[PATCH 0/7] OneNAND OMAP patches (resent)

[PATCH 0/7] OneNAND OMAP patches (resent)

[Adrian Hunter]

Hi

(resent with corrected email for Artem)

Here are some OneNAND and related OMAP patches.

Adrian Hunter (7):
      mtd: OneNAND: OMAP2/3: add support for command line partitioning
      mtd: OneNAND: add enable / disable methods to onenand_chip
      mtd: OneNAND: OMAP2/3: prevent regulator sleeping while OneNAND is in use
      OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds
      OMAP2/3: OneNAND: add 104MHz support
      OMAP2/3: OneNAND: add platform data callback for PM constraints
      mtd: OneNAND: lighten scary initial bad block messages

 arch/arm/mach-omap2/gpmc-nand.c           |    2 +-
 arch/arm/mach-omap2/gpmc-onenand.c        |   19 +++++--
 arch/arm/mach-omap2/gpmc.c                |   12 ++++-
 arch/arm/mach-omap2/usb-tusb6010.c        |    4 +-
 arch/arm/plat-omap/include/plat/gpmc.h    |    9 ++--
 arch/arm/plat-omap/include/plat/onenand.h |    9 +++
 drivers/mtd/onenand/omap2.c               |   85 +++++++++++++++++++++++------
 drivers/mtd/onenand/onenand_base.c        |   26 ++++++----
 drivers/mtd/onenand/onenand_bbt.c         |    4 +-
 include/linux/mtd/onenand.h               |    2 +
 10 files changed, 131 insertions(+), 41 deletions(-) 
 
Regards
Adrian

[PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Adrian Hunter]

>From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Wed, 30 Dec 2009 07:40:16 +0100
Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

Add the ability to parse MTD partition information from the
kernel command line.

Note that a pointless BUG_ON is removed, as are redundant
calls to 'del_mtd_partitions()' and 'del_mtd_device()'
because they are also done by 'onenand_release()'.

Finally note that 'add_mtd_device()' returns 1 on failure
so the error condition was incorrect.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/omap2.c |   27 ++++++++++++---------------
 1 files changed, 12 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index da25a90..324402f 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -65,6 +65,10 @@ struct omap2_onenand {
 	int (*setup)(void __iomem *base, int freq);
 };
 
+#ifdef CONFIG_MTD_PARTITIONS
+static const char *part_probes[] = { "cmdlinepart", NULL,  };
+#endif
+
 static void omap2_onenand_dma_cb(int lch, u16 ch_status, void *data)
 {
 	struct omap2_onenand *c = data;
@@ -730,13 +734,15 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 	}
 
 #ifdef CONFIG_MTD_PARTITIONS
-	if (pdata->parts != NULL)
-		r = add_mtd_partitions(&c->mtd, pdata->parts,
-				       pdata->nr_parts);
+	r = parse_mtd_partitions(&c->mtd, part_probes, &c->parts, 0);
+	if (r > 0)
+		r = add_mtd_partitions(&c->mtd, c->parts, r);
+	else if (pdata->parts != NULL)
+		r = add_mtd_partitions(&c->mtd, pdata->parts, pdata->nr_parts);
 	else
 #endif
 		r = add_mtd_device(&c->mtd);
-	if (r < 0)
+	if (r)
 		goto err_release_onenand;
 
 	platform_set_drvdata(pdev, c);
@@ -760,6 +766,7 @@ err_release_mem_region:
 err_free_cs:
 	gpmc_cs_free(c->gpmc_cs);
 err_kfree:
+	kfree(c->parts);
 	kfree(c);
 
 	return r;
@@ -769,17 +776,6 @@ static int __devexit omap2_onenand_remove(struct platform_device *pdev)
 {
 	struct omap2_onenand *c = dev_get_drvdata(&pdev->dev);
 
-	BUG_ON(c == NULL);
-
-#ifdef CONFIG_MTD_PARTITIONS
-	if (c->parts)
-		del_mtd_partitions(&c->mtd);
-	else
-		del_mtd_device(&c->mtd);
-#else
-	del_mtd_device(&c->mtd);
-#endif
-
 	onenand_release(&c->mtd);
 	if (c->dma_channel != -1)
 		omap_free_dma(c->dma_channel);
@@ -792,6 +788,7 @@ static int __devexit omap2_onenand_remove(struct platform_device *pdev)
 	iounmap(c->onenand.base);
 	release_mem_region(c->phys_base, ONENAND_IO_SIZE);
 	gpmc_cs_free(c->gpmc_cs);
+	kfree(c->parts);
 	kfree(c);
 
 	return 0;
-- 
1.7.0.4

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Artem Bityutskiy]

On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
> From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Wed, 30 Dec 2009 07:40:16 +0100
> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
> 
> Add the ability to parse MTD partition information from the
> kernel command line.
> 
> Note that a pointless BUG_ON is removed, as are redundant
> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
> because they are also done by 'onenand_release()'.
> 
> Finally note that 'add_mtd_device()' returns 1 on failure
> so the error condition was incorrect.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>

I've pushed patches 1 and 7 to l2-mtd-2.6, but patch #2 is not
applicable. I think this is because of changes Kyungmin made recently.
Could you please re-send patches 2 and 3 on top of the l2 tree?

I guess I can apply 2 and 3 now. Patch 4 should wait for comments from
Tony, AFAICS. Patches 5 and 6 are dependent, and hence should wait as
well, AFAICS.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Adrian Hunter]

On 15/12/10 16:04, ext Artem Bityutskiy wrote:
> On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
>>  From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
>> From: Adrian Hunter<adrian.hunter@nokia.com>
>> Date: Wed, 30 Dec 2009 07:40:16 +0100
>> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
>>
>> Add the ability to parse MTD partition information from the
>> kernel command line.
>>
>> Note that a pointless BUG_ON is removed, as are redundant
>> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
>> because they are also done by 'onenand_release()'.
>>
>> Finally note that 'add_mtd_device()' returns 1 on failure
>> so the error condition was incorrect.
>>
>> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
>
> I've pushed patches 1 and 7 to l2-mtd-2.6, but patch #2 is not
> applicable. I think this is because of changes Kyungmin made recently.
> Could you please re-send patches 2 and 3 on top of the l2 tree?

I just tried to apply them to l2 and they applied fine as is.
Has something changed?  Do I have the wrong tree?

	git://git.infradead.org/users/dedekind/l2-mtd-2.6.git

>
> I guess I can apply 2 and 3 now. Patch 4 should wait for comments from
> Tony, AFAICS. Patches 5 and 6 are dependent, and hence should wait as
> well, AFAICS.
>

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Artem Bityutskiy]

On Wed, 2011-01-05 at 13:02 +0200, Adrian Hunter wrote:
> On 15/12/10 16:04, ext Artem Bityutskiy wrote:
> > On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
> >>  From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
> >> From: Adrian Hunter<adrian.hunter@nokia.com>
> >> Date: Wed, 30 Dec 2009 07:40:16 +0100
> >> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
> >>
> >> Add the ability to parse MTD partition information from the
> >> kernel command line.
> >>
> >> Note that a pointless BUG_ON is removed, as are redundant
> >> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
> >> because they are also done by 'onenand_release()'.
> >>
> >> Finally note that 'add_mtd_device()' returns 1 on failure
> >> so the error condition was incorrect.
> >>
> >> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
> >
> > I've pushed patches 1 and 7 to l2-mtd-2.6, but patch #2 is not
> > applicable. I think this is because of changes Kyungmin made recently.
> > Could you please re-send patches 2 and 3 on top of the l2 tree?
> 
> I just tried to apply them to l2 and they applied fine as is.
> Has something changed?  Do I have the wrong tree?

Just tried to apply patch 2 which you re-sent on "Wed, 05 Jan 2011
13:02:40 +0200"

[dedekind@koala l2-mtd-2.6]$ git am -i -s ~/tmp/adrian.mbox 
Commit Body is:
--------------------------
mtd: OneNAND: add enable / disable methods to onenand_chip

 From ece28a7fdec36fb40d60d35a5639326871c162f6 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 19 Feb 2010 15:39:52 +0100
Subject: [PATCH V2 2/7] mtd: OneNAND: add enable / disable methods to
onenand_chip

Add enable / disable methods called from get_device() /
release_device().
These can be used, for example, to allow the driver to prevent the
voltage
regulator from being put to sleep while OneNAND is in use.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
--------------------------
Apply? [y]es/[n]o/[e]dit/[v]iew patch/[a]ccept all e
Commit Body is:
--------------------------
mtd: OneNAND: add enable / disable methods to onenand_chip

 From ece28a7fdec36fb40d60d35a5639326871c162f6 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 19 Feb 2010 15:39:52 +0100
Subject: [PATCH V2 2/7] mtd: OneNAND: add enable / disable methods to
onenand_chip

Add enable / disable methods called from get_device() /
release_device().
These can be used, for example, to allow the driver to prevent the
voltage
regulator from being put to sleep while OneNAND is in use.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
--------------------------
Apply? [y]es/[n]o/[e]dit/[v]iew patch/[a]ccept all y
Applying: mtd: OneNAND: add enable / disable methods to onenand_chip
error: patch failed: drivers/mtd/onenand/onenand_base.c:948
error: drivers/mtd/onenand/onenand_base.c: patch does not apply
error: patch failed: include/linux/mtd/onenand.h:118
error: include/linux/mtd/onenand.h: patch does not apply
Patch failed at 0001 mtd: OneNAND: add enable / disable methods to
onenand_chip
When you have resolved this problem run "git am --resolved".
If you would prefer to skip this patch, instead run "git am --skip".
To restore the original branch and stop patching run "git am --abort".
[dedekind@koala l2-mtd-2.6]$ patch -p1 < .git/rebase-apply/patch 
patching file drivers/mtd/onenand/onenand_base.c
Hunk #1 FAILED at 948.
Hunk #2 FAILED at 974.
2 out of 2 hunks FAILED -- saving rejects to file
drivers/mtd/onenand/onenand_base.c.rej
patching file include/linux/mtd/onenand.h
Hunk #1 FAILED at 118.
1 out of 1 hunk FAILED -- saving rejects to file
include/linux/mtd/onenand.h.rej
[dedekind@koala l2-mtd-2.6]$ cat drivers/mtd/onenand/onenand_base.c.rej
--- drivers/mtd/onenand/onenand_base.c
+++ drivers/mtd/onenand/onenand_base.c
@@ -948,6 +948,8 @@
                if (this->state == FL_READY) {
                        this->state = new_state;
                        spin_unlock(&this->chip_lock);
+                       if (new_state != FL_PM_SUSPENDED &&
this->enable)
+                               this->enable(mtd);
                        break;
                }
                if (new_state == FL_PM_SUSPENDED) {
@@ -974,6 +976,8 @@
  {
        struct onenand_chip *this = mtd->priv;
  
+       if (this->state != FL_PM_SUSPENDED && this->disable)
+               this->disable(mtd);
        /* Release the chip */
        spin_lock(&this->chip_lock);
        this->state = FL_READY;
[dedekind@koala l2-mtd-2.6]$ cat include/linux/mtd/onenand.h.rej
--- include/linux/mtd/onenand.h
+++ include/linux/mtd/onenand.h
@@ -118,6 +118,8 @@
        int (*chip_probe)(struct mtd_info *mtd);
        int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
        int (*scan_bbt)(struct mtd_info *mtd);
+       int (*enable)(struct mtd_info *mtd);
+       int (*disable)(struct mtd_info *mtd);
  
        struct completion       complete;
        int                     irq;
[dedekind@koala l2-mtd-2.6]$ git log
commit 5fca82e7d3827249e840393fb28d19404d337fbf
Author: Aleksandr Koltsoff <aleksandr.koltsoff@ebts.fi>
Date:   Tue Jan 4 10:42:35 2011 +0200

    mtd: m25p80: Fix JEDEC ID for AT26DF321
    
    The last byte of the ID should be zero for this chip. Was added in
    commit d0e8c47c58575b9131e786edb488fd029eba443e . Reported by Tomi
    Varjo.
    
    Signed-off-by: Aleksandr Koltsoff <aleksandr.koltsoff@ebts.fi>

-- 
Best Regards,
Artem Bityutskiy (Битюцкий Артём)

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Adrian Hunter]

[-- Attachment #1: Type: text/plain, Size: 5913 bytes --]

On 05/01/11 14:12, ext Artem Bityutskiy wrote:
> On Wed, 2011-01-05 at 13:02 +0200, Adrian Hunter wrote:
>> On 15/12/10 16:04, ext Artem Bityutskiy wrote:
>>> On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
>>>>    From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
>>>> From: Adrian Hunter<adrian.hunter@nokia.com>
>>>> Date: Wed, 30 Dec 2009 07:40:16 +0100
>>>> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
>>>>
>>>> Add the ability to parse MTD partition information from the
>>>> kernel command line.
>>>>
>>>> Note that a pointless BUG_ON is removed, as are redundant
>>>> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
>>>> because they are also done by 'onenand_release()'.
>>>>
>>>> Finally note that 'add_mtd_device()' returns 1 on failure
>>>> so the error condition was incorrect.
>>>>
>>>> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
>>>
>>> I've pushed patches 1 and 7 to l2-mtd-2.6, but patch #2 is not
>>> applicable. I think this is because of changes Kyungmin made recently.
>>> Could you please re-send patches 2 and 3 on top of the l2 tree?
>>
>> I just tried to apply them to l2 and they applied fine as is.
>> Has something changed?  Do I have the wrong tree?
>
> Just tried to apply patch 2 which you re-sent on "Wed, 05 Jan 2011
> 13:02:40 +0200"

I guess the patches got munched.  Here they are tarred

>
> [dedekind@koala l2-mtd-2.6]$ git am -i -s ~/tmp/adrian.mbox
> Commit Body is:
> --------------------------
> mtd: OneNAND: add enable / disable methods to onenand_chip
>
>   From ece28a7fdec36fb40d60d35a5639326871c162f6 Mon Sep 17 00:00:00 2001
> From: Adrian Hunter<adrian.hunter@nokia.com>
> Date: Fri, 19 Feb 2010 15:39:52 +0100
> Subject: [PATCH V2 2/7] mtd: OneNAND: add enable / disable methods to
> onenand_chip
>
> Add enable / disable methods called from get_device() /
> release_device().
> These can be used, for example, to allow the driver to prevent the
> voltage
> regulator from being put to sleep while OneNAND is in use.
>
> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
> Signed-off-by: Artem Bityutskiy<Artem.Bityutskiy@nokia.com>
> --------------------------
> Apply? [y]es/[n]o/[e]dit/[v]iew patch/[a]ccept all e
> Commit Body is:
> --------------------------
> mtd: OneNAND: add enable / disable methods to onenand_chip
>
>   From ece28a7fdec36fb40d60d35a5639326871c162f6 Mon Sep 17 00:00:00 2001
> From: Adrian Hunter<adrian.hunter@nokia.com>
> Date: Fri, 19 Feb 2010 15:39:52 +0100
> Subject: [PATCH V2 2/7] mtd: OneNAND: add enable / disable methods to
> onenand_chip
>
> Add enable / disable methods called from get_device() /
> release_device().
> These can be used, for example, to allow the driver to prevent the
> voltage
> regulator from being put to sleep while OneNAND is in use.
>
> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
> Signed-off-by: Artem Bityutskiy<Artem.Bityutskiy@nokia.com>
> --------------------------
> Apply? [y]es/[n]o/[e]dit/[v]iew patch/[a]ccept all y
> Applying: mtd: OneNAND: add enable / disable methods to onenand_chip
> error: patch failed: drivers/mtd/onenand/onenand_base.c:948
> error: drivers/mtd/onenand/onenand_base.c: patch does not apply
> error: patch failed: include/linux/mtd/onenand.h:118
> error: include/linux/mtd/onenand.h: patch does not apply
> Patch failed at 0001 mtd: OneNAND: add enable / disable methods to
> onenand_chip
> When you have resolved this problem run "git am --resolved".
> If you would prefer to skip this patch, instead run "git am --skip".
> To restore the original branch and stop patching run "git am --abort".
> [dedekind@koala l2-mtd-2.6]$ patch -p1<  .git/rebase-apply/patch
> patching file drivers/mtd/onenand/onenand_base.c
> Hunk #1 FAILED at 948.
> Hunk #2 FAILED at 974.
> 2 out of 2 hunks FAILED -- saving rejects to file
> drivers/mtd/onenand/onenand_base.c.rej
> patching file include/linux/mtd/onenand.h
> Hunk #1 FAILED at 118.
> 1 out of 1 hunk FAILED -- saving rejects to file
> include/linux/mtd/onenand.h.rej
> [dedekind@koala l2-mtd-2.6]$ cat drivers/mtd/onenand/onenand_base.c.rej
> --- drivers/mtd/onenand/onenand_base.c
> +++ drivers/mtd/onenand/onenand_base.c
> @@ -948,6 +948,8 @@
>                  if (this->state == FL_READY) {
>                          this->state = new_state;
>                          spin_unlock(&this->chip_lock);
> +                       if (new_state != FL_PM_SUSPENDED&&
> this->enable)
> +                               this->enable(mtd);
>                          break;
>                  }
>                  if (new_state == FL_PM_SUSPENDED) {
> @@ -974,6 +976,8 @@
>    {
>          struct onenand_chip *this = mtd->priv;
>
> +       if (this->state != FL_PM_SUSPENDED&&  this->disable)
> +               this->disable(mtd);
>          /* Release the chip */
>          spin_lock(&this->chip_lock);
>          this->state = FL_READY;
> [dedekind@koala l2-mtd-2.6]$ cat include/linux/mtd/onenand.h.rej
> --- include/linux/mtd/onenand.h
> +++ include/linux/mtd/onenand.h
> @@ -118,6 +118,8 @@
>          int (*chip_probe)(struct mtd_info *mtd);
>          int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
>          int (*scan_bbt)(struct mtd_info *mtd);
> +       int (*enable)(struct mtd_info *mtd);
> +       int (*disable)(struct mtd_info *mtd);
>
>          struct completion       complete;
>          int                     irq;
> [dedekind@koala l2-mtd-2.6]$ git log
> commit 5fca82e7d3827249e840393fb28d19404d337fbf
> Author: Aleksandr Koltsoff<aleksandr.koltsoff@ebts.fi>
> Date:   Tue Jan 4 10:42:35 2011 +0200
>
>      mtd: m25p80: Fix JEDEC ID for AT26DF321
>
>      The last byte of the ID should be zero for this chip. Was added in
>      commit d0e8c47c58575b9131e786edb488fd029eba443e . Reported by Tomi
>      Varjo.
>
>      Signed-off-by: Aleksandr Koltsoff<aleksandr.koltsoff@ebts.fi>
>


[-- Attachment #2: huh.tar.gz --]
[-- Type: application/x-gzip, Size: 2065 bytes --]

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Artem Bityutskiy]

On Wed, 2011-01-05 at 14:24 +0200, Adrian Hunter wrote:
> On 05/01/11 14:12, ext Artem Bityutskiy wrote:
> > On Wed, 2011-01-05 at 13:02 +0200, Adrian Hunter wrote:
> >> On 15/12/10 16:04, ext Artem Bityutskiy wrote:
> >>> On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
> >>>>    From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
> >>>> From: Adrian Hunter<adrian.hunter@nokia.com>
> >>>> Date: Wed, 30 Dec 2009 07:40:16 +0100
> >>>> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
> >>>>
> >>>> Add the ability to parse MTD partition information from the
> >>>> kernel command line.
> >>>>
> >>>> Note that a pointless BUG_ON is removed, as are redundant
> >>>> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
> >>>> because they are also done by 'onenand_release()'.
> >>>>
> >>>> Finally note that 'add_mtd_device()' returns 1 on failure
> >>>> so the error condition was incorrect.
> >>>>
> >>>> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
> >>>
> >>> I've pushed patches 1 and 7 to l2-mtd-2.6, but patch #2 is not
> >>> applicable. I think this is because of changes Kyungmin made recently.
> >>> Could you please re-send patches 2 and 3 on top of the l2 tree?
> >>
> >> I just tried to apply them to l2 and they applied fine as is.
> >> Has something changed?  Do I have the wrong tree?
> >
> > Just tried to apply patch 2 which you re-sent on "Wed, 05 Jan 2011
> > 13:02:40 +0200"
> 
> I guess the patches got munched.  Here they are tarred

Pushed to l2 tree both (patches 2 and 3).

-- 
Best Regards,
Artem Bityutskiy (Битюцкий Артём)

Re: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning

[Artem Bityutskiy]

On Mon, 2010-12-13 at 14:20 +0200, Adrian Hunter wrote:
> From d27a88c5f9f18a7c38784c085889c3e1053340cf Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Wed, 30 Dec 2009 07:40:16 +0100
> Subject: [PATCH 1/7] mtd: OneNAND: OMAP2/3: add support for command line partitioning
> 
> Add the ability to parse MTD partition information from the
> kernel command line.
> 
> Note that a pointless BUG_ON is removed, as are redundant
> calls to 'del_mtd_partitions()' and 'del_mtd_device()'
> because they are also done by 'onenand_release()'.
> 
> Finally note that 'add_mtd_device()' returns 1 on failure
> so the error condition was incorrect.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>

Pushed this one to l2-mtd-2.6.git, thank you.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

[PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

[Adrian Hunter]

>From 16b760999cb79f9cf71728c9253f1399717be63a Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 19 Feb 2010 15:39:52 +0100
Subject: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

Add enable / disable methods called from get_device() / release_device().
These can be used, for example, to allow the driver to prevent the voltage
regulator from being put to sleep while OneNAND is in use.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/onenand_base.c |    4 ++++
 include/linux/mtd/onenand.h        |    2 ++
 2 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index c38bf9c..bde274f 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -948,6 +948,8 @@ static int onenand_get_device(struct mtd_info *mtd, int new_state)
 		if (this->state == FL_READY) {
 			this->state = new_state;
 			spin_unlock(&this->chip_lock);
+			if (this->enable)
+				this->enable(mtd);
 			break;
 		}
 		if (new_state == FL_PM_SUSPENDED) {
@@ -974,6 +976,8 @@ static void onenand_release_device(struct mtd_info *mtd)
 {
 	struct onenand_chip *this = mtd->priv;
 
+	if (this->state != FL_PM_SUSPENDED && this->disable)
+		this->disable(mtd);
 	/* Release the chip */
 	spin_lock(&this->chip_lock);
 	this->state = FL_READY;
diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
index 6da3fe3..ae418e4 100644
--- a/include/linux/mtd/onenand.h
+++ b/include/linux/mtd/onenand.h
@@ -118,6 +118,8 @@ struct onenand_chip {
 	int (*chip_probe)(struct mtd_info *mtd);
 	int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
 	int (*scan_bbt)(struct mtd_info *mtd);
+	int (*enable)(struct mtd_info *mtd);
+	int (*disable)(struct mtd_info *mtd);
 
 	struct completion	complete;
 	int			irq;
-- 
1.7.0.4

Re: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

[Kyungmin Park]

Hi,

Now I'm used the clock gating for OneNAND instead of regulator.
so I think the mechanism is similar.

On Mon, Dec 13, 2010 at 9:20 PM, Adrian Hunter <adrian.hunter@nokia.com> wrote:
> From 16b760999cb79f9cf71728c9253f1399717be63a Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Fri, 19 Feb 2010 15:39:52 +0100
> Subject: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip
>
> Add enable / disable methods called from get_device() / release_device().
> These can be used, for example, to allow the driver to prevent the voltage
> regulator from being put to sleep while OneNAND is in use.
>
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
> ---
>  drivers/mtd/onenand/onenand_base.c |    4 ++++
>  include/linux/mtd/onenand.h        |    2 ++
>  2 files changed, 6 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
> index c38bf9c..bde274f 100644
> --- a/drivers/mtd/onenand/onenand_base.c
> +++ b/drivers/mtd/onenand/onenand_base.c
> @@ -948,6 +948,8 @@ static int onenand_get_device(struct mtd_info *mtd, int new_state)
>                if (this->state == FL_READY) {
>                        this->state = new_state;
>                        spin_unlock(&this->chip_lock);
> +                       if (this->enable)
> +                               this->enable(mtd);
>                        break;
>                }
>                if (new_state == FL_PM_SUSPENDED) {

I put the code the end of function like:

static int onenand_get_device(struct mtd_info *mtd, int new_state)
{
        struct onenand_chip *this = mtd->priv;
        DECLARE_WAITQUEUE(wait, current);

        /*
         * Grab the lock and see if the device is available
         */
        while (1) {
                spin_lock(&this->chip_lock);
                if (this->state == FL_READY) {
                        this->state = new_state;
                        spin_unlock(&this->chip_lock);
                        break;
                }
                if (new_state == FL_PM_SUSPENDED) {
                        spin_unlock(&this->chip_lock);
                        return (this->state == FL_PM_SUSPENDED) ? 0 : -EAGAIN;
                }
                set_current_state(TASK_UNINTERRUPTIBLE);
                add_wait_queue(&this->wq, &wait);
                spin_unlock(&this->chip_lock);
                schedule();
                remove_wait_queue(&this->wq, &wait);
        }

        if (this->clk)
                clk_enable(this->clk);

        return 0;
}


> @@ -974,6 +976,8 @@ static void onenand_release_device(struct mtd_info *mtd)
>  {
>        struct onenand_chip *this = mtd->priv;
>
> +       if (this->state != FL_PM_SUSPENDED && this->disable)
> +               this->disable(mtd);
Need to check the SUSPENDED at here? when enter the suspend get_device
is called.
release device is called when resume. so no need to check
FL_PM_SUSPENDED in my case.

>        /* Release the chip */
>        spin_lock(&this->chip_lock);
>        this->state = FL_READY;

and put the same as like:

static void onenand_release_device(struct mtd_info *mtd)
{
        struct onenand_chip *this = mtd->priv;

        /* Release the chip */
        spin_lock(&this->chip_lock);
        this->state = FL_READY;
        wake_up(&this->wq);
        spin_unlock(&this->chip_lock);

        if (this->clk)
                clk_disable(this->clk);
}

Thank you,
Kyungmin Park

> diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
> index 6da3fe3..ae418e4 100644
> --- a/include/linux/mtd/onenand.h
> +++ b/include/linux/mtd/onenand.h
> @@ -118,6 +118,8 @@ struct onenand_chip {
>        int (*chip_probe)(struct mtd_info *mtd);
>        int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
>        int (*scan_bbt)(struct mtd_info *mtd);
> +       int (*enable)(struct mtd_info *mtd);
> +       int (*disable)(struct mtd_info *mtd);
>
>        struct completion       complete;
>        int                     irq;
> --
> 1.7.0.4
>
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>

Re: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

[Adrian Hunter]

On 14/12/10 02:17, Kyungmin Park wrote:
> Hi,
>
> Now I'm used the clock gating for OneNAND instead of regulator.
> so I think the mechanism is similar.

I am not sure what you are trying to say here.  We need the regulator
enable / disable to stop the regulator going to sleep (because it cannot
supply enough current when it is asleep) and also we need to call down to
the board level to set latency constraints.  These things seem pretty
specific to the OMAP driver hence the new enable() / disable() methods.

We do not need any clock gating.  The OneNAND clock is dealt with by
the OMAP GPMC OneNAND controller and GPMC has auto-idle / smart-idle
facilities that let OMAP gate clocks and power it off when it is not
being used.

>
> On Mon, Dec 13, 2010 at 9:20 PM, Adrian Hunter<adrian.hunter@nokia.com>  wrote:
>>  From 16b760999cb79f9cf71728c9253f1399717be63a Mon Sep 17 00:00:00 2001
>> From: Adrian Hunter<adrian.hunter@nokia.com>
>> Date: Fri, 19 Feb 2010 15:39:52 +0100
>> Subject: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip
>>
>> Add enable / disable methods called from get_device() / release_device().
>> These can be used, for example, to allow the driver to prevent the voltage
>> regulator from being put to sleep while OneNAND is in use.
>>
>> Signed-off-by: Adrian Hunter<adrian.hunter@nokia.com>
>> ---
>>   drivers/mtd/onenand/onenand_base.c |    4 ++++
>>   include/linux/mtd/onenand.h        |    2 ++
>>   2 files changed, 6 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
>> index c38bf9c..bde274f 100644
>> --- a/drivers/mtd/onenand/onenand_base.c
>> +++ b/drivers/mtd/onenand/onenand_base.c
>> @@ -948,6 +948,8 @@ static int onenand_get_device(struct mtd_info *mtd, int new_state)
>>                 if (this->state == FL_READY) {
>>                         this->state = new_state;
>>                         spin_unlock(&this->chip_lock);
>> +                       if (this->enable)
>> +                               this->enable(mtd);
>>                         break;
>>                 }
>>                 if (new_state == FL_PM_SUSPENDED) {
>
> I put the code the end of function like:
>
> static int onenand_get_device(struct mtd_info *mtd, int new_state)
> {
>          struct onenand_chip *this = mtd->priv;
>          DECLARE_WAITQUEUE(wait, current);
>
>          /*
>           * Grab the lock and see if the device is available
>           */
>          while (1) {
>                  spin_lock(&this->chip_lock);
>                  if (this->state == FL_READY) {
>                          this->state = new_state;
>                          spin_unlock(&this->chip_lock);
>                          break;
>                  }
>                  if (new_state == FL_PM_SUSPENDED) {
>                          spin_unlock(&this->chip_lock);
>                          return (this->state == FL_PM_SUSPENDED) ? 0 : -EAGAIN;
>                  }
>                  set_current_state(TASK_UNINTERRUPTIBLE);
>                  add_wait_queue(&this->wq,&wait);
>                  spin_unlock(&this->chip_lock);
>                  schedule();
>                  remove_wait_queue(&this->wq,&wait);
>          }
>
>          if (this->clk)
>                  clk_enable(this->clk);
>
>          return 0;
> }
>
>
>> @@ -974,6 +976,8 @@ static void onenand_release_device(struct mtd_info *mtd)
>>   {
>>         struct onenand_chip *this = mtd->priv;
>>
>> +       if (this->state != FL_PM_SUSPENDED&&  this->disable)
>> +               this->disable(mtd);
> Need to check the SUSPENDED at here? when enter the suspend get_device
> is called.
> release device is called when resume. so no need to check
> FL_PM_SUSPENDED in my case.

onenand-get_device should have had:

	if (new_state != FL_PM_SUSPENDED && this->enable)
		this->enable(mtd);

I will resend the patch.


>
>>         /* Release the chip */
>>         spin_lock(&this->chip_lock);
>>         this->state = FL_READY;
>
> and put the same as like:
>
> static void onenand_release_device(struct mtd_info *mtd)
> {
>          struct onenand_chip *this = mtd->priv;
>
>          /* Release the chip */
>          spin_lock(&this->chip_lock);
>          this->state = FL_READY;
>          wake_up(&this->wq);
>          spin_unlock(&this->chip_lock);
>
>          if (this->clk)
>                  clk_disable(this->clk);
> }
>
> Thank you,
> Kyungmin Park
>
>> diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
>> index 6da3fe3..ae418e4 100644
>> --- a/include/linux/mtd/onenand.h
>> +++ b/include/linux/mtd/onenand.h
>> @@ -118,6 +118,8 @@ struct onenand_chip {
>>         int (*chip_probe)(struct mtd_info *mtd);
>>         int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
>>         int (*scan_bbt)(struct mtd_info *mtd);
>> +       int (*enable)(struct mtd_info *mtd);
>> +       int (*disable)(struct mtd_info *mtd);
>>
>>         struct completion       complete;
>>         int                     irq;
>> --
>> 1.7.0.4
>>
>>
>> ______________________________________________________
>> Linux MTD discussion mailing list
>> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>>
>

Re: [PATCH 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

[Adrian Hunter]

 From ece28a7fdec36fb40d60d35a5639326871c162f6 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 19 Feb 2010 15:39:52 +0100
Subject: [PATCH V2 2/7] mtd: OneNAND: add enable / disable methods to onenand_chip

Add enable / disable methods called from get_device() / release_device().
These can be used, for example, to allow the driver to prevent the voltage
regulator from being put to sleep while OneNAND is in use.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
  drivers/mtd/onenand/onenand_base.c |    4 ++++
  include/linux/mtd/onenand.h        |    2 ++
  2 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index c38bf9c..5639e43 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -948,6 +948,8 @@ static int onenand_get_device(struct mtd_info *mtd, int new_state)
  		if (this->state == FL_READY) {
  			this->state = new_state;
  			spin_unlock(&this->chip_lock);
+			if (new_state != FL_PM_SUSPENDED && this->enable)
+				this->enable(mtd);
  			break;
  		}
  		if (new_state == FL_PM_SUSPENDED) {
@@ -974,6 +976,8 @@ static void onenand_release_device(struct mtd_info *mtd)
  {
  	struct onenand_chip *this = mtd->priv;
  
+	if (this->state != FL_PM_SUSPENDED && this->disable)
+		this->disable(mtd);
  	/* Release the chip */
  	spin_lock(&this->chip_lock);
  	this->state = FL_READY;
diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
index 6da3fe3..ae418e4 100644
--- a/include/linux/mtd/onenand.h
+++ b/include/linux/mtd/onenand.h
@@ -118,6 +118,8 @@ struct onenand_chip {
  	int (*chip_probe)(struct mtd_info *mtd);
  	int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
  	int (*scan_bbt)(struct mtd_info *mtd);
+	int (*enable)(struct mtd_info *mtd);
+	int (*disable)(struct mtd_info *mtd);
  
  	struct completion	complete;
  	int			irq;
-- 
1.7.0.4

[PATCH 3/7] mtd: OneNAND: OMAP2/3: prevent regulator sleeping while OneNAND is in use

[Adrian Hunter]

>From 0781d7d541e0c93e2ef60fbe0a21f08f6228e68a Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 19 Feb 2010 15:39:53 +0100
Subject: [PATCH 3/7] mtd: OneNAND: OMAP2/3: prevent regulator sleeping while OneNAND is in use

Prevent OneNAND's voltage regulator from going to sleep while
OneNAND is in use, by explicitly enabling and disabling the
regulator as appropriate.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 arch/arm/plat-omap/include/plat/onenand.h |    1 +
 drivers/mtd/onenand/omap2.c               |   41 ++++++++++++++++++++++++++++-
 2 files changed, 41 insertions(+), 1 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/onenand.h b/arch/arm/plat-omap/include/plat/onenand.h
index 72f433d..affe87e 100644
--- a/arch/arm/plat-omap/include/plat/onenand.h
+++ b/arch/arm/plat-omap/include/plat/onenand.h
@@ -23,6 +23,7 @@ struct omap_onenand_platform_data {
 	int                     (*onenand_setup)(void __iomem *, int freq);
 	int			dma_channel;
 	u8			flags;
+	u8			regulator_can_sleep;
 };
 
 #define ONENAND_MAX_PARTITIONS 8
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index 324402f..1a07bfc 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -35,6 +35,7 @@
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
 #include <linux/slab.h>
+#include <linux/regulator/consumer.h>
 
 #include <asm/mach/flash.h>
 #include <plat/gpmc.h>
@@ -63,6 +64,7 @@ struct omap2_onenand {
 	int dma_channel;
 	int freq;
 	int (*setup)(void __iomem *base, int freq);
+	struct regulator *regulator;
 };
 
 #ifdef CONFIG_MTD_PARTITIONS
@@ -601,6 +603,30 @@ static void omap2_onenand_shutdown(struct platform_device *pdev)
 	memset((__force void *)c->onenand.base, 0, ONENAND_BUFRAM_SIZE);
 }
 
+static int omap2_onenand_enable(struct mtd_info *mtd)
+{
+	int ret;
+	struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd);
+
+	ret = regulator_enable(c->regulator);
+	if (ret != 0)
+		dev_err(&c->pdev->dev, "cant enable regulator\n");
+
+	return ret;
+}
+
+static int omap2_onenand_disable(struct mtd_info *mtd)
+{
+	int ret;
+	struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd);
+
+	ret = regulator_disable(c->regulator);
+	if (ret != 0)
+		dev_err(&c->pdev->dev, "cant disable regulator\n");
+
+	return ret;
+}
+
 static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 {
 	struct omap_onenand_platform_data *pdata;
@@ -715,8 +741,18 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 		}
 	}
 
+	if (pdata->regulator_can_sleep) {
+		c->regulator = regulator_get(&pdev->dev, "vonenand");
+		if (IS_ERR(c->regulator)) {
+			dev_err(&pdev->dev,  "Failed to get regulator\n");
+			goto err_release_dma;
+		}
+		c->onenand.enable = omap2_onenand_enable;
+		c->onenand.disable = omap2_onenand_disable;
+	}
+
 	if ((r = onenand_scan(&c->mtd, 1)) < 0)
-		goto err_release_dma;
+		goto err_release_regulator;
 
 	switch ((c->onenand.version_id >> 4) & 0xf) {
 	case 0:
@@ -751,6 +787,8 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 
 err_release_onenand:
 	onenand_release(&c->mtd);
+err_release_regulator:
+	regulator_put(c->regulator);
 err_release_dma:
 	if (c->dma_channel != -1)
 		omap_free_dma(c->dma_channel);
@@ -777,6 +815,7 @@ static int __devexit omap2_onenand_remove(struct platform_device *pdev)
 	struct omap2_onenand *c = dev_get_drvdata(&pdev->dev);
 
 	onenand_release(&c->mtd);
+	regulator_put(c->regulator);
 	if (c->dma_channel != -1)
 		omap_free_dma(c->dma_channel);
 	omap2_onenand_shutdown(pdev);
-- 
1.7.0.4

[PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

[Adrian Hunter]

>From c52e0cdeedffb7a0f9818ad3d6296d1f51b8669d Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 9 Dec 2010 10:48:27 +0200
Subject: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

The calculations done with sync_clk are anyway in picoseconds
and switching to picoseconds allows sync_clk values that are
not a whole number of nanoseconds - which is sometimes the
case.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 arch/arm/mach-omap2/gpmc-nand.c        |    2 +-
 arch/arm/mach-omap2/gpmc-onenand.c     |   10 +++++-----
 arch/arm/mach-omap2/gpmc.c             |   12 +++++++++++-
 arch/arm/mach-omap2/usb-tusb6010.c     |    4 ++--
 arch/arm/plat-omap/include/plat/gpmc.h |    9 +++++----
 5 files changed, 24 insertions(+), 13 deletions(-)

diff --git a/arch/arm/mach-omap2/gpmc-nand.c b/arch/arm/mach-omap2/gpmc-nand.c
index 7222096..2bb29c1 100644
--- a/arch/arm/mach-omap2/gpmc-nand.c
+++ b/arch/arm/mach-omap2/gpmc-nand.c
@@ -41,7 +41,7 @@ static int omap2_nand_gpmc_retime(void)
 		return 0;
 
 	memset(&t, 0, sizeof(t));
-	t.sync_clk = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->sync_clk);
+	t.sync_clk = gpmc_nand_data->gpmc_t->sync_clk;
 	t.cs_on = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->cs_on);
 	t.adv_on = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->adv_on);
 
diff --git a/arch/arm/mach-omap2/gpmc-onenand.c b/arch/arm/mach-omap2/gpmc-onenand.c
index 7bb6922..1db606c 100644
--- a/arch/arm/mach-omap2/gpmc-onenand.c
+++ b/arch/arm/mach-omap2/gpmc-onenand.c
@@ -174,7 +174,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 
 	switch (freq) {
 	case 83:
-		min_gpmc_clk_period = 12; /* 83 MHz */
+		min_gpmc_clk_period = 12000; /* 83 MHz */
 		t_ces   = 5;
 		t_avds  = 4;
 		t_avdh  = 2;
@@ -183,7 +183,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 		t_rdyo  = 9;
 		break;
 	case 66:
-		min_gpmc_clk_period = 15; /* 66 MHz */
+		min_gpmc_clk_period = 15000; /* 66 MHz */
 		t_ces   = 6;
 		t_avds  = 5;
 		t_avdh  = 2;
@@ -192,7 +192,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 		t_rdyo  = 11;
 		break;
 	default:
-		min_gpmc_clk_period = 18; /* 54 MHz */
+		min_gpmc_clk_period = 18500; /* 54 MHz */
 		t_ces   = 7;
 		t_avds  = 7;
 		t_avdh  = 7;
@@ -271,8 +271,8 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 		t.wr_cycle  = t.rd_cycle;
 		if (cpu_is_omap34xx()) {
 			t.wr_data_mux_bus = gpmc_ticks_to_ns(fclk_offset +
-					gpmc_ns_to_ticks(min_gpmc_clk_period +
-					t_rdyo));
+					gpmc_ps_to_ticks(min_gpmc_clk_period +
+					t_rdyo * 1000));
 			t.wr_access = t.access;
 		}
 	} else {
diff --git a/arch/arm/mach-omap2/gpmc.c b/arch/arm/mach-omap2/gpmc.c
index f46933b..1b7b3e7 100644
--- a/arch/arm/mach-omap2/gpmc.c
+++ b/arch/arm/mach-omap2/gpmc.c
@@ -168,6 +168,16 @@ unsigned int gpmc_ns_to_ticks(unsigned int time_ns)
 	return (time_ns * 1000 + tick_ps - 1) / tick_ps;
 }
 
+unsigned int gpmc_ps_to_ticks(unsigned int time_ps)
+{
+	unsigned long tick_ps;
+
+	/* Calculate in picosecs to yield more exact results */
+	tick_ps = gpmc_get_fclk_period();
+
+	return (time_ps + tick_ps - 1) / tick_ps;
+}
+
 unsigned int gpmc_ticks_to_ns(unsigned int ticks)
 {
 	return ticks * gpmc_get_fclk_period() / 1000;
@@ -235,7 +245,7 @@ int gpmc_cs_calc_divider(int cs, unsigned int sync_clk)
 	int div;
 	u32 l;
 
-	l = sync_clk * 1000 + (gpmc_get_fclk_period() - 1);
+	l = sync_clk + (gpmc_get_fclk_period() - 1);
 	div = l / gpmc_get_fclk_period();
 	if (div > 4)
 		return -1;
diff --git a/arch/arm/mach-omap2/usb-tusb6010.c b/arch/arm/mach-omap2/usb-tusb6010.c
index 64a0112..1e998ea 100644
--- a/arch/arm/mach-omap2/usb-tusb6010.c
+++ b/arch/arm/mach-omap2/usb-tusb6010.c
@@ -120,8 +120,8 @@ static int tusb_set_sync_mode(unsigned sysclk_ps, unsigned fclk_ps)
 	t.adv_on = next_clk(t.cs_on, t_scsnh_advnh - 7000, fclk_ps);
 
 	/* GPMC_CLK rate = fclk rate / div */
-	t.sync_clk = 12 /* 11.1 nsec */;
-	tmp = (t.sync_clk * 1000 + fclk_ps - 1) / fclk_ps;
+	t.sync_clk = 11100 /* 11.1 nsec */;
+	tmp = (t.sync_clk + fclk_ps - 1) / fclk_ps;
 	if (tmp > 4)
 		return -ERANGE;
 	if (tmp <= 0)
diff --git a/arch/arm/plat-omap/include/plat/gpmc.h b/arch/arm/plat-omap/include/plat/gpmc.h
index 9fd99b9..85ded59 100644
--- a/arch/arm/plat-omap/include/plat/gpmc.h
+++ b/arch/arm/plat-omap/include/plat/gpmc.h
@@ -80,12 +80,12 @@
 #define GPMC_PREFETCH_STATUS_COUNT(val)	(val & 0x00003fff)
 
 /*
- * Note that all values in this struct are in nanoseconds, while
- * the register values are in gpmc_fck cycles.
+ * Note that all values in this struct are in nanoseconds except sync_clk
+ * (which is in picoseconds), while the register values are in gpmc_fck cycles.
  */
 struct gpmc_timings {
-	/* Minimum clock period for synchronous mode */
-	u16 sync_clk;
+	/* Minimum clock period for synchronous mode (in picoseconds) */
+	u32 sync_clk;
 
 	/* Chip-select signal timings corresponding to GPMC_CS_CONFIG2 */
 	u16 cs_on;		/* Assertion time */
@@ -117,6 +117,7 @@ struct gpmc_timings {
 };
 
 extern unsigned int gpmc_ns_to_ticks(unsigned int time_ns);
+extern unsigned int gpmc_ps_to_ticks(unsigned int time_ps);
 extern unsigned int gpmc_ticks_to_ns(unsigned int ticks);
 extern unsigned int gpmc_round_ns_to_ticks(unsigned int time_ns);
 extern unsigned long gpmc_get_fclk_period(void);
-- 
1.7.0.4

Re: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

[Tony Lindgren]

* Adrian Hunter <adrian.hunter@nokia.com> [101213 04:21]:
> From c52e0cdeedffb7a0f9818ad3d6296d1f51b8669d Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Thu, 9 Dec 2010 10:48:27 +0200
> Subject: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds
> 
> The calculations done with sync_clk are anyway in picoseconds
> and switching to picoseconds allows sync_clk values that are
> not a whole number of nanoseconds - which is sometimes the
> case.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
> ---
>  arch/arm/mach-omap2/gpmc-nand.c        |    2 +-
>  arch/arm/mach-omap2/gpmc-onenand.c     |   10 +++++-----
>  arch/arm/mach-omap2/gpmc.c             |   12 +++++++++++-
>  arch/arm/mach-omap2/usb-tusb6010.c     |    4 ++--
>  arch/arm/plat-omap/include/plat/gpmc.h |    9 +++++----
>  5 files changed, 24 insertions(+), 13 deletions(-)

Hmm I think I already replied to this.. But looks like not so here's
a new attempt..

This one you should send to linux-omap and linux-arm-kernel and
Cc Felipe Balbi so we can make sure the timing changes work
with tusb6010. Then we can ack it an it can go via the MTD
list.

Regards,

Tony

Re: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

[Adrian Hunter]

>From c52e0cdeedffb7a0f9818ad3d6296d1f51b8669d Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 9 Dec 2010 10:48:27 +0200
Subject: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

The calculations done with sync_clk are anyway in picoseconds
and switching to picoseconds allows sync_clk values that are
not a whole number of nanoseconds - which is sometimes the
case.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
  arch/arm/mach-omap2/gpmc-nand.c        |    2 +-
  arch/arm/mach-omap2/gpmc-onenand.c     |   10 +++++-----
  arch/arm/mach-omap2/gpmc.c             |   12 +++++++++++-
  arch/arm/mach-omap2/usb-tusb6010.c     |    4 ++--
  arch/arm/plat-omap/include/plat/gpmc.h |    9 +++++----
  5 files changed, 24 insertions(+), 13 deletions(-)

diff --git a/arch/arm/mach-omap2/gpmc-nand.c b/arch/arm/mach-omap2/gpmc-nand.c
index 7222096..2bb29c1 100644
--- a/arch/arm/mach-omap2/gpmc-nand.c
+++ b/arch/arm/mach-omap2/gpmc-nand.c
@@ -41,7 +41,7 @@ static int omap2_nand_gpmc_retime(void)
  		return 0;
  
  	memset(&t, 0, sizeof(t));
-	t.sync_clk = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->sync_clk);
+	t.sync_clk = gpmc_nand_data->gpmc_t->sync_clk;
  	t.cs_on = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->cs_on);
  	t.adv_on = gpmc_round_ns_to_ticks(gpmc_nand_data->gpmc_t->adv_on);
  
diff --git a/arch/arm/mach-omap2/gpmc-onenand.c b/arch/arm/mach-omap2/gpmc-onenand.c
index 7bb6922..1db606c 100644
--- a/arch/arm/mach-omap2/gpmc-onenand.c
+++ b/arch/arm/mach-omap2/gpmc-onenand.c
@@ -174,7 +174,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
  
  	switch (freq) {
  	case 83:
-		min_gpmc_clk_period = 12; /* 83 MHz */
+		min_gpmc_clk_period = 12000; /* 83 MHz */
  		t_ces   = 5;
  		t_avds  = 4;
  		t_avdh  = 2;
@@ -183,7 +183,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
  		t_rdyo  = 9;
  		break;
  	case 66:
-		min_gpmc_clk_period = 15; /* 66 MHz */
+		min_gpmc_clk_period = 15000; /* 66 MHz */
  		t_ces   = 6;
  		t_avds  = 5;
  		t_avdh  = 2;
@@ -192,7 +192,7 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
  		t_rdyo  = 11;
  		break;
  	default:
-		min_gpmc_clk_period = 18; /* 54 MHz */
+		min_gpmc_clk_period = 18500; /* 54 MHz */
  		t_ces   = 7;
  		t_avds  = 7;
  		t_avdh  = 7;
@@ -271,8 +271,8 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
  		t.wr_cycle  = t.rd_cycle;
  		if (cpu_is_omap34xx()) {
  			t.wr_data_mux_bus = gpmc_ticks_to_ns(fclk_offset +
-					gpmc_ns_to_ticks(min_gpmc_clk_period +
-					t_rdyo));
+					gpmc_ps_to_ticks(min_gpmc_clk_period +
+					t_rdyo * 1000));
  			t.wr_access = t.access;
  		}
  	} else {
diff --git a/arch/arm/mach-omap2/gpmc.c b/arch/arm/mach-omap2/gpmc.c
index f46933b..1b7b3e7 100644
--- a/arch/arm/mach-omap2/gpmc.c
+++ b/arch/arm/mach-omap2/gpmc.c
@@ -168,6 +168,16 @@ unsigned int gpmc_ns_to_ticks(unsigned int time_ns)
  	return (time_ns * 1000 + tick_ps - 1) / tick_ps;
  }
  
+unsigned int gpmc_ps_to_ticks(unsigned int time_ps)
+{
+	unsigned long tick_ps;
+
+	/* Calculate in picosecs to yield more exact results */
+	tick_ps = gpmc_get_fclk_period();
+
+	return (time_ps + tick_ps - 1) / tick_ps;
+}
+
  unsigned int gpmc_ticks_to_ns(unsigned int ticks)
  {
  	return ticks * gpmc_get_fclk_period() / 1000;
@@ -235,7 +245,7 @@ int gpmc_cs_calc_divider(int cs, unsigned int sync_clk)
  	int div;
  	u32 l;
  
-	l = sync_clk * 1000 + (gpmc_get_fclk_period() - 1);
+	l = sync_clk + (gpmc_get_fclk_period() - 1);
  	div = l / gpmc_get_fclk_period();
  	if (div > 4)
  		return -1;
diff --git a/arch/arm/mach-omap2/usb-tusb6010.c b/arch/arm/mach-omap2/usb-tusb6010.c
index 64a0112..1e998ea 100644
--- a/arch/arm/mach-omap2/usb-tusb6010.c
+++ b/arch/arm/mach-omap2/usb-tusb6010.c
@@ -120,8 +120,8 @@ static int tusb_set_sync_mode(unsigned sysclk_ps, unsigned fclk_ps)
  	t.adv_on = next_clk(t.cs_on, t_scsnh_advnh - 7000, fclk_ps);
  
  	/* GPMC_CLK rate = fclk rate / div */
-	t.sync_clk = 12 /* 11.1 nsec */;
-	tmp = (t.sync_clk * 1000 + fclk_ps - 1) / fclk_ps;
+	t.sync_clk = 11100 /* 11.1 nsec */;
+	tmp = (t.sync_clk + fclk_ps - 1) / fclk_ps;
  	if (tmp > 4)
  		return -ERANGE;
  	if (tmp <= 0)
diff --git a/arch/arm/plat-omap/include/plat/gpmc.h b/arch/arm/plat-omap/include/plat/gpmc.h
index 9fd99b9..85ded59 100644
--- a/arch/arm/plat-omap/include/plat/gpmc.h
+++ b/arch/arm/plat-omap/include/plat/gpmc.h
@@ -80,12 +80,12 @@
  #define GPMC_PREFETCH_STATUS_COUNT(val)	(val & 0x00003fff)
  
  /*
- * Note that all values in this struct are in nanoseconds, while
- * the register values are in gpmc_fck cycles.
+ * Note that all values in this struct are in nanoseconds except sync_clk
+ * (which is in picoseconds), while the register values are in gpmc_fck cycles.
   */
  struct gpmc_timings {
-	/* Minimum clock period for synchronous mode */
-	u16 sync_clk;
+	/* Minimum clock period for synchronous mode (in picoseconds) */
+	u32 sync_clk;
  
  	/* Chip-select signal timings corresponding to GPMC_CS_CONFIG2 */
  	u16 cs_on;		/* Assertion time */
@@ -117,6 +117,7 @@ struct gpmc_timings {
  };
  
  extern unsigned int gpmc_ns_to_ticks(unsigned int time_ns);
+extern unsigned int gpmc_ps_to_ticks(unsigned int time_ps);
  extern unsigned int gpmc_ticks_to_ns(unsigned int ticks);
  extern unsigned int gpmc_round_ns_to_ticks(unsigned int time_ns);
  extern unsigned long gpmc_get_fclk_period(void);
-- 1.7.0.4

Re: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds

[Tony Lindgren]

* Adrian Hunter <adrian.hunter@nokia.com> [101213 04:21]:
> From c52e0cdeedffb7a0f9818ad3d6296d1f51b8669d Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Thu, 9 Dec 2010 10:48:27 +0200
> Subject: [PATCH 4/7] OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds
> 
> The calculations done with sync_clk are anyway in picoseconds
> and switching to picoseconds allows sync_clk values that are
> not a whole number of nanoseconds - which is sometimes the

Can you please post patches 4 - 5 of this series as separate
patches and Cc linux-omap and linux-arm-kernel mailing lists?

This is assuming patches 4 - 5 do not have dependencies to
the other patches in the series..

Thanks,

Tony

[PATCH 5/7] OMAP2/3: OneNAND: add 104MHz support

[Adrian Hunter]

>From ab0260a2cc3b17b98fa918dee34e129c3ff2b9b6 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 9 Dec 2010 11:22:50 +0200
Subject: [PATCH 5/7] OMAP2/3: OneNAND: add 104MHz support

Add GPMC timings for 104MHz OneNAND.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 arch/arm/mach-omap2/gpmc-onenand.c |    9 +++++++++
 drivers/mtd/onenand/omap2.c        |    3 +++
 2 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/arch/arm/mach-omap2/gpmc-onenand.c b/arch/arm/mach-omap2/gpmc-onenand.c
index 1db606c..3a7d25f 100644
--- a/arch/arm/mach-omap2/gpmc-onenand.c
+++ b/arch/arm/mach-omap2/gpmc-onenand.c
@@ -173,6 +173,15 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 	}
 
 	switch (freq) {
+	case 104:
+		min_gpmc_clk_period = 9600; /* 104 MHz */
+		t_ces   = 3;
+		t_avds  = 4;
+		t_avdh  = 2;
+		t_ach   = 3;
+		t_aavdh = 6;
+		t_rdyo  = 9;
+		break;
 	case 83:
 		min_gpmc_clk_period = 12000; /* 83 MHz */
 		t_ces   = 5;
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index 1a07bfc..ac31f46 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -767,6 +767,9 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 	case 3:
 		c->freq = 83;
 		break;
+	case 4:
+		c->freq = 104;
+		break;
 	}
 
 #ifdef CONFIG_MTD_PARTITIONS
-- 
1.7.0.4

[PATCH 6/7] OMAP2/3: OneNAND: add platform data callback for PM constraints

[Adrian Hunter]

>From b93b44d7463b3a3f6d5d1021681596de1881a62b Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 9 Dec 2010 12:56:24 +0200
Subject: [PATCH 6/7] OMAP2/3: OneNAND: add platform data callback for PM constraints

Make it possible for boards to set PM constraints such as maximum wakeup
latency.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 arch/arm/plat-omap/include/plat/onenand.h |    8 ++++++++
 drivers/mtd/onenand/omap2.c               |   16 +++++++++++++++-
 2 files changed, 23 insertions(+), 1 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/onenand.h b/arch/arm/plat-omap/include/plat/onenand.h
index affe87e..1d07119 100644
--- a/arch/arm/plat-omap/include/plat/onenand.h
+++ b/arch/arm/plat-omap/include/plat/onenand.h
@@ -15,12 +15,20 @@
 #define ONENAND_SYNC_READ	(1 << 0)
 #define ONENAND_SYNC_READWRITE	(1 << 1)
 
+enum {
+	ONENAND_PM_CONSTRAINT_OFF,
+	ONENAND_PM_CONSTRAINT_ERASING,
+	ONENAND_PM_CONSTRAINT_OTHER,
+};
+
 struct omap_onenand_platform_data {
 	int			cs;
 	int			gpio_irq;
 	struct mtd_partition	*parts;
 	int			nr_parts;
 	int                     (*onenand_setup)(void __iomem *, int freq);
+	void			(*set_pm_constraints)(struct device *dev,
+						      int on);
 	int			dma_channel;
 	u8			flags;
 	u8			regulator_can_sleep;
diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index ac31f46..866b9f9 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -64,6 +64,7 @@ struct omap2_onenand {
 	int dma_channel;
 	int freq;
 	int (*setup)(void __iomem *base, int freq);
+	void (*set_pm_constraints)(struct device *dev, int on);
 	struct regulator *regulator;
 };
 
@@ -605,13 +606,21 @@ static void omap2_onenand_shutdown(struct platform_device *pdev)
 
 static int omap2_onenand_enable(struct mtd_info *mtd)
 {
-	int ret;
+	int ret, on;
 	struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd);
 
 	ret = regulator_enable(c->regulator);
 	if (ret != 0)
 		dev_err(&c->pdev->dev, "cant enable regulator\n");
 
+	if (c->set_pm_constraints) {
+		if (c->onenand.state == FL_ERASING)
+			on = ONENAND_PM_CONSTRAINT_ERASING;
+		else
+			on = ONENAND_PM_CONSTRAINT_OTHER;
+		c->set_pm_constraints(&c->pdev->dev, on);
+	}
+
 	return ret;
 }
 
@@ -620,6 +629,9 @@ static int omap2_onenand_disable(struct mtd_info *mtd)
 	int ret;
 	struct omap2_onenand *c = container_of(mtd, struct omap2_onenand, mtd);
 
+	if (c->set_pm_constraints)
+		c->set_pm_constraints(&c->pdev->dev, ONENAND_PM_CONSTRAINT_OFF);
+
 	ret = regulator_disable(c->regulator);
 	if (ret != 0)
 		dev_err(&c->pdev->dev, "cant disable regulator\n");
@@ -682,6 +694,8 @@ static int __devinit omap2_onenand_probe(struct platform_device *pdev)
 		c->setup = pdata->onenand_setup;
 	}
 
+	c->set_pm_constraints = pdata->set_pm_constraints;
+
 	if (c->gpio_irq) {
 		if ((r = gpio_request(c->gpio_irq, "OneNAND irq")) < 0) {
 			dev_err(&pdev->dev,  "Failed to request GPIO%d for "
-- 
1.7.0.4

[PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Adrian Hunter]

>From 5fa7a6377683aabd617f78518cd6cd49eae1679e Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 10 Dec 2010 12:04:20 +0200
Subject: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

Initial bad blocks are normal but the messages look like
errors.  Make the messages less scary, make the main
message an informational message not a warning, make the
message displaying registers a debug message and include
the address there instead of in the informational message.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/onenand_base.c |   22 ++++++++++++----------
 drivers/mtd/onenand/onenand_bbt.c  |    4 ++--
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index bde274f..5065fc7 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1488,8 +1488,7 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 {
 	struct onenand_chip *this = mtd->priv;
 	unsigned long timeout;
-	unsigned int interrupt;
-	unsigned int ctrl;
+	unsigned int interrupt, ctrl, ecc, addr1, addr8;
 
 	/* The 20 msec is enough */
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -1501,25 +1500,28 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	/* To get correct interrupt status in timeout case */
 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
+	addr1 = this->read_word(this->base + ONENAND_REG_START_ADDRESS1);
+	addr8 = this->read_word(this->base + ONENAND_REG_START_ADDRESS8);
 
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = onenand_read_ecc(this);
+		ecc = onenand_read_ecc(this);
 		if (ecc & ONENAND_ECC_2BIT_ALL) {
-			printk(KERN_WARNING "%s: ecc error = 0x%04x, "
-				"controller error 0x%04x\n",
-				__func__, ecc, ctrl);
+			printk(KERN_DEBUG "%s: ecc 0x%04x ctrl 0x%04x "
+			       "intr 0x%04x addr1 %#x addr8 %#x\n",
+			       __func__, ecc, ctrl, interrupt, addr1, addr8);
 			return ONENAND_BBT_READ_ECC_ERROR;
 		}
 	} else {
-		printk(KERN_ERR "%s: read timeout! ctrl=0x%04x intr=0x%04x\n",
-			__func__, ctrl, interrupt);
+		printk(KERN_ERR "%s: read timeout! ctrl 0x%04x "
+		       "intr 0x%04x addr1 %#x addr8 %#x\n",
+		       __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_FATAL_ERROR;
 	}
 
 	/* Initial bad block case: 0x2400 or 0x0400 */
 	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_DEBUG "%s: controller error = 0x%04x\n",
-			__func__, ctrl);
+		printk(KERN_DEBUG "%s: ctrl 0x%04x intr 0x%04x addr1 %#x "
+		       "addr8 %#x\n", __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_ERROR;
 	}
 
diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
index 905209b..fc2c16a 100644
--- a/drivers/mtd/onenand/onenand_bbt.c
+++ b/drivers/mtd/onenand/onenand_bbt.c
@@ -101,8 +101,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 			if (ret || check_short_pattern(&buf[j * scanlen],
 					       scanlen, this->writesize, bd)) {
 				bbm->bbt[i >> 3] |= 0x03 << (i & 0x6);
-				printk(KERN_WARNING "Bad eraseblock %d at 0x%08x\n",
-					i >> 1, (unsigned int) from);
+				printk(KERN_INFO "OneNAND eraseblock %d is an "
+					"initial bad block\n", i >> 1);
 				mtd->ecc_stats.badblocks++;
 				break;
 			}
-- 
1.7.0.4

Re: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Kyungmin Park]

Acked-by: Kyungmin Park <kyungmin.park@samsung.com>

On Mon, Dec 13, 2010 at 9:21 PM, Adrian Hunter <adrian.hunter@nokia.com> wrote:
> From 5fa7a6377683aabd617f78518cd6cd49eae1679e Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Fri, 10 Dec 2010 12:04:20 +0200
> Subject: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages
>
> Initial bad blocks are normal but the messages look like
> errors.  Make the messages less scary, make the main
> message an informational message not a warning, make the
> message displaying registers a debug message and include
> the address there instead of in the informational message.
>
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
> ---
>  drivers/mtd/onenand/onenand_base.c |   22 ++++++++++++----------
>  drivers/mtd/onenand/onenand_bbt.c  |    4 ++--
>  2 files changed, 14 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
> index bde274f..5065fc7 100644
> --- a/drivers/mtd/onenand/onenand_base.c
> +++ b/drivers/mtd/onenand/onenand_base.c
> @@ -1488,8 +1488,7 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
>  {
>        struct onenand_chip *this = mtd->priv;
>        unsigned long timeout;
> -       unsigned int interrupt;
> -       unsigned int ctrl;
> +       unsigned int interrupt, ctrl, ecc, addr1, addr8;
>
>        /* The 20 msec is enough */
>        timeout = jiffies + msecs_to_jiffies(20);
> @@ -1501,25 +1500,28 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
>        /* To get correct interrupt status in timeout case */
>        interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
>        ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
> +       addr1 = this->read_word(this->base + ONENAND_REG_START_ADDRESS1);
> +       addr8 = this->read_word(this->base + ONENAND_REG_START_ADDRESS8);
>
>        if (interrupt & ONENAND_INT_READ) {
> -               int ecc = onenand_read_ecc(this);
> +               ecc = onenand_read_ecc(this);
>                if (ecc & ONENAND_ECC_2BIT_ALL) {
> -                       printk(KERN_WARNING "%s: ecc error = 0x%04x, "
> -                               "controller error 0x%04x\n",
> -                               __func__, ecc, ctrl);
> +                       printk(KERN_DEBUG "%s: ecc 0x%04x ctrl 0x%04x "
> +                              "intr 0x%04x addr1 %#x addr8 %#x\n",
> +                              __func__, ecc, ctrl, interrupt, addr1, addr8);
>                        return ONENAND_BBT_READ_ECC_ERROR;
>                }
>        } else {
> -               printk(KERN_ERR "%s: read timeout! ctrl=0x%04x intr=0x%04x\n",
> -                       __func__, ctrl, interrupt);
> +               printk(KERN_ERR "%s: read timeout! ctrl 0x%04x "
> +                      "intr 0x%04x addr1 %#x addr8 %#x\n",
> +                      __func__, ctrl, interrupt, addr1, addr8);
>                return ONENAND_BBT_READ_FATAL_ERROR;
>        }
>
>        /* Initial bad block case: 0x2400 or 0x0400 */
>        if (ctrl & ONENAND_CTRL_ERROR) {
> -               printk(KERN_DEBUG "%s: controller error = 0x%04x\n",
> -                       __func__, ctrl);
> +               printk(KERN_DEBUG "%s: ctrl 0x%04x intr 0x%04x addr1 %#x "
> +                      "addr8 %#x\n", __func__, ctrl, interrupt, addr1, addr8);
>                return ONENAND_BBT_READ_ERROR;
>        }
>
> diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
> index 905209b..fc2c16a 100644
> --- a/drivers/mtd/onenand/onenand_bbt.c
> +++ b/drivers/mtd/onenand/onenand_bbt.c
> @@ -101,8 +101,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
>                        if (ret || check_short_pattern(&buf[j * scanlen],
>                                               scanlen, this->writesize, bd)) {
>                                bbm->bbt[i >> 3] |= 0x03 << (i & 0x6);
> -                               printk(KERN_WARNING "Bad eraseblock %d at 0x%08x\n",
> -                                       i >> 1, (unsigned int) from);
> +                               printk(KERN_INFO "OneNAND eraseblock %d is an "
> +                                       "initial bad block\n", i >> 1);
>                                mtd->ecc_stats.badblocks++;
>                                break;
>                        }
> --
> 1.7.0.4
>
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>

Re: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Artem Bityutskiy]

On Mon, 2010-12-13 at 14:21 +0200, Adrian Hunter wrote:
> From 5fa7a6377683aabd617f78518cd6cd49eae1679e Mon Sep 17 00:00:00 2001
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Fri, 10 Dec 2010 12:04:20 +0200
> Subject: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages
> 
> Initial bad blocks are normal but the messages look like
> errors.  Make the messages less scary, make the main
> message an informational message not a warning, make the
> message displaying registers a debug message and include
> the address there instead of in the informational message.
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>

Pushed this one to l2-mtd-2.6.git, thanks.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

[PATCH 0/7] OneNAND OMAP patches

[Adrian Hunter]

Hi

Here are some OneNAND and related OMAP patches.

Adrian Hunter (7):
      mtd: OneNAND: OMAP2/3: add support for command line partitioning
      mtd: OneNAND: add enable / disable methods to onenand_chip
      mtd: OneNAND: OMAP2/3: prevent regulator sleeping while OneNAND is in use
      OMAP2/3: GPMC: put sync_clk value in picoseconds instead of nanoseconds
      OMAP2/3: OneNAND: add 104MHz support
      OMAP2/3: OneNAND: add platform data callback for PM constraints
      mtd: OneNAND: lighten scary initial bad block messages

 arch/arm/mach-omap2/gpmc-nand.c           |    2 +-
 arch/arm/mach-omap2/gpmc-onenand.c        |   19 +++++--
 arch/arm/mach-omap2/gpmc.c                |   12 ++++-
 arch/arm/mach-omap2/usb-tusb6010.c        |    4 +-
 arch/arm/plat-omap/include/plat/gpmc.h    |    9 ++--
 arch/arm/plat-omap/include/plat/onenand.h |    9 +++
 drivers/mtd/onenand/omap2.c               |   85 +++++++++++++++++++++++------
 drivers/mtd/onenand/onenand_base.c        |   26 ++++++----
 drivers/mtd/onenand/onenand_bbt.c         |    4 +-
 include/linux/mtd/onenand.h               |    2 +
 10 files changed, 131 insertions(+), 41 deletions(-) 
 
Regards
Adrian

[PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Adrian Hunter]

>From 5fa7a6377683aabd617f78518cd6cd49eae1679e Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Fri, 10 Dec 2010 12:04:20 +0200
Subject: [PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

Initial bad blocks are normal but the messages look like
errors.  Make the messages less scary, make the main
message an informational message not a warning, make the
message displaying registers a debug message and include
the address there instead of in the informational message.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/onenand_base.c |   22 ++++++++++++----------
 drivers/mtd/onenand/onenand_bbt.c  |    4 ++--
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index bde274f..5065fc7 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1488,8 +1488,7 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 {
 	struct onenand_chip *this = mtd->priv;
 	unsigned long timeout;
-	unsigned int interrupt;
-	unsigned int ctrl;
+	unsigned int interrupt, ctrl, ecc, addr1, addr8;
 
 	/* The 20 msec is enough */
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -1501,25 +1500,28 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	/* To get correct interrupt status in timeout case */
 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
+	addr1 = this->read_word(this->base + ONENAND_REG_START_ADDRESS1);
+	addr8 = this->read_word(this->base + ONENAND_REG_START_ADDRESS8);
 
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = onenand_read_ecc(this);
+		ecc = onenand_read_ecc(this);
 		if (ecc & ONENAND_ECC_2BIT_ALL) {
-			printk(KERN_WARNING "%s: ecc error = 0x%04x, "
-				"controller error 0x%04x\n",
-				__func__, ecc, ctrl);
+			printk(KERN_DEBUG "%s: ecc 0x%04x ctrl 0x%04x "
+			       "intr 0x%04x addr1 %#x addr8 %#x\n",
+			       __func__, ecc, ctrl, interrupt, addr1, addr8);
 			return ONENAND_BBT_READ_ECC_ERROR;
 		}
 	} else {
-		printk(KERN_ERR "%s: read timeout! ctrl=0x%04x intr=0x%04x\n",
-			__func__, ctrl, interrupt);
+		printk(KERN_ERR "%s: read timeout! ctrl 0x%04x "
+		       "intr 0x%04x addr1 %#x addr8 %#x\n",
+		       __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_FATAL_ERROR;
 	}
 
 	/* Initial bad block case: 0x2400 or 0x0400 */
 	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_DEBUG "%s: controller error = 0x%04x\n",
-			__func__, ctrl);
+		printk(KERN_DEBUG "%s: ctrl 0x%04x intr 0x%04x addr1 %#x "
+		       "addr8 %#x\n", __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_ERROR;
 	}
 
diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
index 905209b..fc2c16a 100644
--- a/drivers/mtd/onenand/onenand_bbt.c
+++ b/drivers/mtd/onenand/onenand_bbt.c
@@ -101,8 +101,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 			if (ret || check_short_pattern(&buf[j * scanlen],
 					       scanlen, this->writesize, bd)) {
 				bbm->bbt[i >> 3] |= 0x03 << (i & 0x6);
-				printk(KERN_WARNING "Bad eraseblock %d at 0x%08x\n",
-					i >> 1, (unsigned int) from);
+				printk(KERN_INFO "OneNAND eraseblock %d is an "
+					"initial bad block\n", i >> 1);
 				mtd->ecc_stats.badblocks++;
 				break;
 			}
-- 
1.7.0.4

[PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Adrian Hunter]

Initial bad blocks are normal but the messages look like
errors.  Make the messages less scary, make the main
message an informational message not a warning, make the
message displaying registers a debug message and include
the address there instead of in the informational message.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/onenand_base.c |   22 ++++++++++++----------
 drivers/mtd/onenand/onenand_bbt.c  |    4 ++--
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index bde274f..5065fc7 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1488,8 +1488,7 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 {
 	struct onenand_chip *this = mtd->priv;
 	unsigned long timeout;
-	unsigned int interrupt;
-	unsigned int ctrl;
+	unsigned int interrupt, ctrl, ecc, addr1, addr8;
 
 	/* The 20 msec is enough */
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -1501,25 +1500,28 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	/* To get correct interrupt status in timeout case */
 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
+	addr1 = this->read_word(this->base + ONENAND_REG_START_ADDRESS1);
+	addr8 = this->read_word(this->base + ONENAND_REG_START_ADDRESS8);
 
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = onenand_read_ecc(this);
+		ecc = onenand_read_ecc(this);
 		if (ecc & ONENAND_ECC_2BIT_ALL) {
-			printk(KERN_WARNING "%s: ecc error = 0x%04x, "
-				"controller error 0x%04x\n",
-				__func__, ecc, ctrl);
+			printk(KERN_DEBUG "%s: ecc 0x%04x ctrl 0x%04x "
+			       "intr 0x%04x addr1 %#x addr8 %#x\n",
+			       __func__, ecc, ctrl, interrupt, addr1, addr8);
 			return ONENAND_BBT_READ_ECC_ERROR;
 		}
 	} else {
-		printk(KERN_ERR "%s: read timeout! ctrl=0x%04x intr=0x%04x\n",
-			__func__, ctrl, interrupt);
+		printk(KERN_ERR "%s: read timeout! ctrl 0x%04x "
+		       "intr 0x%04x addr1 %#x addr8 %#x\n",
+		       __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_FATAL_ERROR;
 	}
 
 	/* Initial bad block case: 0x2400 or 0x0400 */
 	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_DEBUG "%s: controller error = 0x%04x\n",
-			__func__, ctrl);
+		printk(KERN_DEBUG "%s: ctrl 0x%04x intr 0x%04x addr1 %#x "
+		       "addr8 %#x\n", __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_ERROR;
 	}
 
diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
index 905209b..fc2c16a 100644
--- a/drivers/mtd/onenand/onenand_bbt.c
+++ b/drivers/mtd/onenand/onenand_bbt.c
@@ -101,8 +101,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 			if (ret || check_short_pattern(&buf[j * scanlen],
 					       scanlen, this->writesize, bd)) {
 				bbm->bbt[i >> 3] |= 0x03 << (i & 0x6);
-				printk(KERN_WARNING "Bad eraseblock %d@0x%08x\n",
-					i >> 1, (unsigned int) from);
+				printk(KERN_INFO "OneNAND eraseblock %d is an "
+					"initial bad block\n", i >> 1);
 				mtd->ecc_stats.badblocks++;
 				break;
 			}
-- 
1.7.0.4

[PATCH 7/7] mtd: OneNAND: lighten scary initial bad block messages

[Adrian Hunter]

Initial bad blocks are normal but the messages look like
errors.  Make the messages less scary, make the main
message an informational message not a warning, make the
message displaying registers a debug message and include
the address there instead of in the informational message.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mtd/onenand/onenand_base.c |   22 ++++++++++++----------
 drivers/mtd/onenand/onenand_bbt.c  |    4 ++--
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index bde274f..5065fc7 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1488,8 +1488,7 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 {
 	struct onenand_chip *this = mtd->priv;
 	unsigned long timeout;
-	unsigned int interrupt;
-	unsigned int ctrl;
+	unsigned int interrupt, ctrl, ecc, addr1, addr8;
 
 	/* The 20 msec is enough */
 	timeout = jiffies + msecs_to_jiffies(20);
@@ -1501,25 +1500,28 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	/* To get correct interrupt status in timeout case */
 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
+	addr1 = this->read_word(this->base + ONENAND_REG_START_ADDRESS1);
+	addr8 = this->read_word(this->base + ONENAND_REG_START_ADDRESS8);
 
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = onenand_read_ecc(this);
+		ecc = onenand_read_ecc(this);
 		if (ecc & ONENAND_ECC_2BIT_ALL) {
-			printk(KERN_WARNING "%s: ecc error = 0x%04x, "
-				"controller error 0x%04x\n",
-				__func__, ecc, ctrl);
+			printk(KERN_DEBUG "%s: ecc 0x%04x ctrl 0x%04x "
+			       "intr 0x%04x addr1 %#x addr8 %#x\n",
+			       __func__, ecc, ctrl, interrupt, addr1, addr8);
 			return ONENAND_BBT_READ_ECC_ERROR;
 		}
 	} else {
-		printk(KERN_ERR "%s: read timeout! ctrl=0x%04x intr=0x%04x\n",
-			__func__, ctrl, interrupt);
+		printk(KERN_ERR "%s: read timeout! ctrl 0x%04x "
+		       "intr 0x%04x addr1 %#x addr8 %#x\n",
+		       __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_FATAL_ERROR;
 	}
 
 	/* Initial bad block case: 0x2400 or 0x0400 */
 	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_DEBUG "%s: controller error = 0x%04x\n",
-			__func__, ctrl);
+		printk(KERN_DEBUG "%s: ctrl 0x%04x intr 0x%04x addr1 %#x "
+		       "addr8 %#x\n", __func__, ctrl, interrupt, addr1, addr8);
 		return ONENAND_BBT_READ_ERROR;
 	}
 
diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
index 905209b..fc2c16a 100644
--- a/drivers/mtd/onenand/onenand_bbt.c
+++ b/drivers/mtd/onenand/onenand_bbt.c
@@ -101,8 +101,8 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 			if (ret || check_short_pattern(&buf[j * scanlen],
 					       scanlen, this->writesize, bd)) {
 				bbm->bbt[i >> 3] |= 0x03 << (i & 0x6);
-				printk(KERN_WARNING "Bad eraseblock %d@0x%08x\n",
-					i >> 1, (unsigned int) from);
+				printk(KERN_INFO "OneNAND eraseblock %d is an "
+					"initial bad block\n", i >> 1);
 				mtd->ecc_stats.badblocks++;
 				break;
 			}
-- 
1.7.0.4