Finally fix watermarks
Subsystems (3)
v5 cover Finally fix watermarks
Next version of https://patchwork.freedesktop.org/series/10276/ .
Notable series-wide changes:
- Added the following patches:
- "drm/i915/skl: Ensure pipes with changed wms get added to the state"
- "drm/i915: Move CRTC updating in atomic_commit into it's own hook"
- "drm/i915/skl: Update DDB values atomically with wms/plane attrs"
With this, we now actually fix all of the issues causing underruns instead of just working around them.
- Since we now are fixing this issue instead of working around it, the following patches were no longer necessary:
- "drm/i915/skl: Always wait for pipes to update after a flush"
- "drm/i915/skl: Fix extra whitespace in skl_flush_wm_values()"
Lyude (5):
drm/i915/skl: Add support for the SAGV, fix underrun hangs drm/i915/skl: Update plane watermarks atomically during plane updates drm/i915/skl: Ensure pipes with changed wms get added to the state drm/i915: Move CRTC updating in atomic_commit into it's own hook drm/i915/skl: Update DDB values atomically with wms/plane attrs
Matt Roper (1):
drm/i915/gen9: Only copy WM results for changed pipes to skl_hw
drivers/gpu/drm/i915/i915_drv.h | 4 + drivers/gpu/drm/i915/i915_reg.h | 5 + drivers/gpu/drm/i915/intel_display.c | 190 +++++++++++--- drivers/gpu/drm/i915/intel_drv.h | 17 ++ drivers/gpu/drm/i915/intel_pm.c | 467 +++++++++++++++++++++++------------ drivers/gpu/drm/i915/intel_sprite.c | 6 + 6 files changed, 496 insertions(+), 193 deletions(-)
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v5 1/6 drm/i915/skl: Add support for the SAGV, fix underrun hangs
Since the watermark calculations for Skylake are still broken, we're apt to hitting underruns very easily under multi-monitor configurations. While it would be lovely if this was fixed, it's not. Another problem that's been coming from this however, is the mysterious issue of underruns causing full system hangs. An easy way to reproduce this with a skylake system:
- Get a laptop with a skylake GPU, and hook up two external monitors to it
- Move the cursor from the built-in LCD to one of the external displays as quickly as you can
- You'll get a few pipe underruns, and eventually the entire system will just freeze.
After doing a lot of investigation and reading through the bspec, I found the existence of the SAGV, which is responsible for adjusting the system agent voltage and clock frequencies depending on how much power we need. According to the bspec:
"The display engine access to system memory is blocked during the
adjustment time. SAGV defaults to enabled. Software must use the GT-driver pcode mailbox to disable SAGV when the display engine is not able to tolerate the blocking time."
The rest of the bspec goes on to explain that software can simply leave the SAGV enabled, and disable it when we use interlaced pipes/have more then one pipe active.
Sure enough, with this patchset the system hangs resulting from pipe underruns on Skylake have completely vanished on my T460s. Additionally,
the bspec mentions turning off the SAGV with more then one pipe enabled
as a workaround for display underruns. While this patch doesn't entirely fix that, it looks like it does improve the situation a little bit so it's likely this is going to be required to make watermarks on Skylake fully functional.
Changes since v5:
- Don't use is_power_of_2. Makes things confusing
- Don't use the old state to figure out whether or not to enable/disable the sagv, use the new one
- Split the loop in skl_disable_sagv into it's own function
Changes since v4:
- Use is_power_of_2 against active_crtcs to check whether we have > 1 pipe enabled
- Fix skl_sagv_get_hw_state(): (temp & 0x1) indicates disabled, 0x0 enabled
- Call skl_sagv_enable/disable() from pre/post-plane updates
Changes since v3:
- Use time_before() to compare timeout to jiffies
Changes since v2:
- Really apply minor style nitpicks to patch this time
Changes since v1:
- Added comments about this probably being one of the requirements to fixing Skylake's watermark issues
- Minor style nitpicks from Matt Roper
- Disable these functions on Broxton, since it doesn't have an SAGV
--- drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/i915_reg.h | 5 ++ drivers/gpu/drm/i915/intel_display.c | 12 ++++ drivers/gpu/drm/i915/intel_drv.h | 2 + drivers/gpu/drm/i915/intel_pm.c | 112 +++++++++++++++++++++++++++++++++++ 5 files changed, 133 insertions(+)
drivers/gpu/drm/i915/i915_drv.h +2 -0
@@ -1962,6 +1962,8 @@ struct drm_i915_private {struct i915_suspend_saved_registers regfile;struct vlv_s0ix_state vlv_s0ix_state; +bool skl_sagv_enabled; +struct {/* * Raw watermark latency values:
drivers/gpu/drm/i915/i915_reg.h +5 -0
@@ -7170,6 +7170,11 @@ enum {#define HSW_PCODE_DE_WRITE_FREQ_REQ 0x17#define DISPLAY_IPS_CONTROL 0x19#define HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL 0x1A +#define GEN9_PCODE_SAGV_CONTROL 0x21 +#define GEN9_SAGV_DISABLE 0x0 +#define GEN9_SAGV_LOW_FREQ 0x1 +#define GEN9_SAGV_HIGH_FREQ 0x2 +#define GEN9_SAGV_DYNAMIC_FREQ 0x3#define GEN6_PCODE_DATA _MMIO(0x138128)#define GEN6_PCODE_FREQ_IA_RATIO_SHIFT 8#define GEN6_PCODE_FREQ_RING_RATIO_SHIFT 16
drivers/gpu/drm/i915/intel_display.c +12 -0
@@ -4565,6 +4565,7 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)struct intel_crtc_state *pipe_config =to_intel_crtc_state (crtc->base.state);struct drm_device *dev = crtc->base.dev; +struct drm_i915_private *dev_priv =to_i915 (dev);struct drm_plane *primary = crtc->base.primary;struct drm_plane_state *old_pri_state =drm_atomic_get_existing_plane_state (old_state, primary);
@@ -4589,6 +4590,9 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state) !old_primary_state->visible))intel_post_enable_primary (&crtc->base); } + +if (hweight32 (dev_priv->active_crtcs) <=1 ) +skl_enable_sagv (dev_priv); }static void intel_pre_plane_update (struct intel_crtc_state *old_crtc_state)
@@ -4649,6 +4653,14 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state) }/* +* SKL workaround: bspec recommends we disable the SAGV when we have +* more then one pipe enabled +*/ +if (pipe_config->base.active && +hweight32 (dev_priv->active_crtcs |drm_crtc_mask (&crtc->base)) >1 ) +skl_disable_sagv (dev_priv); + +/* * If we're doing a modeset, we're done. No need to do any pre-vblank * watermark programming here. */
drivers/gpu/drm/i915/intel_drv.h +2 -0
@@ -1709,6 +1709,8 @@ void ilk_wm_get_hw_state(struct drm_device *dev);void skl_wm_get_hw_state (struct drm_device *dev);void skl_ddb_get_hw_state (struct drm_i915_private *dev_priv,struct skl_ddb_allocation *ddb/* out */ ); +int skl_enable_sagv (struct drm_i915_private *dev_priv); +int skl_disable_sagv (struct drm_i915_private *dev_priv);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);
drivers/gpu/drm/i915/intel_pm.c +112 -0
124 lines not on this page · raw message
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v5 2/6 drm/i915/gen9: Only copy WM results for changed pipes to skl_hw
In reply to LyudeFrom: Matt Roper <[email protected]>
When we write watermark values to the hardware, those values are stored in dev_priv->wm.skl_hw. However with recent watermark changes, the results structure we're copying from only contains valid watermark and DDB values for the pipes that are actually changing; the values for other pipes remain 0. Thus a blind copy of the entire skl_wm_values structure will clobber the values for unchanged pipes...we need to be more selective and only copy over the values for the changing pipes.
This mistake was hidden until recently due to another bug that caused us to erroneously re-calculate watermarks for all active pipes rather than changing pipes. Only when that bug was fixed was the impact of this bug discovered (e.g., modesets failing with "Requested display configuration exceeds system watermark limitations" messages and leaving watermarks non-functional, even ones initiated by intel_fbdev_restore_mode).
Changes since v1:
- Add a function for copying a pipe's wm values (skl_copy_wm_for_pipe()) so we can reuse this later
--- drivers/gpu/drm/i915/intel_pm.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/intel_pm.c +26 -2
@@ -4064,6 +4064,24 @@ skl_compute_ddb(struct drm_atomic_state *state)return 0 ; } +static void +skl_copy_wm_for_pipe (struct skl_wm_values *dst, +struct skl_wm_values *src, +enum pipe pipe) +{ + dst->wm_linetime[pipe] = src->wm_linetime[pipe]; +memcpy (dst->plane[pipe], src->plane[pipe], +sizeof (dst->plane[pipe])); +memcpy (dst->plane_trans[pipe], src->plane_trans[pipe], +sizeof (dst->plane_trans[pipe])); + + dst->ddb.pipe[pipe] = src->ddb.pipe[pipe]; +memcpy (dst->ddb.y_plane[pipe], src->ddb.y_plane[pipe], +sizeof (dst->ddb.y_plane[pipe])); +memcpy (dst->ddb.plane[pipe], src->ddb.plane[pipe], +sizeof (dst->ddb.plane[pipe])); +} +static int skl_compute_wm (struct drm_atomic_state *state) {
@@ -4136,8 +4154,10 @@ static void skl_update_wm(struct drm_crtc *crtc)struct drm_device *dev = crtc->dev;struct drm_i915_private *dev_priv =to_i915 (dev);struct skl_wm_values *results = &dev_priv->wm.skl_results; +struct skl_wm_values *hw_vals = &dev_priv->wm.skl_hw;struct intel_crtc_state *cstate =to_intel_crtc_state (crtc->state);struct skl_pipe_wm *pipe_wm = &cstate->wm.skl.optimal; +int pipe;if ((results->dirty_pipes &drm_crtc_mask (crtc)) ==0 )return ;
@@ -4149,8 +4169,12 @@ static void skl_update_wm(struct drm_crtc *crtc)skl_write_wm_values (dev_priv, results);skl_flush_wm_values (dev_priv, results); -/* store the new configuration */ - dev_priv->wm.skl_hw = *results; +/* +* Store the new configuration (but only for the pipes that have +* changed; the other values weren't recomputed). +*/ +for_each_pipe_masked (dev_priv, pipe, results->dirty_pipes) +skl_copy_wm_for_pipe (hw_vals, results, pipe);mutex_unlock (&dev_priv->wm.wm_mutex); }
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v5 3/6 drm/i915/skl: Update plane watermarks atomically during plane updates
In reply to LyudeThanks to Ville for suggesting this as a potential solution to pipe underruns on Skylake.
On Skylake all of the registers for configuring planes, including the registers for configuring their watermarks, are double buffered. New values written to them won't take effect until said registers are "armed", which is done by writing to the PLANE_SURF (or in the case of cursor planes, the CURBASE register) register.
With this in mind, up until now we've been updating watermarks on skl like this:
non-modeset {
- calculate (during atomic check phase)
- finish_atomic_commit:
- intel_pre_plane_update:
- intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
- drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}
or
modeset {
- calculate (during atomic check phase)
- finish_atomic_commit:
- crtc_enable:
- intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
- drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}
Now we update watermarks atomically like this:
non-modeset {
- calculate (during atomic check phase)
- finish_atomic_commit:
- intel_pre_plane_update:
- intel_update_watermarks() (wm values aren't written yet)
- drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}
modeset {
- calculate (during atomic check phase)
- finish_atomic_commit:
- crtc_enable:
- intel_update_watermarks() (actual wm values aren't written yet)
- drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}
So this patch moves all of the watermark writes into the right place; inside of the vblank evasion where we update all of the registers for each plane. While this patch doesn't fix everything, it does allow us to update the watermark values in the way the hardware expects us to.
Changes since original patch series:
- Remove mutex_lock/mutex_unlock since they don't do anything and we're not touching global state
- Move skl_write_cursor_wm/skl_write_plane_wm functions into intel_pm.c, make externally visible
- Add skl_write_plane_wm calls to skl_update_plane
- Fix conditional for for loop in skl_write_plane_wm (level < max_level should be level <= max_level)
- Make diagram in commit more accurate to what's actually happening
- Add Fixes:
Changes since v1:
- Use IS_GEN9() instead of IS_SKYLAKE() since these fixes apply to more then just Skylake
- Update description to make it clear this patch doesn't fix everything
- Check if pipes were actually changed before writing watermarks
Changes since v2:
- Write PIPE_WM_LINETIME during vblank evasion
Changes since v3:
- Rebase against new SAGV patch changes
Changes since v4:
- Add a parameter to choose what skl_wm_values struct to use when writing new plane watermarks
--- drivers/gpu/drm/i915/intel_display.c | 8 +++++ drivers/gpu/drm/i915/intel_drv.h | 5 ++++ drivers/gpu/drm/i915/intel_pm.c | 58 ++++++++++++++++++++++++++---------- drivers/gpu/drm/i915/intel_sprite.c | 6 ++++ 4 files changed, 61 insertions(+), 16 deletions(-)
drivers/gpu/drm/i915/intel_display.c +8 -0
@@ -2980,6 +2980,7 @@ static void skylake_update_primary_plane(struct drm_plane *plane,struct intel_crtc *intel_crtc =to_intel_crtc (crtc_state->base.crtc);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;u32 plane_ctl, stride_div, stride;u32 tile_height, plane_offset, plane_size;
@@ -3031,6 +3032,9 @@ static void skylake_update_primary_plane(struct drm_plane *plane, intel_crtc->adjusted_x = x_offset; intel_crtc->adjusted_y = y_offset; +if (wm->dirty_pipes &drm_crtc_mask (&intel_crtc->base)) +skl_write_plane_wm (intel_crtc, wm,0 ); +I915_WRITE (PLANE_CTL (pipe,0 ), plane_ctl);I915_WRITE (PLANE_OFFSET (pipe,0 ), plane_offset);I915_WRITE (PLANE_SIZE (pipe,0 ), plane_size);
@@ -10243,9 +10247,13 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base,struct drm_device *dev = crtc->dev;struct drm_i915_private *dev_priv =to_i915 (dev);struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;uint32_t cntl =0 ; +if (IS_GEN9 (dev_priv) && wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_cursor_wm (intel_crtc, wm); +if (plane_state && plane_state->visible) { cntl =MCURSOR_GAMMA_ENABLE ;switch (plane_state->base.crtc_w) {
drivers/gpu/drm/i915/intel_drv.h +5 -0
@@ -1711,6 +1711,11 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, struct skl_ddb_allocation *ddb /* out */);int skl_enable_sagv (struct drm_i915_private *dev_priv);int skl_disable_sagv (struct drm_i915_private *dev_priv); +void skl_write_cursor_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm); +void skl_write_plane_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm, +int plane);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);
drivers/gpu/drm/i915/intel_pm.c +42 -16
76 lines not on this page · raw message
drivers/gpu/drm/i915/intel_sprite.c +6 -0
@@ -203,6 +203,9 @@ skl_update_plane(struct drm_plane *drm_plane,struct intel_plane *intel_plane =to_intel_plane (drm_plane);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results; +struct drm_crtc *crtc = crtc_state->base.crtc; +struct intel_crtc *intel_crtc =to_intel_crtc (crtc);const int pipe = intel_plane->pipe;const int plane = intel_plane->plane +1 ;u32 plane_ctl, stride_div, stride;
@@ -238,6 +241,9 @@ skl_update_plane(struct drm_plane *drm_plane, crtc_w--; crtc_h--; +if (wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_plane_wm (intel_crtc, wm, plane); +if (key->flags) {I915_WRITE (PLANE_KEYVAL (pipe, plane), key->min_value);I915_WRITE (PLANE_KEYMAX (pipe, plane), key->max_value);
Signature · 6 lines
2.7.4 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
PATCH v5 4/6 drm/i915/skl: Ensure pipes with changed wms get added to the state
In reply to LyudeIf we're enabling a pipe, we'll need to modify the watermarks on all other active pipes. Since those pipes won't be added to the state on their own, we need to add them ourselves.
--- drivers/gpu/drm/i915/intel_pm.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/gpu/drm/i915/intel_pm.c +4 -0
@@ -4085,6 +4085,10 @@ skl_compute_ddb(struct drm_atomic_state *state) ret =skl_allocate_pipe_ddb (cstate, ddb);if (ret)return ret; + + ret =drm_atomic_add_affected_planes (state, &intel_crtc->base); +if (ret) +return ret; }return 0 ;
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v5 5/6 drm/i915: Move CRTC updating in atomic_commit into it's own hook
In reply to LyudeSince we have to write ddb allocations at the same time as we do other plane updates, we're going to need to be able to control the order in which we execute modesets on each pipe. The easiest way to do this is to just factor this section of intel_atomic_commit_tail() (intel_atomic_commit() for stable branches) into it's own function, and add an appropriate display function hook for it.
Based off of Matt Rope's suggestions
[omitting CC for stable, since this patch will need to be changed for such backports first]
Cc: Ville Syrjälä <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Radhakrishna Sripada <[email protected]>
Cc: Hans de Goede <[email protected]>
Cc: Matt Roper <[email protected]>
--- drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/intel_display.c | 74 +++++++++++++++++++++++++----------- 2 files changed, 54 insertions(+), 22 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +2 -0
@@ -632,6 +632,8 @@ struct drm_i915_display_funcs { struct intel_crtc_state *crtc_state);void (*crtc_enable)(struct drm_crtc *crtc);void (*crtc_disable)(struct drm_crtc *crtc); +void (*update_crtcs)(struct drm_atomic_state *state, +unsigned int *crtc_vblank_mask);void (*audio_codec_enable)(struct drm_connector *connector,struct intel_encoder *encoder,const struct drm_display_mode *adjusted_mode);
drivers/gpu/drm/i915/intel_display.c +52 -22
102 lines not on this page · raw message
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v5 6/6 drm/i915/skl: Update DDB values atomically with wms/plane attrs
In reply to LyudeNow that we can hook into update_crtcs and control the order in which we update CRTCs at each modeset, we can finish the final step of fixing Skylake's watermark handling by performing DDB updates at the same time as plane updates and watermark updates.
The first major change in this patch is skl_update_crtcs(), which handles ensuring that we order each CRTC update in our atomic commits properly so that they honor the DDB flush order.
The second major change in this patch is the order in which we flush the pipes. While the previous order may have worked, it can't be used in this approach since it no longer will do the right thing. For example, using the old ddb flush order:
We have pipes A, B, and C enabled, and we're disabling C. Initial ddb allocation looks like this:
| A | B |xxxxxxx|
Since we're performing the ddb updates after performing any CRTC disablements in intel_atomic_commit_tail(), the space to the right of pipe B is unallocated.
1. Flush pipes with new allocation contained into old space. None apply, so we skip this
2. Flush pipes having their allocation reduced, but overlapping with a previous allocation. None apply, so we also skip this
3. Flush pipes that got more space allocated. This applies to A and B, giving us the following update order: A, B
This is wrong, since updating pipe A first will cause it to overlap with B and potentially burst into flames. Our new order (see the code comments for details) would update the pipes in the proper order: B, A.
As well, we calculate the order for each DDB update during the check phase, and reference it later in the commit phase when we hit skl_update_crtcs().
This long overdue patch fixes the rest of the underruns on Skylake.
[omitting CC for stable, since this patch will need to be changed for such backports first]
Cc: Ville Syrjälä <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Radhakrishna Sripada <[email protected]>
Cc: Hans de Goede <[email protected]>
Cc: Matt Roper <[email protected]>
--- drivers/gpu/drm/i915/intel_display.c | 100 ++++++++++-- drivers/gpu/drm/i915/intel_drv.h | 10 ++ drivers/gpu/drm/i915/intel_pm.c | 287 ++++++++++++++++------------------- 3 files changed, 231 insertions(+), 166 deletions(-)
drivers/gpu/drm/i915/intel_display.c +88 -12
@@ -12909,16 +12909,23 @@ static void verify_wm_state(struct drm_crtc *crtc, hw_entry->start, hw_entry->end); } -/* cursor */ - hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR ]; - sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR ]; - -if (!skl_ddb_entry_equal (hw_entry, sw_entry)) { -DRM_ERROR ("mismatch in DDB state pipe %c cursor " -"(expected (%u,%u), found (%u,%u))\n" , -pipe_name (pipe), - sw_entry->start, sw_entry->end, - hw_entry->start, hw_entry->end); +/* +* cursor +* If the cursor plane isn't active, we may not have updated it's ddb +* allocation. In that case since the ddb allocation will be updated +* once the plane becomes visible, we can skip this check +*/ +if (intel_crtc->cursor_addr) { + hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR ]; + sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR ]; + +if (!skl_ddb_entry_equal (hw_entry, sw_entry)) { +DRM_ERROR ("mismatch in DDB state pipe %c cursor " +"(expected (%u,%u), found (%u,%u))\n" , +pipe_name (pipe), + sw_entry->start, sw_entry->end, + hw_entry->start, hw_entry->end); + } } }
@@ -13670,6 +13677,72 @@ static void intel_update_crtcs(struct drm_atomic_state *state, } } +static inline void +skl_do_ddb_step (struct drm_atomic_state *state, +enum skl_ddb_step step) +{ +struct intel_atomic_state *intel_state =to_intel_atomic_state (state); +struct drm_crtc *crtc; +struct drm_crtc_state *old_crtc_state; +unsigned int crtc_vblank_mask;/* unused */ +int i; + +for_each_crtc_in_state (state, crtc, old_crtc_state, i) { +struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct intel_crtc_state *cstate = +to_intel_crtc_state (crtc->state); +bool vblank_wait = false; + +if (cstate->wm.skl.ddb_realloc != step || !crtc->state->active) +continue ; + +/* +* If we're changing the ddb allocation of this pipe to make +* room for another pipe, we have to wait for the pipe's ddb +* allocations to actually update by waiting for a vblank. +* Otherwise we risk the next pipe updating before this pipe +* finishes, resulting in the pipe fetching from ddb space for +* the wrong pipe. +* +* However, if we know we don't have any more pipes to move +* around, we can skip this wait and the new ddb allocation +* will take effect at the start of the next vblank. +*/ +switch (step) { +case SKL_DDB_STEP_NO_OVERLAP : +case SKL_DDB_STEP_OVERLAP : +if (step != intel_state->last_ddb_step) + vblank_wait = true; + +/* drop through */ +case SKL_DDB_STEP_FINAL : +DRM_DEBUG_KMS ( +"Updating [CRTC:%d:pipe %c] for DDB step %d\n" , + crtc->base.id,pipe_name (intel_crtc->pipe), + step); + +case SKL_DDB_STEP_NONE : +break ; + } + +intel_update_crtc (crtc, state, old_crtc_state, + &crtc_vblank_mask); + +if (vblank_wait) +intel_wait_for_vblank (state->dev, intel_crtc->pipe); + } +} + +static void skl_update_crtcs (struct drm_atomic_state *state, +unsigned int *crtc_vblank_mask) +{ +struct intel_atomic_state *intel_state =to_intel_atomic_state (state); +enum skl_ddb_step step; + +for (step =0 ; step <= intel_state->last_ddb_step; step++) +skl_do_ddb_step (state, step); +} +static void intel_atomic_commit_tail (struct drm_atomic_state *state) {struct drm_device *dev = state->dev;
@@ -15230,8 +15303,6 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) dev_priv->display.crtc_disable = i9xx_crtc_disable; } - dev_priv->display.update_crtcs = intel_update_crtcs; -/* Returns the core display clock speed */ if (IS_SKYLAKE (dev_priv) ||IS_KABYLAKE (dev_priv)) dev_priv->display.get_display_clock_speed =
@@ -15321,6 +15392,11 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) skl_modeset_calc_cdclk; } +if (dev_priv->info.gen >=9 ) + dev_priv->display.update_crtcs = skl_update_crtcs; +else + dev_priv->display.update_crtcs = intel_update_crtcs; +switch (INTEL_INFO (dev_priv)->gen) {case 2 : dev_priv->display.queue_flip = intel_gen2_queue_flip;
drivers/gpu/drm/i915/intel_drv.h +10 -0
@@ -334,6 +334,7 @@ struct intel_atomic_state {/* Gen9+ only */ struct skl_wm_values wm_results; +int last_ddb_step; };struct intel_plane_state {
@@ -437,6 +438,13 @@ struct skl_pipe_wm {uint32_t linetime; }; +enum skl_ddb_step { +SKL_DDB_STEP_NONE =0 , +SKL_DDB_STEP_NO_OVERLAP , +SKL_DDB_STEP_OVERLAP , +SKL_DDB_STEP_FINAL +}; + struct intel_crtc_wm_state { union { struct {
@@ -467,6 +475,8 @@ struct intel_crtc_wm_state {/* minimum block allocation */ uint16_t minimum_blocks[I915_MAX_PLANES ];uint16_t minimum_y_blocks[I915_MAX_PLANES ]; + +enum skl_ddb_step ddb_realloc; } skl; };
drivers/gpu/drm/i915/intel_pm.c +133 -154
366 lines not on this page · raw message
Signature · 6 lines
2.7.4 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [v5,1/6] drm/i915/skl: Add support for the SAGV, fix underrun hangs
In reply to LyudeHi,
Quoted 123 lines · On 08/02/2016 08:52 PM, [email protected] wrote:
Since the watermark calculations for Skylake are still broken, we're apt to hitting underruns very easily under multi-monitor configurations. While it would be lovely if this was fixed, it's not. Another problem that's been coming from this however, is the mysterious issue of underruns causing full system hangs. An easy way to reproduce this with a skylake system:
- Get a laptop with a skylake GPU, and hook up two external monitors to it
- Move the cursor from the built-in LCD to one of the external displays as quickly as you can
- You'll get a few pipe underruns, and eventually the entire system will just freeze.After doing a lot of investigation and reading through the bspec, I found the existence of the SAGV, which is responsible for adjusting the system agent voltage and clock frequencies depending on how much power we need. According to the bspec:
"The display engine access to system memory is blocked during the
adjustment time. SAGV defaults to enabled. Software must use the GT-driver pcode mailbox to disable SAGV when the display engine is not able to tolerate the blocking time."The rest of the bspec goes on to explain that software can simply leave the SAGV enabled, and disable it when we use interlaced pipes/have more then one pipe active.
Sure enough, with this patchset the system hangs resulting from pipe underruns on Skylake have completely vanished on my T460s. Additionally,
the bspec mentions turning off the SAGV with more then one pipe enabled
as a workaround for display underruns. While this patch doesn't entirely fix that, it looks like it does improve the situation a little bit so it's likely this is going to be required to make watermarks on Skylake fully functional.Changes since v5:
- Don't useis_power_of_2. Makes things confusing
- Don't use the old state to figure out whether or not to enable/disable the sagv, use the new one
- Split the loop inskl_disable_sagvinto it's own function
Changes since v4:
- Useis_power_of_2againstactive_crtcsto check whether we have > 1 pipe enabled
- Fixskl_sagv_get_hw_state(): (temp &0x1) indicates disabled,0x0enabled
- Call skl_sagv_enable/disable() from pre/post-plane updates
Changes since v3:
- Usetime_before()to compare timeout to jiffies
Changes since v2:
- Really apply minor style nitpicks to patch this time
Changes since v1:
- Added comments about this probably being one of the requirements to fixing Skylake's watermark issues
- Minor style nitpicks from Matt Roper
- Disable these functions on Broxton, since it doesn't have an SAGVReviewed-by Matt Roper <[email protected]>Signed-off-by Lyude <[email protected]>Cc Daniel Vetter <[email protected]>Cc Ville Syrjälä <[email protected]>--- drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/i915_reg.h | 5 ++ drivers/gpu/drm/i915/intel_display.c | 12 ++++ drivers/gpu/drm/i915/intel_drv.h | 2 + drivers/gpu/drm/i915/intel_pm.c | 112 +++++++++++++++++++++++++++++++++++ 5 files changed, 133 insertions(+)drivers/gpu/drm/i915/i915_drv.h +2
-0@@ -1962,6 +1962,8 @@ struct drm_i915_private {struct i915_suspend_saved_registers regfile;struct vlv_s0ix_state vlv_s0ix_state; +bool skl_sagv_enabled; +struct {/* * Raw watermark latency values: drivers/gpu/drm/i915/i915_reg.h +5
-0@@ -7170,6 +7170,11 @@ enum {#define HSW_PCODE_DE_WRITE_FREQ_REQ 0x17#define DISPLAY_IPS_CONTROL 0x19#define HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL 0x1A +#define GEN9_PCODE_SAGV_CONTROL 0x21 +#define GEN9_SAGV_DISABLE 0x0 +#define GEN9_SAGV_LOW_FREQ 0x1 +#define GEN9_SAGV_HIGH_FREQ 0x2 +#define GEN9_SAGV_DYNAMIC_FREQ 0x3#define GEN6_PCODE_DATA _MMIO(0x138128)#define GEN6_PCODE_FREQ_IA_RATIO_SHIFT 8#define GEN6_PCODE_FREQ_RING_RATIO_SHIFT 16drivers/gpu/drm/i915/intel_display.c +4
-0@@ -4565,6 +4565,7 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)struct intel_crtc_state *pipe_config =to_intel_crtc_state (crtc->base.state);struct drm_device *dev = crtc->base.dev; +struct drm_i915_private *dev_priv =to_i915 (dev);struct drm_plane *primary = crtc->base.primary;struct drm_plane_state *old_pri_state =drm_atomic_get_existing_plane_state (old_state, primary);@@ -4589,6 +4590,9 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state) !old_primary_state->visible))intel_post_enable_primary (&crtc->base); } + +if (hweight32 (dev_priv->active_crtcs) <=1 ) +skl_enable_sagv (dev_priv); }static void intel_pre_plane_update (struct intel_crtc_state *old_crtc_state)@@ -4649,6 +4653,14 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state) }
drivers/gpu/drm/i915/intel_display.c static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state) · line 4663
Quoted 10 lines
drivers/gpu/drm/i915/intel_display.c +8
-0/* +* SKL workaround: bspec recommends we disable the SAGV when we have +* more then one pipe enabled +*/ +if (pipe_config->base.active && +hweight32 (dev_priv->active_crtcs |drm_crtc_mask (&crtc->base)) >1 ) +skl_disable_sagv (dev_priv); + +/*
Why | drm_crtc_mask? Also if it is a mask I would expect & not | > Last why are you using drm_crtc_mask here and not in post_plane_update ?
Regards,
Hans
Quoted 139 lines
drivers/gpu/drm/i915/intel_display.c +0
-0* If we're doing a modeset, we're done. No need to do any pre-vblank * watermark programming here. */ drivers/gpu/drm/i915/intel_drv.h +2
-0@@ -1709,6 +1709,8 @@ void ilk_wm_get_hw_state(struct drm_device *dev);void skl_wm_get_hw_state (struct drm_device *dev);void skl_ddb_get_hw_state (struct drm_i915_private *dev_priv,struct skl_ddb_allocation *ddb/* out */ ); +int skl_enable_sagv (struct drm_i915_private *dev_priv); +int skl_disable_sagv (struct drm_i915_private *dev_priv);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);drivers/gpu/drm/i915/intel_pm.c +110
-0@@ -2884,6 +2884,116 @@ skl_wm_plane_id(const struct intel_plane *plane) }static void +skl_sagv_get_hw_state (struct drm_i915_private *dev_priv) +{ +u32 temp; +int ret; + +if (IS_BROXTON (dev_priv)) +return ; + +mutex_lock (&dev_priv->rps.hw_lock); + ret =sandybridge_pcode_read (dev_priv,GEN9_PCODE_SAGV_CONTROL , &temp); +mutex_unlock (&dev_priv->rps.hw_lock); + +if (!ret) { + dev_priv->skl_sagv_enabled = !(temp &0x1 ); + }else { +/* +* If for some reason we can't access the SAGV state, follow +* the bspec and assume it's enabled +*/ +DRM_ERROR ("Failed to get SAGV state, assuming enabled\n" ); + dev_priv->skl_sagv_enabled = true; + } +} + +/* +* SAGV dynamically adjusts the system agent voltage and clock frequencies +* depending on power and performance requirements. The display engine access +* to system memory is blocked during the adjustment time. Having this enabled +* in multi-pipe configurations can cause issues (such as underruns causing +* full system hangs), and the bspec also suggests that software disable it +* when more then one pipe is enabled. +*/ +int +skl_enable_sagv (struct drm_i915_private *dev_priv) +{ +int ret; + +if (IS_BROXTON (dev_priv)) +return 0 ; +if (dev_priv->skl_sagv_enabled) +return 0 ; + +mutex_lock (&dev_priv->rps.hw_lock); +DRM_DEBUG_KMS ("Enabling the SAGV\n" ); + + ret =sandybridge_pcode_write (dev_priv,GEN9_PCODE_SAGV_CONTROL , +GEN9_SAGV_DYNAMIC_FREQ ); +if (!ret) + dev_priv->skl_sagv_enabled = true; +else +DRM_ERROR ("Failed to enable the SAGV\n" ); + +/* We don't need to wait for SAGV when enabling */ +mutex_unlock (&dev_priv->rps.hw_lock); +return ret; +} + +static int +skl_do_sagv_disable (struct drm_i915_private *dev_priv) +{ +int ret; +uint32_t temp; + + ret =sandybridge_pcode_write (dev_priv,GEN9_PCODE_SAGV_CONTROL , +GEN9_SAGV_DISABLE ); +if (ret) { +DRM_ERROR ("Failed to disable the SAGV\n" ); +return ret; + } + + ret =sandybridge_pcode_read (dev_priv,GEN9_PCODE_SAGV_CONTROL , + &temp); +if (ret) { +DRM_ERROR ("Failed to check the status of the SAGV\n" ); +return ret; + } + +return temp &0x1 ; +} + +int +skl_disable_sagv (struct drm_i915_private *dev_priv) +{ +int ret, result; + +if (IS_BROXTON (dev_priv)) +return 0 ; +if (!dev_priv->skl_sagv_enabled) +return 0 ; + +mutex_lock (&dev_priv->rps.hw_lock); +DRM_DEBUG_KMS ("Disabling the SAGV\n" ); + +/* bspec says to keep retrying for at least 1 ms */ + ret =wait_for (result =skl_do_sagv_disable (dev_priv),1 ); +mutex_unlock (&dev_priv->rps.hw_lock); + +if (ret == -ETIMEDOUT ) +DRM_ERROR ("Request to disable SAGV timed out\n" ); +else { +if (result ==1 ) + dev_priv->skl_sagv_enabled = false; + + ret = result; + } + +return ret; +} + +static void skl_ddb_get_pipe_allocation_limits(struct drm_device *dev, const struct intel_crtc_state *cstate, struct skl_ddb_entry *alloc, /* out */
drivers/gpu/drm/i915/intel_pm.c void skl_wm_get_hw_state(struct drm_device *dev) · line 4353
Quoted 8 lines
drivers/gpu/drm/i915/intel_pm.c +2
-0@@ -4236,6 +4346,8 @@ void skl_wm_get_hw_state(struct drm_device *dev)/* Easy/common case; just sanitize DDB now if everything off */ memset (ddb,0 ,sizeof (*ddb)); } + +skl_sagv_get_hw_state (dev_priv); }static void ilk_pipe_wm_get_hw_state (struct drm_crtc *crtc)
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: [PATCH v5 1/6] drm/i915/skl: Add support for the SAGV, fix underrun hangs
In reply to LyudeQuoted 112 lines · Lyude
Since the watermark calculations for Skylake are still broken, we're apt to hitting underruns very easily under multi-monitor configurations. While it would be lovely if this was fixed, it's not. Another problem that's been coming from this however, is the mysterious issue of underruns causing full system hangs. An easy way to reproduce this with a skylake system:
- Get a laptop with a skylake GPU, and hook up two external monitors to it
- Move the cursor from the built-in LCD to one of the external displays as quickly as you can
- You'll get a few pipe underruns, and eventually the entire system will just freeze.After doing a lot of investigation and reading through the bspec, I found the existence of the SAGV, which is responsible for adjusting the system agent voltage and clock frequencies depending on how much power we need. According to the bspec:
"The display engine access to system memory is blocked during the
adjustment time. SAGV defaults to enabled. Software must use the GT-driver pcode mailbox to disable SAGV when the display engine is not able to tolerate the blocking time."The rest of the bspec goes on to explain that software can simply leave the SAGV enabled, and disable it when we use interlaced pipes/have more then one pipe active.
Sure enough, with this patchset the system hangs resulting from pipe underruns on Skylake have completely vanished on my T460s. Additionally,
the bspec mentions turning off the SAGV with more then one pipe enabled
as a workaround for display underruns. While this patch doesn't entirely fix that, it looks like it does improve the situation a little bit so it's likely this is going to be required to make watermarks on Skylake fully functional.Changes since v5:
- Don't useis_power_of_2. Makes things confusing
- Don't use the old state to figure out whether or not to enable/disable the sagv, use the new one
- Split the loop inskl_disable_sagvinto it's own function
Changes since v4:
- Useis_power_of_2againstactive_crtcsto check whether we have > 1 pipe enabled
- Fixskl_sagv_get_hw_state(): (temp &0x1) indicates disabled,0x0enabled
- Call skl_sagv_enable/disable() from pre/post-plane updates
Changes since v3:
- Usetime_before()to compare timeout to jiffies
Changes since v2:
- Really apply minor style nitpicks to patch this time
Changes since v1:
- Added comments about this probably being one of the requirements to fixing Skylake's watermark issues
- Minor style nitpicks from Matt Roper
- Disable these functions on Broxton, since it doesn't have an SAGVReviewed-by Matt Roper <[email protected]>Signed-off-by Lyude <[email protected]>Cc Daniel Vetter <[email protected]>Cc Ville Syrjälä <[email protected]>--- drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/i915_reg.h | 5 ++ drivers/gpu/drm/i915/intel_display.c | 12 ++++ drivers/gpu/drm/i915/intel_drv.h | 2 + drivers/gpu/drm/i915/intel_pm.c | 112 +++++++++++++++++++++++++++++++++++ 5 files changed, 133 insertions(+)drivers/gpu/drm/i915/i915_drv.h +2
-0@@ -1962,6 +1962,8 @@ struct drm_i915_private {struct i915_suspend_saved_registers regfile;struct vlv_s0ix_state vlv_s0ix_state; +bool skl_sagv_enabled; +struct {/* * Raw watermark latency values: drivers/gpu/drm/i915/i915_reg.h +5
-0@@ -7170,6 +7170,11 @@ enum {#define HSW_PCODE_DE_WRITE_FREQ_REQ 0x17#define DISPLAY_IPS_CONTROL 0x19#define HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL 0x1A +#define GEN9_PCODE_SAGV_CONTROL 0x21 +#define GEN9_SAGV_DISABLE 0x0 +#define GEN9_SAGV_LOW_FREQ 0x1 +#define GEN9_SAGV_HIGH_FREQ 0x2 +#define GEN9_SAGV_DYNAMIC_FREQ 0x3#define GEN6_PCODE_DATA _MMIO(0x138128)#define GEN6_PCODE_FREQ_IA_RATIO_SHIFT 8#define GEN6_PCODE_FREQ_RING_RATIO_SHIFT 16drivers/gpu/drm/i915/intel_display.c +1
-0@@ -4565,6 +4565,7 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state)struct intel_crtc_state *pipe_config =to_intel_crtc_state (crtc->base.state);struct drm_device *dev = crtc->base.dev; +struct drm_i915_private *dev_priv =to_i915 (dev);struct drm_plane *primary = crtc->base.primary;struct drm_plane_state *old_pri_state =drm_atomic_get_existing_plane_state (old_state, primary);
drivers/gpu/drm/i915/intel_display.c static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state) · line 4595
Quoted 6 lines
drivers/gpu/drm/i915/intel_display.c +3
-0@@ -4589,6 +4590,9 @@ static void intel_post_plane_update(struct intel_crtc_state *old_crtc_state) !old_primary_state->visible))intel_post_enable_primary (&crtc->base); } + +if (hweight32 (dev_priv->active_crtcs) <=1 ) +skl_enable_sagv (dev_priv);
There might be a slightly better place to handle this; see comment below.
drivers/gpu/drm/i915/intel_display.c static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state) · line 4661
Quoted 13 lines
drivers/gpu/drm/i915/intel_display.c +6
-0}static void intel_pre_plane_update (struct intel_crtc_state *old_crtc_state)@@ -4649,6 +4653,14 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state) }/* +* SKL workaround: bspec recommends we disable the SAGV when we have +* more then one pipe enabled +*/ +if (pipe_config->base.active && +hweight32 (dev_priv->active_crtcs |drm_crtc_mask (&crtc->base)) >1 ) +skl_disable_sagv (dev_priv);
As Hans pointed out, this doesn't look right. I'm guessing you're trying to guard against the case where there's no intersection between the starting pipe usage and the final pipe usage? E.g., only pipe A active before commit, only pipe B active after commit, and you're worried there might be a brief point where both A and B are on together? I don't think this should really matter since we power everything old-changing down before powering anything new-changing up. So I think all we really need to care about for enabling/disabling is how many CRTC's there are in the final state.
It seems like a more natural place to place to handle enable/disable would be in the 'if (intel_state->modeset)' blocks of intel_atomic_commit_tail() (since only a modeset operation could change the number of CRTC's in use to trigger a SAGV toggle). That also has the slight benefit of only getting run once for the atomic transaction rather than once for each CRTC (and you can just do a simple test of hweight(intel_state->active_crtcs) to figure out how many CRTC's are going to be on after the transaction completes.
Quoted 116 lines
drivers/gpu/drm/i915/intel_display.c +2
-0+ +/* * If we're doing a modeset, we're done. No need to do any pre-vblank * watermark programming here. */ drivers/gpu/drm/i915/intel_drv.h +2
-0@@ -1709,6 +1709,8 @@ void ilk_wm_get_hw_state(struct drm_device *dev);void skl_wm_get_hw_state (struct drm_device *dev);void skl_ddb_get_hw_state (struct drm_i915_private *dev_priv,struct skl_ddb_allocation *ddb/* out */ ); +int skl_enable_sagv (struct drm_i915_private *dev_priv); +int skl_disable_sagv (struct drm_i915_private *dev_priv);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);drivers/gpu/drm/i915/intel_pm.c +90
-0@@ -2884,6 +2884,116 @@ skl_wm_plane_id(const struct intel_plane *plane) }static void +skl_sagv_get_hw_state (struct drm_i915_private *dev_priv) +{ +u32 temp; +int ret; + +if (IS_BROXTON (dev_priv)) +return ; + +mutex_lock (&dev_priv->rps.hw_lock); + ret =sandybridge_pcode_read (dev_priv,GEN9_PCODE_SAGV_CONTROL , &temp); +mutex_unlock (&dev_priv->rps.hw_lock); + +if (!ret) { + dev_priv->skl_sagv_enabled = !(temp &0x1 ); + }else { +/* +* If for some reason we can't access the SAGV state, follow +* the bspec and assume it's enabled +*/ +DRM_ERROR ("Failed to get SAGV state, assuming enabled\n" ); + dev_priv->skl_sagv_enabled = true; + } +} + +/* +* SAGV dynamically adjusts the system agent voltage and clock frequencies +* depending on power and performance requirements. The display engine access +* to system memory is blocked during the adjustment time. Having this enabled +* in multi-pipe configurations can cause issues (such as underruns causing +* full system hangs), and the bspec also suggests that software disable it +* when more then one pipe is enabled. +*/ +int +skl_enable_sagv (struct drm_i915_private *dev_priv) +{ +int ret; + +if (IS_BROXTON (dev_priv)) +return 0 ; +if (dev_priv->skl_sagv_enabled) +return 0 ; + +mutex_lock (&dev_priv->rps.hw_lock); +DRM_DEBUG_KMS ("Enabling the SAGV\n" ); + + ret =sandybridge_pcode_write (dev_priv,GEN9_PCODE_SAGV_CONTROL , +GEN9_SAGV_DYNAMIC_FREQ ); +if (!ret) + dev_priv->skl_sagv_enabled = true; +else +DRM_ERROR ("Failed to enable the SAGV\n" ); + +/* We don't need to wait for SAGV when enabling */ +mutex_unlock (&dev_priv->rps.hw_lock); +return ret; +} + +static int +skl_do_sagv_disable (struct drm_i915_private *dev_priv) +{ +int ret; +uint32_t temp; + + ret =sandybridge_pcode_write (dev_priv,GEN9_PCODE_SAGV_CONTROL , +GEN9_SAGV_DISABLE ); +if (ret) { +DRM_ERROR ("Failed to disable the SAGV\n" ); +return ret; + } + + ret =sandybridge_pcode_read (dev_priv,GEN9_PCODE_SAGV_CONTROL , + &temp); +if (ret) { +DRM_ERROR ("Failed to check the status of the SAGV\n" ); +return ret; + } + +return temp &0x1 ; +} + +int +skl_disable_sagv (struct drm_i915_private *dev_priv) +{ +int ret, result; + +if (IS_BROXTON (dev_priv)) +return 0 ; +if (!dev_priv->skl_sagv_enabled) +return 0 ; +
drivers/gpu/drm/i915/intel_pm.c skl_wm_plane_id(const struct intel_plane *plane) · line 2986
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +10
-0+mutex_lock (&dev_priv->rps.hw_lock); +DRM_DEBUG_KMS ("Disabling the SAGV\n" ); + +/* bspec says to keep retrying for at least 1 ms */ + ret =wait_for (result =skl_do_sagv_disable (dev_priv),1 ); +mutex_unlock (&dev_priv->rps.hw_lock); + +if (ret == -ETIMEDOUT ) +DRM_ERROR ("Request to disable SAGV timed out\n" ); +else {
Minor style nitpick; if either branch of an if/else needs braces, they both need braces.
drivers/gpu/drm/i915/intel_pm.c skl_wm_plane_id(const struct intel_plane *plane) · line 2987
Quoted 1 line
drivers/gpu/drm/i915/intel_pm.c +1
-0+if (result ==1 )
You've got a few uses of 0x1 as a magic number that are correct, but seem counterintuitive to someone not looking at the bspec (1=off). Maybe replacing these with a #define might help clarify slightly?
Matt
Quoted 24 lines
drivers/gpu/drm/i915/intel_pm.c +11
-0+ dev_priv->skl_sagv_enabled = false; + + ret = result; + } + +return ret; +} + +static void skl_ddb_get_pipe_allocation_limits(struct drm_device *dev, const struct intel_crtc_state *cstate, struct skl_ddb_entry *alloc, /* out */@@ -4236,6 +4346,8 @@ void skl_wm_get_hw_state(struct drm_device *dev)/* Easy/common case; just sanitize DDB now if everything off */ memset (ddb,0 ,sizeof (*ddb)); } + +skl_sagv_get_hw_state (dev_priv); }static void ilk_pipe_wm_get_hw_state (struct drm_crtc *crtc)Signature · 1 line
2.7.4
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v5 3/6] drm/i915/skl: Update plane watermarks atomically during plane updates
In reply to LyudeQuoted 176 lines · Lyude
Thanks to Ville for suggesting this as a potential solution to pipe underruns on Skylake.
On Skylake all of the registers for configuring planes, including the registers for configuring their watermarks, are double buffered. New values written to them won't take effect until said registers are "armed", which is done by writing to the
PLANE_SURF(or in the case of cursor planes, the CURBASE register) register.With this in mind, up until now we've been updating watermarks on skl like this:
non-modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-intel_pre_plane_update:
-intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}or
modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-crtc_enable:
-intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}Now we update watermarks atomically like this:
non-modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-intel_pre_plane_update:
-intel_update_watermarks()(wm values aren't written yet)
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-crtc_enable:
-intel_update_watermarks()(actual wm values aren't written yet)
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}So this patch moves all of the watermark writes into the right place; inside of the vblank evasion where we update all of the registers for each plane. While this patch doesn't fix everything, it does allow us to update the watermark values in the way the hardware expects us to.
Changes since original patch series:
- Remove mutex_lock/mutex_unlock since they don't do anything and we're not touching global state
- Move skl_write_cursor_wm/skl_write_plane_wm functions intointel_pm.c, make externally visible
- Addskl_write_plane_wmcalls toskl_update_plane
- Fix conditional for for loop inskl_write_plane_wm(level <max_levelshould be level <=max_level)
- Make diagram in commit more accurate to what's actually happening
- Add Fixes:Changes since v1:
- UseIS_GEN9()instead ofIS_SKYLAKE()since these fixes apply to more then just Skylake
- Update description to make it clear this patch doesn't fix everything
- Check if pipes were actually changed before writing watermarksChanges since v2:
- WritePIPE_WM_LINETIMEduring vblank evasionChanges since v3:
- Rebase against new SAGV patch changesChanges since v4:
- Add a parameter to choose whatskl_wm_valuesstruct to use when writing new plane watermarksFixes 2d41c0b59afc ("drm/i915/skl: SKL Watermark Computation")Signed-off-by Lyude <[email protected]>Cc Ville Syrjälä <[email protected]>Cc Daniel Vetter <[email protected]>Cc Radhakrishna Sripada <[email protected]>Cc Hans de Goede <[email protected]>Cc Matt Roper <[email protected]>--- drivers/gpu/drm/i915/intel_display.c | 8 +++++ drivers/gpu/drm/i915/intel_drv.h | 5 ++++ drivers/gpu/drm/i915/intel_pm.c | 58 ++++++++++++++++++++++++++---------- drivers/gpu/drm/i915/intel_sprite.c | 6 ++++ 4 files changed, 61 insertions(+), 16 deletions(-)drivers/gpu/drm/i915/intel_display.c +8
-0@@ -2980,6 +2980,7 @@ static void skylake_update_primary_plane(struct drm_plane *plane,struct intel_crtc *intel_crtc =to_intel_crtc (crtc_state->base.crtc);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;u32 plane_ctl, stride_div, stride;u32 tile_height, plane_offset, plane_size;@@ -3031,6 +3032,9 @@ static void skylake_update_primary_plane(struct drm_plane *plane, intel_crtc->adjusted_x = x_offset; intel_crtc->adjusted_y = y_offset; +if (wm->dirty_pipes &drm_crtc_mask (&intel_crtc->base)) +skl_write_plane_wm (intel_crtc, wm,0 ); +I915_WRITE (PLANE_CTL (pipe,0 ), plane_ctl);I915_WRITE (PLANE_OFFSET (pipe,0 ), plane_offset);I915_WRITE (PLANE_SIZE (pipe,0 ), plane_size);@@ -10243,9 +10247,13 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base,struct drm_device *dev = crtc->dev;struct drm_i915_private *dev_priv =to_i915 (dev);struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;uint32_t cntl =0 ; +if (IS_GEN9 (dev_priv) && wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_cursor_wm (intel_crtc, wm); +if (plane_state && plane_state->visible) { cntl =MCURSOR_GAMMA_ENABLE ;switch (plane_state->base.crtc_w) {drivers/gpu/drm/i915/intel_drv.h +5
-0@@ -1711,6 +1711,11 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, struct skl_ddb_allocation *ddb /* out */);int skl_enable_sagv (struct drm_i915_private *dev_priv);int skl_disable_sagv (struct drm_i915_private *dev_priv); +void skl_write_cursor_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm); +void skl_write_plane_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm, +int plane);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);drivers/gpu/drm/i915/intel_pm.c +4
-0@@ -3798,6 +3798,47 @@ static void skl_ddb_entry_write(struct drm_i915_private *dev_priv,I915_WRITE (reg,0 ); } +void skl_write_plane_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm, +int plane) +{
drivers/gpu/drm/i915/intel_pm.c static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, · line 3814
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +10
-0+struct drm_crtc *crtc = &intel_crtc->base; +struct drm_device *dev = crtc->dev; +struct drm_i915_private *dev_priv =to_i915 (dev); +int level, max_level =ilk_wm_max_level (dev); +enum pipe pipe = intel_crtc->pipe; + +if (!(wm->dirty_pipes &drm_crtc_mask (crtc))) +return ; + +I915_WRITE (PIPE_WM_LINETIME (pipe), wm->wm_linetime[pipe]);
I think I mentioned this on the previous version, but if we program this here then it only gets programmed when a plane is being updated. It's possible that we could have the linetime value change due to a scaler update or a modeset, but without corresponding plane updates (e.g., maybe all planes are off). So we probably need to make sure this value gets written outside the plane updates (but still under vblank evasion).
drivers/gpu/drm/i915/intel_pm.c static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, · line 3820
Quoted 6 lines
drivers/gpu/drm/i915/intel_pm.c +6
-0+ +for (level =0 ; level <= max_level; level++) { +I915_WRITE (PLANE_WM (pipe, plane, level), + wm->plane[pipe][plane][level]); + } +I915_WRITE (PLANE_WM_TRANS (pipe, plane), wm->plane_trans[pipe][plane]);
I notice that you moved the DDB update into skl_write_cursor_wm() down below, but didn't move the plane DDB update in here. I realize you remedy that in patch #6 where you finally fix up the flushing order, but the inconsistency between how cursor vs !cursor planes are handled seems confusing. Can we either move the plane DDB update here in this patch (we don't flush properly yet, but we're really no worse off than we already are), or can we defer the move of the cursor DDB to patch 6 to keep them consistent?
Matt
Quoted 82 lines
drivers/gpu/drm/i915/intel_pm.c +22
-16+} + +void skl_write_cursor_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm) +{ +struct drm_crtc *crtc = &intel_crtc->base; +struct drm_device *dev = crtc->dev; +struct drm_i915_private *dev_priv =to_i915 (dev); +int level, max_level =ilk_wm_max_level (dev); +enum pipe pipe = intel_crtc->pipe; + +for (level =0 ; level <= max_level; level++) { +I915_WRITE (CUR_WM (pipe, level), + wm->plane[pipe][PLANE_CURSOR ][level]); + } +I915_WRITE (CUR_WM_TRANS (pipe), wm->plane_trans[pipe][PLANE_CURSOR ]); + +skl_ddb_entry_write (dev_priv,CUR_BUF_CFG (pipe), + &wm->ddb.plane[pipe][PLANE_CURSOR ]); +} +static void skl_write_wm_values (struct drm_i915_private *dev_priv,const struct skl_wm_values *new) {@@ -3805,7 +3846,7 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,struct intel_crtc *crtc;for_each_intel_crtc (dev, crtc) { -int i, level, max_level =ilk_wm_max_level (dev); +int i;enum pipe pipe = crtc->pipe;if ((new->dirty_pipes &drm_crtc_mask (&crtc->base)) ==0 )@@ -3813,21 +3854,6 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,if (!crtc->active)continue ; -I915_WRITE (PIPE_WM_LINETIME (pipe), new->wm_linetime[pipe]); - -for (level =0 ; level <= max_level; level++) { -for (i =0 ; i <intel_num_planes (crtc); i++) -I915_WRITE (PLANE_WM (pipe, i, level), - new->plane[pipe][i][level]); -I915_WRITE (CUR_WM (pipe, level), - new->plane[pipe][PLANE_CURSOR ][level]); - } -for (i =0 ; i <intel_num_planes (crtc); i++) -I915_WRITE (PLANE_WM_TRANS (pipe, i), - new->plane_trans[pipe][i]); -I915_WRITE (CUR_WM_TRANS (pipe), - new->plane_trans[pipe][PLANE_CURSOR ]); -for (i =0 ; i <intel_num_planes (crtc); i++) { skl_ddb_entry_write(dev_priv, PLANE_BUF_CFG(pipe, i),drivers/gpu/drm/i915/intel_sprite.c +6
-0@@ -203,6 +203,9 @@ skl_update_plane(struct drm_plane *drm_plane,struct intel_plane *intel_plane =to_intel_plane (drm_plane);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results; +struct drm_crtc *crtc = crtc_state->base.crtc; +struct intel_crtc *intel_crtc =to_intel_crtc (crtc);const int pipe = intel_plane->pipe;const int plane = intel_plane->plane +1 ;u32 plane_ctl, stride_div, stride;@@ -238,6 +241,9 @@ skl_update_plane(struct drm_plane *drm_plane, crtc_w--; crtc_h--; +if (wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_plane_wm (intel_crtc, wm, plane); +if (key->flags) {I915_WRITE (PLANE_KEYVAL (pipe, plane), key->min_value);I915_WRITE (PLANE_KEYMAX (pipe, plane), key->max_value);Signature · 1 line
2.7.4
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: [PATCH v5 4/6] drm/i915/skl: Ensure pipes with changed wms get added to the state
In reply to LyudeQuoted 3 lines · Lyude
If we're enabling a pipe, we'll need to modify the watermarks on all other active pipes. Since those pipes won't be added to the state on their own, we need to add them ourselves.
All pipes (crtc's) are already added to the state if we have a change in active pipes. I think what you meant to write was "...we'll need to modify the watermarks on all active *planes*. Since those *planes* won't..."
Aside from the commit message, I believe the logic is correct, so you can consider this
once you reword it.
Matt
Quoted 30 lines
Signed-off-by Lyude <[email protected]>Cc Ville Syrjälä <[email protected]>Cc Daniel Vetter <[email protected]>Cc Radhakrishna Sripada <[email protected]>Cc Hans de Goede <[email protected]>Cc Matt Roper <[email protected]>--- drivers/gpu/drm/i915/intel_pm.c | 4 ++++ 1 file changed, 4 insertions(+)drivers/gpu/drm/i915/intel_pm.c +4
-0@@ -4085,6 +4085,10 @@ skl_compute_ddb(struct drm_atomic_state *state) ret =skl_allocate_pipe_ddb (cstate, ddb);if (ret)return ret; + + ret =drm_atomic_add_affected_planes (state, &intel_crtc->base); +if (ret) +return ret; }return 0 ;Signature · 1 line
2.7.4
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v5 5/6] drm/i915: Move CRTC updating in atomic_commit into it's own hook
In reply to LyudeQuoted 60 lines · Lyude
Since we have to write ddb allocations at the same time as we do other plane updates, we're going to need to be able to control the order in which we execute modesets on each pipe. The easiest way to do this is to just factor this section of
intel_atomic_commit_tail()(intel_atomic_commit()for stable branches) into it's own function, and add an appropriate display function hook for it.Based off of Matt Rope's suggestions
Signed-off-by Lyude <[email protected]>[omitting CC for stable, since this patch will need to be changed for such backports first]
Cc: Ville Syrjälä <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Radhakrishna Sripada <[email protected]>
Cc: Hans de Goede <[email protected]>
Cc: Matt Roper <[email protected]>--- drivers/gpu/drm/i915/i915_drv.h | 2 + drivers/gpu/drm/i915/intel_display.c | 74 +++++++++++++++++++++++++----------- 2 files changed, 54 insertions(+), 22 deletions(-)drivers/gpu/drm/i915/i915_drv.h +2
-0@@ -632,6 +632,8 @@ struct drm_i915_display_funcs { struct intel_crtc_state *crtc_state);void (*crtc_enable)(struct drm_crtc *crtc);void (*crtc_disable)(struct drm_crtc *crtc); +void (*update_crtcs)(struct drm_atomic_state *state, +unsigned int *crtc_vblank_mask);void (*audio_codec_enable)(struct drm_connector *connector,struct intel_encoder *encoder,const struct drm_display_mode *adjusted_mode);drivers/gpu/drm/i915/intel_display.c +17
-0@@ -13624,6 +13624,52 @@ static bool needs_vblank_wait(struct intel_crtc_state *crtc_state)return false; } +static void intel_update_crtc (struct drm_crtc *crtc, +struct drm_atomic_state *state, +struct drm_crtc_state *old_crtc_state, +unsigned int *crtc_vblank_mask) +{ +struct drm_device *dev = crtc->dev; +struct drm_i915_private *dev_priv =to_i915 (dev); +struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct intel_crtc_state *pipe_config =to_intel_crtc_state (crtc->state); +bool modeset =needs_modeset (crtc->state); + +if (modeset) { +update_scanline_offset (intel_crtc); + dev_priv->display.crtc_enable (crtc); + }else { +intel_pre_plane_update (to_intel_crtc_state (old_crtc_state)); + }
drivers/gpu/drm/i915/intel_display.c static bool needs_vblank_wait(struct intel_crtc_state *crtc_state) · line 13653
Quoted 10 lines
drivers/gpu/drm/i915/intel_display.c +10
-0+ +if (drm_atomic_get_existing_plane_state (state, crtc->primary)) { +intel_fbc_enable ( + intel_crtc, pipe_config, +to_intel_plane_state (crtc->primary->state)); + } + +drm_atomic_helper_commit_planes_on_crtc (old_crtc_state); + +if (pipe_config->base.active &&needs_vblank_wait (pipe_config))
You can drop the pipe_config->base.active since pipe_config->base.active is just another way of writing crtc->state->active, which you tested before calling this function.
Other than that,
Quoted 78 lines
drivers/gpu/drm/i915/intel_display.c +25
-22+ *crtc_vblank_mask |=drm_crtc_mask (crtc); +} + +static void intel_update_crtcs (struct drm_atomic_state *state, +unsigned int *crtc_vblank_mask) +{ +struct drm_crtc *crtc; +struct drm_crtc_state *old_crtc_state; +int i; + +for_each_crtc_in_state (state, crtc, old_crtc_state, i) { +if (!crtc->state->active) +continue ; + +intel_update_crtc (crtc, state, old_crtc_state, + crtc_vblank_mask); + } +} +static void intel_atomic_commit_tail (struct drm_atomic_state *state) {struct drm_device *dev = state->dev;@@ -13715,17 +13761,9 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)intel_modeset_verify_disabled (dev); } -/* Now enable the clocks, plane, pipe, and connectors that we set up. */ +/* Complete the events for pipes that have now been disabled */ for_each_crtc_in_state (state, crtc, old_crtc_state, i) { -struct intel_crtc *intel_crtc =to_intel_crtc (crtc);bool modeset =needs_modeset (crtc->state); -struct intel_crtc_state *pipe_config = -to_intel_crtc_state (crtc->state); - -if (modeset && crtc->state->active) { -update_scanline_offset (to_intel_crtc (crtc)); - dev_priv->display.crtc_enable (crtc); - }/* Complete events for now disable pipes here. */ if (modeset && !crtc->state->active && crtc->state->event) {@@ -13735,21 +13773,11 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) crtc->state->event =NULL ; } - -if (!modeset) -intel_pre_plane_update (to_intel_crtc_state (old_crtc_state)); - -if (crtc->state->active && -drm_atomic_get_existing_plane_state (state, crtc->primary)) -intel_fbc_enable (intel_crtc, pipe_config,to_intel_plane_state (crtc->primary->state)); - -if (crtc->state->active) -drm_atomic_helper_commit_planes_on_crtc (old_crtc_state); - -if (pipe_config->base.active &&needs_vblank_wait (pipe_config)) - crtc_vblank_mask |=1 << i; } +/* Now enable the clocks, plane, pipe, and connectors that we set up. */ + dev_priv->display.update_crtcs (state, &crtc_vblank_mask); +/* FIXME: We should call drm_atomic_helper_commit_hw_done() here * already, but still need the state for the delayed optimization. To * fix this: @@ -15202,6 +15230,8 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) dev_priv->display.crtc_disable = i9xx_crtc_disable; } + dev_priv->display.update_crtcs = intel_update_crtcs; +/* Returns the core display clock speed */ if (IS_SKYLAKE (dev_priv) ||IS_KABYLAKE (dev_priv)) dev_priv->display.get_display_clock_speed =Signature · 1 line
2.7.4
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Re: [PATCH v5 6/6] drm/i915/skl: Update DDB values atomically with wms/plane attrs
In reply to LyudeQuoted 141 lines · Lyude
Now that we can hook into
update_crtcsand control the order in which we update CRTCs at each modeset, we can finish the final step of fixing Skylake's watermark handling by performing DDB updates at the same time as plane updates and watermark updates.The first major change in this patch is
skl_update_crtcs(), which handles ensuring that we order each CRTC update in our atomic commits properly so that they honor the DDB flush order.The second major change in this patch is the order in which we flush the pipes. While the previous order may have worked, it can't be used in this approach since it no longer will do the right thing. For example, using the old ddb flush order:
We have pipes A, B, and C enabled, and we're disabling C. Initial ddb allocation looks like this:
| A | B |xxxxxxx|
Since we're performing the ddb updates after performing any CRTC disablements in
intel_atomic_commit_tail(), the space to the right of pipe B is unallocated.1. Flush pipes with new allocation contained into old space. None apply, so we skip this
2. Flush pipes having their allocation reduced, but overlapping with a previous allocation. None apply, so we also skip this
3. Flush pipes that got more space allocated. This applies to A and B, giving us the following update order: A, BThis is wrong, since updating pipe A first will cause it to overlap with B and potentially burst into flames. Our new order (see the code comments for details) would update the pipes in the proper order: B, A.
As well, we calculate the order for each DDB update during the check phase, and reference it later in the commit phase when we hit
skl_update_crtcs().This long overdue patch fixes the rest of the underruns on Skylake.
Fixes 0e8fb7ba7ca5 ("drm/i915/skl: Flush the WM configuration")Fixes 8211bd5bdf5e ("drm/i915/skl: Program the DDB allocation")Signed-off-by Lyude <[email protected]>[omitting CC for stable, since this patch will need to be changed for such backports first]
Cc: Ville Syrjälä <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Radhakrishna Sripada <[email protected]>
Cc: Hans de Goede <[email protected]>
Cc: Matt Roper <[email protected]>--- drivers/gpu/drm/i915/intel_display.c | 100 ++++++++++-- drivers/gpu/drm/i915/intel_drv.h | 10 ++ drivers/gpu/drm/i915/intel_pm.c | 287 ++++++++++++++++------------------- 3 files changed, 231 insertions(+), 166 deletions(-)drivers/gpu/drm/i915/intel_display.c +60
-10@@ -12909,16 +12909,23 @@ static void verify_wm_state(struct drm_crtc *crtc, hw_entry->start, hw_entry->end); } -/* cursor */ - hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR ]; - sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR ]; - -if (!skl_ddb_entry_equal (hw_entry, sw_entry)) { -DRM_ERROR ("mismatch in DDB state pipe %c cursor " -"(expected (%u,%u), found (%u,%u))\n" , -pipe_name (pipe), - sw_entry->start, sw_entry->end, - hw_entry->start, hw_entry->end); +/* +* cursor +* If the cursor plane isn't active, we may not have updated it's ddb +* allocation. In that case since the ddb allocation will be updated +* once the plane becomes visible, we can skip this check +*/ +if (intel_crtc->cursor_addr) { + hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR ]; + sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR ]; + +if (!skl_ddb_entry_equal (hw_entry, sw_entry)) { +DRM_ERROR ("mismatch in DDB state pipe %c cursor " +"(expected (%u,%u), found (%u,%u))\n" , +pipe_name (pipe), + sw_entry->start, sw_entry->end, + hw_entry->start, hw_entry->end); + } } }@@ -13670,6 +13677,72 @@ static void intel_update_crtcs(struct drm_atomic_state *state, } } +static inline void +skl_do_ddb_step (struct drm_atomic_state *state, +enum skl_ddb_step step) +{ +struct intel_atomic_state *intel_state =to_intel_atomic_state (state); +struct drm_crtc *crtc; +struct drm_crtc_state *old_crtc_state; +unsigned int crtc_vblank_mask;/* unused */ +int i; + +for_each_crtc_in_state (state, crtc, old_crtc_state, i) { +struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct intel_crtc_state *cstate = +to_intel_crtc_state (crtc->state); +bool vblank_wait = false; + +if (cstate->wm.skl.ddb_realloc != step || !crtc->state->active) +continue ; + +/* +* If we're changing the ddb allocation of this pipe to make +* room for another pipe, we have to wait for the pipe's ddb +* allocations to actually update by waiting for a vblank. +* Otherwise we risk the next pipe updating before this pipe +* finishes, resulting in the pipe fetching from ddb space for +* the wrong pipe. +* +* However, if we know we don't have any more pipes to move +* around, we can skip this wait and the new ddb allocation +* will take effect at the start of the next vblank. +*/ +switch (step) { +case SKL_DDB_STEP_NO_OVERLAP : +case SKL_DDB_STEP_OVERLAP : +if (step != intel_state->last_ddb_step) + vblank_wait = true; + +/* drop through */ +case SKL_DDB_STEP_FINAL : +DRM_DEBUG_KMS ( +"Updating [CRTC:%d:pipe %c] for DDB step %d\n" , + crtc->base.id,pipe_name (intel_crtc->pipe), + step);
drivers/gpu/drm/i915/intel_display.c static void intel_update_crtcs(struct drm_atomic_state *state, · line 13732
Quoted 10 lines
drivers/gpu/drm/i915/intel_display.c +10
-0+ +case SKL_DDB_STEP_NONE : +break ; + } + +intel_update_crtc (crtc, state, old_crtc_state, + &crtc_vblank_mask); + +if (vblank_wait) +intel_wait_for_vblank (state->dev, intel_crtc->pipe);
We only technically need to wait at the end of a step, not on each individual CRTC within a step. Since we only have three pipes total, that usually winds up being the same thing, but the one case we can optimize slightly is when we re-enable pipe B while A and C are already on.
AAAAAACCCCCC (initial state)
AAAA____CCCC (step 1: reduce both A and C)
AAAABBBBCCCC (final step)We just need to wait until both A and C have had a vblank (as intel_atomic_wait_for_vblanks() does) rather than doing each sequentially.
Quoted 332 lines
drivers/gpu/drm/i915/intel_display.c +18
-2+ } +} + +static void skl_update_crtcs (struct drm_atomic_state *state, +unsigned int *crtc_vblank_mask) +{ +struct intel_atomic_state *intel_state =to_intel_atomic_state (state); +enum skl_ddb_step step; + +for (step =0 ; step <= intel_state->last_ddb_step; step++) +skl_do_ddb_step (state, step); +} +static void intel_atomic_commit_tail (struct drm_atomic_state *state) {struct drm_device *dev = state->dev;@@ -15230,8 +15303,6 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) dev_priv->display.crtc_disable = i9xx_crtc_disable; } - dev_priv->display.update_crtcs = intel_update_crtcs; -/* Returns the core display clock speed */ if (IS_SKYLAKE (dev_priv) ||IS_KABYLAKE (dev_priv)) dev_priv->display.get_display_clock_speed =@@ -15321,6 +15392,11 @@ void intel_init_display_hooks(struct drm_i915_private *dev_priv) skl_modeset_calc_cdclk; } +if (dev_priv->info.gen >=9 ) + dev_priv->display.update_crtcs = skl_update_crtcs; +else + dev_priv->display.update_crtcs = intel_update_crtcs; +switch (INTEL_INFO (dev_priv)->gen) {case 2 : dev_priv->display.queue_flip = intel_gen2_queue_flip;drivers/gpu/drm/i915/intel_drv.h +10
-0@@ -334,6 +334,7 @@ struct intel_atomic_state {/* Gen9+ only */ struct skl_wm_values wm_results; +int last_ddb_step; };struct intel_plane_state {@@ -437,6 +438,13 @@ struct skl_pipe_wm {uint32_t linetime; }; +enum skl_ddb_step { +SKL_DDB_STEP_NONE =0 , +SKL_DDB_STEP_NO_OVERLAP , +SKL_DDB_STEP_OVERLAP , +SKL_DDB_STEP_FINAL +}; + struct intel_crtc_wm_state { union { struct {@@ -467,6 +475,8 @@ struct intel_crtc_wm_state {/* minimum block allocation */ uint16_t minimum_blocks[I915_MAX_PLANES ];uint16_t minimum_y_blocks[I915_MAX_PLANES ]; + +enum skl_ddb_step ddb_realloc; } skl; };drivers/gpu/drm/i915/intel_pm.c +53
-152@@ -3818,6 +3818,11 @@ void skl_write_plane_wm(struct intel_crtc *intel_crtc, wm->plane[pipe][plane][level]); }I915_WRITE (PLANE_WM_TRANS (pipe, plane), wm->plane_trans[pipe][plane]); + +skl_ddb_entry_write (dev_priv,PLANE_BUF_CFG (pipe, plane), + &wm->ddb.plane[pipe][plane]); +skl_ddb_entry_write (dev_priv,PLANE_NV12_BUF_CFG (pipe, plane), + &wm->ddb.y_plane[pipe][plane]); } void skl_write_cursor_wm(struct intel_crtc *intel_crtc,@@ -3839,168 +3844,46 @@ void skl_write_cursor_wm(struct intel_crtc *intel_crtc, &wm->ddb.plane[pipe][PLANE_CURSOR]); } -static void skl_write_wm_values (struct drm_i915_private *dev_priv, -const struct skl_wm_values *new) -{ -struct drm_device *dev = &dev_priv->drm; -struct intel_crtc *crtc; - -for_each_intel_crtc (dev, crtc) { -int i; -enum pipe pipe = crtc->pipe; - -if ((new->dirty_pipes &drm_crtc_mask (&crtc->base)) ==0 ) -continue ; -if (!crtc->active) -continue ; - -for (i =0 ; i <intel_num_planes (crtc); i++) { -skl_ddb_entry_write (dev_priv, -PLANE_BUF_CFG (pipe, i), - &new->ddb.plane[pipe][i]); -skl_ddb_entry_write (dev_priv, -PLANE_NV12_BUF_CFG (pipe, i), - &new->ddb.y_plane[pipe][i]); - } - -skl_ddb_entry_write (dev_priv,CUR_BUF_CFG (pipe), - &new->ddb.plane[pipe][PLANE_CURSOR ]); - } -} - -/* -* When setting up a new DDB allocation arrangement, we need to correctly -* sequence the times at which the new allocations for the pipes are taken into -* account or we'll have pipes fetching from space previously allocated to -* another pipe. -* -* Roughly the sequence looks like: -* 1. re-allocate the pipe(s) with the allocation being reduced and not -* overlapping with a previous light-up pipe (another way to put it is: -* pipes with their new allocation strickly included into their old ones). -* 2. re-allocate the other pipes that get their allocation reduced -* 3. allocate the pipes having their allocation increased -* -* Steps 1. and 2. are here to take care of the following case: -* - Initially DDB looks like this: -* | B | C | -* - enable pipe A. -* - pipe B has a reduced DDB allocation that overlaps with the old pipe C -* allocation -* | A | B | C | -* -* We need to sequence the re-allocation: C, B, A (and not B, C, A). -*/ - -static void -skl_wm_flush_pipe (struct drm_i915_private *dev_priv,enum pipe pipe,int pass) +static bool +skl_ddb_allocation_equals (const struct skl_ddb_allocation *old, +const struct skl_ddb_allocation *new, +enum pipe pipe) { -int plane; - -DRM_DEBUG_KMS ("flush pipe %c (pass %d)\n" ,pipe_name (pipe), pass); - -for_each_plane (dev_priv, pipe, plane) { -I915_WRITE (PLANE_SURF (pipe, plane), -I915_READ (PLANE_SURF (pipe, plane))); - } -I915_WRITE (CURBASE (pipe),I915_READ (CURBASE (pipe))); +return new->pipe[pipe].start == old->pipe[pipe].start && + new->pipe[pipe].end == old->pipe[pipe].end; }static bool -skl_ddb_allocation_included (const struct skl_ddb_allocation *old, +skl_ddb_allocation_overlaps (struct drm_atomic_state *state, +const struct skl_ddb_allocation *old,const struct skl_ddb_allocation *new,enum pipe pipe) { -uint16_t old_size, new_size; - - old_size =skl_ddb_entry_size (&old->pipe[pipe]); - new_size =skl_ddb_entry_size (&new->pipe[pipe]); - -return old_size != new_size && - new->pipe[pipe].start >= old->pipe[pipe].start && - new->pipe[pipe].end <= old->pipe[pipe].end; -} - -static void skl_flush_wm_values (struct drm_i915_private *dev_priv, -struct skl_wm_values *new_values) -{ -struct drm_device *dev = &dev_priv->drm; -struct skl_ddb_allocation *cur_ddb, *new_ddb; -bool reallocated[I915_MAX_PIPES ] = {}; -struct intel_crtc *crtc; -enum pipe pipe; - - new_ddb = &new_values->ddb; - cur_ddb = &dev_priv->wm.skl_hw.ddb; - -/* -* First pass: flush the pipes with the new allocation contained into -* the old space. -* -* We'll wait for the vblank on those pipes to ensure we can safely -* re-allocate the freed space without this pipe fetching from it. -*/ -for_each_intel_crtc (dev, crtc) { -if (!crtc->active) -continue ; - - pipe = crtc->pipe; - -if (!skl_ddb_allocation_included (cur_ddb, new_ddb, pipe)) -continue ; - -skl_wm_flush_pipe (dev_priv, pipe,1 ); -intel_wait_for_vblank (dev, pipe); - - reallocated[pipe] = true; - } - - -/* -* Second pass: flush the pipes that are having their allocation -* reduced, but overlapping with a previous allocation. -* -* Here as well we need to wait for the vblank to make sure the freed -* space is not used anymore. -*/ -for_each_intel_crtc (dev, crtc) { -if (!crtc->active) -continue ; - - pipe = crtc->pipe; - -if (reallocated[pipe]) -continue ; - -if (skl_ddb_entry_size (&new_ddb->pipe[pipe]) < -skl_ddb_entry_size (&cur_ddb->pipe[pipe])) { -skl_wm_flush_pipe (dev_priv, pipe,2 ); -intel_wait_for_vblank (dev, pipe); - reallocated[pipe] = true; - } - } - -/* -* Third pass: flush the pipes that got more space allocated. -* -* We don't need to actively wait for the update here, next vblank -* will just get more DDB space with the correct WM values. -*/ -for_each_intel_crtc (dev, crtc) { -if (!crtc->active) -continue ; +struct drm_device *dev = state->dev; +struct intel_crtc *intel_crtc; +enum pipe otherp; - pipe = crtc->pipe; +for_each_intel_crtc (dev, intel_crtc) { + otherp = intel_crtc->pipe;/* -* At this point, only the pipes more space than before are -* left to re-allocate. +* When checking for overlaps, we don't want to: +* - Compare against ourselves +* - Compare against pipes that will be disabled in step 0 +* - Compare against pipes that won't be enabled until step 3 */ -if (reallocated[pipe]) +if (otherp == pipe || !new->pipe[otherp].end || + !old->pipe[otherp].end)continue ; -skl_wm_flush_pipe (dev_priv, pipe,3 ); +if ((new->pipe[pipe].start >= old->pipe[otherp].start && + new->pipe[pipe].start < old->pipe[otherp].end) || + (old->pipe[otherp].start >= new->pipe[pipe].start && + old->pipe[otherp].start < new->pipe[pipe].end)) +return true; } + +return false; } static int skl_update_pipe_wm(struct drm_crtc_state *cstate,@@ -4043,8 +3926,10 @@ skl_compute_ddb(struct drm_atomic_state *state)struct drm_device *dev = state->dev;struct drm_i915_private *dev_priv =to_i915 (dev);struct intel_atomic_state *intel_state =to_intel_atomic_state (state); +struct intel_crtc_state *cstate;struct intel_crtc *intel_crtc; -struct skl_ddb_allocation *ddb = &intel_state->wm_results.ddb; +struct skl_ddb_allocation *old_ddb = &dev_priv->wm.skl_hw.ddb; +struct skl_ddb_allocation *new_ddb = &intel_state->wm_results.ddb;uint32_t realloc_pipes =pipes_modified (state);int ret;@@ -4076,13 +3961,11 @@ skl_compute_ddb(struct drm_atomic_state *state) }for_each_intel_crtc_mask (dev, intel_crtc, realloc_pipes) { -struct intel_crtc_state *cstate; - cstate =intel_atomic_get_crtc_state (state, intel_crtc);if (IS_ERR (cstate))return PTR_ERR (cstate); - ret =skl_allocate_pipe_ddb (cstate, ddb); + ret =skl_allocate_pipe_ddb (cstate, new_ddb);if (ret)return ret;@@ -4091,6 +3974,73 @@ skl_compute_ddb(struct drm_atomic_state *state)return ret; } +/* +* When setting up a new DDB allocation arrangement, we need to +* correctly sequence the times at which the new allocations for the +* pipes are taken into account or we'll have pipes fetching from space +* previously allocated to another pipe. +* +* Roughly the final sequence we want looks like this: +* 1. Disable any pipes we're not going to be using anymore +* 2. Reallocate all of the active pipes whose new ddb allocations +* won't overlap with another active pipe's ddb allocation. +* 3. Reallocate remaining active pipes, if any. +* 4. Enable any new pipes, if any. +* +* Example: +* Initially DDB looks like this: +* | B | C | +* And the final DDB should look like this: +* | B | C | A |
drivers/gpu/drm/i915/intel_pm.c skl_compute_ddb(struct drm_atomic_state *state) · line 4004
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +10
-0+* +* 1. We're not disabling any pipes, so do nothing on this step. +* 2. Pipe B's new allocation wouldn't overlap with pipe C, however +* pipe C's new allocation does overlap with pipe B's current +* allocation. Reallocate B first so the DDB looks like this: +* | B |xx| C | +* 3. Now we can safely reallocate pipe C to it's new location: +* | B | C |xxxxx| +* 4. Enable any remaining pipes, in this case A +* | B | C | A |
It's worth noting that this sequence will work fine for the three pipe case we have in gen9 since when adding or removing a pipe we never have a situation where multiple pipes are processed in the 'overlap' step (step 3). But if some future platform decides to add more pipes while maintaining the same general DDB programming style, this will break down since we'll need to ensure intelligent ordering of the pipes being processed in step 3. E.g., with naive ordering of step 3 we could have
AAABBBCCCDDD -> AAAABBBBCCCC:
AAABBBCCC___ (step 1)
AAABBB__CCCC (step 2)
AAA!BB__CCCC (step 3a, woops)
Quoted 96 lines
drivers/gpu/drm/i915/intel_pm.c +60
-2+* +* As well, between every pipe reallocation we have to wait for a +* vblank on the pipe so that we ensure it's new allocation has taken +* effect by the time we start moving the next pipe. This can be +* skipped on the last step we need to perform, which is why we keep +* track of that information here. For example, if we've reallocated +* all the pipes that need changing by the time we reach step 3, we can +* finish without waiting for the pipes we changed in step 3 to update. +*/ +for_each_intel_crtc_mask (dev, intel_crtc, realloc_pipes) { +enum pipe pipe = intel_crtc->pipe; +enum skl_ddb_step step; + + cstate =intel_atomic_get_crtc_state (state, intel_crtc); +if (IS_ERR (cstate)) +return PTR_ERR (cstate); + +/* Step 1: Pipes we're disabling / haven't changed */ +if (skl_ddb_allocation_equals (old_ddb, new_ddb, pipe) || + new_ddb->pipe[pipe].end ==0 ) { + step =SKL_DDB_STEP_NONE ; +/* Step 2-3: Active pipes we're reallocating */ + }else if (old_ddb->pipe[pipe].end !=0 ) { +if (skl_ddb_allocation_overlaps (state, old_ddb, new_ddb, + pipe)) + step =SKL_DDB_STEP_OVERLAP ; +else + step =SKL_DDB_STEP_NO_OVERLAP ; +/* Step 4: Pipes we're enabling */ + }else { + step =SKL_DDB_STEP_FINAL ; + } + + cstate->wm.skl.ddb_realloc = step; + +if (step > intel_state->last_ddb_step) + intel_state->last_ddb_step = step; + } +return 0 ; }@@ -4115,10 +4065,13 @@ skl_copy_wm_for_pipe(struct skl_wm_values *dst,static int skl_compute_wm (struct drm_atomic_state *state) { +struct drm_i915_private *dev_priv =to_i915 (state->dev);struct drm_crtc *crtc;struct drm_crtc_state *cstate;struct intel_atomic_state *intel_state =to_intel_atomic_state (state);struct skl_wm_values *results = &intel_state->wm_results; +struct skl_ddb_allocation *old_ddb = &dev_priv->wm.skl_hw.ddb; +struct skl_ddb_allocation *new_ddb = &results->ddb;struct skl_pipe_wm *pipe_wm;bool changed = false;int ret, i;@@ -4157,7 +4110,10 @@ skl_compute_wm(struct drm_atomic_state *state)struct intel_crtc *intel_crtc =to_intel_crtc (crtc);struct intel_crtc_state *intel_cstate =to_intel_crtc_state (cstate); +enum skl_ddb_step step; +enum pipe pipe; + pipe = intel_crtc->pipe; pipe_wm = &intel_cstate->wm.skl.optimal; ret =skl_update_pipe_wm (cstate, &results->ddb, pipe_wm, &changed);@@ -4172,7 +4128,18 @@ skl_compute_wm(struct drm_atomic_state *state)continue ; intel_cstate->update_wm_pre = true; + step = intel_cstate->wm.skl.ddb_realloc;skl_compute_wm_results (crtc->dev, pipe_wm, results, intel_crtc); + +if (!skl_ddb_entry_equal (&old_ddb->pipe[pipe], + &new_ddb->pipe[pipe])) { +DRM_DEBUG_KMS ( +"DDB changes for [CRTC:%d:pipe %c]: (%3d - %3d) -> (%3d - %3d) on step %d\n" , + intel_crtc->base.base.id,pipe_name (pipe), + old_ddb->pipe[pipe].start, old_ddb->pipe[pipe].end, + new_ddb->pipe[pipe].start, new_ddb->pipe[pipe].end, + step); + } }return 0 ;@@ -4196,8 +4163,20 @@ static void skl_update_wm(struct drm_crtc *crtc)mutex_lock (&dev_priv->wm.wm_mutex); -skl_write_wm_values (dev_priv, results); -skl_flush_wm_values (dev_priv, results); +/* +* If this pipe isn't active already, we're going to be enabling it +* very soon. Since it's safe to update these while the pipe's shut off, +* just do so here. Already active pipes will have their watermarks
drivers/gpu/drm/i915/intel_pm.c static void skl_update_wm(struct drm_crtc *crtc) · line 4179
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +10
-0+* updated once we update their planes. +*/ +if (!intel_crtc->active) { +int plane; + +for (plane =0 ; plane <intel_num_planes (intel_crtc); plane++) +skl_write_plane_wm (intel_crtc, results, plane); + +skl_write_cursor_wm (intel_crtc, results); + }
Do we actually need this? Even if the pipe is in the process of being enabled, all of these registers should still be programmed at the same time as the plane registers, right? Those register writes are based on crtc->state->active rather than crtc->active so they should happen when they need to.
On the other hand, if we write these when the pipe is off and isn't in the process of being enabled, then we may not have the power well turned on, so we'll get unclaimed register warnings.
Matt
Quoted 6 lines
drivers/gpu/drm/i915/intel_pm.c +0
-0/* * Store the new configuration (but only for the pipes that have Signature · 1 line
2.7.4
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Matt Roper
Reviewed-byRe: [PATCH v5 3/6] drm/i915/skl: Update plane watermarks atomically during plane updates
In reply to Matt RoperQuoted 98 lines · Matt Roper
Quoted 97 lines · Lyude
Thanks to Ville for suggesting this as a potential solution to pipe underruns on Skylake.
On Skylake all of the registers for configuring planes, including the registers for configuring their watermarks, are double buffered. New values written to them won't take effect until said registers are "armed", which is done by writing to the
PLANE_SURF(or in the case of cursor planes, the CURBASE register) register.With this in mind, up until now we've been updating watermarks on skl like this:
non-modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-intel_pre_plane_update:
-intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}or
modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-crtc_enable:
-intel_update_watermarks()
- {vblank happens; new watermarks + old plane values => underrun }
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- end vblank evasion
}Now we update watermarks atomically like this:
non-modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-intel_pre_plane_update:
-intel_update_watermarks()(wm values aren't written yet)
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}modeset {
- calculate (during atomic check phase)
-finish_atomic_commit:
-crtc_enable:
-intel_update_watermarks()(actual wm values aren't written yet)
-drm_atomic_helper_commit_planes_on_crtc:
- start vblank evasion
- write new plane registers
- write new wm values
- end vblank evasion
}So this patch moves all of the watermark writes into the right place; inside of the vblank evasion where we update all of the registers for each plane. While this patch doesn't fix everything, it does allow us to update the watermark values in the way the hardware expects us to.
Changes since original patch series:
- Remove mutex_lock/mutex_unlock since they don't do anything and we're not touching global state
- Move skl_write_cursor_wm/skl_write_plane_wm functions intointel_pm.c, make externally visible
- Addskl_write_plane_wmcalls toskl_update_plane
- Fix conditional for for loop inskl_write_plane_wm(level <max_levelshould be level <=max_level)
- Make diagram in commit more accurate to what's actually happening
- Add Fixes:Changes since v1:
- UseIS_GEN9()instead ofIS_SKYLAKE()since these fixes apply to more then just Skylake
- Update description to make it clear this patch doesn't fix everything
- Check if pipes were actually changed before writing watermarksChanges since v2:
- WritePIPE_WM_LINETIMEduring vblank evasionChanges since v3:
- Rebase against new SAGV patch changesChanges since v4:
- Add a parameter to choose whatskl_wm_valuesstruct to use when writing new plane watermarks
Quoted 8 lines
Quoted 8 lines
Fixes 2d41c0b59afc ("drm/i915/skl: SKL Watermark Computation")Signed-off-by Lyude <[email protected]>Cc Ville Syrjälä <[email protected]>Cc Daniel Vetter <[email protected]>Cc Radhakrishna Sripada <[email protected]>Cc Hans de Goede <[email protected]>Cc Matt Roper <[email protected]>
I imagine we'll eventually probably want to create a new display vfunc to handle platform-specific pipe-level stuff that needs to happen under vblank evasion (like the scalers and linetime WM we have today) to keep the code clean; maybe add a TODO comment to that effect?
Anyway, this looks good enough for now, so
Quoted 197 lines
Quoted 71 lines
--- drivers/gpu/drm/i915/intel_display.c | 8 +++++ drivers/gpu/drm/i915/intel_drv.h | 5 ++++ drivers/gpu/drm/i915/intel_pm.c | 58 ++++++++++++++++++++++++++---------- drivers/gpu/drm/i915/intel_sprite.c | 6 ++++ 4 files changed, 61 insertions(+), 16 deletions(-)drivers/gpu/drm/i915/intel_display.c +8
-0@@ -2980,6 +2980,7 @@ static void skylake_update_primary_plane(struct drm_plane *plane,struct intel_crtc *intel_crtc =to_intel_crtc (crtc_state->base.crtc);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;u32 plane_ctl, stride_div, stride;u32 tile_height, plane_offset, plane_size;@@ -3031,6 +3032,9 @@ static void skylake_update_primary_plane(struct drm_plane *plane, intel_crtc->adjusted_x = x_offset; intel_crtc->adjusted_y = y_offset; +if (wm->dirty_pipes &drm_crtc_mask (&intel_crtc->base)) +skl_write_plane_wm (intel_crtc, wm,0 ); +I915_WRITE (PLANE_CTL (pipe,0 ), plane_ctl);I915_WRITE (PLANE_OFFSET (pipe,0 ), plane_offset);I915_WRITE (PLANE_SIZE (pipe,0 ), plane_size);@@ -10243,9 +10247,13 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base,struct drm_device *dev = crtc->dev;struct drm_i915_private *dev_priv =to_i915 (dev);struct intel_crtc *intel_crtc =to_intel_crtc (crtc); +struct skl_wm_values *wm = &dev_priv->wm.skl_results;int pipe = intel_crtc->pipe;uint32_t cntl =0 ; +if (IS_GEN9 (dev_priv) && wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_cursor_wm (intel_crtc, wm); +if (plane_state && plane_state->visible) { cntl =MCURSOR_GAMMA_ENABLE ;switch (plane_state->base.crtc_w) {drivers/gpu/drm/i915/intel_drv.h +5
-0@@ -1711,6 +1711,11 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, struct skl_ddb_allocation *ddb /* out */);int skl_enable_sagv (struct drm_i915_private *dev_priv);int skl_disable_sagv (struct drm_i915_private *dev_priv); +void skl_write_cursor_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm); +void skl_write_plane_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm, +int plane);uint32_t ilk_pipe_pixel_rate (const struct intel_crtc_state *pipe_config);bool ilk_disable_lp_wm (struct drm_device *dev);int sanitize_rc6_option (struct drm_i915_private *dev_priv,int enable_rc6);drivers/gpu/drm/i915/intel_pm.c +4
-0@@ -3798,6 +3798,47 @@ static void skl_ddb_entry_write(struct drm_i915_private *dev_priv,I915_WRITE (reg,0 ); } +void skl_write_plane_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm, +int plane) +{drivers/gpu/drm/i915/intel_pm.c static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, · line 3814
Quoted 10 lines
drivers/gpu/drm/i915/intel_pm.c +10
-0+struct drm_crtc *crtc = &intel_crtc->base; +struct drm_device *dev = crtc->dev; +struct drm_i915_private *dev_priv =to_i915 (dev); +int level, max_level =ilk_wm_max_level (dev); +enum pipe pipe = intel_crtc->pipe; + +if (!(wm->dirty_pipes &drm_crtc_mask (crtc))) +return ; + +I915_WRITE (PIPE_WM_LINETIME (pipe), wm->wm_linetime[pipe]);I think I mentioned this on the previous version, but if we program this here then it only gets programmed when a plane is being updated. It's possible that we could have the linetime value change due to a scaler update or a modeset, but without corresponding plane updates (e.g., maybe all planes are off). So we probably need to make sure this value gets written outside the plane updates (but still under vblank evasion).
drivers/gpu/drm/i915/intel_pm.c static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, · line 3820
Quoted 6 lines
drivers/gpu/drm/i915/intel_pm.c +6
-0+ +for (level =0 ; level <= max_level; level++) { +I915_WRITE (PLANE_WM (pipe, plane, level), + wm->plane[pipe][plane][level]); + } +I915_WRITE (PLANE_WM_TRANS (pipe, plane), wm->plane_trans[pipe][plane]);I notice that you moved the DDB update into
skl_write_cursor_wm()down below, but didn't move the plane DDB update in here. I realize you remedy that in patch #6 where you finally fix up the flushing order, but the inconsistency between how cursor vs !cursor planes are handled seems confusing. Can we either move the plane DDB update here in this patch (we don't flush properly yet, but we're really no worse off than we already are), or can we defer the move of the cursor DDB to patch 6 to keep them consistent?Matt
Quoted 82 lines
drivers/gpu/drm/i915/intel_pm.c +22
-16+} + +void skl_write_cursor_wm (struct intel_crtc *intel_crtc, +const struct skl_wm_values *wm) +{ +struct drm_crtc *crtc = &intel_crtc->base; +struct drm_device *dev = crtc->dev; +struct drm_i915_private *dev_priv =to_i915 (dev); +int level, max_level =ilk_wm_max_level (dev); +enum pipe pipe = intel_crtc->pipe; + +for (level =0 ; level <= max_level; level++) { +I915_WRITE (CUR_WM (pipe, level), + wm->plane[pipe][PLANE_CURSOR ][level]); + } +I915_WRITE (CUR_WM_TRANS (pipe), wm->plane_trans[pipe][PLANE_CURSOR ]); + +skl_ddb_entry_write (dev_priv,CUR_BUF_CFG (pipe), + &wm->ddb.plane[pipe][PLANE_CURSOR ]); +} +static void skl_write_wm_values (struct drm_i915_private *dev_priv,const struct skl_wm_values *new) {@@ -3805,7 +3846,7 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,struct intel_crtc *crtc;for_each_intel_crtc (dev, crtc) { -int i, level, max_level =ilk_wm_max_level (dev); +int i;enum pipe pipe = crtc->pipe;if ((new->dirty_pipes &drm_crtc_mask (&crtc->base)) ==0 )@@ -3813,21 +3854,6 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,if (!crtc->active)continue ; -I915_WRITE (PIPE_WM_LINETIME (pipe), new->wm_linetime[pipe]); - -for (level =0 ; level <= max_level; level++) { -for (i =0 ; i <intel_num_planes (crtc); i++) -I915_WRITE (PLANE_WM (pipe, i, level), - new->plane[pipe][i][level]); -I915_WRITE (CUR_WM (pipe, level), - new->plane[pipe][PLANE_CURSOR ][level]); - } -for (i =0 ; i <intel_num_planes (crtc); i++) -I915_WRITE (PLANE_WM_TRANS (pipe, i), - new->plane_trans[pipe][i]); -I915_WRITE (CUR_WM_TRANS (pipe), - new->plane_trans[pipe][PLANE_CURSOR ]); -for (i =0 ; i <intel_num_planes (crtc); i++) { skl_ddb_entry_write(dev_priv, PLANE_BUF_CFG(pipe, i),drivers/gpu/drm/i915/intel_sprite.c +6
-0@@ -203,6 +203,9 @@ skl_update_plane(struct drm_plane *drm_plane,struct intel_plane *intel_plane =to_intel_plane (drm_plane);struct drm_framebuffer *fb = plane_state->base.fb;struct drm_i915_gem_object *obj =intel_fb_obj (fb); +struct skl_wm_values *wm = &dev_priv->wm.skl_results; +struct drm_crtc *crtc = crtc_state->base.crtc; +struct intel_crtc *intel_crtc =to_intel_crtc (crtc);const int pipe = intel_plane->pipe;const int plane = intel_plane->plane +1 ;u32 plane_ctl, stride_div, stride;@@ -238,6 +241,9 @@ skl_update_plane(struct drm_plane *drm_plane, crtc_w--; crtc_h--; +if (wm->dirty_pipes &drm_crtc_mask (crtc)) +skl_write_plane_wm (intel_crtc, wm, plane); +if (key->flags) {I915_WRITE (PLANE_KEYVAL (pipe, plane), key->min_value);I915_WRITE (PLANE_KEYMAX (pipe, plane), key->max_value);Signature · 1 line
2.7.4Signature · 5 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795
Signature · 9 lines
Matt Roper Graphics Software Engineer IoTG Platform Enabling & Development Intel Corporation (916) 356-2795 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v5 3/6] drm/i915/skl: Update plane watermarks atomically during plane updates
On Wed, Aug 03, 2016 at 02:14:53PM -0700, Matt Roper wrote:
...
Quoted 9 lines
I imagine we'll eventually probably want to create a new display vfunc to handle platform-specific pipe-level stuff that needs to happen under vblank evasion (like the scalers and linetime WM we have today) to keep the code clean; maybe add a TODO comment to that effect?
Anyway, this looks good enough for now, so
Reviewed-by Matt Roper <[email protected]>
Woops, meant to add this reply to your v6 version of the patch obviously...
Matt
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx