[PATCH 0/6] OMAPDSS: Misc fixes and cleanups

The first patch in this series is a follow up on the previously posted series
'OMAPDSS: APPLY: Treat overlay manager timings as shadow registers'. It is
required for HDMI and DPI interfaces to work properly, it ensures manager
timings are applied in the set_timing() ops for these interfaces.

The next 3 patches remove some unnecessary usage of omap_dss_device pointer in
DISPC and APPLY.

The last 2 patches are miscellaneous fixes and are self explanatory.

Reference tree containing this series:

git://gitorious.org/~boddob/linux-omap-dss2/archit-dss2-clone.git mgr_timing_and_fixes

Tested on OMAP4 SDP.

Archit Taneja (6):
  OMAPDSS: DPI/HDMI: Apply manager timings even if panel is disabled
  OMAPDSS: APPLY: Remove an unnecessary omap_dss_device pointer
  OMAPDSS: DISPC: Remove omap_dss_device pointer usage from
    dispc_mgr_pclk_rate()
  OMAPDSS: DISPC: Remove usage of dispc_mgr_get_device()
  OMAPDSS: Fix DSI_FCLK clock source selection
  OMAPDSS: DISPC: Remove Fake VSYNC support

 drivers/video/omap2/dss/Kconfig |    9 ------
 drivers/video/omap2/dss/apply.c |    3 --
 drivers/video/omap2/dss/dispc.c |   61 +++++++++++++-------------------------
 drivers/video/omap2/dss/dpi.c   |    2 +
 drivers/video/omap2/dss/dsi.c   |    4 --
 drivers/video/omap2/dss/dss.c   |    5 ++-
 drivers/video/omap2/dss/dss.h   |    1 -
 drivers/video/omap2/dss/hdmi.c  |    2 +
 8 files changed, 28 insertions(+), 59 deletions(-)

-- 
1.7.5.4

The HDMI and DPI interfaces use their 'set_timing' functions to take in a new
set of timings. If the panel is disabled, they do not disable and re-enable
the interface. Currently, the manager timings are applied in hdmi_power_on()
and dpi_set_mode() respectively, these are not called by set_timings if the
panel is disabled.

When checking overlay and manager data, the DSS driver uses the last applied
manager timings, and not the timings stored in omap_dss_device struct. Hence,
there is a need to apply the new manager timings even if the panel is disabled.

Apply the manager timings if the panel is disabled. Eventually, there should be
one common place where the timings are applied independent of the state of the
panel.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/dpi.c  |    2 ++
 drivers/video/omap2/dss/hdmi.c |    2 ++
 2 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/video/omap2/dss/dpi.c b/drivers/video/omap2/dss/dpi.c
index 8127f46..1650050 100644
--- a/drivers/video/omap2/dss/dpi.c
+++ b/drivers/video/omap2/dss/dpi.c

@@ -298,6 +298,8 @@ void dpi_set_timings(struct omap_dss_device *dssdev,
 
 		dispc_runtime_put();
 		dss_runtime_put();
+	} else {
+		dss_mgr_set_timings(dssdev->manager, timings);
 	}
 }
 EXPORT_SYMBOL(dpi_set_timings);

diff --git a/drivers/video/omap2/dss/hdmi.c b/drivers/video/omap2/dss/hdmi.c
index 8d4ff8c..32ad712 100644
--- a/drivers/video/omap2/dss/hdmi.c
+++ b/drivers/video/omap2/dss/hdmi.c

@@ -435,6 +435,8 @@ void omapdss_hdmi_display_set_timing(struct omap_dss_device *dssdev)
 		r = hdmi_power_on(dssdev);
 		if (r)
 			DSSERR("failed to power on device\n");
+	} else {
+		dss_mgr_set_timings(dssdev->manager, &dssdev->panel.timings);
 	}
 }
 

-- 
1.7.5.4

The omap_dss_device pointer declared in dss_ovl_setup_fifo() isn't used. Remove
the pointer variable declaration and it's assignment.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/apply.c |    3 ---
 1 files changed, 0 insertions(+), 3 deletions(-)

diff --git a/drivers/video/omap2/dss/apply.c b/drivers/video/omap2/dss/apply.c
index 052d9a2..756c31a 100644
--- a/drivers/video/omap2/dss/apply.c
+++ b/drivers/video/omap2/dss/apply.c

@@ -977,14 +977,11 @@ static void dss_ovl_setup_fifo(struct omap_overlay *ovl,
 		bool use_fifo_merge)
 {
 	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	struct omap_dss_device *dssdev;
 	u32 fifo_low, fifo_high;
 
 	if (!op->enabled && !op->enabling)
 		return;
 
-	dssdev = ovl->manager->device;
-
 	dispc_ovl_compute_fifo_thresholds(ovl->id, &fifo_low, &fifo_high,
 			use_fifo_merge);
 

-- 
1.7.5.4

The pixel clock rate for the TV manager is calculated by checking the device
type connected to the manager, and then requesting the VENC/HDMI interface for
the pixel clock rate.

Remove the use of omap_dss_device pointer from here by checking which interface
generates the pixel clock by reading the DSS_CTRL.VENC_HDMI_SWITCH bit.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/dispc.c |   11 ++++++-----
 1 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c
index 49015b8..c198cc8 100644
--- a/drivers/video/omap2/dss/dispc.c
+++ b/drivers/video/omap2/dss/dispc.c

@@ -2643,13 +2643,14 @@ unsigned long dispc_mgr_pclk_rate(enum omap_channel channel)
 
 		return r / pcd;
 	} else {
-		struct omap_dss_device *dssdev -			dispc_mgr_get_device(channel);
+		enum dss_hdmi_venc_clk_source_select source;
 
-		switch (dssdev->type) {
-		case OMAP_DISPLAY_TYPE_VENC:
+		source = dss_get_hdmi_venc_clk_source();
+
+		switch (source) {
+		case DSS_VENC_TV_CLK:
 			return venc_get_pixel_clock();
-		case OMAP_DISPLAY_TYPE_HDMI:
+		case DSS_HDMI_M_PCLK:
 			return hdmi_get_pixel_clock();
 		default:
 			BUG();

-- 
1.7.5.4

The functions calc_fclk_five_taps() and check_horiz_timing_omap3() use the
function dispc_mgr_get_device() to get the omap_dss_device pointer to which
the manager is connected, the width of the panel is derived from that.

Replace this by using dss_mgr_get_device() which returns the manager's timing
stored in it's private data in APPLY. This contains the latest timings applied
to the manager. Remove the function dispc_mgr_get_device() as it isn't used
any more.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/dispc.c |   29 +++++++++++++++--------------
 1 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c
index c198cc8..ef131cd 100644
--- a/drivers/video/omap2/dss/dispc.c
+++ b/drivers/video/omap2/dss/dispc.c

@@ -413,14 +413,6 @@ static inline bool dispc_mgr_is_lcd(enum omap_channel channel)
 		return false;
 }
 
-static struct omap_dss_device *dispc_mgr_get_device(enum omap_channel channel)
-{
-	struct omap_overlay_manager *mgr -		omap_dss_get_overlay_manager(channel);
-
-	return mgr ? mgr->device : NULL;
-}
-
 u32 dispc_mgr_get_vsync_irq(enum omap_channel channel)
 {
 	switch (channel) {

@@ -1665,14 +1657,17 @@ static int check_horiz_timing_omap3(enum omap_channel channel, u16 pos_x,
 		u16 width, u16 height, u16 out_width, u16 out_height)
 {
 	int DS = DIV_ROUND_UP(height, out_height);
-	struct omap_dss_device *dssdev = dispc_mgr_get_device(channel);
-	struct omap_video_timings t = dssdev->panel.timings;
+	struct omap_overlay_manager *mgr;
+	struct omap_video_timings *t;
 	unsigned long nonactive, lclk, pclk;
 	static const u8 limits[3] = { 8, 10, 20 };
 	u64 val, blank;
 	int i;
 
-	nonactive = t.x_res + t.hfp + t.hsw + t.hbp - out_width;
+	mgr = omap_dss_get_overlay_manager(channel);
+	t = dss_mgr_get_timings(mgr);
+
+	nonactive = t->x_res + t->hfp + t->hsw + t->hbp - out_width;
 	pclk = dispc_mgr_pclk_rate(channel);
 	if (dispc_mgr_is_lcd(channel))
 		lclk = dispc_mgr_lclk_rate(channel);

@@ -1684,7 +1679,7 @@ static int check_horiz_timing_omap3(enum omap_channel channel, u16 pos_x,
 		i++;
 	if (out_width < width)
 		i++;
-	blank = div_u64((u64)(t.hbp + t.hsw + t.hfp) * lclk, pclk);
+	blank = div_u64((u64)(t->hbp + t->hsw + t->hfp) * lclk, pclk);
 	DSSDBG("blanking period + ppl = %llu (limit = %u)\n", blank, limits[i]);
 	if (blank <= limits[i])
 		return -EINVAL;

@@ -1725,8 +1720,14 @@ static unsigned long calc_core_clk_five_taps(enum omap_channel channel,
 		return (unsigned long) pclk;
 
 	if (height > out_height) {
-		struct omap_dss_device *dssdev = dispc_mgr_get_device(channel);
-		unsigned int ppl = dssdev->panel.timings.x_res;
+		struct omap_overlay_manager *mgr;
+		struct omap_video_timings *mgr_timings;
+		unsigned int ppl;
+
+		mgr = omap_dss_get_overlay_manager(channel);
+		mgr_timings = dss_mgr_get_timings(mgr);
+
+		ppl = mgr_timings->x_res;
 
 		tmp = pclk * height * out_width;
 		do_div(tmp, 2 * out_height * ppl);

-- 
1.7.5.4

The wrong bit field was being updated in DSS_CTRL when trying to configure the
clock source of DSI2 functional clock. Use the correct bit field based on the
dsi module number.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/dss.c |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/video/omap2/dss/dss.c b/drivers/video/omap2/dss/dss.c
index e731aa4..e212acb 100644
--- a/drivers/video/omap2/dss/dss.c
+++ b/drivers/video/omap2/dss/dss.c

@@ -335,7 +335,7 @@ void dss_select_dsi_clk_source(int dsi_module,
 		enum omap_dss_clk_source clk_src)
 {
 	struct platform_device *dsidev;
-	int b;
+	int b, pos;
 
 	switch (clk_src) {
 	case OMAP_DSS_CLK_SRC_FCK:

@@ -357,7 +357,8 @@ void dss_select_dsi_clk_source(int dsi_module,
 		BUG();
 	}
 
-	REG_FLD_MOD(DSS_CONTROL, b, 1, 1);	/* DSI_CLK_SWITCH */
+	pos = dsi_module = 0 ? 1 : 10;
+	REG_FLD_MOD(DSS_CONTROL, b, pos, pos);	/* DSIx_CLK_SWITCH */
 
 	dss.dsi_clk_source[dsi_module] = clk_src;
 }

-- 
1.7.5.4

Fake VSYNC support is a hack and has some bugs in it. It isn't used by any user
of DSS. Remove Fake VSYNC support. For DSI command mode and RFBI panels, a user
of DSS should wait for the completion of a frame by using the panel driver's
sync op.

Signed-off-by: Archit Taneja <redacted>
---
 drivers/video/omap2/dss/Kconfig |    9 ---------
 drivers/video/omap2/dss/dispc.c |   21 ---------------------
 drivers/video/omap2/dss/dsi.c   |    4 ----
 drivers/video/omap2/dss/dss.h   |    1 -
 4 files changed, 0 insertions(+), 35 deletions(-)

diff --git a/drivers/video/omap2/dss/Kconfig b/drivers/video/omap2/dss/Kconfig
index 7be7c06..9aaf324 100644
--- a/drivers/video/omap2/dss/Kconfig
+++ b/drivers/video/omap2/dss/Kconfig

@@ -90,15 +90,6 @@ config OMAP2_DSS_DSI
 
 	  See http://www.mipi.org/ for DSI spesifications.
 
-config OMAP2_DSS_FAKE_VSYNC
-	bool "Fake VSYNC irq from manual update displays"
-	default n
-	help
-	  If this is selected, DSI will generate a fake DISPC VSYNC interrupt
-	  when DSI has sent a frame. This is only needed with DSI or RFBI
-	  displays using manual mode, and you want VSYNC to, for example,
-	  time animation.
-
 config OMAP2_DSS_MIN_FCK_PER_PCK
 	int "Minimum FCK/PCK ratio (for scaling)"
 	range 0 32

diff --git a/drivers/video/omap2/dss/dispc.c b/drivers/video/omap2/dss/dispc.c
index ef131cd..89d54c5 100644
--- a/drivers/video/omap2/dss/dispc.c
+++ b/drivers/video/omap2/dss/dispc.c

@@ -3443,27 +3443,6 @@ int omap_dispc_wait_for_irq_interruptible_timeout(u32 irqmask,
 	return 0;
 }
 
-#ifdef CONFIG_OMAP2_DSS_FAKE_VSYNC
-void dispc_fake_vsync_irq(void)
-{
-	u32 irqstatus = DISPC_IRQ_VSYNC;
-	int i;
-
-	WARN_ON(!in_interrupt());
-
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		struct omap_dispc_isr_data *isr_data;
-		isr_data = &dispc.registered_isr[i];
-
-		if (!isr_data->isr)
-			continue;
-
-		if (isr_data->mask & irqstatus)
-			isr_data->isr(isr_data->arg, irqstatus);
-	}
-}
-#endif
-
 static void _omap_dispc_initialize_irq(void)
 {
 	unsigned long flags;

diff --git a/drivers/video/omap2/dss/dsi.c b/drivers/video/omap2/dss/dsi.c
index db73598..95bc996 100644
--- a/drivers/video/omap2/dss/dsi.c
+++ b/drivers/video/omap2/dss/dsi.c

@@ -4152,10 +4152,6 @@ static void dsi_framedone_irq_callback(void *data, u32 mask)
 	__cancel_delayed_work(&dsi->framedone_timeout_work);
 
 	dsi_handle_framedone(dsidev, 0);
-
-#ifdef CONFIG_OMAP2_DSS_FAKE_VSYNC
-	dispc_fake_vsync_irq();
-#endif
 }
 
 int omap_dsi_update(struct omap_dss_device *dssdev, int channel,

diff --git a/drivers/video/omap2/dss/dss.h b/drivers/video/omap2/dss/dss.h
index f2a51c6..6699197 100644
--- a/drivers/video/omap2/dss/dss.h
+++ b/drivers/video/omap2/dss/dss.h

@@ -399,7 +399,6 @@ void dispc_dump_clocks(struct seq_file *s);
 void dispc_dump_irqs(struct seq_file *s);
 void dispc_dump_regs(struct seq_file *s);
 void dispc_irq_handler(void);
-void dispc_fake_vsync_irq(void);
 
 int dispc_runtime_get(void);
 void dispc_runtime_put(void);

-- 
1.7.5.4

On Mon, 2012-05-07 at 16:51 +0530, Archit Taneja wrote:

The first patch in this series is a follow up on the previously posted series
'OMAPDSS: APPLY: Treat overlay manager timings as shadow registers'. It is
required for HDMI and DPI interfaces to work properly, it ensures manager
timings are applied in the set_timing() ops for these interfaces.

The next 3 patches remove some unnecessary usage of omap_dss_device pointer in
DISPC and APPLY.

The last 2 patches are miscellaneous fixes and are self explanatory.

Reference tree containing this series:

git://gitorious.org/~boddob/linux-omap-dss2/archit-dss2-clone.git mgr_timing_and_fixes

Tested on OMAP4 SDP.

Archit Taneja (6):
  OMAPDSS: DPI/HDMI: Apply manager timings even if panel is disabled
  OMAPDSS: APPLY: Remove an unnecessary omap_dss_device pointer
  OMAPDSS: DISPC: Remove omap_dss_device pointer usage from
    dispc_mgr_pclk_rate()
  OMAPDSS: DISPC: Remove usage of dispc_mgr_get_device()
  OMAPDSS: Fix DSI_FCLK clock source selection
  OMAPDSS: DISPC: Remove Fake VSYNC support

The first four patches seem to be related (or at least based) to the
set_timings series (and the first patch you already had included in
later version).

Should I take the two last patches and apply them already, as they are
fine and don't depend on anything? You could add the first 4 patches
into the set_timings series.

 Tomi

On Tuesday 08 May 2012 05:28 PM, Tomi Valkeinen wrote:

On Mon, 2012-05-07 at 16:51 +0530, Archit Taneja wrote:

quoted

The first patch in this series is a follow up on the previously posted series
'OMAPDSS: APPLY: Treat overlay manager timings as shadow registers'. It is
required for HDMI and DPI interfaces to work properly, it ensures manager
timings are applied in the set_timing() ops for these interfaces.

The next 3 patches remove some unnecessary usage of omap_dss_device pointer in
DISPC and APPLY.

The last 2 patches are miscellaneous fixes and are self explanatory.

Reference tree containing this series:

git://gitorious.org/~boddob/linux-omap-dss2/archit-dss2-clone.git mgr_timing_and_fixes

Tested on OMAP4 SDP.

Archit Taneja (6):
   OMAPDSS: DPI/HDMI: Apply manager timings even if panel is disabled
   OMAPDSS: APPLY: Remove an unnecessary omap_dss_device pointer
   OMAPDSS: DISPC: Remove omap_dss_device pointer usage from
     dispc_mgr_pclk_rate()
   OMAPDSS: DISPC: Remove usage of dispc_mgr_get_device()
   OMAPDSS: Fix DSI_FCLK clock source selection
   OMAPDSS: DISPC: Remove Fake VSYNC support

The first four patches seem to be related (or at least based) to the
set_timings series (and the first patch you already had included in
later version).

Should I take the two last patches and apply them already, as they are
fine and don't depend on anything? You could add the first 4 patches
into the set_timings series.

Yes, you can apply the last 2 patches.

Thanks,
Archit

  Tomi

On Mon, May 7, 2012 at 4:21 AM, Archit Taneja [off-list ref] wrote:

The first patch in this series is a follow up on the previously posted
series
'OMAPDSS: APPLY: Treat overlay manager timings as shadow registers'. It is
required for HDMI and DPI interfaces to work properly, it ensures manager
timings are applied in the set_timing() ops for these interfaces.

The next 3 patches remove some unnecessary usage of omap_dss_device
pointer in
DISPC and APPLY.

The last 2 patches are miscellaneous fixes and are self explanatory.

Reference tree containing this series:

git://gitorious.org/~boddob/linux-omap-dss2/archit-dss2-clone.git
mgr_timing_and_fixes

Tested on OMAP4 SDP.

Archit Taneja (6):
 OMAPDSS: DPI/HDMI: Apply manager timings even if panel is disabled
 OMAPDSS: APPLY: Remove an unnecessary omap_dss_device pointer
 OMAPDSS: DISPC: Remove omap_dss_device pointer usage from
   dispc_mgr_pclk_rate()
 OMAPDSS: DISPC: Remove usage of dispc_mgr_get_device()
 OMAPDSS: Fix DSI_FCLK clock source selection
 OMAPDSS: DISPC: Remove Fake VSYNC support

 drivers/video/omap2/dss/Kconfig |    9 ------
 drivers/video/omap2/dss/apply.c |    3 --
 drivers/video/omap2/dss/dispc.c |   61
+++++++++++++-------------------------
 drivers/video/omap2/dss/dpi.c   |    2 +
 drivers/video/omap2/dss/dsi.c   |    4 --
 drivers/video/omap2/dss/dss.c   |    5 ++-
 drivers/video/omap2/dss/dss.h   |    1 -
 drivers/video/omap2/dss/hdmi.c  |    2 +
 8 files changed, 28 insertions(+), 59 deletions(-)

I did a build and boot of each one of the patches on the
mgr_timing_and_fixes branch including these on:

Beagleboard-xM rev C1
Beagleboard rev B4
Mistral AM37x-EVM
Zoom AM3517 EVM
Zoom AM1808 EVM

Tested-by: Russ.Dill@ti.com

On Wednesday 09 May 2012 05:08 AM, Russ Dill wrote:

On Mon, May 7, 2012 at 4:21 AM, Archit Taneja[off-list ref]  wrote:

quoted

The first patch in this series is a follow up on the previously posted
series
'OMAPDSS: APPLY: Treat overlay manager timings as shadow registers'. It is
required for HDMI and DPI interfaces to work properly, it ensures manager
timings are applied in the set_timing() ops for these interfaces.

The next 3 patches remove some unnecessary usage of omap_dss_device
pointer in
DISPC and APPLY.

The last 2 patches are miscellaneous fixes and are self explanatory.

Reference tree containing this series:

git://gitorious.org/~boddob/linux-omap-dss2/archit-dss2-clone.git
mgr_timing_and_fixes

Tested on OMAP4 SDP.

Archit Taneja (6):
  OMAPDSS: DPI/HDMI: Apply manager timings even if panel is disabled
  OMAPDSS: APPLY: Remove an unnecessary omap_dss_device pointer
  OMAPDSS: DISPC: Remove omap_dss_device pointer usage from
    dispc_mgr_pclk_rate()
  OMAPDSS: DISPC: Remove usage of dispc_mgr_get_device()
  OMAPDSS: Fix DSI_FCLK clock source selection
  OMAPDSS: DISPC: Remove Fake VSYNC support

  drivers/video/omap2/dss/Kconfig |    9 ------
  drivers/video/omap2/dss/apply.c |    3 --
  drivers/video/omap2/dss/dispc.c |   61
+++++++++++++-------------------------
  drivers/video/omap2/dss/dpi.c   |    2 +
  drivers/video/omap2/dss/dsi.c   |    4 --
  drivers/video/omap2/dss/dss.c   |    5 ++-
  drivers/video/omap2/dss/dss.h   |    1 -
  drivers/video/omap2/dss/hdmi.c  |    2 +
  8 files changed, 28 insertions(+), 59 deletions(-)

I did a build and boot of each one of the patches on the
mgr_timing_and_fixes branch including these on:

Beagleboard-xM rev C1
Beagleboard rev B4
Mistral AM37x-EVM
Zoom AM3517 EVM
Zoom AM1808 EVM

Thanks! I'll be posting out a newer version of the series with some 
minor changes. But the content should be more or less the same.

Thanks,
Archit

Tested-by: Russ.Dill@ti.com