All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings
@ 2018-11-27 20:53 Linus Walleij
  2018-11-27 20:53 ` [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control Linus Walleij
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Linus Walleij @ 2018-11-27 20:53 UTC (permalink / raw)
  To: David Woodhouse, Brian Norris, Boris Brezillon, Marek Vasut,
	Richard Weinberger, linux-mtd
  Cc: Linus Walleij

The simple_map_init() may need to be called with some
function pointers already assigned for complex mappings,
just bail out if complex handlers have already been
assigned.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
ChangeLog v2->v3:
- Move check for pre-initialized simple map into
  physmap-core.c
ChangeLog v1->v2:
- Rebase on latest MTD development branch
- Use a new approach as the code changed under me
---
 drivers/mtd/maps/physmap-core.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/maps/physmap-core.c b/drivers/mtd/maps/physmap-core.c
index e8c3b250d842..ecca7784de06 100644
--- a/drivers/mtd/maps/physmap-core.c
+++ b/drivers/mtd/maps/physmap-core.c
@@ -514,10 +514,16 @@ static int physmap_flash_probe(struct platform_device *dev)
 			err = physmap_addr_gpios_map_init(&info->maps[i]);
 			if (err)
 				goto err_out;
-		} else {
-			simple_map_init(&info->maps[i]);
 		}
 
+		/*
+		 * Only use the simple_map implementation if map hooks are not
+		 * implemented. Since map->read() is mandatory checking for its
+		 * presence is enough.
+		 */
+		if (!info->maps[i].read)
+			simple_map_init(&info->maps[i]);
+
 		if (info->probe_type) {
 			info->mtds[i] = do_map_probe(info->probe_type,
 						     &info->maps[i]);
-- 
2.19.1

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control
  2018-11-27 20:53 [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Linus Walleij
@ 2018-11-27 20:53 ` Linus Walleij
  2018-12-07  9:10   ` [2/2,v3] " Boris Brezillon
  2018-12-02  8:47 ` [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Boris Brezillon
  2018-12-07  9:10 ` [1/2,v3] " Boris Brezillon
  2 siblings, 1 reply; 6+ messages in thread
From: Linus Walleij @ 2018-11-27 20:53 UTC (permalink / raw)
  To: David Woodhouse, Brian Norris, Boris Brezillon, Marek Vasut,
	Richard Weinberger, linux-mtd
  Cc: Linus Walleij

This enables the complex mapping for the Gemini and kicks in
custom read/write functions that will wrap the existing
simple functions in calls to enable/disable the parallel
flash pins using pin controls.

This is necessary on some hardware such as the D-Link
DIR-685 where all flash pins are patched in/out at the same
time, but some of the flash pins are in practice unused by
the flash and have anyway been reused as GPIO.

This concerns specifically CE1 on the Gemini. There is only
one flash chip, so only CE0 is used, and the line for CE1
has been reused as chip select for the emulated SPI port
connected to the display. If we try to use the same lines
for flash and GPIO at the same time, one of them will loose:
the GPIO line will disappear because it gets disconnected
from the pin when the flash group is muxed out.

Fix this by introducing two pin control states named simply
"enabled" and "disabled" and only enable the flash lines
when absolutely necessary (during read/write/copy). This
way, they are available for GPIO at all other times and
the display works.

Collect all the state variables in a struct named
struct gemini_flash and allocate this struct at probe
time.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
ChangeLog v2->v3:
- Resend
ChangeLog v1->v2:
- Rebase on latest MTD development branch
---
 drivers/mtd/maps/Kconfig          |   1 +
 drivers/mtd/maps/physmap-gemini.c | 110 +++++++++++++++++++++++++++++-
 2 files changed, 110 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index 2edd8bcdbe1c..e0cf869c8544 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -88,6 +88,7 @@ config MTD_PHYSMAP_GEMINI
 	bool "Cortina Gemini OF-based physical memory map handling"
 	depends on MTD_PHYSMAP_OF
 	depends on MFD_SYSCON
+	select MTD_COMPLEX_MAPPINGS
 	default ARCH_GEMINI
 	help
 	  This provides some extra DT physmap parsing for the Gemini
diff --git a/drivers/mtd/maps/physmap-gemini.c b/drivers/mtd/maps/physmap-gemini.c
index 1cf128a0526d..60775b208fc9 100644
--- a/drivers/mtd/maps/physmap-gemini.c
+++ b/drivers/mtd/maps/physmap-gemini.c
@@ -10,9 +10,11 @@
 #include <linux/of.h>
 #include <linux/of_device.h>
 #include <linux/mtd/map.h>
+#include <linux/mtd/xip.h>
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 #include <linux/bitops.h>
+#include <linux/pinctrl/consumer.h>
 #include "physmap-gemini.h"
 
 /*
@@ -44,6 +46,82 @@
 
 #define FLASH_PARALLEL_HIGH_PIN_CNT	(1 << 20)	/* else low pin cnt */
 
+static const struct of_device_id syscon_match[] = {
+	{ .compatible = "cortina,gemini-syscon" },
+	{ },
+};
+
+struct gemini_flash {
+	struct device *dev;
+	struct pinctrl *p;
+	struct pinctrl_state *enabled_state;
+	struct pinctrl_state *disabled_state;
+};
+
+/* Static local state */
+static struct gemini_flash *gf;
+
+static void gemini_flash_enable_pins(void)
+{
+	int ret;
+
+	if (IS_ERR(gf->enabled_state))
+		return;
+	ret = pinctrl_select_state(gf->p, gf->enabled_state);
+	if (ret)
+		dev_err(gf->dev, "failed to enable pins\n");
+}
+
+static void gemini_flash_disable_pins(void)
+{
+	int ret;
+
+	if (IS_ERR(gf->disabled_state))
+		return;
+	ret = pinctrl_select_state(gf->p, gf->disabled_state);
+	if (ret)
+		dev_err(gf->dev, "failed to disable pins\n");
+}
+
+static map_word __xipram gemini_flash_map_read(struct map_info *map,
+					       unsigned long ofs)
+{
+	map_word __xipram ret;
+
+	gemini_flash_enable_pins();
+	ret = inline_map_read(map, ofs);
+	gemini_flash_disable_pins();
+
+	return ret;
+}
+
+static void __xipram gemini_flash_map_write(struct map_info *map,
+					    const map_word datum,
+					    unsigned long ofs)
+{
+	gemini_flash_enable_pins();
+	inline_map_write(map, datum, ofs);
+	gemini_flash_disable_pins();
+}
+
+static void __xipram gemini_flash_map_copy_from(struct map_info *map,
+						void *to, unsigned long from,
+						ssize_t len)
+{
+	gemini_flash_enable_pins();
+	inline_map_copy_from(map, to, from, len);
+	gemini_flash_disable_pins();
+}
+
+static void __xipram gemini_flash_map_copy_to(struct map_info *map,
+					      unsigned long to,
+					      const void *from, ssize_t len)
+{
+	gemini_flash_enable_pins();
+	inline_map_copy_to(map, to, from, len);
+	gemini_flash_disable_pins();
+}
+
 int of_flash_probe_gemini(struct platform_device *pdev,
 			  struct device_node *np,
 			  struct map_info *map)
@@ -57,6 +135,11 @@ int of_flash_probe_gemini(struct platform_device *pdev,
 	if (!of_device_is_compatible(np, "cortina,gemini-flash"))
 		return 0;
 
+	gf = devm_kzalloc(dev, sizeof(*gf), GFP_KERNEL);
+	if (!gf)
+		return -ENOMEM;
+	gf->dev = dev;
+
 	rmap = syscon_regmap_lookup_by_phandle(np, "syscon");
 	if (IS_ERR(rmap)) {
 		dev_err(dev, "no syscon\n");
@@ -91,7 +174,32 @@ int of_flash_probe_gemini(struct platform_device *pdev,
 				 map->bankwidth * 8);
 	}
 
-	dev_info(&pdev->dev, "initialized Gemini-specific physmap control\n");
+	gf->p = devm_pinctrl_get(dev);
+	if (IS_ERR(gf->p)) {
+		dev_err(dev, "no pinctrl handle\n");
+		ret = PTR_ERR(gf->p);
+		return ret;
+	}
+
+	gf->enabled_state = pinctrl_lookup_state(gf->p, "enabled");
+	if (IS_ERR(gf->enabled_state))
+		dev_err(dev, "no enabled pin control state\n");
+
+	gf->disabled_state = pinctrl_lookup_state(gf->p, "disabled");
+	if (IS_ERR(gf->enabled_state)) {
+		dev_err(dev, "no disabled pin control state\n");
+	} else {
+		ret = pinctrl_select_state(gf->p, gf->disabled_state);
+		if (ret)
+			dev_err(gf->dev, "failed to disable pins\n");
+	}
+
+	map->read = gemini_flash_map_read;
+	map->write = gemini_flash_map_write;
+	map->copy_from = gemini_flash_map_copy_from;
+	map->copy_to = gemini_flash_map_copy_to;
+
+	dev_info(dev, "initialized Gemini-specific physmap control\n");
 
 	return 0;
 }
-- 
2.19.1

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings
  2018-11-27 20:53 [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Linus Walleij
  2018-11-27 20:53 ` [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control Linus Walleij
@ 2018-12-02  8:47 ` Boris Brezillon
  2018-12-04 13:09   ` Linus Walleij
  2018-12-07  9:10 ` [1/2,v3] " Boris Brezillon
  2 siblings, 1 reply; 6+ messages in thread
From: Boris Brezillon @ 2018-12-02  8:47 UTC (permalink / raw)
  To: Linus Walleij
  Cc: David Woodhouse, Brian Norris, Boris Brezillon, Marek Vasut,
	Richard Weinberger, linux-mtd

On Tue, 27 Nov 2018 21:53:57 +0100
Linus Walleij <linus.walleij@linaro.org> wrote:

> The simple_map_init() may need to be called with some
> function pointers already assigned for complex mappings,
> just bail out if complex handlers have already been
> assigned.

The commit message no longer matches what the patch does.

How about:

"
mtd: maps: physmap: Leave assigned complex mappings

SoC-specific drivers might provide their own map->xxx() implementation,
and calling simple_map_init() unconditionally will override those
implementations.

Make sure map->read is NULL before calling simple_map_init().
"

If you're okay with this new commit message, I can fix it when applying.

Regards,

Boris

> 
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
> ChangeLog v2->v3:
> - Move check for pre-initialized simple map into
>   physmap-core.c
> ChangeLog v1->v2:
> - Rebase on latest MTD development branch
> - Use a new approach as the code changed under me
> ---
>  drivers/mtd/maps/physmap-core.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/mtd/maps/physmap-core.c b/drivers/mtd/maps/physmap-core.c
> index e8c3b250d842..ecca7784de06 100644
> --- a/drivers/mtd/maps/physmap-core.c
> +++ b/drivers/mtd/maps/physmap-core.c
> @@ -514,10 +514,16 @@ static int physmap_flash_probe(struct platform_device *dev)
>  			err = physmap_addr_gpios_map_init(&info->maps[i]);
>  			if (err)
>  				goto err_out;
> -		} else {
> -			simple_map_init(&info->maps[i]);
>  		}
>  
> +		/*
> +		 * Only use the simple_map implementation if map hooks are not
> +		 * implemented. Since map->read() is mandatory checking for its
> +		 * presence is enough.
> +		 */
> +		if (!info->maps[i].read)
> +			simple_map_init(&info->maps[i]);
> +
>  		if (info->probe_type) {
>  			info->mtds[i] = do_map_probe(info->probe_type,
>  						     &info->maps[i]);

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings
  2018-12-02  8:47 ` [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Boris Brezillon
@ 2018-12-04 13:09   ` Linus Walleij
  0 siblings, 0 replies; 6+ messages in thread
From: Linus Walleij @ 2018-12-04 13:09 UTC (permalink / raw)
  To: Boris Brezillon
  Cc: David Woodhouse, Brian Norris, Boris BREZILLON, Mark Vasut,
	Richard Weinberger, linux-mtd

On Sun, Dec 2, 2018 at 9:48 AM Boris Brezillon
<boris.brezillon@bootlin.com> wrote:
> On Tue, 27 Nov 2018 21:53:57 +0100
> Linus Walleij <linus.walleij@linaro.org> wrote:
>
> > The simple_map_init() may need to be called with some
> > function pointers already assigned for complex mappings,
> > just bail out if complex handlers have already been
> > assigned.
>
> The commit message no longer matches what the patch does.
>
> How about:
>
> "
> mtd: maps: physmap: Leave assigned complex mappings
>
> SoC-specific drivers might provide their own map->xxx() implementation,
> and calling simple_map_init() unconditionally will override those
> implementations.
>
> Make sure map->read is NULL before calling simple_map_init().
> "
>
> If you're okay with this new commit message, I can fix it when applying.

Sure, go ahead Boris, thanks!

Yours,
Linus Walleij

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [2/2,v3] mtd: physmap_of_gemini: Handle pin control
  2018-11-27 20:53 ` [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control Linus Walleij
@ 2018-12-07  9:10   ` Boris Brezillon
  0 siblings, 0 replies; 6+ messages in thread
From: Boris Brezillon @ 2018-12-07  9:10 UTC (permalink / raw)
  To: Linus Walleij, David Woodhouse, Brian Norris, Boris Brezillon,
	Marek Vasut, Richard Weinberger, linux-mtd
  Cc: Boris Brezillon

On Tue, 2018-11-27 at 20:53:58 UTC, Linus Walleij wrote:
> This enables the complex mapping for the Gemini and kicks in
> custom read/write functions that will wrap the existing
> simple functions in calls to enable/disable the parallel
> flash pins using pin controls.
> 
> This is necessary on some hardware such as the D-Link
> DIR-685 where all flash pins are patched in/out at the same
> time, but some of the flash pins are in practice unused by
> the flash and have anyway been reused as GPIO.
> 
> This concerns specifically CE1 on the Gemini. There is only
> one flash chip, so only CE0 is used, and the line for CE1
> has been reused as chip select for the emulated SPI port
> connected to the display. If we try to use the same lines
> for flash and GPIO at the same time, one of them will loose:
> the GPIO line will disappear because it gets disconnected
> from the pin when the flash group is muxed out.
> 
> Fix this by introducing two pin control states named simply
> "enabled" and "disabled" and only enable the flash lines
> when absolutely necessary (during read/write/copy). This
> way, they are available for GPIO at all other times and
> the display works.
> 
> Collect all the state variables in a struct named
> struct gemini_flash and allocate this struct at probe
> time.
> 
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

Applied to http://git.infradead.org/linux-mtd.git mtd/next, thanks.

Boris

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [1/2,v3] mtd: maps: Leave assigned complex mappings
  2018-11-27 20:53 [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Linus Walleij
  2018-11-27 20:53 ` [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control Linus Walleij
  2018-12-02  8:47 ` [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Boris Brezillon
@ 2018-12-07  9:10 ` Boris Brezillon
  2 siblings, 0 replies; 6+ messages in thread
From: Boris Brezillon @ 2018-12-07  9:10 UTC (permalink / raw)
  To: Linus Walleij, David Woodhouse, Brian Norris, Boris Brezillon,
	Marek Vasut, Richard Weinberger, linux-mtd
  Cc: Boris Brezillon

On Tue, 2018-11-27 at 20:53:57 UTC, Linus Walleij wrote:
> The simple_map_init() may need to be called with some
> function pointers already assigned for complex mappings,
> just bail out if complex handlers have already been
> assigned.
> 
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

Applied to http://git.infradead.org/linux-mtd.git mtd/next, thanks.

Boris

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2018-12-07  9:11 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-11-27 20:53 [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Linus Walleij
2018-11-27 20:53 ` [PATCH 2/2 v3] mtd: physmap_of_gemini: Handle pin control Linus Walleij
2018-12-07  9:10   ` [2/2,v3] " Boris Brezillon
2018-12-02  8:47 ` [PATCH 1/2 v3] mtd: maps: Leave assigned complex mappings Boris Brezillon
2018-12-04 13:09   ` Linus Walleij
2018-12-07  9:10 ` [1/2,v3] " Boris Brezillon

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.