Skip to contentSkip to conversation

drm/bridge: samsung-dsim: Set P divider based on min/max of fin pll

PATCH Superseded Adam Ford 2 participants 8 messages Dec 12, 2023 – Feb 8, 2024

Subsystems (4)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Adam Ford

author
#0 lore raw

PATCH 1/2 drm/bridge: samsung-dsim: Set P divider based on min/max of fin pll

The P divider should be set based on the min and max values of the fin pll which may vary between different platforms. These ranges are defined per platform, but hard-coded values were used instead which resulted in a smaller range available on the i.MX8M[MNP] than what was possible.

Fixes 846307185f0f ("drm/bridge: samsung-dsim: update PLL reference clock")
drivers/gpu/drm/bridge/samsung-dsim.c +2 -2
index be5914caa17d..239d253a7d71 100644
@@ -573,8 +573,8 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi,
 	u16 _m, best_m;
 	u8 _s, best_s;
 
-	p_min = DIV_ROUND_UP(fin, (12 * MHZ));
-	p_max = fin / (6 * MHZ);
+	p_min = DIV_ROUND_UP(fin, (driver_data->pll_fin_max * MHZ));
+	p_max = fin / (driver_data->pll_fin_min * MHZ);
 
 	for (_p = p_min; _p <= p_max; ++_p) {
 		for (_s = 0; _s <= 5; ++_s) {
Signature · 1 line
2.40.1
author
#1 lore raw

PATCH 2/2 drm/bridge: samsung-dsim: Fix porch calcalcuation rounding

When using video sync pulses, the HFP, HBP, and HSA are divided between the available lanes if there is more than one lane. For certain timings and lane configurations, the HFP may not be evenly divisible. If the HFP is rounded down, it ends up being too small which can cause some monitors to not sync properly. In these instances, adjust htotal and hsync to round the HFP up, and recalculate the htotal.

drivers/gpu/drm/bridge/samsung-dsim.c +21 -0
index 239d253a7d71..f5795da1d8bb 100644
@@ -1628,6 +1628,27 @@ static int samsung_dsim_atomic_check(struct drm_bridge *bridge,
 		adjusted_mode->flags |= (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
 	}
 
+	/*
+	 * When using video sync pulses, the HFP, HBP, and HSA are divided between
+	 * the available lanes if there is more than one lane.  For certain
+	 * timings and lane configurations, the HFP may not be evenly divisible.
+	 * If the HFP is rounded down, it ends up being too small which can cause
+	 * some monitors to not sync properly. In these instances, adjust htotal
+	 * and hsync to round the HFP up, and recalculate the htotal. Through trial
+	 * and error, it appears that the HBP and HSA do not appearto need the same
+	 * correction that HFP does.
+	 */
+	if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE && dsi->lanes > 1) {
+		int hfp = adjusted_mode->hsync_start - adjusted_mode->hdisplay;
+		int remainder = hfp % dsi->lanes;
+
+		if (remainder) {
+			adjusted_mode->hsync_start += remainder;
+			adjusted_mode->hsync_end   += remainder;
+			adjusted_mode->htotal      += remainder;
+		}
+	}
+
 	return 0;
 }
 
Signature · 1 line
2.40.1
author
#2 lore raw

Re: [PATCH 2/2] drm/bridge: samsung-dsim: Fix porch calcalcuation rounding

Quoted 8 lines · Adam Ford

When using video sync pulses, the HFP, HBP, and HSA are divided between the available lanes if there is more than one lane. For certain timings and lane configurations, the HFP may not be evenly divisible. If the HFP is rounded down, it ends up being too small which can cause some monitors to not sync properly. In these instances, adjust htotal and hsync to round the HFP up, and recalculate the htotal.

For anyone who's following this, I added a note which I apparently forgot to save:

This adds support for 720p60 in the i.MX8M Plus.

NXP uses a look-up table in their downstream code to accomplish this. Using this calculation, the driver can adjust without the need for a complicated table and should be flexible for different timings and resolutions depending on the monitor.

I don't have a DSI analyzer, and this appears to only work on i.MX8M Plus but not Mini and Nano for some reason, despite their having a similar DSI bridge.

When Frieder teste this, he reported no changes on the Kontrol BL
i.MX8MM: "So at least there is no negative impact in this case"

If someone else has an i.MX8MP, I would appreciate any feedback.

thanks

adam

Quoted 38 lines
drivers/gpu/drm/bridge/samsung-dsim.c +21 -1
index 239d253a7d71..f5795da1d8bb 100644
@@ -1628,6 +1628,27 @@ static int samsung_dsim_atomic_check(struct drm_bridge *bridge,
                adjusted_mode->flags |= (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
        }
 
+       /*
+        * When using video sync pulses, the HFP, HBP, and HSA are divided between
+        * the available lanes if there is more than one lane.  For certain
+        * timings and lane configurations, the HFP may not be evenly divisible.
+        * If the HFP is rounded down, it ends up being too small which can cause
+        * some monitors to not sync properly. In these instances, adjust htotal
+        * and hsync to round the HFP up, and recalculate the htotal. Through trial
+        * and error, it appears that the HBP and HSA do not appearto need the same
+        * correction that HFP does.
+        */
+       if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE && dsi->lanes > 1) {
+               int hfp = adjusted_mode->hsync_start - adjusted_mode->hdisplay;
+               int remainder = hfp % dsi->lanes;
+
+               if (remainder) {
+                       adjusted_mode->hsync_start += remainder;
+                       adjusted_mode->hsync_end   += remainder;
+                       adjusted_mode->htotal      += remainder;
+               }
+       }
+
        return 0;
 }
 
--

2.40.1

author
#3 lore raw

Re: [PATCH 2/2] drm/bridge: samsung-dsim: Fix porch calcalcuation rounding

In reply to Adam Ford
Quoted 10 lines · Adam Ford

When using video sync pulses, the HFP, HBP, and HSA are divided between the available lanes if there is more than one lane. For certain timings and lane configurations, the HFP may not be evenly divisible. If the HFP is rounded down, it ends up being too small which can cause some monitors to not sync properly. In these instances, adjust htotal and hsync to round the HFP up, and recalculate the htotal.

Gentle nudge on this one. Basically this fixes an issue with the 8MP, but it's still unknown why it doesn't work on 8MM or 8MN, but Frieder confirmed there are no regressions on 8MM or 8MN.

adam

Quoted 36 lines
drivers/gpu/drm/bridge/samsung-dsim.c +21 -1
index 239d253a7d71..f5795da1d8bb 100644
@@ -1628,6 +1628,27 @@ static int samsung_dsim_atomic_check(struct drm_bridge *bridge,
                adjusted_mode->flags |= (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
        }
 
+       /*
+        * When using video sync pulses, the HFP, HBP, and HSA are divided between
+        * the available lanes if there is more than one lane.  For certain
+        * timings and lane configurations, the HFP may not be evenly divisible.
+        * If the HFP is rounded down, it ends up being too small which can cause
+        * some monitors to not sync properly. In these instances, adjust htotal
+        * and hsync to round the HFP up, and recalculate the htotal. Through trial
+        * and error, it appears that the HBP and HSA do not appearto need the same
+        * correction that HFP does.
+        */
+       if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE && dsi->lanes > 1) {
+               int hfp = adjusted_mode->hsync_start - adjusted_mode->hdisplay;
+               int remainder = hfp % dsi->lanes;
+
+               if (remainder) {
+                       adjusted_mode->hsync_start += remainder;
+                       adjusted_mode->hsync_end   += remainder;
+                       adjusted_mode->htotal      += remainder;
+               }
+       }
+
        return 0;
 }
 
--

2.40.1

#4 lore raw

Re: [PATCH 2/2] drm/bridge: samsung-dsim: Fix porch calcalcuation rounding

Quoted 9 lines · Adam Ford
Quoted 8 lines · Adam Ford

When using video sync pulses, the HFP, HBP, and HSA are divided between the available lanes if there is more than one lane. For certain timings and lane configurations, the HFP may not be evenly divisible. If the HFP is rounded down, it ends up being too small which can cause some monitors to not sync properly. In these instances, adjust htotal and hsync to round the HFP up, and recalculate the htotal.

Quoted 6 lines
Quoted 2 lines

Gentle nudge on this one. Basically this fixes an issue with the 8MP, but it's still unknown why it doesn't work on 8MM or 8MN, but Frieder confirmed there are no regressions on 8MM or 8MN.

I only tested two specific display setups on i.MX8MM. So of course I can't confirm the absence of regressions in general.

Anyway, I think this should be applied.

author
#5 lore raw

Re: [PATCH 2/2] drm/bridge: samsung-dsim: Fix porch calcalcuation rounding

Quoted 11 lines · Frieder Schrempf
Quoted 9 lines · Adam Ford
Quoted 8 lines · Adam Ford

When using video sync pulses, the HFP, HBP, and HSA are divided between the available lanes if there is more than one lane. For certain timings and lane configurations, the HFP may not be evenly divisible. If the HFP is rounded down, it ends up being too small which can cause some monitors to not sync properly. In these instances, adjust htotal and hsync to round the HFP up, and recalculate the htotal.

Quoted 7 lines
Quoted 6 lines
Quoted 2 lines

Gentle nudge on this one. Basically this fixes an issue with the 8MP, but it's still unknown why it doesn't work on 8MM or 8MN, but Frieder confirmed there are no regressions on 8MM or 8MN.

Inki,

Is there something you need which is holding this back? It's been nearly two months since I posted the initial patch.

Thank you,

adam

Quoted 4 lines

I only tested two specific display setups on i.MX8MM. So of course I can't confirm the absence of regressions in general.

Anyway, I think this should be applied.

Reviewed-by Tested-by on v1 1/2
#6 lore raw
In reply to Adam Ford

Hi Adam,

Quoted 14 lines · Adam Ford

The P divider should be set based on the min and max values of the fin pll which may vary between different platforms. These ranges are defined per platform, but hard-coded values were used instead which resulted in a smaller range available on the i.MX8M[MNP] than what was possible.

Fixes 846307185f0f ("drm/bridge: samsung-dsim: update PLL reference clock")
drivers/gpu/drm/bridge/samsung-dsim.c +0 -0
index be5914caa17d..239d253a7d71 100644

drivers/gpu/drm/bridge/samsung-dsim.c static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi, · line 577

Quoted 7 lines
drivers/gpu/drm/bridge/samsung-dsim.c +2 -2
@@ -573,8 +573,8 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi,
 	u16 _m, best_m;
 	u8 _s, best_s;
 
-	p_min = DIV_ROUND_UP(fin, (12 * MHZ));
-	p_max = fin / (6 * MHZ);
+	p_min = DIV_ROUND_UP(fin, (driver_data->pll_fin_max * MHZ));
+	p_max = fin / (driver_data->pll_fin_min * MHZ);

I did some tinkering with the PLL settings the other day and this is literally one of the things I came up with.

So I'm happy to provide:

Regarding the P divider, I'm also wondering if there is an upper limit for the p-value (not for the resulting fin_pll) that we should take into account, too. The problem is that we have conflicts in the documentation (again) so we don't really know what the correct limit would be.

There are the following ranges given in the RMs:

* 1..63 (i.MX8MM RM 13.7.8.18.4)
* 1..33 (i.MX8MM RM 13.7.10.1)
* 1..63 (i.MX8MP RM 13.2.3.1.5.2)
* 1..63 (i.MX8MP RM 13.7.2.4)

Unfortunately there are similar discrepancies for the other parameters and limits.

Thanks
Frieder

Quoted 3 lines
drivers/gpu/drm/bridge/samsung-dsim.c +0 -0
 	for (_p = p_min; _p <= p_max; ++_p) {
 		for (_s = 0; _s <= 5; ++_s) {
author
#7 lore raw
Quoted 29 lines · Frieder Schrempf

Hi Adam,

Quoted 14 lines · Adam Ford

The P divider should be set based on the min and max values of the fin pll which may vary between different platforms. These ranges are defined per platform, but hard-coded values were used instead which resulted in a smaller range available on the i.MX8M[MNP] than what was possible.

Fixes 846307185f0f ("drm/bridge: samsung-dsim: update PLL reference clock")
drivers/gpu/drm/bridge/samsung-dsim.c +0 -0
index be5914caa17d..239d253a7d71 100644

drivers/gpu/drm/bridge/samsung-dsim.c static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi, · line 577

Quoted 7 lines
drivers/gpu/drm/bridge/samsung-dsim.c +2 -2
@@ -573,8 +573,8 @@ static unsigned long samsung_dsim_pll_find_pms(struct samsung_dsim *dsi,
      u16 _m, best_m;
      u8 _s, best_s;
 
-     p_min = DIV_ROUND_UP(fin, (12 * MHZ));
-     p_max = fin / (6 * MHZ);
+     p_min = DIV_ROUND_UP(fin, (driver_data->pll_fin_max * MHZ));
+     p_max = fin / (driver_data->pll_fin_min * MHZ);

I did some tinkering with the PLL settings the other day and this is literally one of the things I came up with.

Quoted 5 lines

So I'm happy to provide:

Thank you!

Quoted 11 lines

Regarding the P divider, I'm also wondering if there is an upper limit for the p-value (not for the resulting fin_pll) that we should take into account, too. The problem is that we have conflicts in the documentation (again) so we don't really know what the correct limit would be.

There are the following ranges given in the RMs:

* 1..63 (i.MX8MM RM 13.7.8.18.4)
* 1..33 (i.MX8MM RM 13.7.10.1)
* 1..63 (i.MX8MP RM 13.2.3.1.5.2)
* 1..63 (i.MX8MP RM 13.7.2.4)

1...63 (i.IMX8MN RM 13.7.2.4)

Quoted 3 lines

Unfortunately there are similar discrepancies for the other parameters and limits.

For what it's worth, I compared these values to the NXP downstream branch [1], and they seemed to indicate the values were as follows:

.p = { .min = 1, .max = 63, },
.m = { .min = 64, .max = 1023, },
.s = { .min = 0, .max = 5, },
.k = { .min = 0, .max = 32768, }, /* abs(k) */ .fin = { .min = 6000, .max = 300000, }, /* in KHz */ .fpref = { .min = 2000, .max = 30000, }, /* in KHz */ .fvco = { .min = 1050000, .max = 2100000, }, /* in KHz */

In a previous commit [2], I mentioned the fact that I reached out to NXP asking about the discrepancies and my NXP Rep and I received the response:

"Yes it is definitely wrong, the one that is part of the NOTE in MIPI_DPHY_M_PLLPMS register table against PMS_P, PMS_M and PMS_S is not correct. I will report this to Doc team, the one customer should be take into account is the Table 13-40 DPHY PLL Parameters and the Note above."

adam

[1] - https://github.com/nxp-imx/linux-imx/blob/lf-6.1.y/drivers/gpu/drm/imx/sec_mipi_pll_1432x.h#L38C1-L47C1
[2] - https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/gpu/drm/bridge/samsung-dsim.c?h=next-20231212&id=54f1a83c72250b182fa7722b0c5f6eb5e769598d

Quoted 8 lines

Thanks
Frieder

Quoted 3 lines
drivers/gpu/drm/bridge/samsung-dsim.c +0 -0
      for (_p = p_min; _p <= p_max; ++_p) {
              for (_s = 0; _s <= 5; ++_s) {