and re-enabling PSR totally every time we believe someone is going to change the front buffer content we don't allow PSR HW tracking to do this job and specially compromising the whole idea of PSR2 case where the HW tracking detect only the damaged area and do a partial screen update. So, from now on, on the platforms that has hw_tracking let's rely more on HW tracking. This also is the case in used by other drivers and more validated by SV teams. So I hope that this will lead us to less misterious bugs. v2: Only do this for platform that actually has hw tracking. v3 from DK Do this only for flips, small gradual changes are better. Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Jose Roberto de Souza <jose.souza@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307033420.3086-3-dhinakaran.pandiyan@intel.com
2018-03-07 10:34:20 +07:00
return;
drm/i915/tgl: Switch between dc3co and dc5 based on display idleness DC3CO is useful power state, when DMC detects PSR2 idle frame while an active video playback, playing 30fps video on 60hz panel is the classic example of this use case. B.Specs:49196 has a restriction to enable DC3CO only for Video Playback. It will be worthy to enable DC3CO after completion of each pageflip and switch back to DC5 when display is idle because driver doesn't differentiate between video playback and a normal pageflip. We will use Frontbuffer flush call tgl_dc3co_flush() to enable DC3CO state only for ORIGIN_FLIP flush call, because DC3CO state has primarily targeted for VPB use case. We are not interested here for frontbuffer invalidates calls because that triggers PSR2 exit, which will explicitly disable DC3CO. DC5 and DC6 saves more power, but can't be entered during video playback because there are not enough idle frames in a row to meet most PSR2 panel deep sleep entry requirement typically 4 frames. As PSR2 existing implementation is using minimum 6 idle frames for deep sleep, it is safer to enable DC5/6 after 6 idle frames (By scheduling a delayed work of 6 idle frames, once DC3CO has been enabled after a pageflip). After manually waiting for 6 idle frames DC5/6 will be enabled and PSR2 deep sleep idle frames will be restored to 6 idle frames, at this point DMC will triggers DC5/6 once PSR2 enters to deep sleep after 6 idle frames. In future when we will enable S/W PSR2 tracking, we can change the PSR2 required deep sleep idle frames to 1 so DMC can trigger the DC5/6 immediately after S/W manual waiting of 6 idle frames get complete. v2: calculated s/w state to switch over dc3co when there is an update. [Imre] Used cancel_delayed_work_sync() in order to avoid any race with already scheduled delayed work. [Imre] v3: Cancel_delayed_work_sync() may blocked the commit work. hence dropping it, dc5_idle_thread() checks the valid wakeref before putting the reference count, which avoids any chances of dropping a zero wakeref. [Imre (IRC)] v4: Used frontbuffer flush mechanism. [Imre] v5: Used psr.pipe to extract frontbuffer busy bits. [Imre] Used cancel_delayed_work_sync() in encoder disable path. [Imre] Used mod_delayed_work() instead of cancelling and scheduling a delayed work. [Imre] Used psr.lock in tgl_dc5_idle_thread() to enable psr2 deep sleep. [Imre] Removed DC5_REQ_IDLE_FRAMES macro. [Imre] v6: Used dc3co_exitline check instead of TGL and dc3co allowed_dc_mask checks, used delayed_work_pending with the psr lock and removed the psr2_deep_slp_disabled flag. [Imre] v7: Code refactoring, moved most of functional code to inte_psr.c [Imre] Using frontbuffer_bits on psr.pipe check instead of busy_frontbuffer_bits. [Imre] Calculating dc3co_exit_delay in intel_psr_enable_locked. [Imre] Cc: Jani Nikula <jani.nikula@intel.com> Cc: Imre Deak <imre.deak@intel.com> Cc: Animesh Manna <animesh.manna@intel.com> Reviewed-by: Imre Deak <imre.deak@intel.com> Signed-off-by: Anshuman Gupta <anshuman.gupta@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191003081738.22101-6-anshuman.gupta@intel.com
2019-10-03 15:17:37 +07:00
}
drm/i915/psr: Use more PSR HW tracking. So far we are using frontbuffer tracking for everything and ignoring that PSR has a HW capable HW tracking for many modern usages of GPU on Core platforms and newer Atom ones. One reason for that is that we were trying to keep same infrastructure in place for VLV/CHV than the rest of platforms. But also because when this infrastructure was created the front-buffer-tracking origin wasn't that good and stable how it is today after Paulo reworked it to attend FBC cases. However this PSR implementation without HW tracking died on gen8LP. And newer platforms are starting to demand more HW tracking specially with PSR2 cases in mind. By disabling and re-enabling PSR totally every time we believe someone is going to change the front buffer content we don't allow PSR HW tracking to do this job and specially compromising the whole idea of PSR2 case where the HW tracking detect only the damaged area and do a partial screen update. So, from now on, on the platforms that has hw_tracking let's rely more on HW tracking. This also is the case in used by other drivers and more validated by SV teams. So I hope that this will lead us to less misterious bugs. v2: Only do this for platform that actually has hw tracking. v3 from DK Do this only for flips, small gradual changes are better. Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Jim Bride <jim.bride@linux.intel.com> Cc: Vathsala Nagaraju <vathsala.nagaraju@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Jose Roberto de Souza <jose.souza@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180307033420.3086-3-dhinakaran.pandiyan@intel.com
2018-03-07 10:34:20 +07:00
mutex_lock(&dev_priv->psr.lock);
if (!dev_priv->psr.enabled) {
mutex_unlock(&dev_priv->psr.lock);
return;
}
frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(dev_priv->psr.pipe);
dev_priv->psr.busy_frontbuffer_bits &= ~frontbuffer_bits;
drm/i915: PSR: Let's rely more on frontbuffer tracking. The ultimate goal here is to remove the dependency we currently have on audio driver power to get PSR working. Since with audio driver runtime PM disabled the Hardware tracking believes graphics is fully active and prevent PSR Entry, or in other words continuously exit PSR. So, the idea is to transfer the PSR exit responsability from the HW tracking to the SW tracking (frontbuffer tracking), who is really mature right now. However with LPSP masked out there might be cases where we could miss exit from HW tracking since it can be relying on this, like a specific case reported at our mailing list who user reported he would miss screen updates if scrolling firefox in a Gnome environment when i915 runtimepm was enabled. So before masking out LPSP again to make us independent from the audio driver we need to make sure that all our cases are coverred from the frontbuffer tracking perspective, where the flush means invalidate and flush. Without this patch for HSW, BDW and SKL we just do the invalidate part when the flush wasn't originated by a page flip because we were trusting the HW tracking for the flip case. So let's rely more on frontbuffer tracking and do the invalidation regardless the origin as expected for all platforms. v2: Improve commit message as suggested by Paulo. v3: Another attempt to let commit message more clear. Cc: Paulo Zanoni <paulo.r.zanoni@intel.com> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com> Reviewed-by: Damien Lespiau damien.lespiau@intel.com Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2015-11-19 02:21:12 +07:00
/* By definition flush = invalidate + flush */
if (frontbuffer_bits)
psr_force_hw_tracking_exit(dev_priv);
if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits)
schedule_work(&dev_priv->psr.work);
mutex_unlock(&dev_priv->psr.lock);
}
/**
* intel_psr_init - Init basic PSR work and mutex.
* @dev_priv: i915 device private
*
* This function is called only once at driver load to initialize basic
* PSR stuff.
*/
void intel_psr_init(struct drm_i915_private *dev_priv)
{
if (!HAS_PSR(dev_priv))
return;
if (!dev_priv->psr.sink_support)
return;
drm/i915/psr: Make PSR registers relative to transcoders PSR registers are a mess, some have the full address while others just have the additional offset from psr_mmio_base. For BDW+ psr_mmio_base is nothing more than TRANSCODER_EDP_OFFSET + 0x800 and using it makes more difficult for people with an PSR register address or PSR register name from from BSpec as i915 also don't match the BSpec names. For HSW psr_mmio_base is _DDI_BUF_CTL_A + 0x800 and PSR registers are only available in DDIA. Other reason to make relative to transcoder is that since BDW every transcoder have PSR registers, so in theory it should be possible to have PSR enabled in a non-eDP transcoder. So for BDW+ we can use _TRANS2() to get the register offset of any PSR register in any transcoder while for HSW we have _HSW_PSR_ADJ that will calculate the register offset for the single PSR instance, noting that we are already guarded about trying to enable PSR in other port than DDIA on HSW by the 'if (dig_port->base.port != PORT_A)' in intel_psr_compute_config(), this check should only be valid for HSW and will be changed in future. PSR2 registers and PSR_EVENT was added after Haswell so that is why _PSR_ADJ() is not used in some macros. The only registers that can not be relative to transcoder are PSR_IMR and PSR_IIR that are not relative to anything, so keeping it hardcoded. That changed for TGL but it will be handled in another patch. Also removing BDW_EDP_PSR_BASE from GVT because it is not used as it is the only PSR register that GVT have. v5: - Macros changed to be more explicit about HSW (Dhinakaran) - Squashed with the patch that added the tran parameter to the macros (Dhinakaran) v6: - Checking for interruption errors after module reload in the transcoder that will be used (Dhinakaran) - Using lowercase to the registers offsets v7: - Removing IS_HASWELL() from registers macros(Jani) Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Zhi Wang <zhi.a.wang@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190820223325.27490-1-jose.souza@intel.com
2019-08-21 05:33:23 +07:00
if (IS_HASWELL(dev_priv))
/*
* HSW don't have PSR registers on the same space as transcoder
* so set this to a value that when subtract to the register
* in transcoder space results in the right offset for HSW
*/
dev_priv->hsw_psr_mmio_adjust = _SRD_CTL_EDP - _HSW_EDP_PSR_BASE;
if (i915_modparams.enable_psr == -1)
if (INTEL_GEN(dev_priv) < 9 || !dev_priv->vbt.psr.enable)
i915_modparams.enable_psr = 0;
/* Set link_standby x link_off defaults */
if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
/* HSW and BDW require workarounds that we don't implement. */
dev_priv->psr.link_standby = false;
else if (INTEL_GEN(dev_priv) < 12)
/* For new platforms up to TGL let's respect VBT back again */
dev_priv->psr.link_standby = dev_priv->vbt.psr.full_link;
INIT_WORK(&dev_priv->psr.work, intel_psr_work);
INIT_DELAYED_WORK(&dev_priv->psr.dc3co_work, tgl_dc3co_disable_work);
mutex_init(&dev_priv->psr.lock);
}
static int psr_get_status_and_error_status(struct intel_dp *intel_dp,
u8 *status, u8 *error_status)
{
struct drm_dp_aux *aux = &intel_dp->aux;
int ret;
ret = drm_dp_dpcd_readb(aux, DP_PSR_STATUS, status);
if (ret != 1)
return ret;
ret = drm_dp_dpcd_readb(aux, DP_PSR_ERROR_STATUS, error_status);
if (ret != 1)
return ret;
*status = *status & DP_PSR_SINK_STATE_MASK;
return 0;
}
static void psr_alpm_check(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
struct drm_dp_aux *aux = &intel_dp->aux;
struct i915_psr *psr = &dev_priv->psr;
u8 val;
int r;
if (!psr->psr2_enabled)
return;
r = drm_dp_dpcd_readb(aux, DP_RECEIVER_ALPM_STATUS, &val);
if (r != 1) {
drm_err(&dev_priv->drm, "Error reading ALPM status\n");
return;
}
if (val & DP_ALPM_LOCK_TIMEOUT_ERROR) {
intel_psr_disable_locked(intel_dp);
psr->sink_not_reliable = true;
drm_dbg_kms(&dev_priv->drm,
"ALPM lock timeout error, disabling PSR\n");
/* Clearing error */
drm_dp_dpcd_writeb(aux, DP_RECEIVER_ALPM_STATUS, val);
}
}
static void psr_capability_changed_check(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
struct i915_psr *psr = &dev_priv->psr;
u8 val;
int r;
r = drm_dp_dpcd_readb(&intel_dp->aux, DP_PSR_ESI, &val);
if (r != 1) {
drm_err(&dev_priv->drm, "Error reading DP_PSR_ESI\n");
return;
}
if (val & DP_PSR_CAPS_CHANGE) {
intel_psr_disable_locked(intel_dp);
psr->sink_not_reliable = true;
drm_dbg_kms(&dev_priv->drm,
"Sink PSR capability changed, disabling PSR\n");
/* Clearing it */
drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_ESI, val);
}
}
void intel_psr_short_pulse(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
struct i915_psr *psr = &dev_priv->psr;
u8 status, error_status;
const u8 errors = DP_PSR_RFB_STORAGE_ERROR |
DP_PSR_VSC_SDP_UNCORRECTABLE_ERROR |
DP_PSR_LINK_CRC_ERROR;
if (!CAN_PSR(dev_priv) || !intel_dp_is_edp(intel_dp))
return;
mutex_lock(&psr->lock);
drm/i915: Allow control of PSR at runtime through debugfs, v6 Currently tests modify i915.enable_psr and then do a modeset cycle to change PSR. We can write a value to i915_edp_psr_debug to force a certain PSR mode without a modeset. To retain compatibility with older userspace, we also still allow the override through the module parameter, and add some tracking to check whether a debugfs mode is specified. Changes since v1: - Rename dev_priv->psr.enabled to .dp, and .hw_configured to .enabled. - Fix i915_psr_debugfs_mode to match the writes to debugfs. - Rename __i915_edp_psr_write to intel_psr_set_debugfs_mode, simplify it and move it to intel_psr.c. This keeps all internals in intel_psr.c - Perform an interruptible wait for hw completion outside of the psr lock, instead of being forced to trywait and return -EBUSY. Changes since v2: - Rebase on top of intel_psr changes. Changes since v3: - Assign psr.dp during init. (dhnkrn) - Add prepared bool, which should be used instead of relying on psr.dp. (dhnkrn) - Fix -EDEADLK handling in debugfs. (dhnkrn) - Clean up waiting for idle in intel_psr_set_debugfs_mode. - Print PSR mode when trying to enable PSR. (dhnkrn) - Move changing psr debug setting to i915_edp_psr_debug_set. (dhnkrn) Changes since v4: - Return error in _set() function. - Change flag values to make them easier to remember. (dhnkrn) - Only assign psr.dp once. (dhnkrn) - Only set crtc_state->has_psr on the crtc with psr.dp. - Fix typo. (dhnkrn) Changes since v5: - Only wait for PSR idle on the PSR connector correctly. (dhnkrn) - Reinstate WARN_ON(drrs.dp) in intel_psr_enable. (dhnkrn) - Remove stray comment. (dhnkrn) - Be silent in intel_psr_compute_config on wrong connector. (dhnkrn) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180809142101.26155-1-maarten.lankhorst@linux.intel.com Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
2018-08-09 21:21:01 +07:00
if (!psr->enabled || psr->dp != intel_dp)
goto exit;
if (psr_get_status_and_error_status(intel_dp, &status, &error_status)) {
drm_err(&dev_priv->drm,
"Error reading PSR status or error status\n");
goto exit;
}
if (status == DP_PSR_SINK_INTERNAL_ERROR || (error_status & errors)) {
intel_psr_disable_locked(intel_dp);
psr->sink_not_reliable = true;
}
if (status == DP_PSR_SINK_INTERNAL_ERROR && !error_status)
drm_dbg_kms(&dev_priv->drm,
"PSR sink internal error, disabling PSR\n");
if (error_status & DP_PSR_RFB_STORAGE_ERROR)
drm_dbg_kms(&dev_priv->drm,
"PSR RFB storage error, disabling PSR\n");
if (error_status & DP_PSR_VSC_SDP_UNCORRECTABLE_ERROR)
drm_dbg_kms(&dev_priv->drm,
"PSR VSC SDP uncorrectable error, disabling PSR\n");
if (error_status & DP_PSR_LINK_CRC_ERROR)
drm_dbg_kms(&dev_priv->drm,
"PSR Link CRC error, disabling PSR\n");
if (error_status & ~errors)
drm_err(&dev_priv->drm,
"PSR_ERROR_STATUS unhandled errors %x\n",
error_status & ~errors);
/* clear status register */
drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_ERROR_STATUS, error_status);
psr_alpm_check(intel_dp);
psr_capability_changed_check(intel_dp);
exit:
mutex_unlock(&psr->lock);
}
bool intel_psr_enabled(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv = dp_to_i915(intel_dp);
bool ret;
if (!CAN_PSR(dev_priv) || !intel_dp_is_edp(intel_dp))
return false;
mutex_lock(&dev_priv->psr.lock);
ret = (dev_priv->psr.dp == intel_dp && dev_priv->psr.enabled);
mutex_unlock(&dev_priv->psr.lock);
return ret;
}
void intel_psr_atomic_check(struct drm_connector *connector,
struct drm_connector_state *old_state,
struct drm_connector_state *new_state)
{
struct drm_i915_private *dev_priv = to_i915(connector->dev);
struct intel_connector *intel_connector;
struct intel_digital_port *dig_port;
struct drm_crtc_state *crtc_state;
if (!CAN_PSR(dev_priv) || !new_state->crtc ||
drm/i915/psr: Force PSR probe only after full initialization Commit 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") was forcing the state compute too earlier causing errors because not everything was initialized, so here moving to the end of i915_driver_modeset_probe() when the display is all initialized. Also fixing the place where it disarm the force probe as during the atomic check phase errors could happen like the ones due locking and it would cause PSR to never be enabled if that happens. Leaving the disarm to the atomic commit phase, intel_psr_enable() or intel_psr_update() will be called even if the current state do not allow PSR to be enabled. v2: Check if intel_dp is null in intel_psr_force_mode_changed_set() v3: Check intel_dp before get dev_priv v4: - renamed intel_psr_force_mode_changed_set() to intel_psr_set_force_mode_changed() - removed the set parameter from intel_psr_set_force_mode_changed() - not calling intel_psr_set_force_mode_changed() from intel_psr_enable/update(), directly setting it after the same checks that intel_psr_set_force_mode_changed() does - moved intel_psr_set_force_mode_changed() arm call to i915_driver_modeset_probe() as it is a better for a PSR call, all the functions calls happening between the old and the new function call will cause issue Fixes: 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1151 Tested-by: Ross Zwisler <zwisler@google.com> Reported-by: Ross Zwisler <zwisler@google.com> Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200221212635.11614-1-jose.souza@intel.com
2020-02-22 04:26:35 +07:00
!dev_priv->psr.force_mode_changed)
return;
intel_connector = to_intel_connector(connector);
dig_port = enc_to_dig_port(intel_attached_encoder(intel_connector));
if (dev_priv->psr.dp != &dig_port->dp)
return;
crtc_state = drm_atomic_get_new_crtc_state(new_state->state,
new_state->crtc);
crtc_state->mode_changed = true;
drm/i915/psr: Force PSR probe only after full initialization Commit 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") was forcing the state compute too earlier causing errors because not everything was initialized, so here moving to the end of i915_driver_modeset_probe() when the display is all initialized. Also fixing the place where it disarm the force probe as during the atomic check phase errors could happen like the ones due locking and it would cause PSR to never be enabled if that happens. Leaving the disarm to the atomic commit phase, intel_psr_enable() or intel_psr_update() will be called even if the current state do not allow PSR to be enabled. v2: Check if intel_dp is null in intel_psr_force_mode_changed_set() v3: Check intel_dp before get dev_priv v4: - renamed intel_psr_force_mode_changed_set() to intel_psr_set_force_mode_changed() - removed the set parameter from intel_psr_set_force_mode_changed() - not calling intel_psr_set_force_mode_changed() from intel_psr_enable/update(), directly setting it after the same checks that intel_psr_set_force_mode_changed() does - moved intel_psr_set_force_mode_changed() arm call to i915_driver_modeset_probe() as it is a better for a PSR call, all the functions calls happening between the old and the new function call will cause issue Fixes: 60c6a14b489b ("drm/i915/display: Force the state compute phase once to enable PSR") Closes: https://gitlab.freedesktop.org/drm/intel/issues/1151 Tested-by: Ross Zwisler <zwisler@google.com> Reported-by: Ross Zwisler <zwisler@google.com> Cc: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Anshuman Gupta <anshuman.gupta@intel.com> Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com> Signed-off-by: José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200221212635.11614-1-jose.souza@intel.com
2020-02-22 04:26:35 +07:00
}
void intel_psr_set_force_mode_changed(struct intel_dp *intel_dp)
{
struct drm_i915_private *dev_priv;
if (!intel_dp)
return;
dev_priv = dp_to_i915(intel_dp);
if (!CAN_PSR(dev_priv) || intel_dp != dev_priv->psr.dp)
return;
dev_priv->psr.force_mode_changed = true;
}